mirror of
https://github.com/lancedb/lancedb.git
synced 2025-12-25 06:19:57 +00:00
45 lines
1.5 KiB
Python
45 lines
1.5 KiB
Python
import glob
|
|
from typing import Iterator
|
|
from pathlib import Path
|
|
|
|
glob_string = "../src/**/*.md"
|
|
excluded_globs = [
|
|
"../src/fts.md",
|
|
"../src/embedding.md",
|
|
"../src/examples/*.md",
|
|
"../src/integrations/voxel51.md",
|
|
"../src/guides/tables.md",
|
|
"../src/python/duckdb.md",
|
|
]
|
|
|
|
python_prefix = "py"
|
|
python_file = ".py"
|
|
python_folder = "python"
|
|
|
|
files = glob.glob(glob_string, recursive=True)
|
|
excluded_files = [f for excluded_glob in excluded_globs for f in glob.glob(excluded_glob, recursive=True)]
|
|
|
|
def yield_lines(lines: Iterator[str], prefix: str, suffix: str):
|
|
in_code_block = False
|
|
# Python code has strict indentation
|
|
strip_length = 0
|
|
for line in lines:
|
|
if line.strip().startswith(prefix + python_prefix):
|
|
in_code_block = True
|
|
strip_length = len(line) - len(line.lstrip())
|
|
elif in_code_block and line.strip().startswith(suffix):
|
|
in_code_block = False
|
|
yield "\n"
|
|
elif in_code_block:
|
|
yield line[strip_length:]
|
|
|
|
for file in filter(lambda file: file not in excluded_files, files):
|
|
with open(file, "r") as f:
|
|
lines = list(yield_lines(iter(f), "```", "```"))
|
|
|
|
if len(lines) > 0:
|
|
out_path = Path(python_folder) / Path(file).name.strip(".md") / (Path(file).name.strip(".md") + python_file)
|
|
print(out_path)
|
|
out_path.parent.mkdir(exist_ok=True, parents=True)
|
|
with open(out_path, "w") as out:
|
|
out.writelines(lines) |