Merge branch 'master' into pprados/02-pymupdf

Author: pprados

docs/scripts/tool_feat_table.py

@@ -86,14 +86,6 @@
         "link": "/docs/integrations/tools/riza",
         "self_hosting": True,
     },
-    "E2B Data Analysis": {
-        "langauges": "Python. In beta: JavaScript, R, Java",
-        "sandbox_lifetime": "24 Hours",
-        "upload": True,
-        "return_results": "Text, Images, Videos",
-        "link": "/docs/integrations/tools/e2b_data_analysis",
-        "self_hosting": True,
-    },
     "Azure Container Apps dynamic sessions": {
         "langauges": "Python",
         "sandbox_lifetime": "1 Hour",

libs/community/langchain_community/document_loaders/parsers/pdf.py

@@ -254,7 +254,11 @@ def _extract_text_from_page(page: pypdf.PageObject) -> str:
                 Document(
                     page_content=_extract_text_from_page(page=page)
                     + self._extract_images_from_page(page),
-                    metadata={"source": blob.source, "page": page_number},
+                    metadata={
+                        "source": blob.source,
+                        "page": page_number,
+                        "page_label": pdf_reader.page_labels[page_number],
+                    },
                     # type: ignore[attr-defined]
                 )
                 for page_number, page in enumerate(pdf_reader.pages)

libs/community/tests/unit_tests/document_loaders/sample_documents/geotopo-komprimiert.pdf

@@ -12,6 +12,10 @@
     Path(__file__).parent.parent
     / "document_loaders/sample_documents/layout-parser-paper.pdf"
 )
+path_to_multi_label_page_numbers_pdf = (
+    Path(__file__).parent.parent
+    / "document_loaders/sample_documents/geotopo-komprimiert.pdf"
+)
 path_to_layout_pdf_txt = (
     Path(__file__).parent.parent.parent
     / "integration_tests/examples/layout-parser-paper-page-1.txt"
@@ -32,6 +36,7 @@ def test_pypdf_loader() -> None:
     assert len(docs) == 16
     for page, doc in enumerate(docs):
         assert doc.metadata["page"] == page
+        assert doc.metadata["page_label"] == str(page + 1)
         assert doc.metadata["source"].endswith("layout-parser-paper.pdf")
         assert len(doc.page_content) > 10
 
@@ -49,6 +54,7 @@ def test_pypdf_loader_with_layout() -> None:
     assert len(docs) == 16
     for page, doc in enumerate(docs):
         assert doc.metadata["page"] == page
+        assert doc.metadata["page_label"] == str(page + 1)
         assert doc.metadata["source"].endswith("layout-parser-paper.pdf")
         assert len(doc.page_content) > 10
 
@@ -60,3 +66,19 @@ def test_pypdf_loader_with_layout() -> None:
     cleaned_first_page = re.sub(r"\x00", "", first_page)
     cleaned_expected = re.sub(r"\x00", "", expected)
     assert cleaned_first_page == cleaned_expected
+
+
+@pytest.mark.requires("pypdf")
+def test_pypdf_loader_with_multi_labled_page_numbers() -> None:
+    """Test PyPDFLoader with a pdf that contains multi-labled page numbers."""
+    loader = PyPDFLoader(str(path_to_multi_label_page_numbers_pdf))
+    docs = loader.load()
+
+    assert len(docs) == 7
+
+    assert docs[0].metadata["page"] == 0
+    assert docs[0].metadata["page_label"] == "i"
+
+    # Since the actual page numbers in this pdf starts from 4th page
+    assert docs[3].metadata["page"] == 3
+    assert docs[3].metadata["page_label"] == "1"

libs/community/tests/unit_tests/document_loaders/test_pdf.py

@@ -259,8 +259,8 @@ packages:
   downloads: 35495
   downloads_updated_at: '2024-12-23T20:10:11.816059+00:00'
 - name: langchain-couchbase
-  path: libs/partners/couchbase
-  repo: langchain-ai/langchain
+  path: .
+  repo: Couchbase-Ecosystem/langchain-couchbase
   downloads: 347
   downloads_updated_at: '2024-12-23T20:10:11.816059+00:00'
 - name: langchain-ollama

libs/packages.yml

@@ -1,42 +1,3 @@
-# langchain-couchbase
+This package has moved!
 
-This package contains the LangChain integration with Couchbase
-
-## Installation
-
-```bash
-pip install -U langchain-couchbase
-```
-
-## Usage
-
-The `CouchbaseVectorStore` class exposes the connection to the Couchbase vector store.
-
-```python
-from langchain_couchbase.vectorstores import CouchbaseVectorStore
-
-from couchbase.cluster import Cluster
-from couchbase.auth import PasswordAuthenticator
-from couchbase.options import ClusterOptions
-from datetime import timedelta
-
-auth = PasswordAuthenticator(username, password)
-options = ClusterOptions(auth)
-connect_string = "couchbases://localhost"
-cluster = Cluster(connect_string, options)
-
-# Wait until the cluster is ready for use.
-cluster.wait_until_ready(timedelta(seconds=5))
-
-embeddings = OpenAIEmbeddings()
-
-vectorstore = CouchbaseVectorStore(
-    cluster=cluster,
-    bucket_name="",
-    scope_name="",
-    collection_name="",
-    embedding=embeddings,
-    index_name="vector-search-index",
-)
-
-```
+https://github.com/Couchbase-Ecosystem/langchain-couchbase/tree/main/langchain_couchbase