refactor: try upgrade regex-automata (#3575)

* refactor: try upgrade regex-automata

Signed-off-by: tison <wander4096@gmail.com>

* try fix

Signed-off-by: tison <wander4096@gmail.com>

* always check match with next_eoi_state

Signed-off-by: tison <wander4096@gmail.com>

* add a guard to prevent over moving the state

Signed-off-by: tison <wander4096@gmail.com>

* tidy

Signed-off-by: tison <wander4096@gmail.com>

---------

Signed-off-by: tison <wander4096@gmail.com>

Cargo.lock

@@ -4378,7 +4378,7 @@ dependencies = [
  "prost 0.12.3",
  "rand",
  "regex",
- "regex-automata 0.2.0",
+ "regex-automata 0.4.3",
  "snafu",
  "tempfile",
  "tokio",
@@ -7801,17 +7801,6 @@ dependencies = [
  "regex-syntax 0.6.29",
 ]
 
-[[package]]
-name = "regex-automata"
-version = "0.2.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e9368763f5a9b804326f3af749e16f9abf378d227bcdee7634b13d8f17793782"
-dependencies = [
- "fst",
- "memchr",
- "regex-syntax 0.6.29",
-]
-
 [[package]]
 name = "regex-automata"
 version = "0.4.3"

Cargo.toml

@@ -125,7 +125,7 @@ prost = "0.12"
 raft-engine = { version = "0.4.1", default-features = false }
 rand = "0.8"
 regex = "1.8"
-regex-automata = { version = "0.2", features = ["transducer"] }
+regex-automata = { version = "0.4" }
 reqwest = { version = "0.11", default-features = false, features = [
     "json",
     "rustls-tls-native-roots",

src/index/src/inverted_index/error.rs

@@ -113,7 +113,7 @@ pub enum Error {
     #[snafu(display("Failed to parse regex DFA"))]
     ParseDFA {
         #[snafu(source)]
-        error: Box<regex_automata::dfa::Error>,
+        error: Box<regex_automata::dfa::dense::BuildError>,
         location: Location,
     },
 

src/index/src/inverted_index/search/fst_apply/intersection_apply.rs

@@ -17,6 +17,10 @@ use std::mem::size_of;
 use fst::map::OpBuilder;
 use fst::{IntoStreamer, Streamer};
 use regex_automata::dfa::dense::DFA;
+use regex_automata::dfa::Automaton;
+use regex_automata::util::primitives::StateID;
+use regex_automata::util::start::Config;
+use regex_automata::Anchored;
 use snafu::{ensure, ResultExt};
 
 use crate::inverted_index::error::{
@@ -32,7 +36,53 @@ pub struct IntersectionFstApplier {
     ranges: Vec<Range>,
 
     /// A list of `Dfa` compiled from regular expression patterns.
-    dfas: Vec<DFA<Vec<u32>>>,
+    dfas: Vec<DfaFstAutomaton>,
+}
+
+#[derive(Debug)]
+struct DfaFstAutomaton(DFA<Vec<u32>>);
+
+impl fst::Automaton for DfaFstAutomaton {
+    type State = StateID;
+
+    #[inline]
+    fn start(&self) -> Self::State {
+        let config = Config::new().anchored(Anchored::No);
+        self.0.start_state(&config).unwrap()
+    }
+
+    #[inline]
+    fn is_match(&self, state: &Self::State) -> bool {
+        self.0.is_match_state(*state)
+    }
+
+    #[inline]
+    fn can_match(&self, state: &Self::State) -> bool {
+        !self.0.is_dead_state(*state)
+    }
+
+    #[inline]
+    fn accept_eof(&self, state: &StateID) -> Option<StateID> {
+        if self.0.is_match_state(*state) {
+            return Some(*state);
+        }
+        Some(self.0.next_eoi_state(*state))
+    }
+
+    #[inline]
+    fn accept(&self, state: &Self::State, byte: u8) -> Self::State {
+        if self.0.is_match_state(*state) {
+            return *state;
+        }
+        self.0.next_state(*state, byte)
+    }
+}
+
+impl IntersectionFstApplier {
+    fn new(ranges: Vec<Range>, dfas: Vec<DFA<Vec<u32>>>) -> Self {
+        let dfas = dfas.into_iter().map(DfaFstAutomaton).collect();
+        Self { ranges, dfas }
+    }
 }
 
 impl FstApplier for IntersectionFstApplier {
@@ -86,7 +136,7 @@ impl FstApplier for IntersectionFstApplier {
 
         size += self.dfas.capacity() * size_of::<DFA<Vec<u32>>>();
         for dfa in &self.dfas {
-            size += dfa.memory_usage();
+            size += dfa.0.memory_usage();
         }
         size
     }
@@ -119,7 +169,7 @@ impl IntersectionFstApplier {
             }
         }
 
-        Ok(Self { dfas, ranges })
+        Ok(Self::new(ranges, dfas))
     }
 }
 
@@ -365,18 +415,15 @@ mod tests {
 
     #[test]
     fn test_intersection_fst_applier_memory_usage() {
-        let applier = IntersectionFstApplier {
-            ranges: vec![],
-            dfas: vec![],
-        };
+        let applier = IntersectionFstApplier::new(vec![], vec![]);
 
         assert_eq!(applier.memory_usage(), 0);
 
         let dfa = DFA::new("^abc$").unwrap();
         assert_eq!(dfa.memory_usage(), 320);
 
-        let applier = IntersectionFstApplier {
-            ranges: vec![Range {
+        let applier = IntersectionFstApplier::new(
+            vec![Range {
                 lower: Some(Bound {
                     value: b"aa".to_vec(),
                     inclusive: true,
@@ -386,9 +433,8 @@ mod tests {
                     inclusive: true,
                 }),
             }],
-            dfas: vec![dfa],
-        };
-
+            vec![dfa],
+        );
         assert_eq!(
             applier.memory_usage(),
             size_of::<Range>() + 4 + size_of::<DFA<Vec<u32>>>() + 320