stache-ai-ocr 0.1.0__tar.gz

stache_ai_ocr-0.1.0/PKG-INFO

@@ -0,0 +1,54 @@
+Metadata-Version: 2.4
+Name: stache-ai-ocr
+Version: 0.1.0
+Summary: OCR support for Stache AI document loaders
+Author: Stache Contributors
+License: MIT
+Project-URL: Homepage, https://github.com/stache-ai/stache-ai
+Project-URL: Repository, https://github.com/stache-ai/stache-ai
+Keywords: stache,ocr,pdf,document-processing
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: stache-ai>=0.1.0
+Requires-Dist: pdfplumber>=0.10.0
+
+# stache-ai-ocr
+
+OCR support for Stache AI document loaders.
+
+Provides a high-priority PDF loader that falls back to OCR for scanned documents.
+
+## Installation
+
+```bash
+pip install stache-ai-ocr
+apt install ocrmypdf  # System dependency required
+```
+
+## Usage
+
+Once installed, the OCR loader automatically registers and takes priority over the basic PDF loader for all PDF files.
+
+The loader will:
+1. First attempt normal text extraction with pdfplumber
+2. If no text is found (scanned PDF), fall back to OCR using ocrmypdf
+3. Gracefully handle missing ocrmypdf (logs warning and returns empty text)
+
+## System Requirements
+
+- **ocrmypdf** system binary must be installed
+  - Ubuntu/Debian: `apt install ocrmypdf`
+  - macOS: `brew install ocrmypdf`
+  - Includes Tesseract OCR engine
+
+## Priority Override
+
+This loader registers with priority 10, overriding the basic PDF loader (priority 0). This ensures OCR is used when available without affecting systems where it's not installed.

stache_ai_ocr-0.1.0/README.md

@@ -0,0 +1,32 @@
+# stache-ai-ocr
+
+OCR support for Stache AI document loaders.
+
+Provides a high-priority PDF loader that falls back to OCR for scanned documents.
+
+## Installation
+
+```bash
+pip install stache-ai-ocr
+apt install ocrmypdf  # System dependency required
+```
+
+## Usage
+
+Once installed, the OCR loader automatically registers and takes priority over the basic PDF loader for all PDF files.
+
+The loader will:
+1. First attempt normal text extraction with pdfplumber
+2. If no text is found (scanned PDF), fall back to OCR using ocrmypdf
+3. Gracefully handle missing ocrmypdf (logs warning and returns empty text)
+
+## System Requirements
+
+- **ocrmypdf** system binary must be installed
+  - Ubuntu/Debian: `apt install ocrmypdf`
+  - macOS: `brew install ocrmypdf`
+  - Includes Tesseract OCR engine
+
+## Priority Override
+
+This loader registers with priority 10, overriding the basic PDF loader (priority 0). This ensures OCR is used when available without affecting systems where it's not installed.

stache_ai_ocr-0.1.0/pyproject.toml

@@ -0,0 +1,40 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "stache-ai-ocr"
+version = "0.1.0"
+readme = "README.md"
+description = "OCR support for Stache AI document loaders"
+license = {text = "MIT"}
+requires-python = ">=3.10"
+authors = [
+    {name = "Stache Contributors"}
+]
+keywords = ["stache", "ocr", "pdf", "document-processing"]
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: MIT License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+]
+dependencies = [
+    "stache-ai>=0.1.0",
+    "pdfplumber>=0.10.0",
+]
+
+[project.entry-points."stache.loader"]
+pdf-ocr = "stache_ai_ocr.loaders:OcrPdfLoader"
+
+[project.urls]
+Homepage = "https://github.com/stache-ai/stache-ai"
+Repository = "https://github.com/stache-ai/stache-ai"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+include = ["stache_ai_ocr*"]

stache_ai_ocr-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

stache_ai_ocr-0.1.0/src/stache_ai_ocr/__init__.py

@@ -0,0 +1,3 @@
+"""OCR support for Stache AI document loaders"""
+
+__version__ = "0.1.0"

stache_ai_ocr-0.1.0/src/stache_ai_ocr/loaders.py

@@ -0,0 +1,84 @@
+"""PDF loader with OCR fallback for scanned documents"""
+
+import logging
+import subprocess
+import tempfile
+from pathlib import Path
+
+from stache_ai.loaders.base import DocumentLoader
+
+logger = logging.getLogger(__name__)
+
+
+class OcrPdfLoader(DocumentLoader):
+    """PDF loader with OCR fallback for scanned documents
+
+    Requires ocrmypdf system binary: apt install ocrmypdf
+    """
+
+    @property
+    def extensions(self) -> list[str]:
+        return ['pdf']
+
+    @property
+    def priority(self) -> int:
+        return 10  # Override basic PdfLoader (priority 0)
+
+    def load(self, file_path: str) -> str:
+        # Lazy import
+        import pdfplumber
+
+        # Try normal extraction first
+        text_parts = []
+        with pdfplumber.open(file_path) as pdf:
+            for page in pdf.pages:
+                text = page.extract_text()
+                if text:
+                    text_parts.append(text)
+
+        # If no text found, try OCR
+        if not text_parts:
+            logger.info(f"No text in PDF, attempting OCR: {file_path}")
+            text_parts = self._ocr_extract(file_path)
+
+        return "\n\n".join(text_parts)
+
+    def _ocr_extract(self, file_path: str) -> list[str]:
+        """Run OCR on PDF and extract text"""
+        import pdfplumber
+
+        text_parts = []
+        tmp_path = None
+
+        try:
+            with tempfile.NamedTemporaryFile(suffix='.pdf', delete=False) as tmp:
+                tmp_path = tmp.name
+
+            result = subprocess.run(
+                ['ocrmypdf', '--skip-text', '--quiet', file_path, tmp_path],
+                capture_output=True,
+                text=True
+            )
+
+            if result.returncode != 0:
+                logger.warning(f"OCR failed: {result.stderr}")
+                return []
+
+            with pdfplumber.open(tmp_path) as pdf:
+                for page in pdf.pages:
+                    text = page.extract_text()
+                    if text:
+                        text_parts.append(text)
+
+        except FileNotFoundError:
+            logger.warning(
+                "ocrmypdf not installed (install with: apt install ocrmypdf), "
+                "skipping OCR"
+            )
+        except Exception as e:
+            logger.warning(f"OCR error: {e}")
+        finally:
+            if tmp_path:
+                Path(tmp_path).unlink(missing_ok=True)
+
+        return text_parts

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/PKG-INFO

@@ -0,0 +1,54 @@
+Metadata-Version: 2.4
+Name: stache-ai-ocr
+Version: 0.1.0
+Summary: OCR support for Stache AI document loaders
+Author: Stache Contributors
+License: MIT
+Project-URL: Homepage, https://github.com/stache-ai/stache-ai
+Project-URL: Repository, https://github.com/stache-ai/stache-ai
+Keywords: stache,ocr,pdf,document-processing
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: stache-ai>=0.1.0
+Requires-Dist: pdfplumber>=0.10.0
+
+# stache-ai-ocr
+
+OCR support for Stache AI document loaders.
+
+Provides a high-priority PDF loader that falls back to OCR for scanned documents.
+
+## Installation
+
+```bash
+pip install stache-ai-ocr
+apt install ocrmypdf  # System dependency required
+```
+
+## Usage
+
+Once installed, the OCR loader automatically registers and takes priority over the basic PDF loader for all PDF files.
+
+The loader will:
+1. First attempt normal text extraction with pdfplumber
+2. If no text is found (scanned PDF), fall back to OCR using ocrmypdf
+3. Gracefully handle missing ocrmypdf (logs warning and returns empty text)
+
+## System Requirements
+
+- **ocrmypdf** system binary must be installed
+  - Ubuntu/Debian: `apt install ocrmypdf`
+  - macOS: `brew install ocrmypdf`
+  - Includes Tesseract OCR engine
+
+## Priority Override
+
+This loader registers with priority 10, overriding the basic PDF loader (priority 0). This ensures OCR is used when available without affecting systems where it's not installed.

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/SOURCES.txt

@@ -0,0 +1,10 @@
+README.md
+pyproject.toml
+src/stache_ai_ocr/__init__.py
+src/stache_ai_ocr/loaders.py
+src/stache_ai_ocr.egg-info/PKG-INFO
+src/stache_ai_ocr.egg-info/SOURCES.txt
+src/stache_ai_ocr.egg-info/dependency_links.txt
+src/stache_ai_ocr.egg-info/entry_points.txt
+src/stache_ai_ocr.egg-info/requires.txt
+src/stache_ai_ocr.egg-info/top_level.txt

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[stache.loader]
+pdf-ocr = stache_ai_ocr.loaders:OcrPdfLoader

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/requires.txt

@@ -0,0 +1,2 @@
+stache-ai>=0.1.0
+pdfplumber>=0.10.0

stache_ai_ocr-0.1.0/src/stache_ai_ocr.egg-info/top_level.txt

@@ -0,0 +1 @@
+stache_ai_ocr