dirsql-plugin-embeddings 0.1.2__tar.gz → 0.1.4__tar.gz

{dirsql_plugin_embeddings-0.1.2 → dirsql_plugin_embeddings-0.1.4}/PKG-INFO

@@ -1,8 +1,9 @@
 Metadata-Version: 2.4
 Name: dirsql-plugin-embeddings
-Version: 0.1.2
+Version: 0.1.4
 Summary: First-party dirsql plugin: semantic search via an OpenAI-compatible embeddings endpoint.
 Requires-Python: >=3.11
+Requires-Dist: pypdf>=6
 Description-Content-Type: text/markdown
 
 # dirsql-plugin-embeddings

dirsql_plugin_embeddings-0.1.2/e2e-attestations/claude-on-file-integration-e2e-lpeicb.json → dirsql_plugin_embeddings-0.1.4/e2e-attestations/claude-699-read-pdf.json

@@ -1,7 +1,7 @@
 {
   "command": "uv run --with sqlite-vec --with-editable . --with-editable ../../packages/python python -m pytest tests/e2e -x -q",
-  "ran_at": 1785323892,
+  "ran_at": 1785327111,
   "exit_code": 0,
-  "commit": "35c1c6929c4a8b517032cc51e0cc6720f43b58f4",
-  "branch": "claude/on-file-integration-e2e-lpeicb"
+  "commit": "b59ad4bf5dce929e2172cde15888918de9d277fd",
+  "branch": "claude/699-read-pdf"
 }

{dirsql_plugin_embeddings-0.1.2 → dirsql_plugin_embeddings-0.1.4}/pyproject.toml

@@ -12,6 +12,9 @@ dynamic = ["version"]
 description = "First-party dirsql plugin: semantic search via an OpenAI-compatible embeddings endpoint."
 readme = "README.md"
 requires-python = ">=3.11"
+# pypdf, not pymupdf: pymupdf is AGPL-3.0, and a runtime dependency of this
+# MIT-licensed package propagates to everyone who installs it.
+dependencies = ["pypdf>=6"]
 
 [tool.hatch.version]
 source = "vcs"

{dirsql_plugin_embeddings-0.1.2 → dirsql_plugin_embeddings-0.1.4}/src/dirsql_plugin_embeddings/on_file/on_file.py

@@ -13,13 +13,13 @@ import sys
 
 from ..embedder import embed
 from .build_rows import build_rows
-from .read_text import read_text
+from .read_content import read_content
 
 
 def on_file(argv: list[str] | None = None) -> int:
     if argv is None:
         argv = sys.argv
     path = argv[1]
-    text = read_text(path)
+    text = read_content(path)
     print(json.dumps(build_rows(path, text, embed(text))))
     return 0

dirsql_plugin_embeddings-0.1.4/src/dirsql_plugin_embeddings/on_file/read_content.py

@@ -0,0 +1,13 @@
+"""The entry point's single read seam: everything the hook reads comes through
+here, so per-format handling has one place to land without touching `on_file`.
+
+Annotations are evaluated at runtime (no ``from __future__ import
+annotations``) so a mutated ``X | None`` union in a signature fails at import
+rather than surviving as an inert string.
+"""
+
+from .read_text import read_text
+
+
+def read_content(path: str) -> str:
+    return read_text(path)

dirsql_plugin_embeddings-0.1.4/src/dirsql_plugin_embeddings/on_file/read_pdf.py

@@ -0,0 +1,16 @@
+"""Extract a PDF's text.
+
+pypdf rather than pymupdf: pymupdf is AGPL-3.0, and a runtime dependency of
+this MIT-licensed package propagates to everyone who installs it.
+
+Annotations are evaluated at runtime (no ``from __future__ import
+annotations``) so a mutated ``X | None`` union in a signature fails at import
+rather than surviving as an inert string.
+"""
+
+from pypdf import PdfReader
+
+
+def read_pdf(path: str) -> str:
+    reader = PdfReader(path)
+    return "\n".join(page.extract_text() for page in reader.pages)