Compare commits

..
Author SHA1 Message Date
Lance Release d49ac46dc5 Bump version: 0.39.0-beta.0 → 0.39.0-beta.1 2026-09-02 05:28:10 +00:00
85 changed files with 1315 additions and 8590 deletions
+1 -1
View File
@@ -1,5 +1,5 @@
[tool.bumpversion]
current_version = "0.39.0-beta.6"
current_version = "0.39.0-beta.1"
parse = """(?x)
(?P<major>0|[1-9]\\d*)\\.
(?P<minor>0|[1-9]\\d*)\\.
+1 -4
View File
@@ -232,10 +232,7 @@ jobs:
ALL_FEATURES=`cargo metadata --format-version=1 --no-deps \
| jq -r '.packages[] | .features | keys | .[]' \
| grep -v s3-test | sort | uniq | paste -s -d "," -`
# Run doctests before test binaries fill the runner disk. Examples are
# already built by the Linux job, so avoid retaining them here.
cargo test --profile ci --features $ALL_FEATURES --locked --doc
cargo test --profile ci --features $ALL_FEATURES --locked --lib --tests
cargo test --profile ci --features $ALL_FEATURES --locked
windows:
strategy:
Generated
+50 -137
View File
@@ -332,34 +332,6 @@ dependencies = [
"num-traits",
]
[[package]]
name = "arrow-flight"
version = "58.4.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "b2dbe34824c639e43136af8f106992792ab456540d54b880bc320a3192502d2e"
dependencies = [
"arrow-arith",
"arrow-array",
"arrow-buffer",
"arrow-cast",
"arrow-data",
"arrow-ipc",
"arrow-ord",
"arrow-row",
"arrow-schema",
"arrow-select",
"arrow-string",
"base64 0.22.1",
"bytes",
"futures",
"once_cell",
"paste",
"prost",
"prost-types",
"tonic",
"tonic-prost",
]
[[package]]
name = "arrow-ipc"
version = "58.4.0"
@@ -1157,7 +1129,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "edca88bc138befd0323b20752846e6587272d3b03b0343c8ea28a6f819e6e71f"
dependencies = [
"async-trait",
"axum-core 0.4.5",
"axum-core",
"bytes",
"futures-util",
"http 1.5.0",
@@ -1166,7 +1138,7 @@ dependencies = [
"hyper 1.9.0",
"hyper-util",
"itoa",
"matchit 0.7.3",
"matchit",
"memchr",
"mime",
"percent-encoding",
@@ -1184,31 +1156,6 @@ dependencies = [
"tracing",
]
[[package]]
name = "axum"
version = "0.8.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "31b698c5f9a010f6573133b09e0de5408834d0c82f8d7475a89fc1867a71cd90"
dependencies = [
"axum-core 0.5.6",
"bytes",
"futures-util",
"http 1.5.0",
"http-body 1.1.0",
"http-body-util",
"itoa",
"matchit 0.8.4",
"memchr",
"mime",
"percent-encoding",
"pin-project-lite",
"serde_core",
"sync_wrapper",
"tower",
"tower-layer",
"tower-service",
]
[[package]]
name = "axum-core"
version = "0.4.5"
@@ -1230,24 +1177,6 @@ dependencies = [
"tracing",
]
[[package]]
name = "axum-core"
version = "0.5.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "08c78f31d7b1291f7ee735c1c6780ccde7785daae9a9206026862dab7d8792d1"
dependencies = [
"bytes",
"futures-core",
"http 1.5.0",
"http-body 1.1.0",
"http-body-util",
"mime",
"pin-project-lite",
"sync_wrapper",
"tower-layer",
"tower-service",
]
[[package]]
name = "backoff"
version = "0.4.0"
@@ -3526,8 +3455,8 @@ checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"
[[package]]
name = "fsst"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"rand 0.9.5",
@@ -4886,8 +4815,8 @@ checksum = "e037a2e1d8d5fdbd49b16a4ea09d5d6401c1f29eca5ff29d03d3824dba16256a"
[[package]]
name = "lance"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arc-swap",
"arrow",
@@ -4959,8 +4888,8 @@ dependencies = [
[[package]]
name = "lance-arrow"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -4982,7 +4911,7 @@ dependencies = [
[[package]]
name = "lance-arrow-scalar"
version = "58.0.0"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -4996,7 +4925,7 @@ dependencies = [
[[package]]
name = "lance-arrow-stats"
version = "58.0.0"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-schema",
@@ -5005,8 +4934,8 @@ dependencies = [
[[package]]
name = "lance-bitpacking"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrayref",
"crunchy",
@@ -5016,8 +4945,8 @@ dependencies = [
[[package]]
name = "lance-core"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -5054,8 +4983,8 @@ dependencies = [
[[package]]
name = "lance-datafusion"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"arrow-array",
@@ -5085,8 +5014,8 @@ dependencies = [
[[package]]
name = "lance-datagen"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"arrow-array",
@@ -5103,8 +5032,8 @@ dependencies = [
[[package]]
name = "lance-derive"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"proc-macro2",
"quote",
@@ -5113,8 +5042,8 @@ dependencies = [
[[package]]
name = "lance-encoding"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-arith",
"arrow-array",
@@ -5147,8 +5076,8 @@ dependencies = [
[[package]]
name = "lance-file"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-arith",
"arrow-array",
@@ -5179,8 +5108,8 @@ dependencies = [
[[package]]
name = "lance-index"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arc-swap",
"arrow",
@@ -5244,8 +5173,8 @@ dependencies = [
[[package]]
name = "lance-index-core"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-schema",
@@ -5267,8 +5196,8 @@ dependencies = [
[[package]]
name = "lance-io"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"arrow-array",
@@ -5308,8 +5237,8 @@ dependencies = [
[[package]]
name = "lance-linalg"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-schema",
@@ -5323,29 +5252,27 @@ dependencies = [
[[package]]
name = "lance-namespace"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"async-trait",
"bytes",
"lance-core",
"lance-namespace-reqwest-client",
"serde",
"serde_json",
"snafu 0.9.0",
]
[[package]]
name = "lance-namespace-impls"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"arrow-ipc",
"arrow-schema",
"async-trait",
"axum 0.7.9",
"axum",
"base64 0.22.1",
"bytes",
"chrono",
@@ -5378,9 +5305,9 @@ dependencies = [
[[package]]
name = "lance-namespace-reqwest-client"
version = "0.12.0"
version = "0.11.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d8d23e54b1634d5bbb434f8dd33dc3c05f6e58d876a9a27b3b4aef58ddbe11af"
checksum = "1d06b1fbb5d41f93bc652b61e2872af92e8a6c5f6b4ce8839a8ecfa05365d359"
dependencies = [
"reqwest 0.12.28",
"serde",
@@ -5392,8 +5319,8 @@ dependencies = [
[[package]]
name = "lance-select"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -5407,8 +5334,8 @@ dependencies = [
[[package]]
name = "lance-table"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow",
"arrow-array",
@@ -5448,8 +5375,8 @@ dependencies = [
[[package]]
name = "lance-testing"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"arrow-array",
"arrow-schema",
@@ -5462,8 +5389,8 @@ dependencies = [
[[package]]
name = "lance-tokenizer"
version = "12.0.0-beta.15"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.15#271a155ba505fd8e1094c095d4ce356707e93866"
version = "12.0.0-beta.11"
source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.11#4a0e26895729feb86d0cb9c09d551bfd619c6472"
dependencies = [
"frostem",
"icu_segmenter",
@@ -5476,7 +5403,7 @@ dependencies = [
[[package]]
name = "lancedb"
version = "0.39.0-beta.5"
version = "0.39.0-beta.0"
dependencies = [
"ahash",
"anyhow",
@@ -5485,7 +5412,6 @@ dependencies = [
"arrow-buffer",
"arrow-cast",
"arrow-data",
"arrow-flight",
"arrow-ipc",
"arrow-ord",
"arrow-schema",
@@ -5541,7 +5467,6 @@ dependencies = [
"polars",
"polars-arrow",
"pprof 0.14.1",
"prost",
"rand 0.9.5",
"random_word",
"regex",
@@ -5558,7 +5483,6 @@ dependencies = [
"test-log",
"tokenizers",
"tokio",
"tonic",
"url",
"urlencoding",
"uuid",
@@ -5567,7 +5491,7 @@ dependencies = [
[[package]]
name = "lancedb-nodejs"
version = "0.39.0-beta.5"
version = "0.39.0-beta.0"
dependencies = [
"arrow-array",
"arrow-buffer",
@@ -5592,7 +5516,7 @@ dependencies = [
[[package]]
name = "lancedb-python"
version = "0.39.0-beta.5"
version = "0.39.0-beta.0"
dependencies = [
"arrow",
"async-trait",
@@ -5616,7 +5540,6 @@ dependencies = [
"serde_json",
"snafu 0.8.9",
"tokio",
"uuid",
]
[[package]]
@@ -5937,12 +5860,6 @@ version = "0.7.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0e7465ac9959cc2b1404e8e2367b43684a6d13790fe23056cc8c6c5a6b7bcb94"
[[package]]
name = "matchit"
version = "0.8.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "47e1ffaa40ddd1f3ed91f717a33c8c0ee23fff369e3aa8772b9605cc1d22f4c3"
[[package]]
name = "matrixmultiply"
version = "0.3.10"
@@ -7818,7 +7735,6 @@ dependencies = [
"pyo3-build-config",
"pyo3-ffi",
"pyo3-macros",
"uuid",
]
[[package]]
@@ -10171,7 +10087,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ac2a5518c70fa84342385732db33fb3f44bc4cc748936eb5833d2df34d6445ef"
dependencies = [
"async-trait",
"axum 0.8.9",
"base64 0.22.1",
"bytes",
"h2 0.4.16",
@@ -10183,11 +10098,9 @@ dependencies = [
"hyper-util",
"percent-encoding",
"pin-project",
"rustls-native-certs",
"socket2 0.6.3",
"sync_wrapper",
"tokio",
"tokio-rustls 0.26.4",
"tokio-stream",
"tower",
"tower-layer",
+15 -17
View File
@@ -13,20 +13,20 @@ categories = ["database-implementations"]
rust-version = "1.91.0"
[workspace.dependencies]
lance = { "version" = "=12.0.0-beta.15", default-features = false, "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-core = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-datagen = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-file = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-io = { "version" = "=12.0.0-beta.15", default-features = false, "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-index = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-linalg = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace-impls = { "version" = "=12.0.0-beta.15", default-features = false, "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-table = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-testing = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-datafusion = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-encoding = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance-arrow = { "version" = "=12.0.0-beta.15", "tag" = "v12.0.0-beta.15", "git" = "https://github.com/lance-format/lance.git" }
lance = { "version" = "=12.0.0-beta.11", default-features = false, "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-core = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-datagen = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-file = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-io = { "version" = "=12.0.0-beta.11", default-features = false, "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-index = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-linalg = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-namespace-impls = { "version" = "=12.0.0-beta.11", default-features = false, "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-table = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-testing = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-datafusion = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-encoding = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lance-arrow = { "version" = "=12.0.0-beta.11", "tag" = "v12.0.0-beta.11", "git" = "https://github.com/lance-format/lance.git" }
lancedb = { path = "rust/lancedb", default-features = false }
ahash = "0.8"
# Note that this one does not include pyarrow
@@ -39,7 +39,6 @@ arrow-ord = "58.0.0"
arrow-schema = "58.0.0"
arrow-select = "58.0.0"
arrow-cast = "58.0.0"
arrow-flight = { version = "58.0.0", features = ["flight-sql-experimental"] }
async-trait = "0"
bytes = "1"
datafusion = { version = "54.0.0", default-features = false }
@@ -72,8 +71,7 @@ serde = "1"
serde_json = "1"
tempfile = "3.5.0"
tokio = { version = "1.23", features = ["rt-multi-thread", "sync"] }
tonic = { version = "0.14", features = ["tls-native-roots", "tls-ring"] }
uuid = { version = "1.7.0", features = ["v4", "v7"] }
uuid = { version = "1.7.0", features = ["v4"] }
chrono = { version = "0.4", default-features = false, features = ["clock"] }
[profile.ci]
+1 -1
View File
@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
<dependency>
<groupId>com.lancedb</groupId>
<artifactId>lancedb-core</artifactId>
<version>0.39.0-beta.6</version>
<version>0.39.0-beta.1</version>
</dependency>
```
+62 -24
View File
@@ -448,6 +448,26 @@ on the returned job to know when cleanup has finished.
***
### getJob()
```ts
abstract getJob(jobId): Promise<null | JobDescription>
```
Describe a single server-side job by id.
Resolves to `null` when the server has no such job.
#### Parameters
* **jobId**: `string`
#### Returns
`Promise`&lt;`null` \| [`JobDescription`](../interfaces/JobDescription.md)&gt;
***
### isOpen()
```ts
@@ -462,6 +482,48 @@ Return true if the connection has not been closed
***
### job()
```ts
abstract job(jobId): Job
```
A [Job](Job.md) handle for a server-side job by id.
The handle is constructed without a server round trip; an unknown id
surfaces when the handle is used. Dropping the handle has no effect on
the job itself.
#### Parameters
* **jobId**: `string`
#### Returns
[`Job`](Job.md)
***
### jobHistory()
```ts
abstract jobHistory(jobId?): Promise<Table<any>>
```
The lifecycle event history of a server-side job, as an Arrow table.
Lists history across all jobs when `jobId` is omitted.
#### Parameters
* **jobId?**: `string`
#### Returns
`Promise`&lt;`Table`&lt;`any`&gt;&gt;
***
### listJobs()
```ts
@@ -586,30 +648,6 @@ A page of table names and an
***
### openJob()
```ts
abstract openJob(jobId): Promise<Job>
```
Open a server-side job by id, returning a handle with its record already
populated. Rejects when the server has no such job, the way
[Connection.openTable](Connection.md#opentable) does for a missing table.
The returned [Job](Job.md) answers for its own state, specification,
result, failure and event history, so there is no separate
connection-level call for any of them.
#### Parameters
* **jobId**: `string`
#### Returns
`Promise`&lt;[`Job`](Job.md)&gt;
***
### openMaterializedView()
```ts
+16 -163
View File
@@ -8,116 +8,28 @@
A handle to an operation that may still be running.
The operation may already be complete when the handle is created.
## Constructors
The detail getters read what the handle last observed. Submitting an
operation returns only a job id, so populating them eagerly would cost an
extra round trip on every call:
### new Job()
- [Job.refresh](Job.md#refresh) and [Job.status](Job.md#status) fetch the whole record.
- [Job.wait](Job.md#wait) records the terminal state it establishes, but not the
rest of the record.
- Everything is null until one of those runs.
```ts
new Job(): Job
```
#### Returns
[`Job`](Job.md)
## Accessors
### creationMs
```ts
get creationMs(): null | number
```
When the job was created, in milliseconds since the epoch.
#### Returns
`null` \| `number`
***
### failure
```ts
get failure(): null | JobFailureInfo
```
Why the job failed, when it failed and the server reports a reason.
#### Returns
`null` \| [`JobFailureInfo`](../interfaces/JobFailureInfo.md)
***
### id
```ts
get id(): null | string
```
Identifies the operation on the server that is running it.
Operations that run in this process have no server id. The value is
opaque: parsing it or storing it to resume the job later is not supported.
#### Returns
`null` \| `string`
***
### jobType
```ts
get jobType(): null | string
```
The job's type, as the server names it. Null for an in-process job, which
has no server-side record.
#### Returns
`null` \| `string`
***
### result
```ts
get result(): any
```
The job-type-specific terminal result. Null until the job succeeds, so a
job that never terminates reports its progress through [Job.events](Job.md#events)
instead.
#### Returns
`any`
***
### spec
```ts
get spec(): any
```
The job-type-specific specification it was submitted with.
#### Returns
`any`
***
### state
```ts
get state(): null | string
```
The last observed lifecycle state, without contacting the backend.
Identifies the operation on the server that is running it. Operations
that run in this process have no server id. The value is opaque.
#### Returns
@@ -139,61 +51,18 @@ Request cancellation. Cancelling a finished operation is a no-op.
***
### events()
```ts
events(options?): Promise<Table<any>>
```
This job's recorded lifecycle events.
Where the getters above report a terminal result only once the job reaches
one, events are written as the job runs and outlive the workers that
produced them. A distributed job records a `claim`/`claim_complete` pair
per unit of work, each carrying `rows_processed`, so a job that never
finishes still accounts for what it did.
The server caps results at 1000 rows by default and 10,000 at most, and
truncates without saying so, so pass `limit` for a job that emits an event
per fragment. `filter` is a SQL-like expression over the `state`,
`updated_by`, `emitted_from`, `emitted_by`, and `claim_entity` columns.
#### Parameters
* **options?**: [`JobEventsOptions`](../interfaces/JobEventsOptions.md)
#### Returns
`Promise`&lt;`Table`&lt;`any`&gt;&gt;
***
### refresh()
```ts
refresh(): Promise<void>
```
Ask the backend for this job's current state, and for a server-side job
its full record, then cache it for the getters above.
#### Returns
`Promise`&lt;`void`&gt;
***
### status()
```ts
status(): Promise<string>
```
The operation's current lifecycle state: "running", "finished", "failed",
or "cancelled".
The operation's current lifecycle state: "running", "finished",
"failed", or "cancelled".
A point snapshot; unlike [Job.wait](Job.md#wait) it does not block or reject on a
terminal failure state. Also refreshes the getters above.
A point snapshot; unlike [Job.wait](Job.md#wait) it does not block or reject
on a terminal failure state. States a newer server reports that this
client version does not know pass through as-is.
#### Returns
@@ -201,22 +70,6 @@ terminal failure state. Also refreshes the getters above.
***
### toString()
```ts
toString(): string
```
Every field the handle currently knows, one per line, with the JSON
payloads indented -- a refresh job's spec and result are the point of
printing it.
#### Returns
`string`
***
### wait()
```ts
+1 -1
View File
@@ -96,7 +96,7 @@
- [IvfFlatOptions](interfaces/IvfFlatOptions.md)
- [IvfPqOptions](interfaces/IvfPqOptions.md)
- [IvfRqOptions](interfaces/IvfRqOptions.md)
- [JobEventsOptions](interfaces/JobEventsOptions.md)
- [JobDescription](interfaces/JobDescription.md)
- [JobFailureInfo](interfaces/JobFailureInfo.md)
- [JobInfo](interfaces/JobInfo.md)
- [ListNamespacesOptions](interfaces/ListNamespacesOptions.md)
+66
View File
@@ -0,0 +1,66 @@
[**@lancedb/lancedb**](../README.md) • **Docs**
***
[@lancedb/lancedb](../globals.md) / JobDescription
# Interface: JobDescription
A described job from `Connection.getJob`.
## Properties
### creationMs
```ts
creationMs: number;
```
When the job was created, in milliseconds since the epoch.
***
### failure?
```ts
optional failure: JobFailureInfo;
```
Why the job failed, when the job is failed and the server reports a
reason.
***
### jobId
```ts
jobId: string;
```
***
### jobType
```ts
jobType: string;
```
***
### specJson?
```ts
optional specJson: string;
```
The job-type-specific specification as a JSON string, when present.
***
### state
```ts
state: string;
```
Lifecycle state: "running", "finished", "failed", or "cancelled".
@@ -1,29 +0,0 @@
[**@lancedb/lancedb**](../README.md) • **Docs**
***
[@lancedb/lancedb](../globals.md) / JobEventsOptions
# Interface: JobEventsOptions
Which of a job's events [Job.events](../classes/Job.md#events) returns.
## Properties
### filter?
```ts
optional filter: string;
```
SQL-like filter over the event columns.
***
### limit?
```ts
optional limit: number;
```
Maximum event rows to return, up to the server maximum of 10,000.
+1 -1
View File
@@ -26,7 +26,7 @@ When the job was created, in milliseconds since the epoch.
jobId: string;
```
The job id -- what `Connection.openJob` and `Connection.cancelJob`
The job id -- what `Connection.getJob` and `Connection.cancelJob`
accept.
***
-73
View File
@@ -28,59 +28,6 @@ is also an [asynchronous API client](#connections-asynchronous).
::: lancedb.Session
## Remote SQL
Submit SQL against a remote LanceDB database through the connection.
The connected database and `default_namespace_path=["public"]` are used for
unqualified tables. Fully qualified references can still query other databases
and namespaces available to the same deployment. `execute_query` returns a
reader as soon as its initial result stream is available. `execute_query_async`
returns a query handle immediately; use it to inspect progress, open a reader,
or cancel the query. The SQL client is initialized by the first query and
retained for the lifetime of the remote connection. Query ids are random,
connection-scoped references rather than encoded SQL or durable resume tokens:
```python
import lancedb
db = lancedb.connect(
"db://analytics",
api_key="ldb_...",
host_override="https://api.example.com",
sql_host_override="grpc+tls://sql.example.com:10026",
)
reader = db.execute_query(
"""
SELECT events.id, accounts.name
FROM analytics.public.events AS events
JOIN users.public.accounts AS accounts ON events.user_id = accounts.id
""",
default_namespace_path=["public"],
)
for batch in reader:
print(batch.num_rows)
query = db.execute_query_async("SELECT * FROM events")
print(query.id)
print(query.describe().status)
for batch in query.reader():
print(batch.num_rows)
# The async connection exposes the same lifecycle without blocking:
# async_db = await lancedb.connect_async(
# "db://analytics",
# api_key="ldb_...",
# host_override="https://api.example.com",
# sql_host_override="grpc+tls://sql.example.com:10026",
# )
# reader = await async_db.execute_query("SELECT * FROM events")
# query = await async_db.execute_query_async("SELECT * FROM events")
# description = await async_db.describe_query(query.id)
# async for batch in await query.reader():
# print(batch.num_rows)
# await query.cancel()
```
## Namespaces (Synchronous)
A namespace-backed connection resolves tables through a
@@ -147,8 +94,6 @@ listing a storage directory.
::: lancedb.functions.OutputMapping
::: lancedb.functions.AssignmentMapping
::: lancedb.functions.FunctionBinding
::: lancedb.functions.RefreshColumnResult
@@ -157,18 +102,6 @@ listing a storage directory.
::: lancedb.job.AsyncJob
::: lancedb.job.JobInfo
::: lancedb.job.JobDescription
::: lancedb.job.JobFailureInfo
::: lancedb.sql.Query
::: lancedb.sql.AsyncQuery
::: lancedb.sql.QueryDescription
## Materialized Views (Synchronous)
::: lancedb.materialized_view.MaterializedView
@@ -316,12 +249,6 @@ still work. Queries return descriptors. Call
::: lancedb.exceptions.MissingColumnError
::: lancedb.exceptions.JobNotFoundError
::: lancedb.exceptions.JobFailedError
::: lancedb.exceptions.JobCancelledError
## Integrations
## Pydantic
+1 -1
View File
@@ -8,7 +8,7 @@
<parent>
<groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId>
<version>0.39.0-beta.6</version>
<version>0.39.0-beta.1</version>
<relativePath>../pom.xml</relativePath>
</parent>
+2 -2
View File
@@ -6,7 +6,7 @@
<groupId>com.lancedb</groupId>
<artifactId>lancedb-parent</artifactId>
<version>0.39.0-beta.6</version>
<version>0.39.0-beta.1</version>
<packaging>pom</packaging>
<name>${project.artifactId}</name>
<description>LanceDB Java SDK Parent POM</description>
@@ -28,7 +28,7 @@
<properties>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<arrow.version>15.0.0</arrow.version>
<lance-core.version>12.0.0-beta.15</lance-core.version>
<lance-core.version>12.0.0-beta.11</lance-core.version>
<spotless.skip>false</spotless.skip>
<spotless.version>2.30.0</spotless.version>
<spotless.java.googlejavaformat.version>1.7</spotless.java.googlejavaformat.version>
+1 -1
View File
@@ -1,7 +1,7 @@
[package]
name = "lancedb-nodejs"
edition.workspace = true
version = "0.39.0-beta.6"
version = "0.39.0-beta.1"
publish = false
license.workspace = true
description.workspace = true
+11 -66
View File
@@ -939,7 +939,6 @@ describe("remote connection jobs surface", () => {
const { tableFromArrays, tableToIPC } = await import("apache-arrow");
const eventsTable = tableFromArrays({ state: ["created", "succeeded"] });
const eventsBody = Buffer.from(tableToIPC(eventsTable, "stream"));
const queryEventsPayloads: Record<string, unknown>[] = [];
await withMockDatabase(
(req, res) => {
@@ -968,16 +967,6 @@ describe("remote connection jobs surface", () => {
);
}
} else if (req.url === "/v1/jobs/describe") {
if (payload["job_id"] === "job-2") {
res
.writeHead(200, { "Content-Type": "application/json" })
.end(
'{"job_id": "job-2", "job_type": "refresh_column", ' +
'"job_state": "DONE", "creation_ms": 2000, ' +
'"result": {"rows_assigned": 1000000}}',
);
return;
}
if (payload["job_id"] !== "job-1") {
res.writeHead(404).end("no such job");
return;
@@ -999,7 +988,6 @@ describe("remote connection jobs surface", () => {
.writeHead(200, { "Content-Type": "application/json" })
.end('{"job_id": "job-1"}');
} else if (req.url === "/v1/jobs/query_events") {
queryEventsPayloads.push(payload);
res
.writeHead(200, {
"Content-Type": "application/vnd.apache.arrow.stream",
@@ -1016,65 +1004,22 @@ describe("remote connection jobs surface", () => {
expect(jobs[0].state).toEqual("running");
expect(jobs[1].state).toEqual("finished");
const description = await db.getJob("job-1");
expect(description?.state).toEqual("failed");
expect(JSON.parse(description?.specJson ?? "")).toEqual({
column: "vec",
});
expect(description?.failure?.message).toEqual("worker died");
expect(await db.getJob("missing")).toBeNull();
expect(await db.cancelJob("job-1")).toBe(true);
expect(await db.cancelJob("missing")).toBe(false);
// Opening a job hands back a populated handle; a missing one rejects.
await expect(db.openJob("missing")).rejects.toThrow("not found");
const finished = await db.openJob("job-2");
expect(finished.state).toEqual("finished");
expect(finished.result).toEqual({
// biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
rows_assigned: 1000000,
});
const history = await db.jobHistory("job-1");
expect(history.numRows).toEqual(2);
const job = await db.openJob("job-1");
const job = db.job("job-1");
expect(job.id).toEqual("job-1");
// openJob already populated the handle; refresh() re-reads it.
expect(job.state).toEqual("failed");
await job.refresh();
expect(job.state).toEqual("failed");
expect(job.jobType).toEqual("create_index");
expect(job.creationMs).toEqual(1000);
expect(job.spec).toEqual({ column: "vec" });
expect(job.result).toBeNull();
expect(job.failure?.message).toEqual("worker died");
// The handle reaches its own events, supplying its job id.
const jobEvents = await job.events({
limit: 500,
filter: "state = 'claim_complete'",
});
expect(jobEvents.numRows).toEqual(2);
expect(queryEventsPayloads.pop()).toEqual({
// biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
job_id: "job-1",
limit: 500,
filter: "state = 'claim_complete'",
});
// Printing lays every known field out on its own line, with the JSON
// payloads indented rather than crammed onto one line.
expect(`${job}`).toEqual(
[
"Job(",
' id="job-1",',
' state="failed",',
' jobType="create_index",',
" creationMs=1000,",
" spec={",
' "column": "vec"',
" },",
" failure={",
' "phase": "execute",',
' "message": "worker died",',
' "retryable": true',
" },",
")",
].join("\n"),
);
expect(await job.status()).toEqual("failed");
await expect(job.wait()).rejects.toThrow("worker died");
},
+2 -2
View File
@@ -281,7 +281,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
numIndices: 0,
numRows: 3,
// Full on-disk size of the two data files, footers and metadata included.
totalBytes: 550,
totalBytes: 684,
});
// Index files count toward totalBytes too (only deletion files and
@@ -289,7 +289,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
await table.createIndex("id", { config: Index.btree() });
const statsWithIndex = await table.stats();
expect(statsWithIndex.numIndices).toBe(1);
expect(statsWithIndex.totalBytes).toBeGreaterThan(550);
expect(statsWithIndex.totalBytes).toBeGreaterThan(684);
});
it("should overwrite data if asked", async () => {
+37 -11
View File
@@ -1,6 +1,7 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
import { tableFromIPC } from "apache-arrow";
import {
Data,
SchemaLike,
@@ -15,7 +16,6 @@ import {
makeEmptyTable,
} from "./arrow";
import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
import { Job } from "./job";
import {
MaterializedView,
MaterializedViewSelect,
@@ -27,6 +27,8 @@ import type {
CreateNamespaceResponse,
DescribeNamespaceResponse,
DropNamespaceResponse,
Job,
JobDescription,
JobInfo,
ListNamespacesResponse,
ListTablesResponse,
@@ -555,19 +557,24 @@ export abstract class Connection {
): Promise<void>;
/**
* Open a server-side job by id, returning a handle with its record already
* populated. Rejects when the server has no such job, the way
* {@link Connection.openTable} does for a missing table.
* A {@link Job} handle for a server-side job by id.
*
* The returned {@link Job} answers for its own state, specification,
* result, failure and event history, so there is no separate
* connection-level call for any of them.
* The handle is constructed without a server round trip; an unknown id
* surfaces when the handle is used. Dropping the handle has no effect on
* the job itself.
*/
abstract openJob(jobId: string): Promise<Job>;
abstract job(jobId: string): Job;
/** List server-side jobs across the database's tables. */
abstract listJobs(): Promise<JobInfo[]>;
/**
* Describe a single server-side job by id.
*
* Resolves to `null` when the server has no such job.
*/
abstract getJob(jobId: string): Promise<JobDescription | null>;
/**
* Request cancellation of a server-side job by id.
*
@@ -575,6 +582,13 @@ export abstract class Connection {
* such job exists. Cancelling an already-terminal job is a no-op success.
*/
abstract cancelJob(jobId: string): Promise<boolean>;
/**
* The lifecycle event history of a server-side job, as an Arrow table.
*
* Lists history across all jobs when `jobId` is omitted.
*/
abstract jobHistory(jobId?: string): Promise<ArrowTable>;
}
/** @hideconstructor */
@@ -855,7 +869,7 @@ export class LocalConnection extends Connection {
}
async dropTableAsync(name: string, namespacePath?: string[]): Promise<Job> {
return new Job(await this.inner.dropTableAsync(name, namespacePath ?? []));
return this.inner.dropTableAsync(name, namespacePath ?? []);
}
async dropAllTables(namespacePath?: string[]): Promise<void> {
@@ -914,17 +928,29 @@ export class LocalConnection extends Connection {
);
}
async openJob(jobId: string): Promise<Job> {
return new Job(await this.inner.openJob(jobId));
job(jobId: string): Job {
return this.inner.job(jobId);
}
async listJobs(): Promise<JobInfo[]> {
return this.inner.listJobs();
}
async getJob(jobId: string): Promise<JobDescription | null> {
return this.inner.getJob(jobId);
}
async cancelJob(jobId: string): Promise<boolean> {
return this.inner.cancelJob(jobId);
}
async jobHistory(jobId?: string): Promise<ArrowTable> {
const buf = await this.inner.jobHistory(jobId);
if (buf.length === 0) {
return new ArrowTable();
}
return tableFromIPC(buf);
}
}
/**
+7 -3
View File
@@ -94,9 +94,13 @@ export {
RenameTableOptions,
} from "./connection";
export { JobFailureInfo, JobInfo, Session } from "./native.js";
export { Job, JobEventsOptions } from "./job";
export {
Job,
JobDescription,
JobFailureInfo,
JobInfo,
Session,
} from "./native.js";
export {
AutoQuery,
-188
View File
@@ -1,188 +0,0 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
import { Table as ArrowTable, tableFromIPC } from "apache-arrow";
import { JobFailureInfo, Job as NativeJob } from "./native";
/** Which of a job's events {@link Job.events} returns. */
export interface JobEventsOptions {
/** Maximum event rows to return, up to the server maximum of 10,000. */
limit?: number;
/** SQL-like filter over the event columns. */
filter?: string;
}
/**
* A handle to an operation that may still be running.
*
* The operation may already be complete when the handle is created.
*
* The detail getters read what the handle last observed. Submitting an
* operation returns only a job id, so populating them eagerly would cost an
* extra round trip on every call:
*
* - {@link Job.refresh} and {@link Job.status} fetch the whole record.
* - {@link Job.wait} records the terminal state it establishes, but not the
* rest of the record.
* - Everything is null until one of those runs.
*
* @hideconstructor
*/
export class Job {
private readonly inner: NativeJob;
constructor(inner: NativeJob) {
this.inner = inner;
}
/**
* Identifies the operation on the server that is running it.
*
* Operations that run in this process have no server id. The value is
* opaque: parsing it or storing it to resume the job later is not supported.
*/
get id(): string | null {
return this.inner.id ?? null;
}
/** The last observed lifecycle state, without contacting the backend. */
get state(): string | null {
return this.inner.state ?? null;
}
/**
* The job's type, as the server names it. Null for an in-process job, which
* has no server-side record.
*/
get jobType(): string | null {
return this.inner.jobType ?? null;
}
/** When the job was created, in milliseconds since the epoch. */
get creationMs(): number | null {
return this.inner.creationMs ?? null;
}
/** The job-type-specific specification it was submitted with. */
// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
get spec(): any | null {
return parseJson(this.inner.specJson);
}
/**
* The job-type-specific terminal result. Null until the job succeeds, so a
* job that never terminates reports its progress through {@link Job.events}
* instead.
*/
// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
get result(): any | null {
return parseJson(this.inner.resultJson);
}
/** Why the job failed, when it failed and the server reports a reason. */
get failure(): JobFailureInfo | null {
return this.inner.failure ?? null;
}
/**
* The operation's current lifecycle state: "running", "finished", "failed",
* or "cancelled".
*
* A point snapshot; unlike {@link Job.wait} it does not block or reject on a
* terminal failure state. Also refreshes the getters above.
*/
async status(): Promise<string> {
return this.inner.status();
}
/** Wait until the operation reaches a terminal state. */
async wait(): Promise<void> {
return this.inner.wait();
}
/** Request cancellation. Cancelling a finished operation is a no-op. */
async cancel(): Promise<void> {
return this.inner.cancel();
}
/**
* Ask the backend for this job's current state, and for a server-side job
* its full record, then cache it for the getters above.
*/
async refresh(): Promise<void> {
return this.inner.refresh();
}
/**
* This job's recorded lifecycle events.
*
* Where the getters above report a terminal result only once the job reaches
* one, events are written as the job runs and outlive the workers that
* produced them. A distributed job records a `claim`/`claim_complete` pair
* per unit of work, each carrying `rows_processed`, so a job that never
* finishes still accounts for what it did.
*
* The server caps results at 1000 rows by default and 10,000 at most, and
* truncates without saying so, so pass `limit` for a job that emits an event
* per fragment. `filter` is a SQL-like expression over the `state`,
* `updated_by`, `emitted_from`, `emitted_by`, and `claim_entity` columns.
*/
async events(options?: JobEventsOptions): Promise<ArrowTable> {
const buf = await this.inner.events(options?.limit, options?.filter);
if (buf.length === 0) {
return new ArrowTable();
}
return tableFromIPC(buf);
}
/**
* Every field the handle currently knows, one per line, with the JSON
* payloads indented -- a refresh job's spec and result are the point of
* printing it.
*/
toString(): string {
if (this.state === null) {
const known = this.id === null ? "" : `id=${JSON.stringify(this.id)}, `;
return `Job(${known}not refreshed)`;
}
const fields: string[] = [];
if (this.id !== null) {
fields.push(`id=${JSON.stringify(this.id)}`);
}
fields.push(`state=${JSON.stringify(this.state)}`);
if (this.jobType !== null) {
fields.push(`jobType=${JSON.stringify(this.jobType)}`);
}
if (this.creationMs !== null) {
fields.push(`creationMs=${this.creationMs}`);
}
for (const [name, value] of [
["spec", this.spec],
["result", this.result],
] as const) {
if (value !== null) {
fields.push(`${name}=${indentJson(value)}`);
}
}
if (this.failure !== null) {
fields.push(`failure=${indentJson(this.failure)}`);
}
return `Job(${fields.map((field) => `\n${REPR_INDENT}${field},`).join("")}\n)`;
}
[Symbol.for("nodejs.util.inspect.custom")](): string {
return this.toString();
}
}
const REPR_INDENT = " ";
// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
function indentJson(value: any): string {
return JSON.stringify(value, null, 4).replace(/\n/g, `\n${REPR_INDENT}`);
}
// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
function parseJson(raw: string | null | undefined): any | null {
return raw === null || raw === undefined ? null : JSON.parse(raw);
}
+9 -11
View File
@@ -19,7 +19,6 @@ import {
import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
import { IndexOptions } from "./indices";
import { Job } from "./job";
import { MergeInsertBuilder } from "./merge";
import {
AddColumnsResult,
@@ -31,6 +30,7 @@ import {
DropColumnsResult,
IndexConfig,
IndexStatistics,
Job,
LsmStats,
Branches as NativeBranches,
OptimizeStats,
@@ -1124,15 +1124,13 @@ export class LocalTable extends Table {
): Promise<Job> {
// biome-ignore lint/suspicious/noExplicitAny: skip
const nativeIndex = (options?.config as any)?.inner;
return new Job(
await this.inner.createIndexAsync(
nativeIndex,
column,
options?.replace,
options?.waitTimeoutSeconds,
options?.name,
options?.train,
),
return await this.inner.createIndexAsync(
nativeIndex,
column,
options?.replace,
options?.waitTimeoutSeconds,
options?.name,
options?.train,
);
}
@@ -1315,7 +1313,7 @@ export class LocalTable extends Table {
}
async refreshColumnAsync(column: string): Promise<Job> {
return new Job(await this.inner.refreshColumnAsync(column));
return await this.inner.refreshColumnAsync(column);
}
async refreshMaterializedView(
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-darwin-arm64",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["darwin"],
"cpu": ["arm64"],
"main": "lancedb.darwin-arm64.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-arm64-gnu",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["arm64"],
"main": "lancedb.linux-arm64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-arm64-musl",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["arm64"],
"main": "lancedb.linux-arm64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-x64-gnu",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["x64"],
"main": "lancedb.linux-x64-gnu.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-linux-x64-musl",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["linux"],
"cpu": ["x64"],
"main": "lancedb.linux-x64-musl.node",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-win32-arm64-msvc",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": [
"win32"
],
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@lancedb/lancedb-win32-x64-msvc",
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"os": ["win32"],
"cpu": ["x64"],
"main": "lancedb.win32-x64-msvc.node",
+3 -3
View File
@@ -11,7 +11,7 @@
"ann"
],
"private": false,
"version": "0.39.0-beta.6",
"version": "0.39.0-beta.1",
"main": "dist/index.js",
"exports": {
".": "./dist/index.js",
@@ -44,7 +44,7 @@
"@biomejs/biome": "^1.7.3",
"@jest/globals": "^29.7.0",
"@napi-rs/cli": "3.7.0",
"@opentelemetry/sdk-metrics": "^2.10.0",
"@opentelemetry/sdk-metrics": "^1.30.0",
"@types/axios": "^0.14.0",
"@types/jest": "^29.1.2",
"@types/node": "22.7.4",
@@ -56,7 +56,7 @@
"eslint": "^8.57.0",
"jest": "^29.7.0",
"shx": "^0.3.4",
"tmp": "^0.2.7",
"tmp": "^0.2.3",
"ts-jest": "^29.1.2",
"typedoc": "0.26.4",
"typedoc-plugin-markdown": "4.2.1",
+443 -599
View File
File diff suppressed because it is too large Load Diff
-38
View File
@@ -16,41 +16,3 @@ allowBuilds:
onnxruntime-node: true
protobufjs: true
sharp: true
minimumReleaseAgeExclude:
- protobufjs@7.5.8
- tmp@0.2.6
- form-data@4.0.6
- tar@7.5.16
- markdown-it@14.1.2
- linkify-it@5.0.1
- js-yaml@3.15.0
- js-yaml@4.1.2
- protobufjs@7.6.1
- protobufjs@7.6.3
- '@babel/core@7.29.1'
- axios@1.18.0
- brace-expansion@2.1.2
- brace-expansion@1.1.16
- js-yaml@4.3.0
- tar@7.5.18
- tar@7.5.19
- tar@7.5.17
- protobufjs@7.6.5
- linkify-it@5.0.2
- sharp@0.35.0
- brace-expansion@1.1.17
- brace-expansion@2.1.3
- brace-expansion@2.1.4
- brace-expansion@1.1.18
- js-yaml@3.15.1
- js-yaml@4.3.1
- tar@7.5.21
- '@opentelemetry/core@2.8.0'
# @huggingface/transformers pins sharp ^0.33.5 and no released version has moved
# past ^0.34.5, all of which inherit the libvips CVEs in GHSA-f88m-g3jw-g9cj.
# Force the patched line. sharp is only reached by transformers' image pipeline,
# which LanceDB's text embedding function never uses.
overrides:
sharp: ^0.35.4
+45 -8
View File
@@ -442,15 +442,13 @@ impl Connection {
self.get_inner()?.drop_all_tables(&ns).await.default_error()
}
/// Open a server-side job by id, returning a handle with its record
/// already populated. Rejects when the server has no such job.
/// A `Job` handle for a server-side job by id.
///
/// The returned handle answers for its own state, specification, result,
/// failure and event history, so there is no separate connection-level
/// call for any of them.
#[napi(catch_unwind)]
pub async fn open_job(&self, job_id: String) -> napi::Result<crate::job::Job> {
let job = self.get_inner()?.open_job(&job_id).await.default_error()?;
/// The handle is constructed without a server round trip; an unknown id
/// surfaces when the handle is used.
#[napi]
pub fn job(&self, job_id: String) -> napi::Result<crate::job::Job> {
let job = self.get_inner()?.job(job_id).default_error()?;
Ok(crate::job::Job::new(job))
}
@@ -461,6 +459,17 @@ impl Connection {
Ok(jobs.into_iter().map(Into::into).collect())
}
/// Describe a single server-side job by id. `null` when the server has
/// no such job.
#[napi(catch_unwind)]
pub async fn get_job(
&self,
job_id: String,
) -> napi::Result<Option<crate::job::JobDescription>> {
let description = self.get_inner()?.get_job(&job_id).await.default_error()?;
Ok(description.map(Into::into))
}
/// Request cancellation of a server-side job by id. Returns true if the
/// server accepted the cancellation, false if no such job exists.
#[napi(catch_unwind)]
@@ -468,6 +477,34 @@ impl Connection {
self.get_inner()?.cancel_job(&job_id).await.default_error()
}
/// The lifecycle event history of a server-side job (all jobs when
/// `job_id` is null), as an Arrow IPC stream buffer. Empty when there is
/// no history.
#[napi(catch_unwind)]
pub async fn job_history(&self, job_id: Option<String>) -> napi::Result<Buffer> {
let batches = self
.get_inner()?
.job_history(job_id.as_deref())
.await
.default_error()?;
let Some(first) = batches.first() else {
return Ok(Buffer::from(Vec::<u8>::new()));
};
let mut out = Vec::new();
let mut writer = arrow_ipc::writer::StreamWriter::try_new(&mut out, &first.schema())
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
for batch in &batches {
writer
.write(batch)
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
}
writer
.finish()
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
drop(writer);
Ok(Buffer::from(out))
}
#[napi(catch_unwind)]
/// Describe a namespace and return its properties.
pub async fn describe_namespace(
+34 -90
View File
@@ -3,9 +3,6 @@
use std::sync::Arc;
use arrow_array::RecordBatch;
use lancedb::job::JobEventsRequest;
use napi::bindgen_prelude::Buffer;
use napi_derive::napi;
use crate::error::NapiErrorExt;
@@ -58,98 +55,12 @@ impl Job {
pub async fn cancel(&self) -> napi::Result<()> {
self.inner.cancel().await.default_error()
}
/// Ask the backend for this job's current state, and for a server-side job
/// its full record, then cache it for the getters below.
///
/// They are all null until this runs, because submitting an operation
/// returns only a job id. {@link Job.status} fetches the whole record too;
/// {@link Job.wait} records only the terminal state it establishes.
#[napi(catch_unwind)]
pub async fn refresh(&self) -> napi::Result<()> {
self.inner.refresh().await.default_error()
}
/// The last observed lifecycle state, without contacting the backend.
#[napi(getter)]
pub fn state(&self) -> Option<String> {
self.inner.state()
}
/// The job's type, as the server names it. Null for an in-process job,
/// which has no server-side record.
#[napi(getter)]
pub fn job_type(&self) -> Option<String> {
self.inner.job_type()
}
/// When the job was created, in milliseconds since the epoch.
#[napi(getter)]
pub fn creation_ms(&self) -> Option<i64> {
self.inner.creation_ms()
}
/// The job-type-specific specification as a JSON string, when present.
#[napi(getter)]
pub fn spec_json(&self) -> Option<String> {
self.inner.spec().map(|spec| spec.to_string())
}
/// The job-type-specific terminal result as a JSON string. Null until the
/// job succeeds, so a job that never terminates reports its progress
/// through {@link Job.events} instead.
#[napi(getter)]
pub fn result_json(&self) -> Option<String> {
self.inner.result().map(|result| result.to_string())
}
/// Why the job failed, when it failed and the server reports a reason.
#[napi(getter)]
pub fn failure(&self) -> Option<JobFailureInfo> {
self.inner.failure().map(|failure| JobFailureInfo {
phase: failure.phase,
message: failure.message,
retryable: failure.retryable,
})
}
/// This job's recorded lifecycle events, as an Arrow IPC stream buffer.
/// The TypeScript wrapper turns it into an Arrow table.
#[napi(catch_unwind)]
pub async fn events(&self, limit: Option<u32>, filter: Option<String>) -> napi::Result<Buffer> {
let batches = self
.inner
.events(JobEventsRequest { limit, filter })
.await
.default_error()?;
batches_to_ipc_buffer(&batches)
}
}
/// Serialise Arrow batches as a single IPC stream for the TypeScript layer.
fn batches_to_ipc_buffer(batches: &[RecordBatch]) -> napi::Result<Buffer> {
let Some(first) = batches.first() else {
return Ok(Buffer::from(Vec::<u8>::new()));
};
let mut out = Vec::new();
let mut writer = arrow_ipc::writer::StreamWriter::try_new(&mut out, &first.schema())
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
for batch in batches {
writer
.write(batch)
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
}
writer
.finish()
.map_err(|e| napi::Error::from_reason(e.to_string()))?;
drop(writer);
Ok(Buffer::from(out))
}
/// A row from `Connection.listJobs`: one server-side job.
#[napi(object)]
pub struct JobInfo {
/// The job id -- what `Connection.openJob` and `Connection.cancelJob`
/// The job id -- what `Connection.getJob` and `Connection.cancelJob`
/// accept.
pub job_id: String,
/// The table the job runs against, without URI or namespace.
@@ -180,3 +91,36 @@ pub struct JobFailureInfo {
pub message: Option<String>,
pub retryable: Option<bool>,
}
/// A described job from `Connection.getJob`.
#[napi(object)]
pub struct JobDescription {
pub job_id: String,
pub job_type: String,
/// Lifecycle state: "running", "finished", "failed", or "cancelled".
pub state: String,
/// When the job was created, in milliseconds since the epoch.
pub creation_ms: i64,
/// The job-type-specific specification as a JSON string, when present.
pub spec_json: Option<String>,
/// Why the job failed, when the job is failed and the server reports a
/// reason.
pub failure: Option<JobFailureInfo>,
}
impl From<lancedb::database::JobDescription> for JobDescription {
fn from(description: lancedb::database::JobDescription) -> Self {
Self {
job_id: description.job_id,
job_type: description.job_type,
state: description.state,
creation_ms: description.creation_ms,
spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
failure: description.failure.map(|failure| JobFailureInfo {
phase: failure.phase,
message: failure.message,
retryable: failure.retryable,
}),
}
}
}
+2 -3
View File
@@ -1,6 +1,6 @@
[package]
name = "lancedb-python"
version = "0.39.0-beta.6"
version = "0.39.0-beta.1"
publish = false
edition.workspace = true
description = "Python bindings for LanceDB"
@@ -28,7 +28,7 @@ env_logger.workspace = true
log.workspace = true
# Maturin enables extension-module mode for Python builds. Keeping it out of
# Cargo features lets Rust unit tests link against libpython.
pyo3 = { version = "0.28", features = ["abi3-py310", "chrono", "uuid"] }
pyo3 = { version = "0.28", features = ["abi3-py310", "chrono"] }
chrono.workspace = true
pyo3-async-runtimes = { version = "0.28", features = [
"attributes",
@@ -40,7 +40,6 @@ serde.workspace = true
serde_json.workspace = true
snafu.workspace = true
tokio.workspace = true
uuid.workspace = true
libc = "0.2"
[build-dependencies]
-1
View File
@@ -139,7 +139,6 @@ include = [
"python/lancedb/exceptions.py",
"python/lancedb/background_loop.py",
"python/lancedb/schema.py",
"python/lancedb/sql.py",
"python/lancedb/remote/__init__.py",
"python/lancedb/remote/errors.py",
"python/lancedb/embeddings/__init__.py",
-18
View File
@@ -22,11 +22,7 @@ from .remote.db import RemoteDBConnection
from .expr import Expr, col, lit, func
from .schema import blob, vector
from .job import AsyncJob, Job
from .sql import AsyncQuery as AsyncSqlQuery
from .sql import Query as SqlQuery
from .sql import QueryDescription
from .functions import (
AssignmentMapping as AssignmentMapping,
FunctionArtifactRequest as FunctionArtifactRequest,
FunctionApplication as FunctionApplication,
FunctionBinding as FunctionBinding,
@@ -105,7 +101,6 @@ def connect(
api_key: Optional[str] = None,
region: str = "us-east-1",
host_override: Optional[str] = None,
sql_host_override: Optional[str] = None,
read_consistency_interval: Optional[timedelta] = None,
request_thread_pool: Optional[Union[int, ThreadPoolExecutor]] = None,
client_config: Union[ClientConfig, Dict[str, Any], None] = None,
@@ -134,9 +129,6 @@ def connect(
The region to use for LanceDB Cloud.
host_override: str, optional
The override url for LanceDB Cloud.
sql_host_override: str, optional
The remote SQL service endpoint override. The client connects lazily when SQL
is first executed and retains that connection.
read_consistency_interval: timedelta, default None
The interval at which to check for updates to the table from other
processes. If None, then consistency is not checked. For performance
@@ -278,7 +270,6 @@ def connect(
api_key,
region,
host_override,
sql_host_override=sql_host_override,
# TODO: remove this (deprecation warning downstream)
request_thread_pool=request_thread_pool,
client_config=client_config,
@@ -421,7 +412,6 @@ def deserialize_conn(
parsed["api_key"],
parsed.get("region", "us-east-1"),
host_override=parsed.get("host_override"),
sql_host_override=parsed.get("sql_host_override"),
client_config=parsed.get("client_config"),
storage_options=storage_options,
)
@@ -435,7 +425,6 @@ async def connect_async(
api_key: Optional[str] = None,
region: str = "us-east-1",
host_override: Optional[str] = None,
sql_host_override: Optional[str] = None,
read_consistency_interval: Optional[timedelta] = None,
client_config: Optional[Union[ClientConfig, Dict[str, Any]]] = None,
storage_options: Optional[Dict[str, str]] = None,
@@ -458,9 +447,6 @@ async def connect_async(
The region to use for LanceDB Cloud.
host_override: str, optional
The override url for LanceDB Cloud.
sql_host_override: str, optional
The remote SQL service endpoint override. The client connects lazily when SQL
is first executed and retains that connection.
read_consistency_interval: timedelta, default None
The interval at which to check for updates to the table from other
processes. If None, then consistency is not checked. For performance
@@ -548,7 +534,6 @@ async def connect_async(
api_key,
region,
host_override,
sql_host_override,
read_consistency_interval_secs,
client_config,
storage_options,
@@ -571,7 +556,6 @@ __all__ = [
"connect_namespace_async",
"AsyncConnection",
"AsyncJob",
"AsyncSqlQuery",
"AsyncLanceNamespaceDBConnection",
"AsyncTable",
"FtsToken",
@@ -586,8 +570,6 @@ __all__ = [
"vector",
"DBConnection",
"Job",
"QueryDescription",
"SqlQuery",
"LanceDBConnection",
"LanceNamespaceDBConnection",
"LsmWriteSpec",
+6 -45
View File
@@ -1,7 +1,6 @@
from datetime import date, datetime, timedelta
from decimal import Decimal
from typing import Dict, List, Optional, Tuple, Any, TypedDict, Union, Literal
from uuid import UUID
import pyarrow as pa
@@ -148,20 +147,17 @@ class Connection(object):
start_after: Optional[str],
limit: Optional[int],
) -> list[str]: ... # Deprecated: Use list_tables instead
async def open_job(self, job_id: str) -> Job: ...
def job(self, job_id: str) -> Job: ...
async def create_function_async(self, request_json: str) -> Job: ...
async def get_function(self, name: str, version: str) -> str: ...
async def list_functions(self) -> List[str]: ...
async def drop_function(self, name: str, version: str) -> bool: ...
async def list_jobs(self) -> List[JobInfo]: ...
async def get_job(self, job_id: str) -> Optional[JobDescription]: ...
async def cancel_job(self, job_id: str) -> bool: ...
async def execute_query_async(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> SqlQuery: ...
async def describe_query(self, query_id: UUID) -> QueryDescription: ...
async def job_history(
self, job_id: Optional[str] = None
) -> List[pa.RecordBatch]: ...
async def create_table(
self,
name: str,
@@ -240,20 +236,9 @@ class BlobFile:
class Job:
@property
def id(self) -> Optional[str]: ...
@property
def _state(self) -> Optional[str]: ...
@property
def _description(self) -> Optional[JobDescription]: ...
async def status(self) -> str: ...
async def wait(self) -> Optional[str]: ...
async def cancel(self) -> None: ...
async def refresh(self) -> None: ...
async def events(
self,
*,
limit: Optional[int] = None,
filter: Optional[str] = None,
) -> pa.Table: ...
class JobInfo:
@property
@@ -285,33 +270,10 @@ class JobDescription:
@property
def creation_ms(self) -> int: ...
@property
def _spec_json(self) -> Optional[str]: ...
@property
def _result_json(self) -> Optional[str]: ...
@property
def spec(self) -> Optional[Any]: ...
@property
def result(self) -> Optional[Any]: ...
def spec_json(self) -> Optional[str]: ...
@property
def failure(self) -> Optional[JobFailureInfo]: ...
class SqlQuery:
@property
def id(self) -> UUID: ...
async def describe(self) -> QueryDescription: ...
async def reader(self) -> RecordBatchStream: ...
async def cancel(self) -> None: ...
class QueryDescription:
@property
def id(self) -> UUID: ...
@property
def status(self) -> str: ...
@property
def progress(self) -> Optional[float]: ...
@property
def expires_at(self) -> Optional[datetime]: ...
class Table:
def name(self) -> str: ...
def __repr__(self) -> str: ...
@@ -490,7 +452,6 @@ async def connect(
api_key: Optional[str],
region: Optional[str],
host_override: Optional[str],
sql_host_override: Optional[str],
read_consistency_interval: Optional[float],
client_config: Optional[Union[ClientConfig, Dict[str, Any]]],
storage_options: Optional[Dict[str, str]],
+62 -93
View File
@@ -19,7 +19,6 @@ from typing import (
Optional,
Union,
)
from uuid import UUID
if sys.version_info >= (3, 12):
from typing import override
@@ -48,9 +47,6 @@ from . import __version__
from ._lancedb import connect as lancedb_connect # type: ignore
from .functions import FunctionVersion, UdfDefinition
from .job import AsyncJob, Job, _typed_job
from .sql import AsyncQuery as AsyncSqlQuery
from .sql import Query as SqlQuery
from .sql import QueryDescription
from .materialized_view import (
AsyncMaterializedView,
MaterializedView,
@@ -72,11 +68,10 @@ import deprecation
if TYPE_CHECKING:
import pyarrow as pa
from .arrow import AsyncRecordBatchReader
from .pydantic import LanceModel
from ._lancedb import Connection as LanceDbConnection
from ._lancedb import JobInfo
from ._lancedb import JobDescription, JobInfo
from .common import DATA, URI
from .embeddings import EmbeddingFunctionConfig
from ._lancedb import Session
@@ -745,23 +740,26 @@ class DBConnection(EnforceOverrides):
"Function catalog operations are not supported for this connection type"
)
def open_job(self, job_id: str) -> Job:
"""Open a server-side job by id, returning a handle with its record
already populated.
def job(self, job_id: str) -> Job:
"""A [Job][lancedb.job.Job] handle for a server-side job by id.
The returned [Job][lancedb.job.Job] answers for its own state,
specification, result, failure and event history, so there is no
separate connection-level call for any of them.
Raises `JobNotFoundError` when the server has no such job, the way
`open_table` does for a missing table.
The handle is constructed without a server round trip; an unknown id
surfaces when the handle is used. Dropping the handle has no effect
on the job itself.
"""
raise NotImplementedError("open_job is not supported for this connection type")
raise NotImplementedError("job is not supported for this connection type")
def list_jobs(self) -> List[JobInfo]:
"""List server-side jobs across the database's tables."""
raise NotImplementedError("list_jobs is not supported for this connection type")
def get_job(self, job_id: str) -> Optional[JobDescription]:
"""Describe a single server-side job by id.
Returns None when the server has no such job.
"""
raise NotImplementedError("get_job is not supported for this connection type")
def cancel_job(self, job_id: str) -> bool:
"""Request cancellation of a server-side job by id.
@@ -773,38 +771,14 @@ class DBConnection(EnforceOverrides):
"cancel_job is not supported for this connection type"
)
def execute_query(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> pa.RecordBatchReader:
"""Execute SQL and return a blocking Arrow reader.
def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
"""The lifecycle event history of a server-side job, as Arrow batches.
This submits through :meth:`execute_query_async` and waits until the
initial result stream is readable. It does not wait for the full query
to finish.
Lists history across all jobs when `job_id` is None.
"""
return self.execute_query_async(
query,
default_namespace_path=default_namespace_path,
).reader()
def execute_query_async(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> SqlQuery:
"""Start executing SQL and return its query handle.
Local connections do not support SQL.
"""
raise NotImplementedError("SQL is not supported for this connection type")
def describe_query(self, query_id: UUID) -> QueryDescription:
"""Describe a submitted SQL query by its connection-scoped id."""
raise NotImplementedError("SQL is not supported for this connection type")
raise NotImplementedError(
"job_history is not supported for this connection type"
)
class LanceDBConnection(DBConnection):
@@ -901,7 +875,6 @@ class LanceDBConnection(DBConnection):
None,
None,
None,
None,
read_consistency_interval_secs,
None,
storage_options,
@@ -1450,11 +1423,14 @@ class LanceDBConnection(DBConnection):
)
@override
def open_job(self, job_id: str) -> Job:
"""Open a server-side job by id. See
[DBConnection.open_job][lancedb.db.DBConnection.open_job].
def job(self, job_id: str) -> Job:
"""A [Job][lancedb.job.Job] handle for a server-side job by id.
The handle is constructed without a server round trip; an unknown id
surfaces when the handle is used. Dropping the handle has no effect
on the job itself.
"""
return Job(LOOP.run(self._conn.open_job(job_id)))
return Job(self._conn.job(job_id))
@override
def create_function_async(self, definition: UdfDefinition) -> Job[FunctionVersion]:
@@ -1478,6 +1454,14 @@ class LanceDBConnection(DBConnection):
"""List server-side jobs across the database's tables."""
return LOOP.run(self._conn.list_jobs())
@override
def get_job(self, job_id: str) -> Optional[JobDescription]:
"""Describe a single server-side job by id.
Returns None when the server has no such job.
"""
return LOOP.run(self._conn.get_job(job_id))
@override
def cancel_job(self, job_id: str) -> bool:
"""Request cancellation of a server-side job by id.
@@ -1488,6 +1472,14 @@ class LanceDBConnection(DBConnection):
"""
return LOOP.run(self._conn.cancel_job(job_id))
@override
def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
"""The lifecycle event history of a server-side job, as Arrow batches.
Lists history across all jobs when `job_id` is None.
"""
return LOOP.run(self._conn.job_history(job_id))
@override
def namespace_client(self) -> LanceNamespace:
"""Get the equivalent namespace client for this connection.
@@ -2258,11 +2250,15 @@ class AsyncConnection(object):
namespace_path = []
await self._inner.drop_all_tables(namespace_path=namespace_path)
async def open_job(self, job_id: str) -> AsyncJob:
"""Open a server-side job by id. See
[DBConnection.open_job][lancedb.db.DBConnection.open_job].
def job(self, job_id: str) -> AsyncJob:
"""An [AsyncJob][lancedb.job.AsyncJob] handle for a server-side job
by id.
The handle is constructed without a server round trip; an unknown id
surfaces when the handle is used. Dropping the handle has no effect
on the job itself.
"""
return AsyncJob(await self._inner.open_job(job_id))
return AsyncJob(self._inner.job(job_id))
async def create_function_async(
self, definition: UdfDefinition
@@ -2302,6 +2298,13 @@ class AsyncConnection(object):
"""List server-side jobs across the database's tables."""
return await self._inner.list_jobs()
async def get_job(self, job_id: str) -> Optional[JobDescription]:
"""Describe a single server-side job by id.
Returns None when the server has no such job.
"""
return await self._inner.get_job(job_id)
async def cancel_job(self, job_id: str) -> bool:
"""Request cancellation of a server-side job by id.
@@ -2311,46 +2314,12 @@ class AsyncConnection(object):
"""
return await self._inner.cancel_job(job_id)
async def execute_query(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> AsyncRecordBatchReader:
"""Execute SQL and return an asynchronous Arrow reader.
async def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
"""The lifecycle event history of a server-side job, as Arrow batches.
This submits through :meth:`execute_query_async` and waits until the
initial result stream is readable. It does not wait for the full query
to finish.
Lists history across all jobs when `job_id` is None.
"""
submitted = await self.execute_query_async(
query,
default_namespace_path=default_namespace_path,
)
return await submitted.reader()
async def execute_query_async(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> AsyncSqlQuery:
"""Start executing SQL and return its query handle.
The database from ``connect_async`` is used for unqualified database
references. The namespace defaults to ``["public"]``. Local
connections raise ``NotImplementedError``.
"""
return AsyncSqlQuery(
await self._inner.execute_query_async(
query,
default_namespace_path=default_namespace_path,
)
)
async def describe_query(self, query_id: UUID) -> QueryDescription:
"""Describe a submitted SQL query by its connection-scoped id."""
return await self._inner.describe_query(query_id)
return await self._inner.job_history(job_id)
async def namespace_client(self) -> LanceNamespace:
"""Get the equivalent namespace client for this connection.
-6
View File
@@ -35,9 +35,3 @@ class JobCancelledError(RuntimeError):
"""Exception raised when an asynchronous job was cancelled."""
pass
class JobNotFoundError(ValueError):
"""Exception raised when opening a job the server does not have."""
pass
+12 -16
View File
@@ -470,7 +470,11 @@ class InputBinding(_RemoteValue):
class OutputMapping(_RemoteValue):
"""One stable result-field mapping."""
"""One stable result-field mapping.
Assignment state is outside the Slice 1 client contract. During the NULL
transition Lance exposes no public cell-flag identifier to persist here.
"""
result_field: str
output_name: str
@@ -480,13 +484,6 @@ class OutputMapping(_RemoteValue):
nullable: bool
class AssignmentMapping(_RemoteValue):
"""Internal physical column preserving flattened struct validity."""
output_name: str
output_field_id: _Int32
class FunctionBinding(_RemoteValue):
"""Immutable Function binding persisted by the Enterprise table service."""
@@ -494,7 +491,6 @@ class FunctionBinding(_RemoteValue):
function: FunctionVersionRef
inputs: tuple[InputBinding, ...]
outputs: tuple[OutputMapping, ...]
assignment: Optional[AssignmentMapping] = None
input_schema: Optional[Mapping[str, Any]] = None
output_schema: Optional[Mapping[str, Any]] = None
@@ -915,6 +911,8 @@ def _function_output(output: pa.DataType | pa.Field | pa.Schema) -> FunctionOutp
if not fields:
raise ValueError("named-struct Function output must contain at least one field")
if any(field.nullable for field in fields):
raise ValueError("Function output fields must be non-nullable")
for field in fields:
_validate_exact_arrow_field(field)
names = [field.name for field in fields]
@@ -926,7 +924,7 @@ def _function_output(output: pa.DataType | pa.Field | pa.Schema) -> FunctionOutp
FunctionResultField(
name=field.name,
arrow_type=_canonical_arrow_field(field),
nullable=field.nullable,
nullable=False,
)
for field in fields
),
@@ -1309,9 +1307,8 @@ def udf(
Input and output signatures are inferred from supported annotations. For
Arrow types annotations cannot express precisely, pass ``input_schema``
and ``output_schema`` together. Scalar outputs must be non-nullable. Every
named-struct field may be nullable; Enterprise preserves the struct's
validity when the result is expanded into sibling columns.
and ``output_schema`` together. Nullable outputs are rejected because V1
uses physical NULL to represent unassigned computed-column rows.
Parameters
----------
@@ -1323,8 +1320,8 @@ def udf(
Explicit input fields in the exact order of the callable parameters.
Must be provided together with ``output_schema``.
output_schema : pyarrow.DataType, pyarrow.Field, or pyarrow.Schema, optional
Explicit scalar or named-struct output. Scalar outputs must be
non-nullable. Must be provided together with ``input_schema``.
Explicit scalar or named-struct output. Must be non-nullable and be
provided together with ``input_schema``.
pip : sequence of str, optional
Pip requirements for the remote environment.
conda : sequence of str, optional
@@ -1390,7 +1387,6 @@ def udf(
__all__ = [
"AssignmentMapping",
"ApplicationInput",
"FunctionApplication",
"FunctionArtifact",
-224
View File
@@ -4,27 +4,15 @@
"""Handles to operations a server may run asynchronously."""
import asyncio
import json
from datetime import timedelta
from typing import Any, Callable, Generic, Optional, TypeVar, cast
import pyarrow as pa
from lancedb.background_loop import LOOP
from . import _lancedb
from ._lancedb import JobDescription, JobFailureInfo, JobInfo
T = TypeVar("T")
__all__ = [
"AsyncJob",
"Job",
"JobDescription",
"JobFailureInfo",
"JobInfo",
]
class AsyncJob(Generic[T]):
"""A handle to an operation that may still be running.
@@ -90,149 +78,6 @@ class AsyncJob(Generic[T]):
return
await self._inner.cancel()
async def refresh(self) -> None:
"""Ask the backend for this job's current state, and for a server-side
job its full record, then cache it for the properties below.
The properties are all `None` until this runs, because submitting an
operation returns only a job id. `status` fetches the whole record too;
`wait` records only the terminal state it establishes.
"""
if self._inner is None:
return
await self._inner.refresh()
@property
def state(self) -> Optional[str]:
"""The last observed lifecycle state, without contacting the backend.
`None` until the handle has talked to it. See :meth:`AsyncJob.refresh`.
"""
if self._inner is None:
return "finished"
return self._inner._state
@property
def job_type(self) -> Optional[str]:
"""The job's type, as the server names it.
`None` for an in-process job, which has no server-side record.
"""
return self._field("job_type")
@property
def creation_ms(self) -> Optional[int]:
"""When the job was created, in milliseconds since the epoch."""
return self._field("creation_ms")
@property
def spec(self) -> Optional[Any]:
"""The job-type-specific specification it was submitted with."""
return self._field("spec")
@property
def result(self) -> Optional[Any]:
"""The job-type-specific terminal result, as reported data rather than
the typed model :meth:`AsyncJob.wait` returns.
`None` until the job succeeds, so a job that never terminates reports
its progress through :meth:`AsyncJob.events` instead.
"""
return self._field("result")
@property
def failure(self) -> Optional[JobFailureInfo]:
"""Why the job failed, when it failed and the server reports a reason."""
return self._field("failure")
@property
def _spec_json(self) -> Optional[str]:
return self._field("_spec_json")
@property
def _result_json(self) -> Optional[str]:
return self._field("_result_json")
def _field(self, name: str) -> Optional[Any]:
description = self._inner._description if self._inner is not None else None
return getattr(description, name) if description is not None else None
async def events(
self,
*,
limit: Optional[int] = None,
filter: Optional[str] = None,
) -> "pa.Table":
"""This job's recorded lifecycle events.
Where the properties above report a terminal result only once the job
reaches one, events are written as the job runs and outlive the workers
that produced them. A distributed job records a `claim`/`claim_complete`
pair per unit of work, each carrying `rows_processed`, so a job that
never finishes still accounts for what it did.
Parameters
----------
limit: int, optional
Maximum event rows to return. The server caps results at 1000 by
default and 10,000 at most, and truncates without saying so, so
pass this for a job that emits an event per fragment.
filter: str, optional
SQL-like expression over the `state`, `updated_by`, `emitted_from`,
`emitted_by`, and `claim_entity` columns, such as
``state = 'claim_complete'``.
"""
if self._inner is None:
raise NotImplementedError(
"job event history is only available for server-side jobs"
)
return await self._inner.events(limit=limit, filter=filter)
def __repr__(self) -> str:
return _job_repr("AsyncJob", self)
_REPR_INDENT = " " * 4
def _repr_payload(value: Any) -> str:
"""Render a job payload as indented JSON, aligned under its field."""
try:
rendered = json.dumps(value, indent=4)
except TypeError:
return repr(value)
return rendered.replace("\n", "\n" + _REPR_INDENT)
def _job_repr(kind: str, job: Any) -> str:
"""Render every field the handle currently knows, omitting the rest.
One field per line, with the JSON payloads indented, because a refresh
job's spec and result are the point of printing it.
"""
state = job.state
if state is None:
# Nothing has been fetched yet, so there is nothing to lay out.
known = f"id={job.id!r}, " if job.id is not None else ""
return f"{kind}({known}not refreshed)"
fields = []
if job.id is not None:
fields.append(f"id={job.id!r}")
fields.append(f"state={state!r}")
for name in ("job_type", "creation_ms"):
value = getattr(job, name)
if value is not None:
fields.append(f"{name}={value!r}")
for name in ("spec", "result"):
value = getattr(job, name)
if value is not None:
fields.append(f"{name}={_repr_payload(value)}")
if job.failure is not None:
fields.append(f"failure={job.failure!r}")
body = "".join(f"\n{_REPR_INDENT}{field}," for field in fields)
return f"{kind}({body}\n)"
class Job(Generic[T]):
"""Synchronous counterpart of `AsyncJob` with the same result type."""
@@ -277,75 +122,6 @@ class Job(Generic[T]):
return
LOOP.run(self._inner.cancel())
def refresh(self) -> None:
"""Ask the backend for this job's current state and record.
See :meth:`AsyncJob.refresh`.
"""
if self._inner is None:
return
LOOP.run(self._inner.refresh())
@property
def state(self) -> Optional[str]:
"""The last observed lifecycle state. See :attr:`AsyncJob.state`."""
return self._inner.state if self._inner is not None else "finished"
@property
def job_type(self) -> Optional[str]:
"""The job's type. See :attr:`AsyncJob.job_type`."""
return self._field("job_type")
@property
def creation_ms(self) -> Optional[int]:
"""When the job was created. See :attr:`AsyncJob.creation_ms`."""
return self._field("creation_ms")
@property
def spec(self) -> Optional[Any]:
"""The job's specification. See :attr:`AsyncJob.spec`."""
return self._field("spec")
@property
def result(self) -> Optional[Any]:
"""The job's terminal result. See :attr:`AsyncJob.result`."""
return self._field("result")
@property
def failure(self) -> Optional[JobFailureInfo]:
"""Why the job failed. See :attr:`AsyncJob.failure`."""
return self._field("failure")
@property
def _spec_json(self) -> Optional[str]:
return self._field("_spec_json")
@property
def _result_json(self) -> Optional[str]:
return self._field("_result_json")
def _field(self, name: str) -> Optional[Any]:
return getattr(self._inner, name) if self._inner is not None else None
def events(
self,
*,
limit: Optional[int] = None,
filter: Optional[str] = None,
) -> "pa.Table":
"""This job's recorded lifecycle events.
See :meth:`AsyncJob.events`.
"""
if self._inner is None:
raise NotImplementedError(
"job event history is only available for server-side jobs"
)
return LOOP.run(self._inner.events(limit=limit, filter=filter))
def __repr__(self) -> str:
return _job_repr("Job", self)
def _typed_job(
inner: "_lancedb.Job", result_decoder: Callable[[str], T]
-35
View File
@@ -12,7 +12,6 @@ from __future__ import annotations
import sys
from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Union
from uuid import UUID
if sys.version_info >= (3, 12):
from typing import override
@@ -49,11 +48,8 @@ from lancedb._lancedb import (
connect_namespace_client as _connect_namespace_client,
)
from lancedb.background_loop import LOOP
from lancedb.arrow import AsyncRecordBatchReader
from lancedb.db import AsyncConnection, DBConnection
from lancedb.job import AsyncJob, Job
from lancedb.sql import AsyncQuery as AsyncSqlQuery
from lancedb.sql import QueryDescription
from lance_namespace import (
LanceNamespace,
connect as namespace_connect,
@@ -1451,37 +1447,6 @@ class AsyncLanceNamespaceDBConnection:
namespace_path=namespace_path, page_token=page_token, limit=limit
)
async def execute_query(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> AsyncRecordBatchReader:
"""Execute SQL when supported by the underlying connection."""
return await self._inner.execute_query(
query,
default_namespace_path=default_namespace_path,
)
async def execute_query_async(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> AsyncSqlQuery:
"""Start executing SQL when supported by the underlying connection.
Namespace-backed local connections do not support SQL.
"""
return await self._inner.execute_query_async(
query,
default_namespace_path=default_namespace_path,
)
async def describe_query(self, query_id: UUID) -> QueryDescription:
"""Describe a submitted SQL query when supported."""
return await self._inner.describe_query(query_id)
async def namespace_client(self) -> LanceNamespace:
"""Get the namespace client for this connection.
+20 -39
View File
@@ -9,7 +9,6 @@ from concurrent.futures import ThreadPoolExecutor
import sys
from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Union
from urllib.parse import urlparse
from uuid import UUID
import warnings
if sys.version_info >= (3, 12):
@@ -26,12 +25,10 @@ from ..common import DATA
from ..db import DBConnection, LOOP
from ..functions import FunctionVersion, UdfDefinition
from ..job import AsyncJob, Job
from ..sql import Query as SqlQuery
from ..sql import QueryDescription
from ..materialized_view import MaterializedView, SelectArg
if TYPE_CHECKING:
from .._lancedb import JobInfo
from .._lancedb import JobDescription, JobInfo
from ..embeddings import EmbeddingFunctionConfig
from lance_namespace import (
LanceNamespace,
@@ -119,7 +116,6 @@ class RemoteDBConnection(DBConnection):
read_timeout: Optional[float] = None,
storage_options: Optional[Dict[str, str]] = None,
read_consistency_interval: Optional[timedelta] = None,
sql_host_override: Optional[str] = None,
):
"""Connect to a remote LanceDB database."""
if isinstance(client_config, dict):
@@ -165,7 +161,6 @@ class RemoteDBConnection(DBConnection):
self.api_key = api_key
self.region = region
self.host_override = host_override
self.sql_host_override = sql_host_override
self.storage_options = storage_options
self.db_name = parsed.netloc
@@ -180,7 +175,6 @@ class RemoteDBConnection(DBConnection):
api_key=api_key,
region=region,
host_override=host_override,
sql_host_override=sql_host_override,
client_config=client_config,
storage_options=storage_options,
read_consistency_interval=read_consistency_interval,
@@ -199,7 +193,6 @@ class RemoteDBConnection(DBConnection):
"api_key": self.api_key,
"region": self.region,
"host_override": self.host_override,
"sql_host_override": self.sql_host_override,
"client_config": _client_config_to_dict(self.client_config),
"storage_options": self.storage_options,
}
@@ -739,11 +732,14 @@ class RemoteDBConnection(DBConnection):
)
@override
def open_job(self, job_id: str) -> Job:
"""Open a server-side job by id. See
[DBConnection.open_job][lancedb.db.DBConnection.open_job].
def job(self, job_id: str) -> Job:
"""A [Job][lancedb.job.Job] handle for a server-side job by id.
The handle is constructed without a server round trip; an unknown id
surfaces when the handle is used. Dropping the handle has no effect
on the job itself.
"""
return Job(LOOP.run(self._conn.open_job(job_id)))
return Job(self._conn.job(job_id))
@override
def create_function_async(self, definition: UdfDefinition) -> Job[FunctionVersion]:
@@ -766,6 +762,14 @@ class RemoteDBConnection(DBConnection):
"""List server-side jobs across the database's tables."""
return LOOP.run(self._conn.list_jobs())
@override
def get_job(self, job_id: str) -> Optional["JobDescription"]:
"""Describe a single server-side job by id.
Returns None when the server has no such job.
"""
return LOOP.run(self._conn.get_job(job_id))
@override
def cancel_job(self, job_id: str) -> bool:
"""Request cancellation of a server-side job by id.
@@ -777,35 +781,12 @@ class RemoteDBConnection(DBConnection):
return LOOP.run(self._conn.cancel_job(job_id))
@override
def execute_query_async(
self,
query: str,
*,
default_namespace_path: Optional[List[str]] = None,
) -> SqlQuery:
"""Start executing SQL through this remote connection.
def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
"""The lifecycle event history of a server-side job, as Arrow batches.
Unqualified tables use this connection's database and the
``["public"]`` namespace by default. Fully qualified table names may
reference other databases available to the same deployment.
Lists history across all jobs when `job_id` is None.
"""
return SqlQuery(
LOOP.run(
self._conn.execute_query_async(
query,
default_namespace_path=default_namespace_path,
)
)
)
@override
def describe_query(self, query_id: UUID) -> QueryDescription:
"""Describe a submitted SQL query by its connection-scoped id."""
return LOOP.run(
self._conn.describe_query(
query_id,
)
)
return LOOP.run(self._conn.job_history(job_id))
@override
def namespace_client(self) -> LanceNamespace:
+1 -4
View File
@@ -177,7 +177,4 @@ class OAuthProvider(HeaderProvider):
if not self._current_token:
raise RuntimeError("Failed to obtain OAuth token")
return {
"Authorization": f"Bearer {self._current_token}",
"x-lancedb-credential-type": "oidc",
}
return {"Authorization": f"Bearer {self._current_token}"}
-88
View File
@@ -1,88 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright The LanceDB Authors
"""Handles to SQL queries running on a remote database."""
from uuid import UUID
import pyarrow as pa
from lancedb.background_loop import LOOP
from . import _lancedb
from .arrow import AsyncRecordBatchReader
QueryDescription = _lancedb.QueryDescription
class AsyncQuery:
"""A handle to a submitted SQL query on an asynchronous connection."""
def __init__(self, inner: "_lancedb.SqlQuery"):
self._inner = inner
@property
def id(self) -> UUID:
"""The stable identifier scoped to the connection that submitted it."""
return self._inner.id
async def describe(self) -> QueryDescription:
"""Get a point-in-time description of the query."""
return await self._inner.describe()
async def reader(self) -> AsyncRecordBatchReader:
"""Wait for the initial result stream and return its Arrow reader.
Results are single-consumer. Calling this method more than once on the
same query raises an error. Later batches are streamed as they become
available without waiting for the full query to finish.
"""
return AsyncRecordBatchReader(await self._inner.reader())
async def cancel(self) -> None:
"""Request cancellation of the query."""
await self._inner.cancel()
class Query:
"""Synchronous counterpart of :class:`AsyncQuery`."""
def __init__(self, inner: AsyncQuery):
self._inner = inner
@property
def id(self) -> UUID:
"""The stable identifier scoped to the connection that submitted it."""
return self._inner.id
def describe(self) -> QueryDescription:
"""Get a point-in-time description of the query."""
return LOOP.run(self._inner.describe())
def reader(self) -> pa.RecordBatchReader:
"""Wait for the initial result stream and return a blocking reader.
Results are single-consumer. Calling this method more than once on the
same query raises an error. Later batches block only until they become
available, without waiting for the full query to finish.
"""
reader = LOOP.run(self._inner.reader())
def next_batch():
try:
return LOOP.run(reader.__anext__())
except StopAsyncIteration:
return None
def batches():
while (batch := next_batch()) is not None:
yield batch
return pa.RecordBatchReader.from_batches(reader.schema, batches())
def cancel(self) -> None:
"""Request cancellation of the query."""
LOOP.run(self._inner.cancel())
__all__ = ["AsyncQuery", "Query", "QueryDescription"]
-9
View File
@@ -1793,9 +1793,6 @@ class Table(ABC):
The result has the same length and order as ``row_ids``. Null blobs
produce null slots; valid empty blobs produce ``b""``.
``_rowid`` values stay valid after compaction when the table has stable
row ids.
Convenience for small payloads. For large values use
:meth:`fetch_blob_files`.
"""
@@ -1813,9 +1810,6 @@ class Table(ABC):
The result has the same length and order as ``requests``; null blobs
produce null slots and empty ranges on non-null blobs produce ``b""``.
``_rowid`` values stay valid after compaction when the table has stable
row ids.
Row IDs can be obtained from a query with ``with_row_id(True)``. This
API is currently supported only by local tables.
"""
@@ -1831,9 +1825,6 @@ class Table(ABC):
``_rowid`` or a ``_lance_row_id`` field on the blob descriptor. Null
rows are ``None``. Remote tables require LanceDB Cloud server 0.5.0 or
newer.
``_rowid`` values stay valid after compaction when the table has stable
row ids.
"""
@abstractmethod
+2 -60
View File
@@ -66,25 +66,6 @@ def _row_ids_by_id(table):
return dict(zip(hits["id"].to_pylist(), hits["_rowid"].to_pylist()))
def _assert_missing_blob_row_ids(exc_info):
message = str(exc_info.value)
assert "row ids" in message
assert "rowaddr" not in message
assert "fragment" not in message
def _assert_fetch_apis_reject_missing_row_ids(table, row_ids):
with pytest.raises(ValueError) as exc_info:
table.fetch_blobs("image", row_ids)
_assert_missing_blob_row_ids(exc_info)
with pytest.raises(ValueError) as exc_info:
table.fetch_blob_files("image", row_ids)
_assert_missing_blob_row_ids(exc_info)
with pytest.raises(ValueError) as exc_info:
table.fetch_blob_ranges("image", [(row_id, 0, 1) for row_id in row_ids])
_assert_missing_blob_row_ids(exc_info)
def test_blob_factory_declares_v2_field():
field = lancedb.blob("image")
assert isinstance(field.type, pa.ExtensionType)
@@ -297,10 +278,7 @@ def test_blob_v2_projection_sources_use_typed_column_name():
def _legacy_v1_table(name):
# Legacy v1 blob columns are only writable at file version <= 2.1.
db = lancedb.connect(
"memory:///", storage_options={"new_table_data_storage_version": "2.1"}
)
db = lancedb.connect("memory:///")
schema = pa.schema(
[
pa.field("id", pa.int64()),
@@ -713,25 +691,6 @@ def test_fetch_blobs_accepts_query_result():
assert {blobs[i].as_py() for i in range(len(blobs))} == {b"gamma"}
def test_fetch_blobs_after_compact_with_stable_row_ids(tmp_path):
db = lancedb.connect(
tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
)
schema = pa.schema([pa.field("id", pa.int64()), lancedb.blob("image")])
table = db.create_table("t", schema=schema)
table.add([{"id": 1, "image": b"frag-one"}])
table.add([{"id": 2, "image": b"frag-two"}])
by_id = _row_ids_by_id(table)
ids = [by_id[1], by_id[2]]
table.optimize()
blobs = table.fetch_blobs("image", ids)
assert blobs.to_pylist() == [b"frag-one", b"frag-two"]
ranges = table.fetch_blob_ranges("image", [(ids[0], 5, 3), (ids[1], 5, 3)])
assert ranges.to_pylist() == [b"one", b"two"]
def test_fetch_blobs_preserves_null_and_empty_values():
table = _blob_table(
"nulls",
@@ -780,25 +739,8 @@ def test_fetch_blob_ranges_validates_requests():
with pytest.raises(ValueError, match="offset \\+ length overflowed"):
table.fetch_blob_ranges("image", [(row_id, 2**64 - 1, 1)])
with pytest.raises(ValueError) as exc_info:
with pytest.raises(ValueError, match="row IDs"):
table.fetch_blob_ranges("image", [(2**64 - 1, 0, 1)])
_assert_missing_blob_row_ids(exc_info)
def test_fetch_blob_apis_reject_missing_fragment_row_addr():
table = _blob_table("missing_frag", [{"id": 1, "image": b"x"}])
live = _row_ids_by_id(table)[1]
_assert_fetch_apis_reject_missing_row_ids(table, [1 << 32, live])
def test_fetch_blob_apis_reject_deleted_row_ids():
table = _blob_table(
"deleted_rows",
[{"id": 1, "image": b"one"}, {"id": 2, "image": b"two"}],
)
by_id = _row_ids_by_id(table)
table.delete("id = 2")
_assert_fetch_apis_reject_missing_row_ids(table, [by_id[2], by_id[1]])
def test_fetch_blob_ranges_empty_requests_returns_empty_array():
@@ -850,43 +850,6 @@ def test_named_struct_function_can_include_a_blob_result_field():
]
def test_named_struct_function_preserves_nullable_result_fields():
@udf(
input_schema=pa.schema([pa.field("value", pa.int64(), nullable=False)]),
output_schema=pa.schema(
[
pa.field("result", pa.int64(), nullable=True),
pa.field("failure_code", pa.int32(), nullable=False),
]
),
)
def nullable_result(value):
return {"result": value, "failure_code": 0}
output = nullable_result.registration_request.signature.output
assert [(field.name, field.nullable) for field in output.fields] == [
("result", True),
("failure_code", False),
]
@udf(
input_schema=pa.schema([pa.field("value", pa.int64(), nullable=False)]),
output_schema=pa.schema(
[
pa.field("result", pa.int64(), nullable=True),
pa.field("failure_code", pa.int32(), nullable=True),
]
),
)
def all_nullable(value):
return {"result": value, "failure_code": None}
assert all(
field.nullable
for field in all_nullable.registration_request.signature.output.fields
)
def test_metadata_marked_blob_field_uses_the_semantic_type():
extension = lancedb.blob("image", nullable=False).type
storage = (
+6 -24
View File
@@ -54,10 +54,7 @@ class TestOAuthProvider:
provider = OAuthProvider(fetcher)
headers = provider.get_headers()
assert headers == {
"Authorization": "Bearer token123",
"x-lancedb-credential-type": "oidc",
}
assert headers == {"Authorization": "Bearer token123"}
assert provider._current_token == "token123"
assert provider._token_expires_at is not None
@@ -76,20 +73,14 @@ class TestOAuthProvider:
# First call
headers1 = provider.get_headers()
assert headers1 == {
"Authorization": "Bearer token1",
"x-lancedb-credential-type": "oidc",
}
assert headers1 == {"Authorization": "Bearer token1"}
# Wait for token to expire
time.sleep(1.1)
# Second call should refresh
headers2 = provider.get_headers()
assert headers2 == {
"Authorization": "Bearer token2",
"x-lancedb-credential-type": "oidc",
}
assert headers2 == {"Authorization": "Bearer token2"}
assert call_count == 2
def test_no_expiry_info(self):
@@ -101,18 +92,12 @@ class TestOAuthProvider:
provider = OAuthProvider(fetcher)
headers = provider.get_headers()
assert headers == {
"Authorization": "Bearer permanent_token",
"x-lancedb-credential-type": "oidc",
}
assert headers == {"Authorization": "Bearer permanent_token"}
assert provider._token_expires_at is None
# Should not refresh on second call
headers2 = provider.get_headers()
assert headers2 == {
"Authorization": "Bearer permanent_token",
"x-lancedb-credential-type": "oidc",
}
assert headers2 == {"Authorization": "Bearer permanent_token"}
def test_missing_access_token(self):
"""Test error handling when access_token is missing."""
@@ -136,10 +121,7 @@ class TestOAuthProvider:
provider = OAuthProvider(fetcher)
headers = provider.get_headers()
assert headers == {
"Authorization": "Bearer sync_token",
"x-lancedb-credential-type": "oidc",
}
assert headers == {"Authorization": "Bearer sync_token"}
class TestClientConfigIntegration:
+1 -7
View File
@@ -193,13 +193,7 @@ class TestNamespaceConnection:
),
)
# Legacy v1 blob columns are only writable at file version <= 2.1.
table = db.create_table(
"blob_table",
data,
namespace_path=["test_ns"],
storage_options={"new_table_data_storage_version": "2.1"},
)
table = db.create_table("blob_table", data, namespace_path=["test_ns"])
df = table.to_pandas(blob_mode="lazy").sort_values("id")
blob = df["blob"].iloc[0]
+10 -38
View File
@@ -40,10 +40,6 @@ from utils import exception_output
from importlib.util import find_spec
# Legacy v1 blob columns are only writable at file version <= 2.1.
LEGACY_BLOB_STORAGE_OPTIONS = {"new_table_data_storage_version": "2.1"}
def _blob_query_data():
return pa.table(
{
@@ -123,17 +119,13 @@ def _assert_blob_bytes_projection(df):
def _blob_query_table(db, name, blob_schema):
if blob_schema == "v1":
return db.create_table(
name, _blob_query_data(), storage_options=LEGACY_BLOB_STORAGE_OPTIONS
)
return db.create_table(name, _blob_query_data())
return _create_blob_v2_query_table(db, name)
async def _blob_query_table_async(db, name, blob_schema):
if blob_schema == "v1":
return await db.create_table(
name, _blob_query_data(), storage_options=LEGACY_BLOB_STORAGE_OPTIONS
)
return await db.create_table(name, _blob_query_data())
return await _create_blob_v2_query_table_async(db, name)
@@ -283,9 +275,7 @@ async def test_query_to_pandas_kwargs(table, table_async):
def test_plain_scan_query_to_pandas_blob_modes(tmp_db, blob_mode):
pytest.importorskip("lance")
table = tmp_db.create_table(
f"test_query_to_pandas_blob_{blob_mode}",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
f"test_query_to_pandas_blob_{blob_mode}", _blob_query_data()
)
df = (
@@ -332,9 +322,7 @@ def test_plain_scan_query_to_pandas_blob_mode_does_not_collect_arrow(
):
pytest.importorskip("lance")
table = tmp_db.create_table(
"test_query_to_pandas_blob_no_arrow_collect",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
"test_query_to_pandas_blob_no_arrow_collect", _blob_query_data()
)
query = table.search().where("id = 1").select(["id", "blob"])
@@ -359,9 +347,7 @@ def test_plain_scan_query_to_pandas_blob_descriptions_flatten_uses_scanner(
):
pytest.importorskip("lance")
table = tmp_db.create_table(
"test_query_to_pandas_blob_desc_flatten",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
"test_query_to_pandas_blob_desc_flatten", _blob_query_data()
)
query = table.search().where("id = 1").select(["id", "blob"])
@@ -379,11 +365,7 @@ def test_plain_scan_query_to_pandas_blob_descriptions_flatten_uses_scanner(
def test_plain_scan_query_to_pandas_scanner_state(tmp_db):
pytest.importorskip("lance")
data = _blob_query_data()
table = tmp_db.create_table(
"test_query_to_pandas_scanner_state",
data.slice(0, 2),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
)
table = tmp_db.create_table("test_query_to_pandas_scanner_state", data.slice(0, 2))
table.add(data.slice(2, 2))
fragments = table.to_lance().get_fragments()
@@ -418,9 +400,7 @@ def test_plain_scan_query_to_pandas_scanner_state(tmp_db):
async def test_async_plain_scan_query_to_pandas_blob_projection(tmp_db_async):
pytest.importorskip("lance")
table = await tmp_db_async.create_table(
"test_async_query_to_pandas_blob_projection",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
"test_async_query_to_pandas_blob_projection", _blob_query_data()
)
lazy_df = await (
@@ -472,9 +452,7 @@ async def test_async_plain_scan_query_to_pandas_blob_mode_does_not_collect_arrow
):
pytest.importorskip("lance")
table = await tmp_db_async.create_table(
"test_async_query_to_pandas_blob_no_arrow_collect",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
"test_async_query_to_pandas_blob_no_arrow_collect", _blob_query_data()
)
query = table.query().where("id = 1").select(["id", "blob"])
@@ -496,11 +474,7 @@ async def test_async_plain_scan_query_to_pandas_blob_mode_does_not_collect_arrow
def test_vector_query_to_pandas_blob_mode_requires_native_path(tmp_db):
pytest.importorskip("lance")
table = tmp_db.create_table(
"test_vector_query_blob_mode",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
)
table = tmp_db.create_table("test_vector_query_blob_mode", _blob_query_data())
with pytest.raises(RuntimeError, match="Lance native pandas conversion"):
table.search([1.0, 0.0]).select(["blob", "vector"]).limit(1).to_pandas(
@@ -511,9 +485,7 @@ def test_vector_query_to_pandas_blob_mode_requires_native_path(tmp_db):
def test_vector_query_to_pandas_blob_descriptions_requires_plain_scan(tmp_db):
pytest.importorskip("lance")
table = tmp_db.create_table(
"test_vector_query_blob_descriptions",
_blob_query_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
"test_vector_query_blob_descriptions", _blob_query_data()
)
with pytest.raises(RuntimeError, match="plain scan query"):
+15 -117
View File
@@ -2467,7 +2467,7 @@ def test_remote_blob_byte_apis_not_supported_on_old_server():
def test_remote_connection_jobs_surface():
from lancedb.exceptions import JobFailedError, JobNotFoundError
from lancedb.exceptions import JobFailedError
schema = pa.schema([("state", pa.string())])
batch = pa.record_batch([pa.array(["created", "done"])], schema=schema)
@@ -2475,7 +2475,6 @@ def test_remote_connection_jobs_surface():
with pa.ipc.new_stream(sink, schema) as writer:
writer.write_batch(batch)
events_body = sink.getvalue().to_pybytes()
query_events_payloads = []
def handler(request):
content_len = int(request.headers.get("Content-Length", 0))
@@ -2513,22 +2512,6 @@ def test_remote_connection_jobs_surface():
request.end_headers()
request.wfile.write(json.dumps(rsp).encode())
elif request.path == "/v1/jobs/describe":
if payload["job_id"] == "job-2":
request.send_response(200)
request.send_header("Content-Type", "application/json")
request.end_headers()
request.wfile.write(
json.dumps(
dict(
job_id="job-2",
job_type="refresh_column",
job_state="DONE",
creation_ms=2000,
result=dict(rows_assigned=1000000, rows_failed=0),
)
).encode()
)
return
if payload["job_id"] != "job-1":
request.send_response(404)
request.end_headers()
@@ -2560,7 +2543,7 @@ def test_remote_connection_jobs_surface():
request.end_headers()
request.wfile.write(b'{"job_id": "job-1"}')
elif request.path == "/v1/jobs/query_events":
query_events_payloads.append(payload)
assert payload["job_id"] == "job-1"
request.send_response(200)
request.send_header("Content-Type", "application/vnd.apache.arrow.stream")
request.end_headers()
@@ -2576,109 +2559,24 @@ def test_remote_connection_jobs_surface():
assert jobs[0].table == "t1"
assert jobs[1].state == "finished"
description = db.get_job("job-1")
assert description.job_type == "create_index"
assert description.state == "failed"
assert json.loads(description.spec_json) == {"column": "vec"}
assert description.failure.message == "worker died"
assert description.failure.retryable is True
assert db.get_job("missing") is None
assert db.cancel_job("job-1") is True
assert db.cancel_job("missing") is False
# Opening a job hands back a populated handle; a missing one fails.
with pytest.raises(JobNotFoundError, match="missing"):
db.open_job("missing")
finished = db.open_job("job-2")
assert finished.state == "finished"
assert finished.result == {"rows_assigned": 1000000, "rows_failed": 0}
batches = db.job_history("job-1")
assert len(batches) == 1
assert batches[0].num_rows == 2
assert batches[0].column("state").to_pylist() == ["created", "done"]
job = db.open_job("job-1")
job = db.job("job-1")
assert job.id == "job-1"
# Opening already populated the handle.
assert job.state == "failed"
assert job.spec == {"column": "vec"}
assert job.failure.message == "worker died"
assert job.status() == "failed"
with pytest.raises(JobFailedError, match="worker died"):
job.wait(timeout=timedelta(seconds=5))
def test_remote_job_handle_reports_its_own_detail():
schema = pa.schema([("state", pa.string())])
batch = pa.record_batch([pa.array(["claim_complete"])], schema=schema)
sink = pa.BufferOutputStream()
with pa.ipc.new_stream(sink, schema) as writer:
writer.write_batch(batch)
events_body = sink.getvalue().to_pybytes()
event_payloads = []
def handler(request):
content_len = int(request.headers.get("Content-Length", 0))
body = request.rfile.read(content_len) if content_len > 0 else b""
payload = json.loads(body) if body else {}
if request.path == "/v1/jobs/describe":
request.send_response(200)
request.send_header("Content-Type", "application/json")
request.end_headers()
request.wfile.write(
json.dumps(
dict(
job_id="job-1",
job_type="refresh_column",
job_state="DONE",
creation_ms=2000,
spec=dict(column="vec"),
result=dict(rows_assigned=1000000),
)
).encode()
)
elif request.path == "/v1/jobs/query_events":
event_payloads.append(payload)
request.send_response(200)
request.send_header("Content-Type", "application/vnd.apache.arrow.stream")
request.end_headers()
request.wfile.write(events_body)
else:
request.send_response(404)
request.end_headers()
with mock_lancedb_connection(handler) as db:
job = db.open_job("job-1")
# Opening populates the handle in the same round trip.
assert job.state == "finished"
job.refresh()
assert job.job_type == "refresh_column"
assert job.creation_ms == 2000
assert job.spec == {"column": "vec"}
assert job.result == {"rows_assigned": 1000000}
assert job.failure is None
# The JSON payloads stay reachable, but as internal APIs.
assert json.loads(job._spec_json) == {"column": "vec"}
assert json.loads(job._result_json) == {"rows_assigned": 1000000}
# print() shows everything the handle knows and nothing it does not.
# print() lays every known field out on its own line, with the JSON
# payloads indented rather than crammed onto one line.
assert repr(job) == "\n".join(
[
"Job(",
" id='job-1',",
" state='finished',",
" job_type='refresh_column',",
" creation_ms=2000,",
" spec={",
' "column": "vec"',
" },",
" result={",
' "rows_assigned": 1000000',
" },",
")",
]
)
# Nothing it does not know shows up.
assert "failure" not in repr(job)
events = job.events(filter="state = 'claim_complete'", limit=500)
assert isinstance(events, pa.Table)
assert events.column("state").to_pylist() == ["claim_complete"]
# The handle supplies job_id; the caller only narrows the query.
assert event_payloads[-1] == {
"job_id": "job-1",
"limit": 500,
"filter": "state = 'claim_complete'",
}
-162
View File
@@ -1,162 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright The LanceDB Authors
from uuid import UUID
import pytest
import pyarrow as pa
import lancedb
from lancedb import _lancedb
from lancedb.arrow import AsyncRecordBatchReader
from lancedb.db import AsyncConnection
from lancedb.remote.db import RemoteDBConnection
from lancedb.sql import AsyncQuery, Query
NIL_QUERY_ID = UUID(int=0)
class FakeNativeQuery:
id = UUID("0198f1b2-c3d4-7e5f-8123-456789abcdef")
async def reader(self):
return pa.table({"value": [1, 2]})
class FakeNativeConnection:
async def execute_query_async(self, query, *, default_namespace_path=None):
return FakeNativeQuery()
class FakeAsyncConnection:
async def execute_query_async(self, query, *, default_namespace_path=None):
return AsyncQuery(FakeNativeQuery())
def remote_connection(sql_host_override=None):
return lancedb.connect(
"db://analytics",
api_key="test-key",
host_override="http://localhost:10024",
sql_host_override=sql_host_override,
)
def test_sql_is_connection_scoped():
assert hasattr(lancedb, "sql")
assert not callable(lancedb.sql)
assert not hasattr(_lancedb, "sql")
assert not hasattr(remote_connection(), "sql")
assert hasattr(remote_connection(), "execute_query")
assert hasattr(remote_connection(), "execute_query_async")
assert hasattr(remote_connection(), "describe_query")
def test_query_id_is_uuid():
query = AsyncQuery(FakeNativeQuery())
assert isinstance(query.id, UUID)
assert Query(query).id == query.id
def test_connection_serializes_sql_host_override():
endpoint = "grpc+tls://sql.example.com:10026"
restored = lancedb.deserialize_conn(
remote_connection(sql_host_override=endpoint).serialize()
)
assert restored.sql_host_override == endpoint
@pytest.mark.asyncio
async def test_async_sql_reader_is_record_batch_stream():
reader = await AsyncQuery(FakeNativeQuery()).reader()
assert isinstance(reader, AsyncRecordBatchReader)
assert (await reader.read_all())[0].column(0).to_pylist() == [1, 2]
def test_sync_sql_reader_is_record_batch_reader():
reader = Query(AsyncQuery(FakeNativeQuery())).reader()
assert isinstance(reader, pa.RecordBatchReader)
assert reader.read_all().column(0).to_pylist() == [1, 2]
def test_execute_query_returns_blocking_reader():
connection = RemoteDBConnection.__new__(RemoteDBConnection)
connection._conn = FakeAsyncConnection()
reader = connection.execute_query("SELECT 1")
assert isinstance(reader, pa.RecordBatchReader)
assert reader.read_all().column(0).to_pylist() == [1, 2]
@pytest.mark.asyncio
async def test_async_execute_query_returns_async_reader():
connection = AsyncConnection(FakeNativeConnection())
reader = await connection.execute_query("SELECT 1")
assert isinstance(reader, AsyncRecordBatchReader)
assert (await reader.read_all())[0].column(0).to_pylist() == [1, 2]
def test_local_connection_rejects_sql(tmp_path):
connection = lancedb.connect(tmp_path)
with pytest.raises(NotImplementedError, match="SQL"):
connection.execute_query("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
connection.execute_query_async("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
connection.describe_query(NIL_QUERY_ID)
@pytest.mark.asyncio
async def test_local_async_connection_rejects_sql(tmp_path):
connection = await lancedb.connect_async(tmp_path)
with pytest.raises(NotImplementedError, match="SQL"):
await connection.execute_query("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
await connection.execute_query_async("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
await connection.describe_query(NIL_QUERY_ID)
@pytest.mark.asyncio
async def test_async_namespace_connection_rejects_sql(tmp_path):
connection = lancedb.connect_namespace_async("dir", {"root": str(tmp_path)})
with pytest.raises(NotImplementedError, match="SQL"):
await connection.execute_query("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
await connection.execute_query_async("SELECT 1")
with pytest.raises(NotImplementedError, match="SQL"):
await connection.describe_query(NIL_QUERY_ID)
def test_describe_query_requires_uuid():
with pytest.raises(TypeError, match="UUID"):
remote_connection().describe_query(str(NIL_QUERY_ID))
@pytest.mark.parametrize(
"default_namespace_path",
["public", ("public",), [1]],
)
def test_execute_query_async_requires_namespace_path_list(default_namespace_path):
with pytest.raises(ValueError, match="default_namespace_path"):
remote_connection().execute_query_async(
"SELECT 1", default_namespace_path=default_namespace_path
)
def test_execute_query_async_rejects_invalid_endpoint():
connection = remote_connection(sql_host_override="invalid://localhost")
with pytest.raises(ValueError, match="sql_host_override"):
connection.execute_query_async("SELECT 1")
@pytest.mark.parametrize(
"default_namespace_path",
[[""], ["café"], ["pub\tlic"], ["events$raw"]],
)
def test_execute_query_async_rejects_invalid_namespace_components(
default_namespace_path,
):
with pytest.raises(ValueError, match="default_namespace_path"):
remote_connection().execute_query_async(
"SELECT 1", default_namespace_path=default_namespace_path
)
+4 -16
View File
@@ -64,23 +64,15 @@ async def _blob_v2_table_async(db: AsyncConnection, name: str):
return table
# Legacy v1 blob columns are only writable at file version <= 2.1.
LEGACY_BLOB_STORAGE_OPTIONS = {"new_table_data_storage_version": "2.1"}
def _blob_table(db: DBConnection, name: str, blob_schema: str):
if blob_schema == "v1":
return db.create_table(
name, data=_blob_test_data(), storage_options=LEGACY_BLOB_STORAGE_OPTIONS
)
return db.create_table(name, data=_blob_test_data())
return _blob_v2_table(db, name)
async def _blob_table_async(db: AsyncConnection, name: str, blob_schema: str):
if blob_schema == "v1":
return await db.create_table(
name, data=_blob_test_data(), storage_options=LEGACY_BLOB_STORAGE_OPTIONS
)
return await db.create_table(name, data=_blob_test_data())
return await _blob_v2_table_async(db, name)
@@ -155,11 +147,7 @@ def test_table_to_pandas_invalid_blob_mode_non_blob_table(tmp_db: DBConnection):
@pytest.mark.parametrize("blob_mode", ["lazy", "bytes", "descriptions"])
def test_table_to_pandas_blob_modes(tmp_db: DBConnection, blob_mode):
pytest.importorskip("lance")
table = tmp_db.create_table(
f"test_to_pandas_blob_{blob_mode}",
_blob_test_data(),
storage_options=LEGACY_BLOB_STORAGE_OPTIONS,
)
table = tmp_db.create_table(f"test_to_pandas_blob_{blob_mode}", _blob_test_data())
df = table.to_pandas(blob_mode=blob_mode)
@@ -3971,7 +3959,7 @@ def test_stats(mem_db: DBConnection):
print(f"{stats=}")
assert stats == {
# Full on-disk size of the data file, footer and metadata included.
"total_bytes": 637,
"total_bytes": 633,
"num_rows": 2,
"num_indices": 0,
"fragment_stats": {
+36 -68
View File
@@ -13,7 +13,11 @@ use crate::{
runtime::future_into_py,
table::Table,
};
use arrow::{datatypes::Schema, ffi_stream::ArrowArrayStreamReader, pyarrow::FromPyArrow};
use arrow::{
datatypes::Schema,
ffi_stream::ArrowArrayStreamReader,
pyarrow::{FromPyArrow, ToPyArrow},
};
use lancedb::{
connection::Connection as LanceConnection,
connection::NamespaceClientPushdownOperation,
@@ -24,7 +28,7 @@ use pyo3::{
Bound, FromPyObject, Py, PyAny, PyRef, PyResult, Python,
exceptions::{PyRuntimeError, PyValueError},
pyclass, pyfunction, pymethods,
types::{PyAnyMethods, PyDict, PyDictMethods, PyList},
types::{PyDict, PyDictMethods, PyList, PyListMethods},
};
#[pyclass]
@@ -82,24 +86,6 @@ impl Connection {
}
}
fn parse_default_namespace_path(path: Option<Bound<'_, PyAny>>) -> PyResult<Vec<String>> {
match path {
Some(path) => {
if !path.is_instance_of::<PyList>() {
return Err(PyValueError::new_err(
"Connection.execute_query_async default_namespace_path must be a list",
));
}
path.extract::<Vec<String>>().map_err(|_| {
PyValueError::new_err(
"Connection.execute_query_async default_namespace_path components must be strings",
)
})
}
None => Ok(vec!["public".to_string()]),
}
}
#[pymethods]
impl Connection {
fn __repr__(&self) -> String {
@@ -122,40 +108,6 @@ impl Connection {
self.get_inner().map(|inner| inner.uri().to_string())
}
#[pyo3(signature = (query, *, default_namespace_path=None))]
pub fn execute_query_async<'a>(
self_: PyRef<'a, Self>,
query: String,
default_namespace_path: Option<Bound<'_, PyAny>>,
) -> PyResult<Bound<'a, PyAny>> {
let inner = self_.get_inner()?.clone();
let default_namespace_path = parse_default_namespace_path(default_namespace_path)?;
future_into_py(self_.py(), async move {
let operation = inner
.execute_query_async(query)
.default_namespace_path(default_namespace_path);
operation
.execute()
.await
.map(crate::sql::Query::new)
.infer_error()
})
}
pub fn describe_query<'a>(
self_: PyRef<'a, Self>,
query_id: uuid::Uuid,
) -> PyResult<Bound<'a, PyAny>> {
let inner = self_.get_inner()?.clone();
future_into_py(self_.py(), async move {
inner
.describe_query(query_id)
.await
.map(crate::sql::QueryDescription::from)
.infer_error()
})
}
#[pyo3(signature = ())]
pub fn get_read_consistency_interval(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.get_inner()?.clone();
@@ -640,12 +592,9 @@ impl Connection {
})
}
pub fn open_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.get_inner()?.clone();
future_into_py(self_.py(), async move {
let job = inner.open_job(&job_id).await.infer_error()?;
Ok(crate::job::Job::new(job))
})
pub fn job(&self, job_id: String) -> PyResult<crate::job::Job> {
let inner = self.get_inner()?.clone();
Ok(crate::job::Job::new(inner.job(job_id).infer_error()?))
}
pub fn create_function_async(
@@ -715,16 +664,42 @@ impl Connection {
})
}
pub fn get_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.get_inner()?.clone();
future_into_py(self_.py(), async move {
let description = inner.get_job(&job_id).await.infer_error()?;
Ok(description.map(crate::job::JobDescription::from))
})
}
pub fn cancel_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.get_inner()?.clone();
future_into_py(self_.py(), async move {
inner.cancel_job(&job_id).await.infer_error()
})
}
#[pyo3(signature = (job_id=None))]
pub fn job_history(
self_: PyRef<'_, Self>,
job_id: Option<String>,
) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.get_inner()?.clone();
future_into_py(self_.py(), async move {
let batches = inner.job_history(job_id.as_deref()).await.infer_error()?;
Python::attach(|py| {
let list = PyList::empty(py);
for batch in batches {
list.append(batch.to_pyarrow(py)?)?;
}
Ok(list.unbind())
})
})
}
}
#[pyfunction]
#[pyo3(signature = (uri, api_key=None, region=None, host_override=None, sql_host_override=None, read_consistency_interval=None, client_config=None, storage_options=None, session=None, manifest_enabled=false, namespace_client_properties=None, oauth_config=None))]
#[pyo3(signature = (uri, api_key=None, region=None, host_override=None, read_consistency_interval=None, client_config=None, storage_options=None, session=None, manifest_enabled=false, namespace_client_properties=None, oauth_config=None))]
#[allow(clippy::too_many_arguments)]
pub fn connect(
py: Python<'_>,
@@ -732,7 +707,6 @@ pub fn connect(
api_key: Option<String>,
region: Option<String>,
host_override: Option<String>,
sql_host_override: Option<String>,
read_consistency_interval: Option<f64>,
client_config: Option<PyClientConfig>,
storage_options: Option<HashMap<String, String>>,
@@ -752,12 +726,6 @@ pub fn connect(
if let Some(host_override) = host_override {
builder = builder.host_override(&host_override);
}
#[cfg(feature = "remote")]
if let Some(sql_host_override) = sql_host_override {
builder = builder.sql_host_override(&sql_host_override);
}
#[cfg(not(feature = "remote"))]
let _ = sql_host_override;
if let Some(read_consistency_interval) = read_consistency_interval {
let read_consistency_interval = Duration::from_secs_f64(read_consistency_interval);
builder = builder.read_consistency_interval(read_consistency_interval);
-6
View File
@@ -114,12 +114,6 @@ impl<T> PythonErrorExt<T> for std::result::Result<T, LanceError> {
.getattr(intern!(py, "JobCancelledError"))?;
Err(PyErr::from_value(cls.call1((err.to_string(),))?))
}),
LanceError::JobNotFound { .. } => Python::attach(|py| {
let cls = py
.import(intern!(py, "lancedb.exceptions"))?
.getattr(intern!(py, "JobNotFoundError"))?;
Err(PyErr::from_value(cls.call1((err.to_string(),))?))
}),
_ => self.runtime_error(),
},
}
+9 -126
View File
@@ -4,50 +4,11 @@
use std::sync::Arc;
use crate::runtime::future_into_py;
use arrow::{
datatypes::Schema,
pyarrow::{IntoPyArrow, Table as PyArrowTable},
};
use lancedb::job::JobEventsRequest;
use pyo3::{
Bound, PyAny, PyRef, PyResult, Python,
exceptions::PyValueError,
pyclass, pymethods,
types::{PyAnyMethods, PyDict, PyDictMethods},
};
use pyo3::{Bound, PyAny, PyRef, PyResult, pyclass, pymethods};
use serde::Serialize;
use crate::error::PythonErrorExt;
const REPR_INDENT: &str = " ";
/// Parse a stored JSON payload into Python data. The bindings carry these as
/// strings because that is what crosses the boundary cheaply; the public
/// Python surface is the parsed form.
fn parse_json_payload<'py>(
py: Python<'py>,
raw: Option<&str>,
) -> PyResult<Option<Bound<'py, PyAny>>> {
match raw {
None => Ok(None),
Some(raw) => Ok(Some(py.import("json")?.call_method1("loads", (raw,))?)),
}
}
/// A payload rendered as indented JSON, aligned under the field that holds it.
fn pretty_json_payload(py: Python<'_>, raw: Option<&str>) -> PyResult<Option<String>> {
let Some(parsed) = parse_json_payload(py, raw)? else {
return Ok(None);
};
let kwargs = PyDict::new(py);
kwargs.set_item("indent", 4)?;
let rendered: String = py
.import("json")?
.call_method("dumps", (parsed,), Some(&kwargs))?
.extract()?;
Ok(Some(rendered.replace('\n', &format!("\n{REPR_INDENT}"))))
}
#[pyclass]
pub struct Job {
inner: Arc<lancedb::Job<std::result::Result<Option<String>, String>>>,
@@ -106,48 +67,6 @@ impl Job {
Ok(())
})
}
pub fn refresh(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.inner.clone();
future_into_py(self_.py(), async move {
inner.refresh().await.infer_error()?;
Ok(())
})
}
/// The last observed lifecycle state, without contacting the backend.
#[getter]
pub fn _state(&self) -> Option<String> {
self.inner.state()
}
/// The last observed server-side record. `None` for an in-process job.
#[getter]
pub fn _description(&self) -> Option<JobDescription> {
self.inner.description().map(JobDescription::from)
}
#[pyo3(signature = (*, limit=None, filter=None))]
pub fn events(
self_: PyRef<'_, Self>,
limit: Option<u32>,
filter: Option<String>,
) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.inner.clone();
let request = JobEventsRequest { limit, filter };
future_into_py(self_.py(), async move {
let batches = inner.events(request).await.infer_error()?;
Python::attach(|py| {
let schema = batches
.first()
.map(|batch| batch.schema())
.unwrap_or_else(|| Arc::new(Schema::empty()));
let table = PyArrowTable::try_new(batches, schema)
.map_err(|err| PyValueError::new_err(err.to_string()))?;
table.into_pyarrow(py).map(|table| table.unbind())
})
})
}
}
/// A row from `Connection.list_jobs`: one server-side job.
@@ -202,7 +121,7 @@ impl JobFailureInfo {
}
}
/// The server-side record behind a `Job` handle.
/// A described job from `Connection.get_job`.
#[pyclass(get_all, skip_from_py_object)]
#[derive(Clone)]
pub struct JobDescription {
@@ -210,49 +129,17 @@ pub struct JobDescription {
job_type: String,
state: String,
creation_ms: i64,
/// Internal: the wire form behind the `spec` property.
_spec_json: Option<String>,
/// Internal: the wire form behind the `result` property.
_result_json: Option<String>,
spec_json: Option<String>,
failure: Option<JobFailureInfo>,
}
#[pymethods]
impl JobDescription {
/// The job-type-specific specification it was submitted with.
#[getter]
fn spec<'py>(&self, py: Python<'py>) -> PyResult<Option<Bound<'py, PyAny>>> {
parse_json_payload(py, self._spec_json.as_deref())
}
/// The job-type-specific terminal result. `None` until the job succeeds.
#[getter]
fn result<'py>(&self, py: Python<'py>) -> PyResult<Option<Bound<'py, PyAny>>> {
parse_json_payload(py, self._result_json.as_deref())
}
fn __repr__(&self, py: Python<'_>) -> PyResult<String> {
let mut fields = vec![
format!("job_id={:?}", self.job_id),
format!("job_type={:?}", self.job_type),
format!("state={:?}", self.state),
format!("creation_ms={}", self.creation_ms),
];
// Lay the payloads out as indented JSON, the same way the `Job` repr
// does, so the two agree on how the same data looks.
for (name, payload) in [("spec", &self._spec_json), ("result", &self._result_json)] {
if let Some(rendered) = pretty_json_payload(py, payload.as_deref())? {
fields.push(format!("{name}={rendered}"));
}
}
if let Some(failure) = &self.failure {
fields.push(format!("failure={}", failure.__repr__()));
}
let body = fields
.iter()
.map(|field| format!("\n{REPR_INDENT}{field},"))
.collect::<String>();
Ok(format!("JobDescription({body}\n)"))
fn __repr__(&self) -> String {
format!(
"JobDescription(job_id={:?}, job_type={:?}, state={:?}, creation_ms={})",
self.job_id, self.job_type, self.state, self.creation_ms
)
}
}
@@ -263,11 +150,7 @@ impl From<lancedb::database::JobDescription> for JobDescription {
job_type: description.job_type,
state: description.state,
creation_ms: description.creation_ms,
_spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
_result_json: description
.result
.filter(|result| !result.is_null())
.map(|result| result.to_string()),
spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
failure: description.failure.map(|failure| JobFailureInfo {
phase: failure.phase,
message: failure.message,
-3
View File
@@ -34,7 +34,6 @@ pub mod permutation;
pub mod query;
pub mod runtime;
pub mod session;
pub mod sql;
pub mod table;
pub mod util;
@@ -51,8 +50,6 @@ pub fn _lancedb(_py: Python, m: &Bound<'_, PyModule>) -> PyResult<()> {
m.add_class::<crate::job::JobInfo>()?;
m.add_class::<crate::job::JobDescription>()?;
m.add_class::<crate::job::JobFailureInfo>()?;
m.add_class::<crate::sql::Query>()?;
m.add_class::<crate::sql::QueryDescription>()?;
m.add_class::<PyBlobFile>()?;
m.add_class::<IndexConfig>()?;
m.add_class::<Query>()?;
-90
View File
@@ -1,90 +0,0 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
use std::sync::Arc;
use chrono::{DateTime, Utc};
use pyo3::{Bound, PyAny, PyRef, PyResult, pyclass, pymethods};
use uuid::Uuid;
use crate::arrow::RecordBatchStream;
use crate::error::PythonErrorExt;
use crate::runtime::future_into_py;
#[pyclass(name = "SqlQuery")]
pub struct Query {
inner: Arc<lancedb::sql::Query>,
}
impl Query {
pub(crate) fn new(inner: lancedb::sql::Query) -> Self {
Self {
inner: Arc::new(inner),
}
}
}
#[pymethods]
impl Query {
#[getter]
pub fn id(&self) -> Uuid {
self.inner.id()
}
pub fn describe(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.inner.clone();
future_into_py(self_.py(), async move {
inner
.describe()
.await
.map(QueryDescription::from)
.infer_error()
})
}
pub fn reader(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.inner.clone();
future_into_py(self_.py(), async move {
let stream = inner.reader().await.infer_error()?;
Ok(RecordBatchStream::new(stream))
})
}
pub fn cancel(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
let inner = self_.inner.clone();
future_into_py(self_.py(), async move {
inner.cancel().await.infer_error()?;
Ok(())
})
}
}
#[pyclass(get_all, skip_from_py_object)]
#[derive(Clone)]
pub struct QueryDescription {
id: Uuid,
status: String,
progress: Option<f64>,
expires_at: Option<DateTime<Utc>>,
}
#[pymethods]
impl QueryDescription {
fn __repr__(&self) -> String {
format!(
"QueryDescription(id={:?}, status={:?}, progress={:?}, expires_at={:?})",
self.id, self.status, self.progress, self.expires_at
)
}
}
impl From<lancedb::sql::QueryDescription> for QueryDescription {
fn from(description: lancedb::sql::QueryDescription) -> Self {
Self {
id: description.id,
status: description.status.to_string(),
progress: description.progress,
expires_at: description.expires_at,
}
}
}
+1 -7
View File
@@ -1,6 +1,6 @@
[package]
name = "lancedb"
version = "0.39.0-beta.6"
version = "0.39.0-beta.1"
edition.workspace = true
description = "LanceDB: A serverless, low-latency vector database for AI applications"
license.workspace = true
@@ -21,8 +21,6 @@ arrow-select = { workspace = true }
arrow-ord = { workspace = true }
arrow-cast = { workspace = true }
arrow-ipc.workspace = true
arrow-flight = { workspace = true, optional = true }
prost = { version = "0.14", optional = true }
chrono = { workspace = true }
datafusion-catalog.workspace = true
datafusion-common.workspace = true
@@ -79,7 +77,6 @@ reqwest = { version = "0.12.0", default-features = false, features = [
"rustls-tls-native-roots",
"stream",
], optional = true }
tonic = { workspace = true, optional = true }
http = { version = "1", optional = true } # Matching what is in reqwest
urlencoding = { version = "2", optional = true }
uuid = { workspace = true, features = ["v5"] }
@@ -148,11 +145,8 @@ huggingface = [
]
dynamodb = ["lance/dynamodb", "aws"]
remote = [
"dep:arrow-flight",
"dep:prost",
"dep:reqwest",
"dep:http",
"dep:tonic",
"dep:urlencoding",
"lance-namespace-impls/rest",
"lance-namespace-impls/rest-adapter",
+8 -58
View File
@@ -7,9 +7,7 @@
//! raw `Binary` / `LargeBinary` into the blob struct layout. Queries return
//! small descriptors, not bytes.
//!
//! Blob tables require Lance file format >= 2.2. `_rowid` values stay valid
//! after compaction when the table has stable row ids. Overwrite is a new
//! create and does not keep the previous table's stable row id setting.
//! Blob tables require Lance file format >= 2.2 and stable row ids at create.
use std::ops::Range;
use std::sync::Arc;
@@ -326,7 +324,6 @@ pub(crate) fn blob_column_names(schema: &Schema) -> Vec<String> {
}
/// Bumps storage format to at least [`LanceFileVersion::V2_2`] for blob schemas.
/// Leaves `enable_stable_row_ids` unchanged.
pub(crate) fn ensure_blob_storage_version(schema: &Schema, params: &mut WriteParams) {
if !has_blob_columns(schema) {
return;
@@ -388,30 +385,6 @@ fn ensure_all_row_ids_resolved(column: &str, requested: usize, resolved: usize)
}
}
/// Lance take reports a missing physical row address as NotSupported or InvalidInput.
fn map_blob_take_error(column: &str, requested: usize, err: lance::Error) -> Error {
let missing_row_addr = match &err {
lance::Error::NotSupported { source, .. } => {
source.to_string().contains("must not target deleted rows")
}
lance::Error::InvalidInput { source, .. } => source
.to_string()
.contains("belongs to non-existent fragment"),
_ => false,
};
if missing_row_addr {
Error::InvalidInput {
message: format!(
"blob read for column '{column}' requested {requested} row ids but some \
do not exist in the table; pass row ids collected from this table"
),
}
} else {
err.into()
}
}
/// Materialize blob-local ranges (same length and order as `requests`, nulls preserved).
pub(crate) async fn take_blob_ranges_aligned(
dataset: &Arc<Dataset>,
@@ -432,8 +405,7 @@ pub(crate) async fn take_blob_ranges_aligned(
.with_row_ids(lance_requests)
.preserve_order(true)
.execute()
.await
.map_err(|err| map_blob_take_error(column, requests.len(), err))?;
.await?;
ensure_all_row_ids_resolved(column, requests.len(), payloads.len())?;
let mut builder = LargeBinaryBuilder::new();
@@ -462,8 +434,7 @@ pub(crate) async fn take_blobs_aligned(
.with_row_ids(row_ids.to_vec())
.preserve_order(true)
.execute()
.await
.map_err(|err| map_blob_take_error(column, row_ids.len(), err))?;
.await?;
ensure_all_row_ids_resolved(column, row_ids.len(), payloads.len())?;
let mut builder = LargeBinaryBuilder::new();
@@ -487,10 +458,7 @@ pub(crate) async fn take_blob_files_aligned(
return Ok(Vec::new());
}
let handles = dataset
.take_blobs(row_ids, column)
.await
.map_err(|err| map_blob_take_error(column, row_ids.len(), err))?;
let handles = dataset.take_blobs(row_ids, column).await?;
ensure_all_row_ids_resolved(column, row_ids.len(), handles.len())?;
Ok(handles
.into_iter()
@@ -532,27 +500,10 @@ mod tests {
fn storage_version_bumps_to_v2_2() {
let mut params = WriteParams::default();
ensure_blob_storage_version(&blob_schema(), &mut params);
let resolved = params
.data_storage_version
.unwrap_or(LanceFileVersion::Stable)
.resolve();
assert_eq!(resolved, ConcreteFileVersion::V2_2);
assert!(!params.enable_stable_row_ids);
}
#[test]
fn storage_version_leaves_stable_row_ids_enabled() {
let mut params = WriteParams {
enable_stable_row_ids: true,
..Default::default()
};
ensure_blob_storage_version(&blob_schema(), &mut params);
assert!(params.enable_stable_row_ids);
let resolved = params
.data_storage_version
.unwrap_or(LanceFileVersion::Stable)
.resolve();
assert_eq!(resolved, ConcreteFileVersion::V2_2);
assert_eq!(
params.data_storage_version.unwrap().resolve(),
ConcreteFileVersion::V2_2
);
}
#[test]
@@ -625,6 +576,5 @@ mod tests {
let mut params = WriteParams::default();
ensure_blob_storage_version(&schema, &mut params);
assert!(params.data_storage_version.is_none());
assert!(!params.enable_stable_row_ids);
}
}
+23 -151
View File
@@ -23,18 +23,15 @@ use crate::connection::create_table::CreateTableBuilder;
use crate::data::scannable::Scannable;
use crate::database::listing::ListingDatabase;
use crate::database::{
CloneTableRequest, Database, DatabaseOptions, JobInfo, OpenTableRequest, ReadConsistency,
TableNamesRequest,
CloneTableRequest, Database, DatabaseOptions, JobDescription, JobInfo, OpenTableRequest,
ReadConsistency, TableNamesRequest,
};
use crate::embeddings::{EmbeddingRegistry, MemoryRegistry};
use crate::error::{Error, Result};
#[cfg(feature = "remote")]
use crate::remote::{
client::ClientConfig,
db::{
OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION,
OPT_REMOTE_SQL_HOST_OVERRIDE,
},
db::{OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION},
};
use lance::io::ObjectStoreParams;
pub use lance_file::version::LanceFileVersion;
@@ -325,43 +322,6 @@ pub struct CloneTableBuilder {
request: CloneTableRequest,
}
/// Builder for asynchronously executing a SQL statement on a remote database.
pub struct ExecuteQueryAsyncBuilder {
parent: Arc<dyn Database>,
query: String,
default_namespace_path: Vec<String>,
}
impl ExecuteQueryAsyncBuilder {
fn new(parent: Arc<dyn Database>, query: String) -> Self {
Self {
parent,
query,
default_namespace_path: vec!["public".to_string()],
}
}
/// Set the namespace used for unqualified table names.
///
/// An empty path is treated as `public`, which is the SQL name for the
/// root Lance namespace.
pub fn default_namespace_path<I, S>(mut self, path: I) -> Self
where
I: IntoIterator<Item = S>,
S: Into<String>,
{
self.default_namespace_path = path.into_iter().map(Into::into).collect();
self
}
/// Start the statement and return its asynchronous query handle.
pub async fn execute(self) -> Result<crate::sql::Query> {
self.parent
.execute_query_async(&self.query, &self.default_namespace_path)
.await
}
}
impl CloneTableBuilder {
fn new(parent: Arc<dyn Database>, target_table_name: String, source_uri: String) -> Self {
Self {
@@ -445,51 +405,6 @@ impl Connection {
&self.internal
}
/// Start executing SQL on a remote LanceDB database.
///
/// The query can reference tables in other databases with SQL dot notation.
/// Use [`ExecuteQueryAsyncBuilder::default_namespace_path`] to avoid qualifying
/// tables in the default namespace. Local connections return
/// [`Error::NotSupported`].
///
/// # Example
///
/// ```no_run
/// # async fn query(db: &lancedb::Connection) -> lancedb::Result<()> {
/// use futures::TryStreamExt;
///
/// let query = db
/// .execute_query_async("SELECT * FROM events LIMIT 10")
/// .default_namespace_path(["public"])
/// .execute()
/// .await?;
/// println!("query id: {}", query.id());
/// let mut batches = query.reader().await?;
/// while let Some(batch) = batches.try_next().await? {
/// println!("received {} rows", batch.num_rows());
/// }
/// # Ok(())
/// # }
/// ```
pub fn execute_query_async(&self, query: impl Into<String>) -> ExecuteQueryAsyncBuilder {
ExecuteQueryAsyncBuilder::new(self.internal.clone(), query.into())
}
/// Describe a submitted SQL query by its connection-scoped id.
///
/// This performs one bounded status poll using state retained by this
/// connection. Running state with a live query handle is not evicted;
/// abandoned state has bounded retention, and server expiration is
/// honored. Terminal state is retained briefly.
/// Query ids are not portable to another connection. Local connections
/// return [`Error::NotSupported`].
pub async fn describe_query(
&self,
query_id: uuid::Uuid,
) -> Result<crate::sql::QueryDescription> {
self.internal.describe_query(query_id).await
}
/// Get the names of all tables in the database
///
/// The names will be returned in lexicographical order (ascending)
@@ -670,34 +585,14 @@ impl Connection {
self.internal.read_consistency().await
}
/// Open a server-side job by id, returning a handle with its record
/// already populated. Fails with [`crate::Error::JobNotFound`] when the
/// server has no such job, the way [`Connection::open_table`] does for a
/// missing table.
/// A [`crate::job::Job`] handle for a server-side job by id, suitable for
/// waiting on or cancelling the job.
///
/// This is the one way in: the returned [`crate::job::Job`] answers for
/// its own state, specification, result, failure and event history, so
/// there is no separate connection-level call for any of them.
///
/// # Example
///
/// ```no_run
/// # use lancedb::job::JobEventsRequest;
/// # async fn open_job(
/// # connection: &lancedb::Connection,
/// # job_id: &str,
/// # ) -> Result<(), Box<dyn std::error::Error>> {
/// let job = connection.open_job(job_id).await?;
/// println!("{:?} {:?}", job.state(), job.result());
/// let done = job
/// .events(JobEventsRequest::default().filter("state = 'claim_complete'"))
/// .await?;
/// println!("{} completions", done.iter().map(|b| b.num_rows()).sum::<usize>());
/// # Ok(())
/// # }
/// ```
pub async fn open_job(&self, job_id: impl AsRef<str>) -> Result<crate::job::Job> {
self.internal.open_job(job_id.as_ref()).await
/// The handle is constructed without a server round trip; an unknown id
/// surfaces when the handle is used. Only server-backed databases support
/// job handles by id.
pub fn job(&self, job_id: impl AsRef<str>) -> Result<crate::job::Job> {
self.internal.job(job_id.as_ref())
}
/// List server-side jobs across the database's tables.
@@ -705,12 +600,24 @@ impl Connection {
self.internal.list_jobs().await
}
/// Describe a single server-side job by id. `None` when the server has no
/// such job.
pub async fn get_job(&self, job_id: impl AsRef<str>) -> Result<Option<JobDescription>> {
self.internal.get_job(job_id.as_ref()).await
}
/// Request cancellation of a server-side job by id. Returns true if the
/// server accepted the cancellation, false if no such job exists.
pub async fn cancel_job(&self, job_id: impl AsRef<str>) -> Result<bool> {
self.internal.cancel_job(job_id.as_ref()).await
}
/// The lifecycle event history of a server-side job (all jobs when
/// `job_id` is `None`), as recorded Arrow batches.
pub async fn job_history(&self, job_id: Option<&str>) -> Result<Vec<RecordBatch>> {
self.internal.job_history(job_id).await
}
/// Drop a table in the database.
///
/// # Arguments
@@ -957,19 +864,6 @@ impl ConnectBuilder {
self
}
/// Set the SQL service host override for a remote connection.
///
/// The SQL client is initialized lazily when the connection first executes
/// SQL and is retained for the connection's lifetime.
#[cfg(feature = "remote")]
pub fn sql_host_override(mut self, sql_host_override: &str) -> Self {
self.request.options.insert(
OPT_REMOTE_SQL_HOST_OVERRIDE.to_string(),
sql_host_override.to_string(),
);
self
}
/// Set the database specific options
///
/// See [crate::database::listing::ListingDatabaseOptions] for the options available for
@@ -1159,7 +1053,6 @@ impl ConnectBuilder {
let mut merged_options = self.request.options.clone();
Self::apply_env_defaults(&ENV_VARS_TO_STORAGE_OPTS, &mut merged_options);
let sql_host_override = merged_options.get(OPT_REMOTE_SQL_HOST_OVERRIDE).cloned();
let options = RemoteDatabaseOptions::parse_from_map(&merged_options)?;
let region = options.region.ok_or_else(|| Error::InvalidInput {
@@ -1201,15 +1094,11 @@ impl ConnectBuilder {
}
let storage_options = StorageOptions(options.storage_options.clone());
let host_overrides = crate::remote::db::RemoteHostOverrides {
rest: options.host_override,
sql: sql_host_override,
};
let internal = Arc::new(crate::remote::db::RemoteDatabase::try_new(
&self.request.uri,
&api_key,
&region,
host_overrides,
options.host_override,
client_config,
storage_options.into(),
self.request.read_consistency_interval,
@@ -1503,23 +1392,6 @@ mod tests {
assert_eq!(tc.connection.uri(), tc.uri);
}
#[tokio::test]
async fn test_local_connection_rejects_sql_queries() {
let directory = tempdir().unwrap();
let connection = connect(directory.path().to_str().unwrap())
.execute()
.await
.unwrap();
assert!(matches!(
connection.execute_query_async("SELECT 1").execute().await,
Err(Error::NotSupported { .. })
));
assert!(matches!(
connection.describe_query(uuid::Uuid::nil()).await,
Err(Error::NotSupported { .. })
));
}
#[cfg(feature = "remote")]
#[test]
fn test_apply_env_defaults() {
+19 -28
View File
@@ -18,6 +18,8 @@ use std::collections::HashMap;
use std::sync::Arc;
use std::time::Duration;
use arrow_array::RecordBatch;
use lance::dataset::ReadParams;
use lance_namespace::LanceNamespace;
use lance_namespace::models::{
@@ -204,8 +206,8 @@ pub enum ReadConsistency {
/// compaction, column refresh, ...).
#[derive(Debug, Clone)]
pub struct JobInfo {
/// The job id -- what [`Database::open_job`] and
/// [`Database::cancel_job`] accept.
/// The job id -- what [`Database::get_job`] and [`Database::cancel_job`]
/// accept.
pub job_id: String,
/// The table the job runs against, without URI or namespace.
pub table: String,
@@ -216,8 +218,8 @@ pub struct JobInfo {
pub created_at_millis: i64,
}
/// The server-side record behind a [`crate::job::Job`] handle: lifecycle
/// state plus the job-type-specific specification and result.
/// A described job from [`Database::get_job`]: lifecycle state plus the
/// job-type-specific specification.
#[derive(Debug, Clone)]
pub struct JobDescription {
pub job_id: String,
@@ -228,10 +230,6 @@ pub struct JobDescription {
pub creation_ms: i64,
/// The job-type-specific specification. Null when the server omits it.
pub spec: serde_json::Value,
/// The job-type-specific terminal result, for job types that define one.
/// `None` until the job succeeds, so a job that never terminates reports
/// its progress through [`crate::job::Job::events`] instead.
pub result: Option<serde_json::Value>,
/// Why the job failed, when the job is failed and the server reports a
/// reason.
pub failure: Option<crate::error::JobFailure>,
@@ -317,37 +315,30 @@ pub trait Database:
async fn drop_function(&self, _name: &str, _version: &str) -> Result<bool> {
function_catalog_not_supported()
}
/// Open a job by id, returning a handle with its record already
/// populated. Fails with [`crate::Error::JobNotFound`] when the server has
/// no such job.
async fn open_job(&self, _job_id: &str) -> Result<crate::job::Job> {
job_op_not_supported("open_job")
/// A [`crate::job::Job`] handle for a server-side job by id, suitable for
/// waiting on or cancelling the job. The handle is constructed without a
/// server round trip; an unknown id surfaces when the handle is used.
fn job(&self, _job_id: &str) -> Result<crate::job::Job> {
job_op_not_supported("job")
}
/// List server-side jobs across the database's tables.
async fn list_jobs(&self) -> Result<Vec<JobInfo>> {
job_op_not_supported("list_jobs")
}
/// Describe a single job by id. `None` when the server has no such job.
async fn get_job(&self, _job_id: &str) -> Result<Option<JobDescription>> {
job_op_not_supported("get_job")
}
/// Request cancellation of a job by id. Returns true if the server
/// accepted the cancellation, false if no such job exists. Cancelling an
/// already-terminal job is a no-op success.
async fn cancel_job(&self, _job_id: &str) -> Result<bool> {
job_op_not_supported("cancel_job")
}
/// Start executing a SQL statement on a remote database.
async fn execute_query_async(
&self,
_query: &str,
_default_namespace_path: &[String],
) -> Result<crate::sql::Query> {
Err(crate::error::Error::NotSupported {
message: "SQL is not supported by this database".to_string(),
})
}
/// Describe a submitted SQL query by its connection-scoped id.
async fn describe_query(&self, _query_id: uuid::Uuid) -> Result<crate::sql::QueryDescription> {
Err(crate::error::Error::NotSupported {
message: "SQL is not supported by this database".to_string(),
})
/// The lifecycle event history of a job (all jobs when `job_id` is
/// `None`), as recorded Arrow batches.
async fn job_history(&self, _job_id: Option<&str>) -> Result<Vec<RecordBatch>> {
job_op_not_supported("job_history")
}
/// Open a table in the database
async fn open_table(&self, request: OpenTableRequest) -> Result<Arc<dyn BaseTable>>;
+2 -1
View File
@@ -18,7 +18,7 @@ use lance_table::io::commit::commit_handler_from_url;
use object_store::local::LocalFileSystem;
use snafu::ResultExt;
use crate::blob::ensure_blob_storage_version;
use crate::blob::{ensure_blob_storage_version, has_blob_columns};
use crate::connection::ConnectRequest;
use crate::database::ReadConsistency;
use crate::database::namespace::LanceNamespaceDatabase;
@@ -827,6 +827,7 @@ impl ListingDatabase {
if let Some(enable_stable_row_ids) = overrides
.enable_stable_row_ids
.or(self.new_table_config.enable_stable_row_ids)
.or(has_blob_columns(&data_schema).then_some(true))
{
write_params.enable_stable_row_ids = enable_stable_row_ids;
}
+2 -6
View File
@@ -3,7 +3,6 @@
//! Namespace-based database implementation that delegates table management to lance-namespace
use lance_datafusion::utils::StreamingWriteSource;
use std::collections::{HashMap, HashSet};
use std::sync::{Arc, Mutex};
@@ -24,7 +23,7 @@ use lance_namespace_impls::ConnectBuilder;
use lance_table::io::commit::CommitHandler;
use lance_table::io::commit::external_manifest::ExternalManifestCommitHandler;
use crate::blob::ensure_blob_storage_version;
use crate::blob::{ensure_blob_storage_version, has_blob_columns};
use crate::connection::NamespaceClientPushdownOperation;
use crate::database::ReadConsistency;
use crate::database::listing::{NewTableConfig, take_request_creation_overrides};
@@ -218,6 +217,7 @@ impl LanceNamespaceDatabase {
if let Some(enable_stable_row_ids) = overrides
.enable_stable_row_ids
.or(self.new_table_config.enable_stable_row_ids)
.or(has_blob_columns(data_schema.as_ref()).then_some(true))
{
params.enable_stable_row_ids = enable_stable_row_ids;
}
@@ -305,10 +305,6 @@ impl Database for LanceNamespaceDatabase {
}
async fn create_table(&self, request: DbCreateTableRequest) -> Result<Arc<dyn BaseTable>> {
// Refuse a bad declaration before the namespace records a table.
crate::table::computed_columns::ensure_declarations_are_planned(
&request.data.arrow_schema(),
)?;
let mut table_id = request.namespace_path.clone();
table_id.push(request.name.clone());
let mut existing_table = None;
-2
View File
@@ -102,8 +102,6 @@ pub enum Error {
},
#[snafu(display("Job{} was cancelled", job_id.as_ref().map(|id| format!(" {id}")).unwrap_or_default()))]
JobCancelled { job_id: Option<String> },
#[snafu(display("Job '{job_id}' was not found"))]
JobNotFound { job_id: String },
// 3rd party / external errors
#[snafu(display("object_store error: {source}"))]
+2 -16
View File
@@ -582,8 +582,8 @@ pub struct InputBinding {
/// Ordered result-field to table-field mapping for a Function binding.
///
/// `nullable` describes the logical Function result. Physical computed-column
/// fields remain nullable while unassigned.
/// Assignment state is not part of the Slice 1 client contract. During the
/// NULL transition there is no public Lance cell-flag identifier to persist.
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct OutputMapping {
pub result_field: String,
@@ -594,14 +594,6 @@ pub struct OutputMapping {
pub nullable: bool,
}
/// Internal physical column preserving the parent validity of a flattened
/// named-struct result.
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct AssignmentMapping {
pub output_name: String,
pub output_field_id: i32,
}
/// Immutable Function binding persisted by the Enterprise table service.
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct FunctionBinding {
@@ -609,8 +601,6 @@ pub struct FunctionBinding {
function: FunctionVersionRef,
inputs: Vec<InputBinding>,
outputs: Vec<OutputMapping>,
#[serde(default, skip_serializing_if = "Option::is_none")]
assignment: Option<AssignmentMapping>,
/// Exact Arrow schema presented to the Function, encoded with the Lance
/// Namespace Arrow JSON representation.
#[serde(default, skip_serializing_if = "Option::is_none")]
@@ -637,10 +627,6 @@ impl FunctionBinding {
&self.outputs
}
pub fn assignment(&self) -> Option<&AssignmentMapping> {
self.assignment.as_ref()
}
pub fn input_schema(&self) -> Option<&Value> {
self.input_schema.as_ref()
}
+12 -247
View File
@@ -3,53 +3,16 @@
//! Handles to operations a server may run asynchronously.
use std::sync::{Arc, RwLock, RwLockReadGuard, RwLockWriteGuard};
use std::sync::Arc;
use arrow_array::RecordBatch;
use async_trait::async_trait;
use serde::{Serialize, de::DeserializeOwned};
use serde_json::Value;
use tokio::sync::watch;
use tokio::task::{AbortHandle, JoinHandle};
use crate::database::JobDescription;
use crate::error::{Error, JobFailure, Result};
/// Which of a job's events [`Job::events`] returns.
///
/// The handle already knows which job to ask about, so this narrows the
/// query rather than naming one.
#[derive(Debug, Clone, Default)]
pub struct JobEventsRequest {
/// Maximum event rows to return. The server applies its own default
/// (1000 rows) and maximum (10,000 rows) when this is `None`, and
/// truncates without saying so, which matters for a job with one event
/// per fragment.
pub limit: Option<u32>,
/// SQL-like filter over the event columns `state`, `updated_by`,
/// `emitted_from`, `emitted_by`, and `claim_entity`. For example
/// `state = 'claim_complete'` selects only per-claim completions.
pub filter: Option<String>,
}
impl JobEventsRequest {
pub fn limit(mut self, limit: u32) -> Self {
self.limit = Some(limit);
self
}
pub fn filter(mut self, filter: impl Into<String>) -> Self {
self.filter = Some(filter.into());
self
}
}
fn job_detail_not_supported<T>(what: &str) -> Result<T> {
Err(Error::NotSupported {
message: format!("{what} is only available for server-side jobs"),
})
}
/// Backend-specific tracking for an asynchronous operation.
#[async_trait]
pub(crate) trait JobHandle: Send + Sync {
@@ -60,15 +23,6 @@ pub(crate) trait JobHandle: Send + Sync {
async fn status(&self) -> Result<String>;
async fn wait(&self) -> Result<TerminalResult>;
async fn cancel(&self) -> Result<()>;
/// The job's full server-side record. Backends that run the operation in
/// this process have none and keep the default.
async fn describe(&self) -> Result<JobDescription> {
job_detail_not_supported("describing a job")
}
/// The job's recorded lifecycle events.
async fn events(&self, _request: JobEventsRequest) -> Result<Vec<RecordBatch>> {
job_detail_not_supported("job event history")
}
}
/// A backend-neutral successful terminal result.
@@ -131,34 +85,16 @@ enum JobInner<T> {
Completed(T),
}
/// What a handle last learned about its job. `state` is separate because an
/// in-process job can report one but has no server-side record behind it.
#[derive(Default)]
struct JobCache {
state: Option<String>,
description: Option<JobDescription>,
}
/// A handle to an operation that may still be running.
///
/// The operation may already be complete when the handle is created. `T` is
/// the endpoint's successful terminal result; unit-result operations use the
/// default `Job<()>`.
///
/// The detail accessors ([`Job::state`], [`Job::job_type`], ...) read what the
/// handle last observed. Submitting an operation returns only a job id, so
/// populating them eagerly would cost an extra round trip on every call:
///
/// - [`Job::refresh`] and [`Job::status`] fetch the whole record.
/// - [`Job::wait`] records the terminal state it establishes, but not the rest
/// of the record; call [`Job::refresh`] for that.
/// - Everything is `None` until one of those runs.
pub struct Job<T = ()>
where
T: Clone + Send + Sync + 'static,
{
inner: JobInner<T>,
cache: RwLock<JobCache>,
}
impl<T> std::fmt::Debug for Job<T>
@@ -166,40 +102,18 @@ where
T: Clone + Send + Sync + 'static,
{
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
let cache = self.cache_read();
let mut out = f.debug_struct("Job");
out.field("id", &self.id())
.field("done", &matches!(self.inner, JobInner::Completed(_)));
if let Some(state) = &cache.state {
out.field("state", state);
}
if let Some(description) = &cache.description {
out.field("job_type", &description.job_type)
.field("creation_ms", &description.creation_ms);
if !description.spec.is_null() {
out.field("spec", &description.spec);
}
if let Some(result) = &description.result {
out.field("result", result);
}
if let Some(failure) = &description.failure {
out.field("failure", failure);
}
}
out.finish()
f.debug_struct("Job")
.field("id", &self.id())
.field("done", &matches!(self.inner, JobInner::Completed(_)))
.finish()
}
}
impl Job<()> {
/// A job whose operation finished before the handle was created. Its
/// state is known without asking anyone, so the cache starts populated.
/// A job whose operation finished before the handle was created.
pub(crate) fn new_done() -> Self {
Self {
inner: JobInner::Completed(()),
cache: RwLock::new(JobCache {
state: Some("finished".to_string()),
description: None,
}),
}
}
@@ -209,21 +123,8 @@ impl Job<()> {
handle,
decode: Arc::new(|_| Ok(())),
},
cache: RwLock::default(),
}
}
/// A handle whose record the caller has already fetched, so the detail
/// accessors answer without a second round trip.
pub(crate) fn opened(handle: Box<dyn JobHandle>, description: JobDescription) -> Self {
let job = Self::new(handle);
{
let mut cache = job.cache_write();
cache.state = Some(description.state.clone());
cache.description = Some(description);
}
job
}
}
impl<T> Job<T>
@@ -237,7 +138,6 @@ where
handle,
decode: Arc::new(TerminalResult::decode::<T>),
},
cache: RwLock::default(),
}
}
}
@@ -269,124 +169,16 @@ where
}
}
fn cache_read(&self) -> RwLockReadGuard<'_, JobCache> {
self.cache.read().unwrap_or_else(|err| err.into_inner())
}
fn cache_write(&self) -> RwLockWriteGuard<'_, JobCache> {
self.cache.write().unwrap_or_else(|err| err.into_inner())
}
/// Asks the backend for this job's current state, and for a server-side
/// job its full record, then caches the answer for the detail accessors.
///
/// In-process operations have no server-side record, so only
/// [`Job::state`] is populated for them.
pub async fn refresh(&self) -> Result<()> {
self.refresh_state().await.map(|_| ())
}
/// Refreshes and reports the state, which every backend can answer.
async fn refresh_state(&self) -> Result<String> {
let JobInner::Handle { handle, .. } = &self.inner else {
let state = "finished".to_string();
self.cache_write().state = Some(state.clone());
return Ok(state);
};
match handle.describe().await {
Ok(description) => {
let state = description.state.clone();
let mut cache = self.cache_write();
cache.state = Some(state.clone());
cache.description = Some(description);
Ok(state)
}
// An in-process job knows its own state and nothing more.
Err(Error::NotSupported { .. }) => {
let state = handle.status().await?;
self.cache_write().state = Some(state.clone());
Ok(state)
}
Err(err) => Err(err),
}
}
/// The operation's current lifecycle state: "running", "finished",
/// "failed", or "cancelled".
///
/// A point snapshot; unlike [`Job::wait`] it does not block, raise on a
/// terminal failure state, or retry. States a newer server reports that
/// this client version does not know pass through as-is. Also refreshes
/// the detail accessors.
/// this client version does not know pass through as-is.
pub async fn status(&self) -> Result<String> {
self.refresh_state().await
}
/// The last lifecycle state this handle observed, without contacting the
/// backend. `None` until the handle has.
pub fn state(&self) -> Option<String> {
self.cache_read().state.clone()
}
/// The whole server-side record this handle last observed. The accessors
/// below read individual fields out of it. `None` for an in-process job,
/// which has no such record.
pub fn description(&self) -> Option<JobDescription> {
self.cache_read().description.clone()
}
/// The job's type, as the server names it. `None` for an in-process job.
pub fn job_type(&self) -> Option<String> {
self.with_description(|description| description.job_type.clone())
}
/// When the job was created, in milliseconds since the epoch. `None` for
/// an in-process job.
pub fn creation_ms(&self) -> Option<i64> {
self.with_description(|description| description.creation_ms)
}
/// The job-type-specific specification it was submitted with.
pub fn spec(&self) -> Option<Value> {
self.with_description(|description| description.spec.clone())
.filter(|spec| !spec.is_null())
}
/// The job-type-specific terminal result, as reported data rather than the
/// typed model [`Job::wait`] returns. `None` until the job succeeds.
pub fn result(&self) -> Option<Value> {
self.with_description(|description| description.result.clone())
.flatten()
}
/// Why the job failed, when it failed and the server reports a reason.
pub fn failure(&self) -> Option<JobFailure> {
self.with_description(|description| description.failure.clone())
.flatten()
}
fn with_description<R>(&self, read: impl FnOnce(&JobDescription) -> R) -> Option<R> {
self.cache_read().description.as_ref().map(read)
}
/// This job's recorded lifecycle events.
///
/// Unlike the detail accessors, which report a terminal result only once
/// the job reaches one, events are written as the job runs and outlive the
/// workers that produced them. A distributed job records a
/// `claim`/`claim_complete` pair per unit of work, each carrying
/// `rows_processed`, so a job that never finishes still accounts for what
/// it did. In-process operations keep no event history.
pub async fn events(&self, request: JobEventsRequest) -> Result<Vec<RecordBatch>> {
match &self.inner {
JobInner::Handle { handle, .. } => handle.events(request).await,
// The operation finished before the handle existed, so there is no
// id to query with even when a server ran it.
JobInner::Completed(_) => Err(Error::NotSupported {
message: "this operation completed before its handle was created, so it \
carries no job id to query events with"
.to_string(),
}),
JobInner::Handle { handle, .. } => handle.status().await,
JobInner::Completed(_) => Ok("finished".to_string()),
}
}
@@ -398,19 +190,8 @@ where
/// [`crate::Error::JobCancelled`] if it was cancelled.
pub async fn wait(&self) -> Result<T> {
match &self.inner {
JobInner::Handle { handle, decode } => {
let settled = handle.wait().await;
// Waiting already established a terminal state; record it so
// the detail accessors do not need another round trip for it.
if let Some(state) = terminal_state(&settled) {
self.cache_write().state = Some(state.to_string());
}
(decode)(settled?)
}
JobInner::Completed(result) => {
self.cache_write().state = Some("finished".to_string());
Ok(result.clone())
}
JobInner::Handle { handle, decode } => (decode)(handle.wait().await?),
JobInner::Completed(result) => Ok(result.clone()),
}
}
@@ -443,36 +224,20 @@ where
U: Clone + Send + Sync + 'static,
F: Fn(T) -> U + Send + Sync + 'static,
{
// The mapped handle tracks the same job, so it inherits what this one
// has already learned about it.
let Self { inner, cache } = self;
match inner {
match self.inner {
JobInner::Handle { handle, decode } => Job {
inner: JobInner::Handle {
handle,
decode: Arc::new(move |result| Ok(map((decode)(result)?))),
},
cache,
},
JobInner::Completed(result) => Job {
inner: JobInner::Completed(map(result)),
cache,
},
}
}
}
/// The lifecycle state a settled [`JobHandle::wait`] implies.
fn terminal_state(settled: &Result<TerminalResult>) -> Option<&'static str> {
match settled {
Ok(_) => Some("finished"),
Err(Error::JobFailed { .. }) => Some("failed"),
Err(Error::JobCancelled { .. }) => Some("cancelled"),
// Anything else is a transport failure, not a verdict on the job.
Err(_) => None,
}
}
/// How an in-process operation ended. Cloneable so every waiter can be given
/// the outcome; [`Error`] is not, so failures share one behind an [`Arc`].
#[derive(Clone)]
-1
View File
@@ -195,7 +195,6 @@ pub mod query;
#[cfg(feature = "remote")]
pub mod remote;
pub mod rerankers;
pub mod sql;
pub mod table;
#[cfg(test)]
pub mod test_utils;
File diff suppressed because it is too large Load Diff
+17 -536
View File
@@ -24,8 +24,8 @@ use std::time::{SystemTime, UNIX_EPOCH};
use arrow_array::cast::AsArray;
use arrow_array::types::UInt64Type;
use arrow_array::{RecordBatch, UInt64Array, new_null_array};
use arrow_schema::{FieldRef, Schema as ArrowSchema, SchemaRef};
use arrow_array::{RecordBatch, UInt64Array};
use arrow_schema::{Schema as ArrowSchema, SchemaRef};
use datafusion::common::ScalarValue;
use datafusion::error::DataFusionError;
use datafusion::physical_plan::SendableRecordBatchStream;
@@ -34,7 +34,7 @@ use datafusion::prelude::{col, lit};
use futures::{StreamExt, TryStreamExt};
use lance::Dataset;
use lance::dataset::mem_wal::DatasetMemWalExt;
use lance::dataset::transaction::{Operation, Transaction, UpdateMode};
use lance::dataset::transaction::{Operation, Transaction};
use lance::dataset::write::delete::DeleteBuilder;
use lance::dataset::write::merge_insert::inserted_rows::{
KeyExistenceFilter, KeyExistenceFilterBuilder, KeyValue,
@@ -51,9 +51,6 @@ use super::{
definition_to_metadata,
};
use crate::database::OpenTableRequest;
use crate::table::computed_columns::{
computed_column_from_field, computed_columns, ensure_declarations_are_planned,
};
use crate::table::{NativeTable, NativeTableExt, Table};
use crate::{Error, Result};
@@ -170,52 +167,30 @@ pub(crate) async fn execute_refresh(
.map(|p| (p.output.clone(), p.expression.clone()))
.collect();
validate_inputs(&source_ds, definition)?;
let (replanned, planned_fields, _renames) = super::plan(
let (replanned, mut planned_fields, _renames) = super::plan(
source_schema,
&definition.source_table,
&definition.source_namespace,
Some(&projections),
&projections,
definition.filter.as_deref(),
definition.limit,
)?;
let mut planned_fields = planned_fields;
planned_fields.push(arrow_schema::Field::new(
SOURCE_ROW_ID_COLUMN,
arrow_schema::DataType::UInt64,
false,
));
// A computed column is not planned from the source: refresh writes it
// NULL and its declaration's owner fills it. Its declaration must still
// be complete, and it must be able to hold NULL.
let physical = ArrowSchema::from(view_ds.schema());
let mut computed = computed_columns(&physical).into_iter().map(|c| c.name);
if let Some(name) = computed.by_ref().find(|name| {
physical
.field_with_name(name)
.is_ok_and(|f| !f.is_nullable())
}) {
return Err(Error::Schema {
message: format!(
"computed column '{name}' of view '{}' cannot hold NULL; recreate the view",
view.name()
),
});
}
ensure_declarations_are_planned(&physical)?;
let physical_planned: Vec<&FieldRef> = physical
let planned_shape: Vec<_> = planned_fields
.iter()
.map(|f| (f.name().clone(), f.data_type().clone(), f.is_nullable()))
.collect();
let physical_shape: Vec<_> = physical
.fields()
.iter()
.filter(|f| computed_column_from_field(f).is_none())
.map(|f| (f.name().clone(), f.data_type().clone(), f.is_nullable()))
.collect();
// A projected column that became nullable at the source still fits the
// view's nullable field; the reverse would not.
let matches = planned_fields.len() == physical_planned.len()
&& planned_fields.iter().zip(&physical_planned).all(|(e, p)| {
e.name() == p.name()
&& e.data_type() == p.data_type()
&& (p.is_nullable() || !e.is_nullable())
});
if !matches {
if planned_shape != physical_shape {
return Err(Error::Schema {
message: format!(
"the stored definition of view '{}' does not produce this \
@@ -254,18 +229,11 @@ pub(crate) async fn execute_refresh(
.get(SOURCE_VERSION_TS_META_KEY)
.and_then(|raw| raw.parse().ok());
// The watermark speaks only for the view state its refresh left behind;
// any other commit on the view since then is drift, except a fill of its
// computed columns, which rewrites nothing refresh certifies.
let recorded_view_version = metadata
// any other commit on the view since then is drift.
let view_intact = metadata
.get(VIEW_VERSION_META_KEY)
.and_then(|raw| raw.parse::<u64>().ok());
let view_intact = match recorded_view_version {
Some(recorded) if recorded == view_ds.version().version => true,
Some(recorded) if recorded < view_ds.version().version => {
only_computed_rewrites_since(&view_ds, recorded).await?
}
_ => false,
};
.and_then(|raw| raw.parse::<u64>().ok())
== Some(view_ds.version().version);
if !full && watermark == Some(source_version) && view_intact && recorded_ts == Some(source_ts) {
return Ok(RefreshMaterializedViewResult {
@@ -1122,69 +1090,6 @@ struct RowScope {
limit: Option<u64>,
}
/// Whether every commit on the view after `recorded` is a fill of its
/// computed columns: a column rewrite or data replacement touching only
/// those fields and neither adding nor removing rows. A version whose
/// transaction cannot be read is not proven, so it counts as drift.
async fn only_computed_rewrites_since(view_ds: &Dataset, recorded: u64) -> Result<bool> {
// A fill may write any field under a computed column, so the whole
// subtree counts, not only the root.
let physical = ArrowSchema::from(view_ds.schema());
fn subtree(field: &lance_core::datatypes::Field, ids: &mut Vec<u32>) {
ids.push(field.id as u32);
for child in &field.children {
subtree(child, ids);
}
}
let mut computed_fields = Vec::new();
for column in computed_columns(&physical) {
if let Some(field) = view_ds.schema().field(&column.name) {
subtree(field, &mut computed_fields);
}
}
if computed_fields.is_empty() {
return Ok(false);
}
for version in recorded + 1..=view_ds.version().version {
let Some(transaction) = view_ds.read_transaction_by_version(version).await? else {
return Ok(false);
};
let fill = match &transaction.operation {
Operation::Update {
removed_fragment_ids,
new_fragments,
fields_modified,
update_mode: Some(UpdateMode::RewriteColumns),
..
} => {
removed_fragment_ids.is_empty()
&& new_fragments.is_empty()
&& !fields_modified.is_empty()
&& fields_modified
.iter()
.all(|field| computed_fields.contains(field))
}
// What `refresh_column` commits for a SQL declaration.
Operation::DataReplacement { replacements } => {
!replacements.is_empty()
&& replacements.iter().all(|group| {
!group.1.fields.is_empty()
&& group
.1
.fields
.iter()
.all(|field| computed_fields.contains(&(*field as u32)))
})
}
_ => false,
};
if !fill {
return Ok(false);
}
}
Ok(true)
}
async fn compute_stream(
source: &Dataset,
definition: &MaterializedViewDefinition,
@@ -1253,10 +1158,6 @@ async fn compute_stream(
let batch = batch.map_err(|e| DataFusionError::External(Box::new(e)))?;
let mut columns = Vec::with_capacity(out_schema.fields().len());
for field in out_schema.fields() {
if computed_column_from_field(field).is_some() {
columns.push(new_null_array(field.data_type(), batch.num_rows()));
continue;
}
let name = if field.name() == SOURCE_ROW_ID_COLUMN {
ROW_ID
} else {
@@ -2867,7 +2768,7 @@ mod tests {
let (conn, source) = db_with_source(vec![1]).await;
let prepared = crate::materialized_view::prepare_declaration(
&source,
Some(&[("x".into(), "x".into()), ("twice".into(), "x * 2".into())]),
&[("x".into(), "x".into()), ("twice".into(), "x * 2".into())],
None,
None,
)
@@ -3231,424 +3132,4 @@ mod tests {
let err = view.refresh().execute().await.unwrap_err();
assert!(err.to_string().contains("source table 'src'"), "{err}");
}
/// A view with a computed column, declared over `people` and refreshed.
async fn refreshed_computed_view(conn: &Connection) -> MaterializedView {
use crate::materialized_view::tests::{computed_field, people, test_binding};
let source = people(conn).await;
let view = crate::materialized_view::prepare_declaration(
&source,
Some(&[
("id".to_string(), "id".to_string()),
("name".to_string(), "name".to_string()),
]),
None,
None,
)
.await
.unwrap()
.with_computed_columns(
vec![(2, computed_field("emb", "fb_1", "name"))],
&[test_binding("fb_1", "name", "emb")],
)
.unwrap()
.create("v")
.await
.unwrap();
let result = view.refresh().execute().await.unwrap();
assert_eq!(result.mode, RefreshMode::Rebuild);
view
}
async fn unfilled(view: &MaterializedView) -> usize {
view.table()
.count_rows(Some("emb IS NULL".to_string()))
.await
.unwrap()
}
async fn append_people(conn: &Connection, ids: Vec<i32>, names: Vec<&str>) {
let batch = record_batch!(("id", Int32, ids), ("name", Utf8, names)).unwrap();
conn.open_table("people")
.execute()
.await
.unwrap()
.add(batch)
.execute()
.await
.unwrap();
}
/// Commit the fill job's shape on the view: a column rewrite of
/// `fields`, touching no rows. The data is left as it is; what matters
/// here is how the next refresh classifies the commit.
async fn commit_column_rewrite(view: &MaterializedView, fields: &[&str]) {
let native = view.table().as_native().unwrap();
native.dataset.reload().await.unwrap();
let dataset = native.dataset.get().await.unwrap().as_ref().clone();
let fields_modified = fields
.iter()
.map(|name| dataset.schema().field(name).unwrap().id as u32)
.collect();
let updated_fragments = dataset
.get_fragments()
.iter()
.map(|fragment| fragment.metadata().clone())
.collect();
let operation = Operation::Update {
removed_fragment_ids: Vec::new(),
updated_fragments,
new_fragments: Vec::new(),
fields_modified,
compacted_sstables: Vec::new(),
fields_for_preserving_frag_bitmap: Vec::new(),
update_mode: Some(UpdateMode::RewriteColumns),
inserted_rows_filter: None,
updated_fragment_offsets: None,
};
let read_version = dataset.version().version;
CommitBuilder::new(WriteDestination::Dataset(Arc::new(dataset)))
.execute(Transaction::new(read_version, operation, None))
.await
.unwrap();
}
/// Refresh never computes a computed column: every row it writes, on a
/// rebuild, an append and a rewrite, carries NULL there, and the
/// declaration survives all three.
#[tokio::test]
async fn test_computed_columns_are_written_null_and_kept() {
let conn = connect("memory://").execute().await.unwrap();
let view = refreshed_computed_view(&conn).await;
assert_eq!(unfilled(&view).await, 3);
append_people(&conn, vec![4, 5], vec!["d", "e"]).await;
let result = view.refresh().execute().await.unwrap();
assert_eq!(result.mode, RefreshMode::Incremental);
assert_eq!(unfilled(&view).await, 5);
conn.open_table("people")
.execute()
.await
.unwrap()
.update()
.column("name", "'z'")
.only_if("id = 1")
.execute()
.await
.unwrap();
view.refresh().execute().await.unwrap();
assert_eq!(unfilled(&view).await, 5);
assert_eq!(read(view.table(), "id").await, vec![1, 2, 3, 4, 5]);
let schema = view.table().schema().await.unwrap();
assert!(
crate::table::computed_columns::function_bindings(&schema)
.unwrap()
.iter()
.any(|b| b.binding_id() == "fb_1"),
"the binding envelope was lost"
);
assert!(
computed_column_from_field(schema.field_with_name("emb").unwrap()).is_some(),
"the declaration was lost"
);
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::NoOp
);
}
/// The fill job's commit rewrites only computed columns. It is the one
/// commit on a view that is not drift: the next refresh carries on from
/// its watermark instead of rebuilding, which would null what the fill
/// just wrote.
#[tokio::test]
async fn test_a_computed_column_fill_is_not_drift() {
let conn = connect("memory://").execute().await.unwrap();
let view = refreshed_computed_view(&conn).await;
commit_column_rewrite(&view, &["emb"]).await;
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::NoOp
);
commit_column_rewrite(&view, &["emb"]).await;
append_people(&conn, vec![4], vec!["d"]).await;
let result = view.refresh().execute().await.unwrap();
assert_eq!(result.mode, RefreshMode::Incremental);
assert_eq!(result.rows_written, 1);
assert_eq!(read(view.table(), "id").await, vec![1, 2, 3, 4]);
}
/// A column rewrite that reaches a projected column is drift like any
/// other write: refresh certifies those columns and must recompute them.
#[tokio::test]
async fn test_a_rewrite_of_a_projected_column_is_drift() {
let conn = connect("memory://").execute().await.unwrap();
let view = refreshed_computed_view(&conn).await;
commit_column_rewrite(&view, &["emb", "name"]).await;
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::Rebuild
);
}
/// The declaration contract is checked before any refresh mutation: a
/// missing binding envelope and a column that lost its declaration both
/// fail closed.
#[tokio::test]
async fn test_a_broken_declaration_is_refused_before_refresh() {
let conn = connect("memory://").execute().await.unwrap();
let view = refreshed_computed_view(&conn).await;
let native = view.table().as_native().unwrap();
let mut dataset = native.dataset.get().await.unwrap().as_ref().clone();
dataset
.update_schema_metadata(vec![(
crate::table::computed_columns::FUNCTION_BINDINGS_META_KEY.to_string(),
None,
)])
.await
.unwrap();
let err = view.refresh().execute().await.unwrap_err().to_string();
assert!(err.contains("references missing binding 'fb_1'"), "{err}");
let conn = connect("memory://").execute().await.unwrap();
let view = refreshed_computed_view(&conn).await;
let native = view.table().as_native().unwrap();
let mut dataset = native.dataset.get().await.unwrap().as_ref().clone();
dataset
.replace_field_metadata(vec![(
dataset.schema().field("emb").unwrap().id as u32,
HashMap::new(),
)])
.await
.unwrap();
let err = view.refresh().execute().await.unwrap_err().to_string();
assert!(err.contains("does not match binding 'fb_1'"), "{err}");
}
/// An input the view does not project is materialized on every refresh
/// path, before the provenance column, with the source's values.
#[tokio::test]
async fn test_internal_inputs_are_materialized_and_refreshed() {
use crate::materialized_view::tests::{computed_field, strict_people, test_binding};
let conn = connect("memory://").execute().await.unwrap();
let source = strict_people(&conn).await;
let mut prepared = crate::materialized_view::prepare_declaration(
&source,
Some(&[("id".to_string(), "id".to_string())]),
None,
None,
)
.await
.unwrap();
let input = prepared.input_column("name").unwrap();
let view = prepared
.with_computed_columns(
vec![(1, computed_field("emb", "fb_1", &input))],
&[test_binding("fb_1", &input, "emb")],
)
.unwrap()
.create("v")
.await
.unwrap();
let names: Vec<String> = view
.table()
.schema()
.await
.unwrap()
.fields()
.iter()
.map(|f| f.name().clone())
.collect();
assert_eq!(names, ["id", "emb", "__input_name", SOURCE_ROW_ID_COLUMN]);
let unfilled_inputs = || async {
view.table()
.count_rows(Some("__input_name IS NULL".to_string()))
.await
.unwrap()
};
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::Rebuild
);
assert_eq!(view.table().count_rows(None).await.unwrap(), 3);
assert_eq!(unfilled_inputs().await, 0);
let more = arrow_array::RecordBatch::try_new(
source.schema().await.unwrap(),
vec![
Arc::new(Int32Array::from(vec![4])),
Arc::new(arrow_array::StringArray::from(vec!["d"])),
],
)
.unwrap();
source.add(more).execute().await.unwrap();
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::Incremental
);
assert_eq!(unfilled_inputs().await, 0);
assert_eq!(
view.table()
.count_rows(Some("__input_name = 'd'".to_string()))
.await
.unwrap(),
1
);
source
.update()
.column("name", "'z'")
.only_if("id = 1")
.execute()
.await
.unwrap();
view.refresh().execute().await.unwrap();
assert_eq!(
view.table()
.count_rows(Some("__input_name = 'z'".to_string()))
.await
.unwrap(),
1
);
assert_eq!(
unfilled(&view).await,
4,
"rewritten and new rows are unfilled"
);
}
/// A SQL declaration is filled by `refresh_column` on the view, which
/// commits a data replacement; the next refresh continues from its
/// watermark and keeps what the fill wrote, and only rows the view added
/// since come back unfilled.
#[tokio::test]
async fn test_a_sql_fill_is_not_drift() {
use crate::materialized_view::tests::{people, sql_field};
let conn = connect("memory://").execute().await.unwrap();
let source = people(&conn).await;
let view = crate::materialized_view::prepare_declaration(
&source,
Some(&[("id".to_string(), "id".to_string())]),
None,
None,
)
.await
.unwrap()
.with_computed_columns(
vec![(
1,
sql_field("next", arrow_schema::DataType::Int32, "id + 1", r#"["id"]"#),
)],
&[],
)
.unwrap()
.create("v")
.await
.unwrap();
let filled = || async {
view.table()
.count_rows(Some("next = id + 1".to_string()))
.await
.unwrap()
};
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::Rebuild
);
assert_eq!(
view.table()
.refresh_column("next")
.await
.unwrap()
.rows_filled,
3
);
assert_eq!(filled().await, 3);
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::NoOp
);
assert_eq!(filled().await, 3);
append_people(&conn, vec![4], vec!["d"]).await;
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::Incremental
);
assert_eq!(filled().await, 3);
assert_eq!(
view.table()
.refresh_column("next")
.await
.unwrap()
.rows_filled,
1
);
assert_eq!(filled().await, 4);
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::NoOp
);
}
/// A fill of a nested computed column writes its child fields; that is
/// still a fill, not drift.
#[tokio::test]
async fn test_a_nested_sql_fill_is_not_drift() {
use crate::materialized_view::tests::{people, sql_field};
let conn = connect("memory://").execute().await.unwrap();
let source = people(&conn).await;
let payload = sql_field(
"payload",
arrow_schema::DataType::Struct(
vec![arrow_schema::Field::new(
"value",
arrow_schema::DataType::Utf8,
true,
)]
.into(),
),
"named_struct('value', name)",
r#"["name"]"#,
);
let view = crate::materialized_view::prepare_declaration(
&source,
Some(&[("name".to_string(), "name".to_string())]),
None,
None,
)
.await
.unwrap()
.with_computed_columns(vec![(1, payload)], &[])
.unwrap()
.create("v")
.await
.unwrap();
view.refresh().execute().await.unwrap();
assert_eq!(
view.table()
.refresh_column("payload")
.await
.unwrap()
.rows_filled,
3
);
assert_eq!(
view.refresh().execute().await.unwrap().mode,
RefreshMode::NoOp
);
assert_eq!(
view.table()
.count_rows(Some("payload.value = name".to_string()))
.await
.unwrap(),
3
);
}
}
-1
View File
@@ -11,7 +11,6 @@ pub(crate) mod db;
pub(crate) mod job;
pub mod oauth;
mod retry;
pub(crate) mod sql;
pub(crate) mod table;
pub(crate) mod util;
+92 -173
View File
@@ -20,20 +20,19 @@ use lance_namespace::models::{
use crate::Error;
use crate::database::{
CloneTableRequest, CreateTableMode, CreateTableRequest, Database, DatabaseOptions, JobInfo,
OpenTableRequest, ReadConsistency, TableNamesRequest,
CloneTableRequest, CreateTableMode, CreateTableRequest, Database, DatabaseOptions,
JobDescription, JobInfo, OpenTableRequest, ReadConsistency, TableNamesRequest,
};
use crate::error::Result;
use crate::function::{FunctionRegistrationRequest, FunctionVersion};
use crate::job::Job;
use crate::remote::job::{RemoteJob, job_state_to_client};
use crate::remote::job::{DescribeJobResponse, RemoteJob, job_state_to_client};
use crate::remote::util::stream_as_body;
use crate::table::BaseTable;
use super::client::{
ClientConfig, HeaderProvider, HttpSend, RequestResultExt, RestfulLanceDbClient, Sender,
};
use super::sql::SqlClient;
use super::table::RemoteTable;
use super::util::parse_server_version;
use super::{ARROW_STREAM_CONTENT_TYPE, extract_job_id};
@@ -98,7 +97,6 @@ pub const OPT_REMOTE_PREFIX: &str = "remote_database_";
pub const OPT_REMOTE_API_KEY: &str = "remote_database_api_key";
pub const OPT_REMOTE_REGION: &str = "remote_database_region";
pub const OPT_REMOTE_HOST_OVERRIDE: &str = "remote_database_host_override";
pub const OPT_REMOTE_SQL_HOST_OVERRIDE: &str = "remote_database_sql_host_override";
// TODO: add support for configuring client config via key/value options
#[derive(Clone, Debug, Default)]
@@ -214,7 +212,6 @@ pub struct RemoteDatabase<S: HttpSend = Sender> {
namespace_context_provider: Option<Arc<dyn DynamicContextProvider>>,
/// TLS configuration for mTLS support
tls_config: Option<super::client::TlsConfig>,
sql_client: Option<SqlClient>,
}
#[derive(Clone)]
@@ -272,35 +269,22 @@ impl DynamicContextProvider for NamespaceHeaderProviderContext {
}
}
pub struct RemoteHostOverrides {
pub rest: Option<String>,
pub sql: Option<String>,
}
impl RemoteDatabase {
pub(crate) fn try_new(
pub fn try_new(
uri: &str,
api_key: &str,
region: &str,
host_overrides: RemoteHostOverrides,
host_override: Option<String>,
client_config: ClientConfig,
options: RemoteOptions,
read_consistency_interval: Option<std::time::Duration>,
) -> Result<Self> {
let parsed = super::client::parse_db_url(uri)?;
let sql_client = SqlClient::new(
parsed.db_name.clone(),
parsed.db_prefix.clone(),
api_key.to_string(),
host_overrides.rest.clone(),
host_overrides.sql,
client_config.clone(),
);
let header_map = RestfulLanceDbClient::<Sender>::default_headers(
api_key,
region,
&parsed.db_name,
host_overrides.rest.is_some(),
host_override.is_some(),
&options,
parsed.db_prefix.as_deref(),
&client_config,
@@ -328,7 +312,7 @@ impl RemoteDatabase {
let client = RestfulLanceDbClient::try_new(
&parsed,
region,
host_overrides.rest,
host_override,
header_map,
client_config.clone(),
read_consistency_interval,
@@ -346,7 +330,6 @@ impl RemoteDatabase {
namespace_headers,
namespace_context_provider,
tls_config: client_config.tls_config,
sql_client: Some(sql_client),
})
}
}
@@ -444,7 +427,6 @@ mod test_utils {
namespace_headers: HashMap::new(),
namespace_context_provider: None,
tls_config: None,
sql_client: None,
}
}
@@ -467,7 +449,6 @@ mod test_utils {
namespace_headers: config.extra_headers.clone(),
namespace_context_provider,
tls_config: config.tls_config.clone(),
sql_client: None,
}
}
}
@@ -671,18 +652,11 @@ impl<S: HttpSend> Database for RemoteDatabase<S> {
Ok(response.dropped)
}
async fn open_job(&self, job_id: &str) -> Result<Job> {
let handle = super::job::RemoteJob::new(self.client.clone(), job_id.to_string());
match crate::job::JobHandle::describe(&handle).await {
Ok(description) => Ok(Job::opened(Box::new(handle), description)),
Err(Error::Http {
status_code: Some(StatusCode::NOT_FOUND),
..
}) => Err(Error::JobNotFound {
job_id: job_id.to_string(),
}),
Err(err) => Err(err),
}
fn job(&self, job_id: &str) -> Result<crate::job::Job> {
Ok(crate::job::Job::new(Box::new(super::job::RemoteJob::new(
self.client.clone(),
job_id.to_string(),
))))
}
async fn list_jobs(&self) -> Result<Vec<JobInfo>> {
@@ -719,6 +693,31 @@ impl<S: HttpSend> Database for RemoteDatabase<S> {
Ok(out)
}
async fn get_job(&self, job_id: &str) -> Result<Option<JobDescription>> {
let req = self
.client
.post("/v1/jobs/describe")
.json(&serde_json::json!({ "job_id": job_id }));
let (request_id, rsp) = self.client.send(req).await?;
let rsp = match self.client.check_response(&request_id, rsp).await {
Ok(rsp) => rsp,
Err(Error::Http {
status_code: Some(StatusCode::NOT_FOUND),
..
}) => return Ok(None),
Err(err) => return Err(err),
};
let body: DescribeJobResponse = rsp.json().await.err_to_http(request_id)?;
Ok(Some(JobDescription {
job_id: body.job_id,
job_type: body.job_type,
state: job_state_to_client(&body.job_state),
creation_ms: body.creation_ms,
spec: body.spec,
failure: body.failure.map(|reported| reported.into_job_failure()),
}))
}
async fn cancel_job(&self, job_id: &str) -> Result<bool> {
let req = self
.client
@@ -735,28 +734,19 @@ impl<S: HttpSend> Database for RemoteDatabase<S> {
}
}
async fn execute_query_async(
&self,
query: &str,
default_namespace_path: &[String],
) -> Result<crate::sql::Query> {
let client = self
.sql_client
.as_ref()
.ok_or_else(|| Error::NotSupported {
message: "SQL is unavailable for this remote database client".to_string(),
})?;
client.submit(query, default_namespace_path).await
}
async fn describe_query(&self, query_id: uuid::Uuid) -> Result<crate::sql::QueryDescription> {
let client = self
.sql_client
.as_ref()
.ok_or_else(|| Error::NotSupported {
message: "SQL is unavailable for this remote database client".to_string(),
})?;
client.describe(query_id).await
async fn job_history(&self, job_id: Option<&str>) -> Result<Vec<arrow_array::RecordBatch>> {
let mut body = serde_json::json!({});
if let Some(job_id) = job_id {
body["job_id"] = serde_json::Value::String(job_id.to_string());
}
let req = self.client.post("/v1/jobs/query_events").json(&body);
let (request_id, rsp) = self.client.send(req).await?;
let rsp = self.client.check_response(&request_id, rsp).await?;
let bytes = rsp.bytes().await.err_to_http(request_id)?;
let reader = arrow_ipc::reader::StreamReader::try_new(std::io::Cursor::new(bytes), None)?;
reader
.collect::<std::result::Result<Vec<_>, _>>()
.map_err(Into::into)
}
async fn table_names(&self, request: TableNamesRequest) -> Result<Vec<String>> {
@@ -1274,7 +1264,6 @@ mod tests {
use crate::{
Connection, Error,
database::CreateTableMode,
job::JobEventsRequest,
remote::{ARROW_STREAM_CONTENT_TYPE, ClientConfig, HeaderProvider, JSON_CONTENT_TYPE},
};
@@ -2623,7 +2612,7 @@ mod tests {
}
#[tokio::test]
async fn test_open_job() {
async fn test_get_job() {
let conn = Connection::new_with_handler(|request| {
assert_eq!(request.method(), &reqwest::Method::POST);
assert_eq!(request.url().path(), "/v1/jobs/describe");
@@ -2637,55 +2626,51 @@ mod tests {
)
.unwrap()
});
// Opening populates the handle, so the accessors answer without a
// second round trip.
let job = conn.open_job("job-1").await.unwrap();
assert_eq!(job.id(), Some("job-1"));
assert_eq!(job.job_type().as_deref(), Some("create_index"));
assert_eq!(job.state().as_deref(), Some("failed"));
assert_eq!(job.creation_ms(), Some(1000));
assert_eq!(job.spec().unwrap()["column"], "vec");
assert!(job.result().is_none());
let failure = job.failure().unwrap();
let job = conn.get_job("job-1").await.unwrap().unwrap();
assert_eq!(job.job_id, "job-1");
assert_eq!(job.job_type, "create_index");
assert_eq!(job.state, "failed");
assert_eq!(job.creation_ms, 1000);
assert_eq!(job.spec["column"], "vec");
let failure = job.failure.unwrap();
assert_eq!(failure.phase.as_deref(), Some("execute"));
assert_eq!(failure.message.as_deref(), Some("worker died"));
assert_eq!(failure.retryable, Some(true));
}
#[tokio::test]
async fn test_open_job_reports_the_terminal_result() {
let conn = Connection::new_with_handler(|_| {
http::Response::builder()
.status(200)
.body(
r#"{"job_id": "job-1", "job_type": "refresh_column", "job_state": "DONE", "creation_ms": 1000, "result": {"rows_assigned": 1000000, "rows_failed": 0}}"#,
)
.unwrap()
});
let job = conn.open_job("job-1").await.unwrap();
assert_eq!(job.state().as_deref(), Some("finished"));
let result = job.result().unwrap();
assert_eq!(result["rows_assigned"], 1_000_000);
assert_eq!(result["rows_failed"], 0);
}
#[tokio::test]
async fn test_open_job_missing_fails() {
async fn test_get_job_missing_is_none() {
let conn = Connection::new_with_handler(|_| {
http::Response::builder()
.status(404)
.body("no such job")
.unwrap()
});
let err = conn.open_job("nope").await.unwrap_err();
assert!(
matches!(&err, Error::JobNotFound { job_id } if job_id == "nope"),
"{err:?}"
);
assert!(conn.get_job("nope").await.unwrap().is_none());
}
#[tokio::test]
async fn test_job_events_scope_to_that_job() {
async fn test_cancel_job() {
let conn = Connection::new_with_handler(|request| {
assert_eq!(request.url().path(), "/v1/jobs/cancel");
http::Response::builder()
.status(200)
.body(r#"{"job_id": "job-1"}"#)
.unwrap()
});
assert!(conn.cancel_job("job-1").await.unwrap());
let conn = Connection::new_with_handler(|_| {
http::Response::builder()
.status(404)
.body("no such job")
.unwrap()
});
assert!(!conn.cancel_job("nope").await.unwrap());
}
#[tokio::test]
async fn test_job_history_parses_arrow_stream() {
let schema = Arc::new(Schema::new(vec![Field::new(
"state",
DataType::Utf8,
@@ -2694,91 +2679,29 @@ mod tests {
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(arrow_array::StringArray::from(vec![
"claim_complete",
"created", "done",
]))],
)
.unwrap();
let mut events = Vec::new();
let mut body = Vec::new();
{
let mut writer =
arrow_ipc::writer::StreamWriter::try_new(&mut events, &schema).unwrap();
let mut writer = arrow_ipc::writer::StreamWriter::try_new(&mut body, &schema).unwrap();
writer.write(&batch).unwrap();
writer.finish().unwrap();
}
let conn = Connection::new_with_handler(move |request| {
let body: serde_json::Value =
serde_json::from_slice(request.body().unwrap().as_bytes().unwrap()).unwrap();
if request.url().path() == "/v1/jobs/describe" {
return http::Response::builder()
.status(200)
.body(
r#"{"job_id": "job-1", "job_type": "refresh_column", "job_state": "IN_PROGRESS", "creation_ms": 1}"#
.as_bytes()
.to_vec(),
)
.unwrap();
}
assert_eq!(request.url().path(), "/v1/jobs/query_events");
// The handle supplies job_id; the caller only narrows the query.
assert_eq!(body["job_id"], "job-1");
assert_eq!(body["limit"], 500);
assert_eq!(body["filter"], "state = 'claim_complete'");
http::Response::builder()
.status(200)
.body(events.clone())
.unwrap()
});
let job = conn.open_job("job-1").await.unwrap();
let batches = job
.events(
JobEventsRequest::default()
.limit(500)
.filter("state = 'claim_complete'"),
)
.await
.unwrap();
assert_eq!(batches.len(), 1);
assert_eq!(batches[0].num_rows(), 1);
}
#[tokio::test]
async fn test_job_events_keep_the_schema_when_nothing_matches() {
let schema = Arc::new(Schema::new(vec![Field::new(
"state",
DataType::Utf8,
false,
)]));
let mut events = Vec::new();
{
let mut writer =
arrow_ipc::writer::StreamWriter::try_new(&mut events, &schema).unwrap();
writer.finish().unwrap();
}
let conn = Connection::new_with_handler(move |request| {
if request.url().path() == "/v1/jobs/describe" {
return http::Response::builder()
.status(200)
.body(
r#"{"job_id": "job-1", "job_type": "refresh_column", "job_state": "IN_PROGRESS", "creation_ms": 1}"#
.as_bytes()
.to_vec(),
)
.unwrap();
}
let body: serde_json::Value =
let req_body: serde_json::Value =
serde_json::from_slice(request.body().unwrap().as_bytes().unwrap()).unwrap();
// Only the job id when the caller narrows nothing.
assert_eq!(body, serde_json::json!({ "job_id": "job-1" }));
assert_eq!(req_body["job_id"], "job-1");
http::Response::builder()
.status(200)
.body(events.clone())
.body(body.clone())
.unwrap()
});
let job = conn.open_job("job-1").await.unwrap();
let batches = job.events(JobEventsRequest::default()).await.unwrap();
let batches = conn.job_history(Some("job-1")).await.unwrap();
assert_eq!(batches.len(), 1);
assert_eq!(batches[0].num_rows(), 0);
assert_eq!(batches[0].schema(), schema);
assert_eq!(batches[0].num_rows(), 2);
}
#[tokio::test]
@@ -2973,9 +2896,7 @@ mod tests {
let polls_ref = polls.clone();
let conn = Connection::new_with_handler(move |request| {
assert_eq!(request.url().path(), "/v1/jobs/describe");
// Two in-progress answers: one for the load, one for the first
// status poll.
let state = if polls_ref.fetch_add(1, Ordering::SeqCst) < 2 {
let state = if polls_ref.fetch_add(1, Ordering::SeqCst) == 0 {
"IN_PROGRESS"
} else {
"DONE"
@@ -2988,13 +2909,11 @@ mod tests {
))
.unwrap()
});
let job = conn.open_job("job-1").await.unwrap();
let job = conn.job("job-1").unwrap();
assert_eq!(job.id(), Some("job-1"));
// Opening already answered the state; no extra call needed for it.
assert_eq!(job.state().as_deref(), Some("running"));
assert_eq!(job.status().await.unwrap(), "running");
job.wait().await.unwrap();
assert_eq!(job.status().await.unwrap(), "finished");
assert!(polls.load(Ordering::SeqCst) >= 4);
assert!(polls.load(Ordering::SeqCst) >= 3);
}
}
+5 -55
View File
@@ -5,15 +5,13 @@
use std::time::Duration;
use arrow_array::RecordBatch;
use async_trait::async_trait;
use tokio::time::sleep;
use serde::Deserialize;
use crate::database::JobDescription;
use crate::error::{Error, JobFailure, Result};
use crate::job::{JobEventsRequest, JobHandle, TerminalResult};
use crate::job::{JobHandle, TerminalResult};
use crate::remote::client::{HttpSend, RequestResultExt, RestfulLanceDbClient};
/// Delay before the second job-state poll; doubles up to [`MAX_POLL_INTERVAL`].
@@ -88,7 +86,7 @@ pub(super) struct DescribeJobResponse {
#[serde(default)]
pub(super) spec: serde_json::Value,
#[serde(default)]
pub(super) result: Option<serde_json::Value>,
result: Option<serde_json::Value>,
#[serde(default)]
pub(super) failure: Option<ReportedFailure>,
}
@@ -112,39 +110,6 @@ impl DescribeJobResponse {
fn into_terminal_result(self, request_id: String) -> TerminalResult {
TerminalResult::remote(self.result, request_id)
}
/// The public description this wire envelope stands for.
pub(super) fn into_description(self) -> JobDescription {
JobDescription {
job_id: self.job_id,
job_type: self.job_type,
state: JobState::from(self.job_state.as_str()).client_label(),
creation_ms: self.creation_ms,
spec: self.spec,
result: self.result,
failure: self.failure.map(ReportedFailure::into_job_failure),
}
}
}
/// One `/v1/jobs/query_events` round trip.
pub(super) async fn fetch_job_events<S: HttpSend>(
client: &RestfulLanceDbClient<S>,
body: serde_json::Value,
) -> Result<Vec<RecordBatch>> {
let request = client.post("/v1/jobs/query_events").json(&body);
let (request_id, response) = client.send(request).await?;
let response = client.check_response(&request_id, response).await?;
let bytes = response.bytes().await.err_to_http(request_id)?;
let reader = arrow_ipc::reader::StreamReader::try_new(std::io::Cursor::new(bytes), None)?;
let schema = reader.schema();
let mut batches = reader.collect::<std::result::Result<Vec<_>, _>>()?;
// A query that matched nothing still describes the event columns.
// Keep that schema so callers can build a typed empty result.
if batches.is_empty() {
batches.push(RecordBatch::new_empty(schema));
}
Ok(batches)
}
pub struct RemoteJob<S: HttpSend> {
@@ -158,7 +123,7 @@ impl<S: HttpSend> RemoteJob<S> {
}
/// One `/v1/jobs/describe` round trip.
async fn fetch_description(&self) -> Result<(String, DescribeJobResponse)> {
async fn describe(&self) -> Result<(String, DescribeJobResponse)> {
let request = self
.client
.post("/v1/jobs/describe")
@@ -183,28 +148,13 @@ impl<S: HttpSend> JobHandle for RemoteJob<S> {
}
async fn status(&self) -> Result<String> {
Ok(self.fetch_description().await?.1.state().client_label())
}
async fn describe(&self) -> Result<JobDescription> {
Ok(self.fetch_description().await?.1.into_description())
}
async fn events(&self, request: JobEventsRequest) -> Result<Vec<RecordBatch>> {
let mut body = serde_json::json!({ "job_id": self.job_id });
if let Some(limit) = request.limit {
body["limit"] = serde_json::Value::from(limit);
}
if let Some(filter) = request.filter {
body["filter"] = serde_json::Value::String(filter);
}
fetch_job_events(&self.client, body).await
Ok(self.describe().await?.1.state().client_label())
}
async fn wait(&self) -> Result<TerminalResult> {
let mut interval = INITIAL_POLL_INTERVAL;
loop {
let (request_id, description) = self.fetch_description().await?;
let (request_id, description) = self.describe().await?;
match description.state() {
JobState::Done => return Ok(description.into_terminal_result(request_id)),
JobState::Failed => {
+5 -7
View File
@@ -466,9 +466,7 @@ impl TokenSource for AzureImdsSource {
/// OAuth header provider that manages the full token lifecycle.
///
/// Implements [`HeaderProvider`] to inject `Authorization: Bearer <token>`
/// headers into every LanceDB request, with automatic token refresh. It also
/// identifies the bearer credential as OIDC so LanceDB's SQL service selects
/// OIDC validation instead of API-key validation.
/// headers into every LanceDB request, with automatic token refresh.
pub struct OAuthHeaderProvider {
token_source: Box<dyn TokenSource>,
token_state: Arc<RwLock<TokenState>>,
@@ -556,10 +554,10 @@ impl OAuthHeaderProvider {
impl HeaderProvider for OAuthHeaderProvider {
async fn get_headers(&self) -> Result<HashMap<String, String>> {
let token = self.get_valid_token().await?;
Ok(HashMap::from([
("authorization".to_string(), format!("Bearer {token}")),
("x-lancedb-credential-type".to_string(), "oidc".to_string()),
]))
Ok(HashMap::from([(
"authorization".to_string(),
format!("Bearer {token}"),
)]))
}
}
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
-77
View File
@@ -264,17 +264,6 @@ impl<S: HttpSend> crate::job::JobHandle for FreshnessJob<S> {
crate::job::JobHandle::status(&self.inner).await
}
async fn describe(&self) -> Result<crate::database::JobDescription> {
crate::job::JobHandle::describe(&self.inner).await
}
async fn events(
&self,
request: crate::job::JobEventsRequest,
) -> Result<Vec<arrow_array::RecordBatch>> {
crate::job::JobHandle::events(&self.inner, request).await
}
async fn wait(&self) -> Result<crate::job::TerminalResult> {
let result = crate::job::JobHandle::wait(&self.inner).await?;
let version = self.version.read().await;
@@ -7993,72 +7982,6 @@ mod tests {
);
}
/// The refresh handle is wrapped for read-freshness tracking, so it has to
/// forward the detail APIs too -- this is the job an operator is holding
/// when a backfill goes quiet.
#[tokio::test]
async fn test_refresh_job_handle_reports_detail_and_events() {
let schema = Arc::new(Schema::new(vec![Field::new(
"state",
DataType::Utf8,
false,
)]));
let batch = RecordBatch::try_new(
schema.clone(),
vec![Arc::new(arrow_array::StringArray::from(vec![
"claim_complete",
]))],
)
.unwrap();
let mut events = Vec::new();
{
let mut writer =
arrow_ipc::writer::StreamWriter::try_new(&mut events, &schema).unwrap();
writer.write(&batch).unwrap();
writer.finish().unwrap();
}
let table = Table::new_with_handler("my_table", move |request| {
match request.url().path() {
"/v1/table/my_table/backfill_column" => http::Response::builder()
.status(202)
.body(br#"{"job_id": "j-42"}"#.to_vec())
.unwrap(),
"/v1/jobs/describe" => http::Response::builder()
.status(200)
.body(
r#"{"job_id": "j-42", "job_type": "refresh_column", "job_state": "IN_PROGRESS", "creation_ms": 7, "spec": {"column": "doubled"}}"#
.as_bytes()
.to_vec(),
)
.unwrap(),
"/v1/jobs/query_events" => {
let body: serde_json::Value =
serde_json::from_slice(request.body().unwrap().as_bytes().unwrap())
.unwrap();
assert_eq!(body["job_id"], "j-42");
http::Response::builder()
.status(200)
.body(events.clone())
.unwrap()
}
other => panic!("unexpected path {other}"),
}
});
let job = table.refresh_column_async("doubled").await.unwrap();
job.refresh().await.unwrap();
assert_eq!(job.state().as_deref(), Some("running"));
assert_eq!(job.job_type().as_deref(), Some("refresh_column"));
assert_eq!(job.creation_ms(), Some(7));
assert_eq!(job.spec().unwrap()["column"], "doubled");
let batches = job
.events(crate::job::JobEventsRequest::default())
.await
.unwrap();
assert_eq!(batches.iter().map(|b| b.num_rows()).sum::<usize>(), 1);
}
#[tokio::test]
async fn test_refresh_submission_uses_add_columns_version_fence() {
let table = Table::new_with_handler("my_table", |request| match request.url().path() {
-124
View File
@@ -1,124 +0,0 @@
// SPDX-License-Identifier: Apache-2.0
// SPDX-FileCopyrightText: Copyright The LanceDB Authors
//! Handles to SQL queries running on a remote database.
use std::{fmt, sync::Arc};
use async_trait::async_trait;
use chrono::{DateTime, Utc};
use uuid::Uuid;
use crate::{Result, arrow::SendableRecordBatchStream};
/// The externally visible lifecycle state of a submitted SQL query.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub enum QueryStatus {
/// The server is still executing the query.
Running,
/// The server has made the complete result available.
Finished,
/// The server accepted cancellation but has not confirmed it yet.
Cancelling,
/// The server confirmed cancellation.
Cancelled,
}
impl fmt::Display for QueryStatus {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
formatter.write_str(match self {
Self::Running => "running",
Self::Finished => "finished",
Self::Cancelling => "cancelling",
Self::Cancelled => "cancelled",
})
}
}
/// A point-in-time description of a submitted SQL query.
#[derive(Clone, Debug, PartialEq)]
pub struct QueryDescription {
/// The stable, connection-scoped identifier assigned when the query was submitted.
pub id: Uuid,
/// The server-visible lifecycle state.
pub status: QueryStatus,
/// Server-reported completion progress, when known. Values are in `[0.0, 1.0]`,
/// with `1.0` meaning complete.
pub progress: Option<f64>,
/// When the server may stop accepting this query's continuation token.
pub expires_at: Option<DateTime<Utc>>,
}
#[async_trait]
pub(crate) trait QueryHandle: Send + Sync {
fn id(&self) -> Uuid;
async fn describe(&self) -> Result<QueryDescription>;
async fn reader(&self) -> Result<SendableRecordBatchStream>;
async fn cancel(&self) -> Result<()>;
}
/// A handle to a submitted SQL query.
///
/// The handle can be inspected, opened as an Arrow reader, or cancelled.
/// Dropping it does not cancel the server-side query.
/// Identifier lookup is scoped to the connection that submitted the query and
/// is not a durable resume mechanism.
pub struct Query {
handle: Arc<dyn QueryHandle>,
}
impl std::fmt::Debug for Query {
fn fmt(&self, formatter: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
formatter
.debug_struct("Query")
.field("id", &self.id())
.finish()
}
}
impl Query {
#[cfg(feature = "remote")]
pub(crate) fn new(handle: Arc<dyn QueryHandle>) -> Self {
Self { handle }
}
/// Return the stable, connection-scoped identifier for this query.
pub fn id(&self) -> Uuid {
self.handle.id()
}
/// Get a point-in-time description of the query.
pub async fn describe(&self) -> Result<QueryDescription> {
self.handle.describe().await
}
/// Wait for the initial result stream and return its Arrow record batches.
///
/// The stream can begin yielding partial results before query execution is
/// complete. It continues polling for newly available result endpoints
/// until the query finishes and all endpoints have been consumed.
///
/// Results are single-consumer. Calling this method more than once on the
/// same handle returns an error.
pub async fn reader(&self) -> Result<SendableRecordBatchStream> {
self.handle.reader().await
}
/// Request cancellation of the query.
pub async fn cancel(&self) -> Result<()> {
self.handle.cancel().await
}
}
#[cfg(test)]
mod tests {
use super::QueryStatus;
#[test]
fn query_status_display_is_stable() {
assert_eq!(QueryStatus::Running.to_string(), "running");
assert_eq!(QueryStatus::Finished.to_string(), "finished");
assert_eq!(QueryStatus::Cancelling.to_string(), "cancelling");
assert_eq!(QueryStatus::Cancelled.to_string(), "cancelled");
}
}
+2 -12
View File
@@ -1192,9 +1192,6 @@ impl Table {
/// valid empty blobs contain empty byte strings. Prefer
/// [`Self::fetch_blob_files`] for large selections.
///
/// `_rowid` values stay valid after compaction when the table has stable
/// row ids.
///
/// ```
/// use arrow_array::UInt64Array;
/// use futures::TryStreamExt;
@@ -1236,9 +1233,6 @@ impl Table {
/// the requests. Null blobs produce null output slots; empty ranges on
/// non-null blobs produce empty byte strings.
///
/// `_rowid` values stay valid after compaction when the table has stable
/// row ids.
///
/// ```
/// use lancedb::blob::BlobRangeRequest;
///
@@ -1277,9 +1271,6 @@ impl Table {
/// Same length and order as `row_ids`. Null rows are `None`. Bytes are not
/// read from disk until a call to [`BlobFile::read`].
///
/// `_rowid` values stay valid after compaction when the table has stable
/// row ids.
///
/// ```
/// # use lancedb::Table;
/// # async fn lazy_read(table: &Table, row_ids: &[u64]) -> Result<(), Box<dyn std::error::Error>> {
@@ -2804,7 +2795,7 @@ impl NativeTable {
namespace_client: Option<Arc<dyn LanceNamespace>>,
pushdown_operations: HashSet<NamespaceClientPushdownOperation>,
) -> Result<Self> {
let batches = computed_columns::admit_create_source(batches)?;
computed_columns::ensure_no_foreign_declarations(batches.arrow_schema().fields())?;
// Default params uses format v1.
let params = params.unwrap_or(WriteParams {
..Default::default()
@@ -2904,7 +2895,6 @@ impl NativeTable {
pushdown_operations: HashSet<NamespaceClientPushdownOperation>,
session: Option<Arc<lance::session::Session>>,
) -> Result<Self> {
let batches = computed_columns::admit_create_source(batches)?;
// Build table_id from namespace + name for the storage options provider
let mut table_id = namespace.clone();
table_id.push(name.to_string());
@@ -5678,7 +5668,7 @@ mod tests {
TableStatistics {
num_rows: 250,
num_indices: 0,
total_bytes: 8969,
total_bytes: 8925,
fragment_stats: FragmentStatistics {
num_fragments: 11,
num_small_fragments: 11,
+63 -519
View File
@@ -21,7 +21,6 @@
//! [`computed_columns`] and [`computed_column_from_field`] read declarations
//! back off a schema.
use futures::StreamExt;
use std::collections::{BTreeSet, HashMap, HashSet};
use std::sync::Arc;
@@ -61,10 +60,6 @@ pub const FUNCTION_BINDING_ID_META_KEY: &str = "computed_column.function.binding
/// Field metadata key holding this sibling's ordered Function output ordinal.
pub const FUNCTION_OUTPUT_ORDINAL_META_KEY: &str = "computed_column.function.output_ordinal";
/// Reserved Function output ordinal for an internal flattened-result
/// assignment column.
pub const FUNCTION_ASSIGNMENT_OUTPUT_ORDINAL: u32 = u32::MAX;
/// Schema metadata key holding all immutable Function bindings.
pub const FUNCTION_BINDINGS_META_KEY: &str = "lancedb::function_bindings";
@@ -317,29 +312,22 @@ pub(crate) fn ensure_supported_function_metadata(schema: &ArrowSchema) -> Result
binding_id
),
})?;
let destination = if output_ordinal == FUNCTION_ASSIGNMENT_OUTPUT_ORDINAL {
binding
.assignment()
.map(|assignment| assignment.output_name.as_str())
} else {
binding
.outputs()
.get(output_ordinal as usize)
.map(|output| output.output_name.as_str())
}
.ok_or_else(|| Error::InvalidInput {
message: format!(
"Function output '{}' has invalid ordinal {}",
field.name(),
output_ordinal
),
})?;
if destination != field.name().as_str() {
let output = binding
.outputs()
.get(output_ordinal as usize)
.ok_or_else(|| Error::InvalidInput {
message: format!(
"Function output '{}' has invalid ordinal {}",
field.name(),
output_ordinal
),
})?;
if output.output_name != field.name().as_str() {
return Err(Error::InvalidInput {
message: format!(
"Function output '{}' does not match binding destination '{}'",
field.name(),
destination
output.output_name
),
});
}
@@ -510,7 +498,6 @@ fn ensure_known_binding_shape(value: &Value) -> Result<()> {
"function",
"inputs",
"outputs",
"assignment",
"input_schema",
"output_schema",
],
@@ -559,13 +546,6 @@ fn ensure_known_binding_shape(value: &Value) -> Result<()> {
"output mapping",
)?;
}
if let Some(assignment) = object.get("assignment") {
reject_unknown_object_fields(
assignment,
&["output_name", "output_field_id"],
"assignment mapping",
)?;
}
Ok(())
}
@@ -760,33 +740,14 @@ fn function_output_field(name: &str, nullable: bool, raw: &str) -> Result<JsonAr
Ok(field)
}
/// Whether two fields describe the same Function output.
///
/// `compare_identity` covers the field's own name and nullability. Struct
/// children carry both as part of the declaration and compare with it on. List
/// children do not: Lance rewrites a list item's name and nullability when it
/// writes, so a stored `fixed_size_list<item: float not null>` comes back as
/// `fixed_size_list<item: float>` and never matches the declaration again.
/// Comparing those by type alone keeps this agreeing with the server, which
/// draws the same distinction and is what accepted the column when it was
/// declared.
fn function_output_field_matches(
expected: &ArrowField,
actual: &ArrowField,
compare_identity: bool,
) -> bool {
if compare_identity
&& (expected.name() != actual.name() || expected.is_nullable() != actual.is_nullable())
{
return false;
}
match (expected.is_blob_v2(), actual.is_blob_v2()) {
(false, false) => function_output_type_matches(expected.data_type(), actual.data_type()),
(true, true) => {
fn function_output_field_matches(expected: &ArrowField, actual: &ArrowField) -> bool {
expected.name() == actual.name()
&& expected.is_nullable() == actual.is_nullable()
&& if expected.is_blob_v2() {
has_supported_blob_v2_layout(expected) && has_supported_blob_v2_layout(actual)
} else {
function_output_type_matches(expected.data_type(), actual.data_type())
}
_ => false,
}
}
fn function_output_type_matches(expected: &DataType, actual: &DataType) -> bool {
@@ -799,19 +760,33 @@ fn function_output_type_matches(expected: &DataType, actual: &DataType) -> bool
&& expected
.iter()
.zip(actual)
.all(|(expected, actual)| function_output_field_matches(expected, actual, true))
.all(|(expected, actual)| function_output_field_matches(expected, actual))
}
(DataType::List(expected), DataType::List(actual))
| (DataType::LargeList(expected), DataType::LargeList(actual)) => {
function_output_field_matches(expected, actual, false)
function_output_field_matches(expected, actual)
}
(
DataType::FixedSizeList(expected, expected_size),
DataType::FixedSizeList(actual, actual_size),
) => expected_size == actual_size && function_output_field_matches(expected, actual, false),
) => expected_size == actual_size && function_output_field_matches(expected, actual),
(DataType::Map(expected, expected_sorted), DataType::Map(actual, actual_sorted)) => {
expected_sorted == actual_sorted
&& function_output_field_matches(expected, actual, true)
expected_sorted == actual_sorted && function_output_field_matches(expected, actual)
}
_ => false,
}
}
fn function_output_type_has_blob(data_type: &DataType) -> bool {
match data_type {
DataType::Struct(fields) => fields
.iter()
.any(|field| field.is_blob_v2() || function_output_type_has_blob(field.data_type())),
DataType::List(field)
| DataType::LargeList(field)
| DataType::FixedSizeList(field, _)
| DataType::Map(field, _) => {
field.is_blob_v2() || function_output_type_has_blob(field.data_type())
}
_ => false,
}
@@ -890,20 +865,23 @@ fn ensure_binding_matches_schema(schema: &ArrowSchema, binding: &FunctionBinding
output.output_name
))
})?;
if field.name() != &output.output_name || !field.is_nullable() {
if field.name() != &output.output_name || !field.is_nullable() || output.nullable {
return Err(invalid_function(format!(
"Function output '{}' no longer matches binding '{}'",
output.output_name,
binding.binding_id()
)));
}
let type_matches = if output.arrow_type == FUNCTION_BLOB_V2_TYPE {
has_supported_blob_v2_layout(field)
let (type_matches, has_semantic_blob) = if output.arrow_type == FUNCTION_BLOB_V2_TYPE {
(has_supported_blob_v2_layout(field), true)
} else {
let expected_type = parse_output_arrow_type(&output.arrow_type)?;
let expected_type = lance_namespace::schema::convert_json_arrow_type(&expected_type)
.map_err(|e| invalid_function(format!("invalid Function output type: {e}")))?;
function_output_type_matches(&expected_type, field.data_type())
(
function_output_type_matches(&expected_type, field.data_type()),
function_output_type_has_blob(&expected_type),
)
};
if !type_matches {
return Err(invalid_function(format!(
@@ -934,71 +912,19 @@ fn ensure_binding_matches_schema(schema: &ArrowSchema, binding: &FunctionBinding
binding.binding_id()
)));
}
// Rebuild from the declaration rather than from the stored field. The
// stored field carries Lance's write-time normalization, which would
// never round-trip back to the schema the binding recorded -- the same
// reason list children compare by type above. Whether the column on
// disk still matches is settled by that comparison, not here.
output_fields.push(function_output_field(
field.name(),
true,
&output.arrow_type,
)?);
}
if let Some(assignment) = binding.assignment() {
if binding
.outputs()
.iter()
.any(|output| output.result_field == WHOLE_RESULT_FIELD)
{
return Err(invalid_function(format!(
"Function binding '{}' cannot attach an assignment column to a whole result",
binding.binding_id()
)));
if has_semantic_blob {
output_fields.push(function_output_field(
field.name(),
true,
&output.arrow_type,
)?);
} else {
let json = lance_namespace::schema::arrow_schema_to_json(&ArrowSchema::new(vec![
ArrowField::new(field.name().clone(), field.data_type().clone(), true),
]))
.map_err(|e| invalid_function(format!("invalid Function output schema: {e}")))?;
output_fields.push(json.fields.into_iter().next().unwrap());
}
let field = schema
.field_with_name(&assignment.output_name)
.map_err(|_| {
invalid_function(format!(
"Function binding '{}' assignment column '{}' is missing",
binding.binding_id(),
assignment.output_name
))
})?;
let metadata = field.metadata();
if field.data_type() != &DataType::Boolean
|| !field.is_nullable()
|| metadata.get(COMPUTED_COLUMN_META_KEY).map(String::as_str) != Some("true")
|| metadata.get(KIND_META_KEY).map(String::as_str) != Some(FUNCTION_KIND)
|| metadata
.get(FUNCTION_BINDING_ID_META_KEY)
.map(String::as_str)
!= Some(binding.binding_id())
|| metadata
.get(FUNCTION_OUTPUT_ORDINAL_META_KEY)
.and_then(|value| value.parse::<u32>().ok())
!= Some(FUNCTION_ASSIGNMENT_OUTPUT_ORDINAL)
{
return Err(invalid_function(format!(
"Function binding '{}' assignment column no longer matches its declaration",
binding.binding_id()
)));
}
let json = lance_namespace::schema::arrow_schema_to_json(&ArrowSchema::new(vec![
ArrowField::new(assignment.output_name.clone(), DataType::Boolean, true),
]))
.map_err(|e| invalid_function(format!("invalid Function assignment schema: {e}")))?;
output_fields.push(json.fields.into_iter().next().unwrap());
} else if binding.outputs().iter().all(|output| output.nullable)
&& binding
.outputs()
.iter()
.all(|output| output.result_field != WHOLE_RESULT_FIELD)
{
return Err(invalid_function(format!(
"Function binding '{}' has no flattened-result assignment column",
binding.binding_id()
)));
}
let output_schema = JsonArrowSchema::new(output_fields);
let output_schema = serde_json::to_value(output_schema).map_err(|e| {
@@ -1155,6 +1081,11 @@ pub(crate) fn plan_function_application(
"named-struct Function result field names must be unique",
));
}
if output.fields.iter().any(|field| field.nullable) {
return Err(invalid_function(
"Function logical outputs must be non-nullable during NULL assignment",
));
}
let unknown = application
.columns()
.keys()
@@ -1176,9 +1107,7 @@ pub(crate) fn plan_function_application(
let fields = output
.fields
.iter()
.map(|field| {
function_output_field(&field.name, field.nullable, &field.arrow_type)
})
.map(|field| function_output_field(&field.name, false, &field.arrow_type))
.collect::<Result<Vec<_>>>()?;
let mut data_type = JsonArrowDataType::new("struct".to_string());
data_type.fields = Some(fields);
@@ -1339,106 +1268,6 @@ pub(crate) fn ensure_batch_writes_no_computed_values(
Ok(())
}
/// Validate every computed-column declaration `schema` carries against the
/// schema itself: every field with declaration metadata is a complete
/// declaration, a SQL declaration re-plans to the field it declares, a
/// Function declaration satisfies the binding contract, and no declaration
/// reads another computed column. What passes here is what `refresh_column`
/// can execute.
pub(crate) fn ensure_declarations_are_planned(schema: &ArrowSchema) -> Result<()> {
let invalid = |message: String| Error::InvalidInput { message };
// A field with any declaration key is a declaration; a partial one is
// not "no declaration", it is a broken one.
for field in schema.fields() {
if field.metadata().keys().any(|k| is_declaration_key(k))
&& computed_column_from_field(field).is_none()
{
return Err(invalid(format!(
"field '{}' carries an incomplete computed-column declaration",
field.name()
)));
}
}
let declared: HashSet<String> = computed_columns(schema)
.into_iter()
.map(|c| c.name)
.collect();
for column in computed_columns(schema) {
let field = schema.field_with_name(&column.name)?;
if !field.is_nullable() {
return Err(invalid(format!(
"computed column '{}' must be nullable until a refresh fills it",
column.name
)));
}
match &column.kind {
ComputedColumnKind::Sql { expression } => {
let others: Vec<ArrowField> = schema
.fields()
.iter()
.filter(|f| f.name() != &column.name)
.map(|f| f.as_ref().clone())
.collect();
let bound = bind(Arc::new(ArrowSchema::new(others)), &column.name, expression)?;
if let Some(input) = bound.roots.iter().find(|r| declared.contains(*r)) {
return Err(invalid(format!(
"computed column '{}' reads computed column '{input}'",
column.name
)));
}
if &bound.data_type != field.data_type() {
return Err(invalid(format!(
"computed column '{}' is declared as {} but its expression yields {}",
column.name,
field.data_type(),
bound.data_type
)));
}
let mut declared_inputs = column.inputs.clone();
declared_inputs.sort();
if declared_inputs != bound.inputs {
return Err(invalid(format!(
"computed column '{}' declares inputs {:?} but its expression reads {:?}",
column.name, declared_inputs, bound.inputs
)));
}
}
ComputedColumnKind::Function { binding_id, .. } => {
// The binding validator resolves each input's leaf; the
// no-computed-input rule is about the root it hangs from.
let bindings = function_bindings(schema)?;
let Some(binding) = bindings.iter().find(|b| b.binding_id() == binding_id) else {
continue; // reported by the binding validator below
};
// Roots come from the canonical path parser: a quoted
// top-level name may itself contain a dot.
if let Some(input) = binding
.inputs()
.iter()
.filter_map(|input| resolve_field_path(schema, &input.field_path).ok())
.map(|resolved| resolved.root.name().as_str())
.find(|r| declared.contains(*r))
{
return Err(invalid(format!(
"computed column '{}' reads computed column '{input}'",
column.name
)));
}
}
ComputedColumnKind::Unrecognized { kind } => {
return Err(Error::NotSupported {
message: format!(
"computed column '{}' is defined by '{kind}', which this version \
of lancedb cannot fill",
column.name
),
});
}
}
}
ensure_supported_function_metadata(schema)
}
/// Reject fields carrying declaration metadata that did not come through
/// [`plan`]. One authority for creation, overwrite and raw transforms.
pub(crate) fn ensure_no_foreign_declarations<'a>(
@@ -1897,54 +1726,6 @@ pub(super) async fn add_foreign_kind(table: &crate::Table, name: &str, kind: &st
.unwrap();
}
/// Admit a table's initial data: every declaration it carries is validated,
/// and the stream refuses any batch with values in a computed column, whose
/// values come from refresh alone. One boundary for every way a table is
/// created.
pub(crate) fn admit_create_source<S: lance_datafusion::utils::StreamingWriteSource>(
batches: S,
) -> Result<UnfilledDeclarations<S>> {
let schema = batches.arrow_schema();
ensure_declarations_are_planned(&schema)?;
let declared = computed_columns(&schema)
.into_iter()
.map(|c| c.name)
.collect();
Ok(UnfilledDeclarations {
inner: batches,
declared,
})
}
/// A write source whose computed columns must arrive unfilled.
pub(crate) struct UnfilledDeclarations<S> {
inner: S,
declared: Vec<String>,
}
impl<S: lance_datafusion::utils::StreamingWriteSource> lance_datafusion::utils::StreamingWriteSource
for UnfilledDeclarations<S>
{
fn arrow_schema(&self) -> SchemaRef {
self.inner.arrow_schema()
}
fn into_stream(self) -> datafusion_physical_plan::SendableRecordBatchStream {
if self.declared.is_empty() {
return self.inner.into_stream();
}
let schema = self.inner.arrow_schema();
let declared = self.declared;
let stream = self.inner.into_stream().map(move |batch| {
let batch = batch?;
ensure_batch_writes_no_computed_values(&declared, &batch)
.map_err(|e| datafusion_common::DataFusionError::External(Box::new(e)))?;
Ok(batch)
});
Box::pin(datafusion_physical_plan::stream::RecordBatchStreamAdapter::new(schema, stream))
}
}
#[cfg(test)]
mod tests {
/// The gate's reproducer: the validator applies the same schema-level
@@ -1963,69 +1744,6 @@ mod tests {
assert!(super::validate_declarations(schema, &declarations).is_err());
}
#[test]
fn list_children_match_by_type_but_struct_children_by_identity() {
use arrow_schema::Field as F;
// Lance rewrites a list item's name and nullability on write, so the
// stored field is no longer identical to what was declared. Comparing
// those by type keeps a table with a vector output usable.
let declared =
DataType::FixedSizeList(Arc::new(F::new("item", DataType::Float32, false)), 4);
let stored = DataType::FixedSizeList(Arc::new(F::new("item", DataType::Float32, true)), 4);
assert!(super::function_output_type_matches(&declared, &stored));
let renamed =
DataType::FixedSizeList(Arc::new(F::new("element", DataType::Float32, true)), 4);
assert!(super::function_output_type_matches(&declared, &renamed));
// The dimension is still part of the declaration.
let resized = DataType::FixedSizeList(Arc::new(F::new("item", DataType::Float32, true)), 8);
assert!(!super::function_output_type_matches(&declared, &resized));
// Struct children keep comparing by name and nullability.
let struct_declared =
DataType::Struct(vec![F::new("changed", DataType::Boolean, false)].into());
let struct_nullable =
DataType::Struct(vec![F::new("changed", DataType::Boolean, true)].into());
let struct_renamed =
DataType::Struct(vec![F::new("altered", DataType::Boolean, false)].into());
assert!(super::function_output_type_matches(
&struct_declared,
&struct_declared
));
assert!(!super::function_output_type_matches(
&struct_declared,
&struct_nullable
));
assert!(!super::function_output_type_matches(
&struct_declared,
&struct_renamed
));
// A list nested inside a struct gets the list rule.
let nested_declared = DataType::Struct(
vec![F::new(
"tokens",
DataType::List(Arc::new(F::new("item", DataType::Utf8, false))),
true,
)]
.into(),
);
let nested_stored = DataType::Struct(
vec![F::new(
"tokens",
DataType::List(Arc::new(F::new("item", DataType::Utf8, true))),
true,
)]
.into(),
);
assert!(super::function_output_type_matches(
&nested_declared,
&nested_stored
));
}
#[test]
fn output_arrow_type_grammar_matches_the_shared_golden() {
let golden: serde_json::Value = serde_json::from_str(include_str!(
@@ -2795,8 +2513,6 @@ mod tests {
);
}
/// A create carries a declaration only if it re-plans completely; this
/// one lacks its inputs and is refused before its forged value matters.
#[tokio::test]
async fn test_create_table_cannot_inject_a_declaration() {
let conn = connect("memory://").execute().await.unwrap();
@@ -2824,7 +2540,7 @@ mod tests {
.await
.unwrap_err();
assert!(
matches!(&err, Error::InvalidInput { message } if message.contains("computed column 'doubled'")),
matches!(&err, Error::InvalidInput { message } if message.contains("computed()")),
"{err:?}"
);
}
@@ -3142,17 +2858,6 @@ mod tests {
&inputs,
));
}
if let Some(assignment) = binding.assignment() {
fields.push(
ArrowField::new(&assignment.output_name, DataType::Boolean, true).with_metadata(
function_computed_column_metadata(
binding.binding_id(),
FUNCTION_ASSIGNMENT_OUTPUT_ORDINAL,
&inputs,
),
),
);
}
ArrowSchema::new(fields)
}
@@ -3170,74 +2875,6 @@ mod tests {
.unwrap();
}
#[test]
fn test_binding_preserves_all_nullable_outputs_with_an_assignment_column() {
let mut raw_binding: Value = serde_json::from_str(include_str!(
"../../tests/fixtures/first_class_functions/v1/remote_function_binding.json"
))
.unwrap();
raw_binding["outputs"][0]["nullable"] = Value::Bool(true);
raw_binding["outputs"][1]["nullable"] = Value::Bool(true);
let without_assignment: FunctionBinding =
serde_json::from_value(raw_binding.clone()).unwrap();
let error = ensure_binding_matches_schema(
&valid_function_binding_schema(true, true, &without_assignment),
&without_assignment,
)
.unwrap_err();
assert!(
error
.to_string()
.contains("flattened-result assignment column")
);
raw_binding["assignment"] = serde_json::json!({
"output_name": "__function_assignment_fb_01K3TEXT",
"output_field_id": -1,
});
raw_binding["output_schema"]["fields"]
.as_array_mut()
.unwrap()
.push(serde_json::json!({
"name": "__function_assignment_fb_01K3TEXT",
"nullable": true,
"type": {"type": "bool"},
}));
let binding: FunctionBinding = serde_json::from_value(raw_binding).unwrap();
ensure_binding_matches_schema(
&valid_function_binding_schema(true, true, &binding),
&binding,
)
.unwrap();
let schema = ArrowSchema::new_with_metadata(
valid_function_binding_schema(true, true, &binding)
.fields()
.to_vec(),
HashMap::from([(
FUNCTION_BINDINGS_META_KEY.to_string(),
function_bindings_metadata(std::slice::from_ref(&binding)).unwrap(),
)]),
);
ensure_supported_function_metadata(&schema).unwrap();
let mut metadata: Value =
serde_json::from_str(schema.metadata().get(FUNCTION_BINDINGS_META_KEY).unwrap())
.unwrap();
metadata["bindings"][0]["assignment"]["future"] = Value::Bool(true);
let future_schema = ArrowSchema::new_with_metadata(
schema.fields().to_vec(),
HashMap::from([(
FUNCTION_BINDINGS_META_KEY.to_string(),
serde_json::to_string(&metadata).unwrap(),
)]),
);
assert!(matches!(
ensure_supported_function_metadata(&future_schema),
Err(Error::NotSupported { .. })
));
}
#[test]
fn test_nullable_function_input_cannot_bind_to_non_nullable_parameter() {
let mut raw_binding: Value = serde_json::from_str(include_str!(
@@ -3457,35 +3094,6 @@ mod tests {
);
}
#[test]
fn test_named_struct_plan_preserves_nullable_result_fields() {
let mut value = serde_json::to_value(named_struct_application("{}")).unwrap();
value["output"]["fields"][0]["nullable"] = Value::Bool(true);
value["output"]["fields"][1]["nullable"] = Value::Bool(true);
let application = FunctionApplication::from_json(&value.to_string()).unwrap();
let expanded =
plan_function_application(&function_input_schema(), &application, None).unwrap();
assert!(
expanded
.output_schema
.fields
.iter()
.all(|field| field.nullable)
);
let whole =
plan_function_application(&function_input_schema(), &application, Some("features"))
.unwrap();
let fields = whole.output_schema.fields[0]
.r#type
.fields
.as_ref()
.unwrap();
assert!(fields[0].nullable);
assert!(fields[1].nullable);
}
#[test]
fn test_blob_function_plans_semantic_input_and_scalar_output() {
let schema = ArrowSchema::new(vec![crate::blob("image", false)]);
@@ -3502,70 +3110,6 @@ mod tests {
assert!(output_schema.field(0).is_blob_v2());
}
#[test]
fn binding_accepts_a_lance_normalized_list_child() {
// The whole guard, not just the type helper: this also reaches the
// output-schema comparison at the end of ensure_binding_matches_schema,
// which used to rebuild the schema from the stored field and so failed
// on exactly the same normalization.
let input = ArrowField::new("value", DataType::Int64, false);
let application = FunctionApplication::from_json(
&serde_json::json!({
"function": {"name": "embed", "version": "fv_embed"},
"inputs": [{
"parameter": "value",
"kind": "column",
"value": {"path": "value"}
}],
"output": {
"kind": "scalar",
"arrow_type": "fixed_size_list<float32, 4>",
"nullable": false
}
})
.to_string(),
)
.unwrap();
let plan = plan_function_application(
&ArrowSchema::new(vec![input.clone()]),
&application,
Some("embedding"),
)
.unwrap();
let binding = binding_from_plan(&plan);
// The declaration says the item is non-nullable; Lance rewrites it to
// nullable on write, so this is what the column looks like on disk.
let stored = DataType::FixedSizeList(
Arc::new(ArrowField::new("item", DataType::Float32, true)),
4,
);
let output = ArrowField::new("embedding", stored, true).with_metadata(
function_computed_column_metadata(binding.binding_id(), 0, &["value".into()]),
);
ensure_binding_matches_schema(&ArrowSchema::new(vec![input.clone(), output]), &binding)
.unwrap();
// A different element type is still a mismatch.
let wrong = ArrowField::new(
"embedding",
DataType::FixedSizeList(
Arc::new(ArrowField::new("item", DataType::Float64, true)),
4,
),
true,
)
.with_metadata(function_computed_column_metadata(
binding.binding_id(),
0,
&["value".into()],
));
assert!(
ensure_binding_matches_schema(&ArrowSchema::new(vec![input, wrong]), &binding).is_err()
);
}
#[test]
fn test_blob_scalar_binding_accepts_full_logical_layout() {
let input = crate::blob("image", false);
+2 -78
View File
@@ -133,7 +133,7 @@ impl NativeTable {
),
});
}
(resolved.canonical_path, resolved.terminal_field)
(resolved.canonical_path, resolved.field)
} else {
Self::resolve_index_field(dataset.schema(), &opts.columns[0])?
};
@@ -439,8 +439,7 @@ mod tests {
use arrow_array::record_batch;
use arrow_array::{
Array, ArrayRef, BinaryArray, BooleanArray, FixedSizeListArray, Float32Array, Int32Array,
LargeBinaryArray, LargeStringArray, ListArray, RecordBatch, StringArray, StructArray,
UInt32Array,
LargeBinaryArray, LargeStringArray, RecordBatch, StringArray, StructArray,
};
use arrow_data::ArrayDataBuilder;
use arrow_schema::{DataType, Field, Schema};
@@ -459,7 +458,6 @@ mod tests {
use crate::query::{ExecutableQuery, QueryBase};
use crate::table::optimize::{CompactionOptions, OptimizeAction};
use lance_index::scalar::FullTextSearchQuery;
use lance_index::scalar::inverted::query::{FtsQuery, MatchQuery};
fn create_fixed_size_list<T: Array>(
values: T,
@@ -601,80 +599,6 @@ mod tests {
assert!(invalid_granularity.is_err());
}
#[tokio::test]
async fn test_nested_list_fts_uses_deepest_document_coordinates() {
let conn = connect("memory://").execute().await.unwrap();
let mut docs = ListBuilder::new(ListBuilder::new(StringBuilder::new()));
docs.values().values().append_value("alpha");
docs.values().values().append_value("beta");
docs.values().append(true);
docs.values().values().append_value("gamma");
docs.values().values().append_value("alpha delta");
docs.values().append(true);
docs.append(true);
docs.values().append(true);
docs.values().values().append_value("alpha");
docs.values().append(true);
docs.append(true);
let batch = RecordBatch::try_from_iter(vec![
("id", Arc::new(Int32Array::from(vec![0, 1])) as ArrayRef),
("docs", Arc::new(docs.finish()) as ArrayRef),
])
.unwrap();
let table = conn.create_table("nested", batch).execute().await.unwrap();
let job = table
.create_index(
&["docs"],
Index::FTS(
FtsIndexBuilder::default()
.document_granularity(DocumentGranularity::ListElement),
),
)
.execute_async()
.await
.unwrap();
job.wait().await.unwrap();
let query = FullTextSearchQuery::new_query(FtsQuery::Match(
MatchQuery::new("alpha".to_string())
.with_column(Some("docs".to_string()))
.with_document_granularity(DocumentGranularity::ListElement),
));
let batches = table
.query()
.full_text_search(query)
.limit(10)
.execute()
.await
.unwrap()
.try_collect::<Vec<_>>()
.await
.unwrap();
let mut hits = Vec::new();
for batch in batches {
let ids = batch["id"].as_any().downcast_ref::<Int32Array>().unwrap();
let coordinates = batch["_doc_index"]
.as_any()
.downcast_ref::<ListArray>()
.unwrap();
for row in 0..batch.num_rows() {
let coordinate = coordinates.value(row);
let coordinate = coordinate.as_any().downcast_ref::<UInt32Array>().unwrap();
hits.push((ids.value(row), coordinate.values().to_vec()));
}
}
hits.sort_unstable();
assert_eq!(
hits,
vec![(0, vec![0, 0]), (0, vec![1, 1]), (1, vec![1, 0])]
);
}
/// Concurrent waiters, and a wait issued after the job settled, all
/// succeed once the build does.
#[tokio::test]
@@ -5,17 +5,12 @@
//!
//! [`super::cast::cast_to_table_schema`] calls [`coerce_blob_expr`].
use std::fmt;
use std::hash::{Hash, Hasher};
use std::sync::Arc;
use arrow_array::{Array, BooleanArray, RecordBatch};
use arrow_schema::{DataType, Field, FieldRef, Fields, Schema};
use arrow_select::nullif::nullif;
use arrow_schema::{DataType, Field, FieldRef, Fields};
use datafusion::functions::core::{get_field, named_struct};
use datafusion_common::ScalarValue;
use datafusion_common::config::ConfigOptions;
use datafusion_expr::ColumnarValue;
use datafusion_physical_expr::ScalarFunctionExpr;
use datafusion_physical_expr::expressions::{CastExpr, Literal};
use datafusion_physical_plan::PhysicalExpr;
@@ -138,102 +133,16 @@ pub(super) fn coerce_blob_expr(
ns_args.push(value);
}
let built: Arc<dyn PhysicalExpr> = Arc::new(ScalarFunctionExpr::new(
let expr: Arc<dyn PhysicalExpr> = Arc::new(ScalarFunctionExpr::new(
&format!("named_struct({})", table_field.name()),
named_struct(),
ns_args,
table_field.clone(),
config.clone(),
));
// `named_struct` always yields a valid struct, so a null input would land
// as a row that set neither `data` nor `uri` -- not an absent blob but a
// malformed one, which Lance rejects on write.
let expr: Arc<dyn PhysicalExpr> = Arc::new(AbsentBlobIsNull {
source: input_expr,
built,
field: table_field.clone(),
});
Ok((expr, table_field.clone()))
}
/// Carries the source column's nullity onto the struct built for it.
///
/// This is its own expression rather than a `CASE` because the projection
/// takes its output field from `return_field`, and the generic implementation
/// rebuilds a bare field -- which would drop the `lance.blob.v2` extension
/// metadata and stop the column being recognised as a blob at all.
#[derive(Debug, Clone)]
struct AbsentBlobIsNull {
source: Arc<dyn PhysicalExpr>,
built: Arc<dyn PhysicalExpr>,
field: FieldRef,
}
impl fmt::Display for AbsentBlobIsNull {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "absent_blob_is_null({}, {})", self.source, self.built)
}
}
impl PartialEq for AbsentBlobIsNull {
fn eq(&self, other: &Self) -> bool {
self.source.eq(&other.source) && self.built.eq(&other.built) && self.field == other.field
}
}
impl Eq for AbsentBlobIsNull {}
impl Hash for AbsentBlobIsNull {
fn hash<H: Hasher>(&self, state: &mut H) {
self.source.hash(state);
self.built.hash(state);
self.field.hash(state);
}
}
impl PhysicalExpr for AbsentBlobIsNull {
fn return_field(&self, _input_schema: &Schema) -> datafusion_common::Result<FieldRef> {
Ok(self.field.clone())
}
fn nullable(&self, _input_schema: &Schema) -> datafusion_common::Result<bool> {
Ok(true)
}
fn evaluate(&self, batch: &RecordBatch) -> datafusion_common::Result<ColumnarValue> {
let rows = batch.num_rows();
let built = self.built.evaluate(batch)?.into_array(rows)?;
let source = self.source.evaluate(batch)?.into_array(rows)?;
let Some(nulls) = source.logical_nulls() else {
return Ok(ColumnarValue::Array(built));
};
// `nullif` nulls the rows the mask marks true, which is where the
// source had no value.
let absent = BooleanArray::new(!nulls.inner(), None);
Ok(ColumnarValue::Array(nullif(built.as_ref(), &absent)?))
}
fn children(&self) -> Vec<&Arc<dyn PhysicalExpr>> {
vec![&self.source, &self.built]
}
fn with_new_children(
self: Arc<Self>,
children: Vec<Arc<dyn PhysicalExpr>>,
) -> datafusion_common::Result<Arc<dyn PhysicalExpr>> {
Ok(Arc::new(Self {
source: children[0].clone(),
built: children[1].clone(),
field: self.field.clone(),
}))
}
fn fmt_sql(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(f, "{self}")
}
}
enum BlobInputShape<'a> {
Bytes,
String,
@@ -404,11 +313,6 @@ mod tests {
let data = image.column_by_name("data").unwrap();
assert!(!data.is_null(0));
assert!(data.is_null(1));
// The row itself has to be null, not merely a struct whose children
// are. A present-but-empty struct set neither `data` nor `uri`, which
// Lance rejects as malformed rather than reading as an absent blob.
assert!(!image.is_null(0));
assert!(image.is_null(1));
}
#[tokio::test]
+4 -5
View File
@@ -227,7 +227,7 @@ pub(crate) fn resolve_arrow_field_path(schema: &Schema, column: &str) -> Result<
pub(crate) struct ResolvedFtsField {
pub canonical_path: String,
pub terminal_field: Field,
pub field: Field,
pub list_depth: usize,
}
@@ -309,7 +309,7 @@ pub(crate) fn resolve_lance_fts_field_path(
);
Ok(ResolvedFtsField {
canonical_path,
terminal_field: Field::from(terminal),
field: Field::from(field),
list_depth,
})
}
@@ -375,7 +375,7 @@ pub(crate) fn resolve_arrow_fts_field_path(
message: format!("Invalid schema: {}", e),
})?;
let resolved = resolve_lance_fts_field_path(&lance_schema, column)?;
Ok((resolved.canonical_path, resolved.terminal_field))
Ok((resolved.canonical_path, resolved.field))
}
pub fn supported_btree_data_type(dtype: &DataType) -> bool {
@@ -647,9 +647,8 @@ mod tests {
Field::new("docs", text_list(), true),
]);
let (path, field) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
let (path, _) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
assert_eq!(path, "docs.content");
assert_eq!(field.data_type(), &DataType::Utf8);
let lance_schema = lance_core::datatypes::Schema::try_from(&schema).unwrap();
let field_id = lance_schema
+28 -99
View File
@@ -19,7 +19,6 @@ use lancedb::{
connect, connect_namespace,
database::listing::{
ListingDatabaseOptions, NewTableConfig, OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS,
OPT_NEW_TABLE_STORAGE_VERSION,
},
query::{ExecutableQuery, QueryBase},
table::{AddDataMode, CompactionOptions, OptimizeAction, OptimizeStats, WriteOptions},
@@ -112,7 +111,7 @@ async fn query_image_struct(table: &Table) -> StructArray {
}
#[tokio::test]
async fn declaring_blob_column_uses_v2_2_and_default_row_ids() -> Result<()> {
async fn declaring_blob_column_bumps_format_and_enables_stable_row_ids() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table = db
@@ -121,12 +120,12 @@ async fn declaring_blob_column_uses_v2_2_and_default_row_ids() -> Result<()> {
.await?;
assert!(supports_blob_v2(storage_format_version(&table).await));
assert!(!uses_stable_row_ids(&table).await);
assert!(uses_stable_row_ids(&table).await);
Ok(())
}
#[tokio::test]
async fn blob_create_honors_disabled_stable_row_ids() -> Result<()> {
async fn explicit_stable_row_id_setting_wins_over_blob_default() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table = db
@@ -147,10 +146,7 @@ async fn non_blob_table_keeps_default_format_and_row_id_setting() -> Result<()>
let schema = Arc::new(Schema::new(vec![Field::new("id", DataType::Int64, false)]));
let table = db.create_empty_table("t", schema).execute().await?;
assert_eq!(
storage_format_version(&table).await,
LanceFileVersion::Stable.resolve()
);
assert!(!supports_blob_v2(storage_format_version(&table).await));
assert!(!uses_stable_row_ids(&table).await);
Ok(())
}
@@ -183,7 +179,7 @@ async fn creating_with_blob_data_bumps_format() -> Result<()> {
let table = db.create_table("t", batch).execute().await?;
assert!(supports_blob_v2(storage_format_version(&table).await));
assert!(!uses_stable_row_ids(&table).await);
assert!(uses_stable_row_ids(&table).await);
assert_eq!(table.count_rows(None).await?, 1);
Ok(())
}
@@ -281,7 +277,7 @@ async fn add_rejects_uncoercible_blob_input() -> Result<()> {
}
#[tokio::test]
async fn connection_disables_stable_row_ids_on_blob_create() -> Result<()> {
async fn connection_level_stable_row_id_setting_wins_over_blob_default() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap())
.storage_option(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS, "false")
@@ -298,7 +294,7 @@ async fn connection_disables_stable_row_ids_on_blob_create() -> Result<()> {
}
#[tokio::test]
async fn namespace_blob_create_uses_v2_2_and_default_row_ids() -> Result<()> {
async fn namespace_create_applies_blob_defaults() -> Result<()> {
let tmp = tempdir().unwrap();
let mut properties = std::collections::HashMap::new();
properties.insert("root".to_string(), tmp.path().to_str().unwrap().to_string());
@@ -308,23 +304,6 @@ async fn namespace_blob_create_uses_v2_2_and_default_row_ids() -> Result<()> {
.execute()
.await?;
assert!(supports_blob_v2(storage_format_version(&table).await));
assert!(!uses_stable_row_ids(&table).await);
Ok(())
}
#[tokio::test]
async fn namespace_create_honors_enabled_stable_row_ids() -> Result<()> {
let tmp = tempdir().unwrap();
let mut properties = std::collections::HashMap::new();
properties.insert("root".to_string(), tmp.path().to_str().unwrap().to_string());
let db = connect_namespace("dir", properties).execute().await?;
let table = db
.create_empty_table("t", blob_table_schema())
.storage_option(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS, "true")
.execute()
.await?;
assert!(supports_blob_v2(storage_format_version(&table).await));
assert!(uses_stable_row_ids(&table).await);
Ok(())
@@ -451,35 +430,6 @@ async fn collect_id_rowid(table: &Table) -> Result<Vec<(i64, u64)>> {
.collect())
}
fn assert_missing_blob_row_ids(err: &Error) {
assert!(matches!(err, Error::InvalidInput { .. }), "got {err:?}");
let message = err.to_string();
assert!(message.contains("row ids"), "{message}");
assert!(!message.contains("rowaddr"), "{message}");
assert!(!message.contains("fragment"), "{message}");
}
async fn assert_fetch_apis_reject_missing_row_ids(table: &Table, row_ids: &[u64]) -> Result<()> {
let err = table.fetch_blobs("image", row_ids).await.unwrap_err();
assert_missing_blob_row_ids(&err);
let err = table.fetch_blob_files("image", row_ids).await.unwrap_err();
assert_missing_blob_row_ids(&err);
let err = table
.fetch_blob_ranges(
"image",
row_ids
.iter()
.copied()
.map(|row_id| BlobRangeRequest::new(row_id, 0, 1)),
)
.await
.unwrap_err();
assert_missing_blob_row_ids(&err);
Ok(())
}
#[tokio::test]
async fn fetch_blobs_round_trips_bytes() -> Result<()> {
let tmp = tempdir().unwrap();
@@ -532,7 +482,7 @@ async fn fetch_blobs_round_trips_nested_blob_column() -> Result<()> {
let table = db.create_table("t", batch).execute().await?;
assert!(supports_blob_v2(storage_format_version(&table).await));
assert!(!uses_stable_row_ids(&table).await);
assert!(uses_stable_row_ids(&table).await);
let ids = collect_row_ids(&table).await?;
let bytes = table.fetch_blobs("info.blob", &ids).await?;
@@ -706,7 +656,8 @@ async fn fetch_blob_ranges_validates_requests() -> Result<()> {
.fetch_blob_ranges("image", [BlobRangeRequest::new(u64::MAX, 0, 1)])
.await
.unwrap_err();
assert_missing_blob_row_ids(&err);
assert!(matches!(&err, Error::InvalidInput { .. }), "got {err:?}");
assert!(err.to_string().contains("row IDs"));
Ok(())
}
@@ -739,21 +690,7 @@ async fn fetch_blobs_out_of_range_id_errors_without_panic() -> Result<()> {
let table = create_inline_blob_table(&db, "t", &[1], &[Some(b"x".as_slice())]).await?;
let err = table.fetch_blobs("image", &[u64::MAX]).await.unwrap_err();
assert_missing_blob_row_ids(&err);
Ok(())
}
#[tokio::test]
async fn fetch_blob_files_rejects_missing_fragment_row_addr() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table = create_inline_blob_table(&db, "t", &[1], &[Some(b"x".as_slice())]).await?;
let err = table
.fetch_blob_files("image", &[1u64 << 32])
.await
.unwrap_err();
assert_missing_blob_row_ids(&err);
assert!(err.to_string().contains("row IDs"));
Ok(())
}
@@ -763,25 +700,24 @@ async fn fetch_blob_apis_reject_mixed_valid_and_missing_row_ids() -> Result<()>
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table = create_inline_blob_table(&db, "t", &[1], &[Some(b"x".as_slice())]).await?;
let row_id = collect_row_ids(&table).await?[0];
let missing_row_addr = 1u64 << 32;
let row_ids = [missing_row_addr, row_id];
assert_fetch_apis_reject_missing_row_ids(&table, &row_ids).await
}
let row_ids = [u64::MAX, row_id];
#[tokio::test]
async fn fetch_blob_apis_reject_deleted_row_ids() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table =
create_inline_blob_table(&db, "t", &[1, 2], &[Some(b"one".as_slice()), Some(b"two")])
.await?;
let pairs = collect_id_rowid(&table).await?;
let deleted_row_addr = pairs.iter().find(|(id, _)| *id == 2).unwrap().1;
let live_row_addr = pairs.iter().find(|(id, _)| *id == 1).unwrap().1;
let err = table.fetch_blobs("image", &row_ids).await.unwrap_err();
assert!(matches!(&err, Error::InvalidInput { .. }), "got {err:?}");
assert!(err.to_string().contains("row IDs"));
table.delete("id = 2").await?;
let err = table.fetch_blob_files("image", &row_ids).await.unwrap_err();
assert!(matches!(&err, Error::InvalidInput { .. }), "got {err:?}");
assert!(err.to_string().contains("row IDs"));
assert_fetch_apis_reject_missing_row_ids(&table, &[deleted_row_addr, live_row_addr]).await
let requests = row_ids.map(|row_id| BlobRangeRequest::new(row_id, 0, 1));
let err = table
.fetch_blob_ranges("image", requests)
.await
.unwrap_err();
assert!(matches!(&err, Error::InvalidInput { .. }), "got {err:?}");
assert!(err.to_string().contains("row IDs"));
Ok(())
}
#[tokio::test]
@@ -813,11 +749,7 @@ async fn fetch_blobs_rejects_unknown_column() -> Result<()> {
#[tokio::test]
async fn fetch_blobs_rejects_legacy_v1_blob_column() -> Result<()> {
let tmp = tempdir().unwrap();
// Legacy v1 blob columns are only writable at file version <= 2.1.
let db = connect(tmp.path().to_str().unwrap())
.storage_options([(OPT_NEW_TABLE_STORAGE_VERSION, "2.1")])
.execute()
.await?;
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let legacy = Field::new("image", DataType::LargeBinary, true).with_metadata(
std::collections::HashMap::from([("lance-encoding:blob".to_string(), "true".to_string())]),
);
@@ -988,10 +920,7 @@ async fn fetch_blobs_after_delete() -> Result<()> {
#[tokio::test]
async fn fetch_blobs_with_precompaction_row_ids_survives_compaction() -> Result<()> {
let tmp = tempdir().unwrap();
let db = connect(tmp.path().to_str().unwrap())
.storage_option(OPT_NEW_TABLE_ENABLE_STABLE_ROW_IDS, "true")
.execute()
.await?;
let db = connect(tmp.path().to_str().unwrap()).execute().await?;
let table = create_inline_blob_table(&db, "t", &[1], &[Some(b"frag-one".as_slice())]).await?;
table
.add(binary_input_batch(&[2], &[Some(b"frag-two".as_slice())]))