docs: generate table for document loaders (#22871)

Co-authored-by: Bagatur <baskaryan@gmail.com>
pull/22900/head
Isaac Francisco 4 months ago committed by GitHub
parent 8e89178047
commit 984c7a9d42
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

@ -38,6 +38,8 @@ generate-files:
$(PYTHON) scripts/model_feat_table.py $(INTERMEDIATE_DIR)
$(PYTHON) scripts/document_loader_feat_table.py $(INTERMEDIATE_DIR)
$(PYTHON) scripts/copy_templates.py $(INTERMEDIATE_DIR)
wget -q https://raw.githubusercontent.com/langchain-ai/langserve/main/README.md -O $(INTERMEDIATE_DIR)/langserve.md

@ -0,0 +1,77 @@
import sys
from pathlib import Path
from langchain_community import document_loaders
from langchain_core.document_loaders.base import BaseLoader
DOCUMENT_LOADER_TEMPLATE = """\
---
sidebar_position: 0
sidebar_class_name: hidden
keywords: [compatibility]
custom_edit_url:
hide_table_of_contents: true
---
# Document loaders
## Features
The following table shows the feature support for all document loaders.
{table}
"""
def get_document_loader_table() -> str:
"""Get the table of document loaders."""
doc_loaders_feat_table = {}
for cm in document_loaders.__all__:
doc_loaders_feat_table[cm] = {}
cls = getattr(document_loaders, cm)
if issubclass(cls, BaseLoader):
for feat in ("aload", "alazy_load", ("lazy_load", "lazy_loading")):
if isinstance(feat, tuple):
feat, name = feat
else:
feat, name = feat, feat
doc_loaders_feat_table[cm][name] = getattr(cls, feat) != getattr(
BaseLoader, feat
)
native_async = (
doc_loaders_feat_table[cm]["aload"]
or doc_loaders_feat_table[cm]["alazy_load"]
)
del doc_loaders_feat_table[cm]["aload"]
del doc_loaders_feat_table[cm]["alazy_load"]
doc_loaders_feat_table[cm]["native_async"] = native_async
doc_loaders_feat_table[cm]["description"] = (cls.__doc__ or "").split("\n")[
0
]
header = ["loader", "description", "lazy_loading", "native_async"]
title = ["Document Loader", "Description", "Lazy loading", "Native async support"]
rows = [title, [":-"] + [":-:"] * (len(title) - 1)]
for loader, feats in sorted(doc_loaders_feat_table.items()):
if not feats:
continue
rows += [
[loader, feats["description"]]
+ ["" if feats.get(h) else "" for h in header[2:]]
]
return "\n".join(["|".join(row) for row in rows])
if __name__ == "__main__":
output_dir = Path(sys.argv[1])
output_integrations_dir = output_dir / "integrations"
output_integrations_dir_doc_loaders = output_integrations_dir / "document_loaders"
output_integrations_dir_doc_loaders.mkdir(parents=True, exist_ok=True)
document_loader_page = DOCUMENT_LOADER_TEMPLATE.format(
table=get_document_loader_table()
)
with open(output_integrations_dir / "document_loaders" / "index.mdx", "w") as f:
f.write(document_loader_page)

@ -184,8 +184,8 @@ module.exports = {
},
],
link: {
type: "generated-index",
slug: "integrations/document_loaders",
type: "doc",
id: "integrations/document_loaders/index",
},
},
{

Loading…
Cancel
Save