mistocr 0.2.1__tar.gz → 0.2.2__tar.gz

{mistocr-0.2.1/mistocr.egg-info → mistocr-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mistocr
-Version: 0.2.1
+Version: 0.2.2
 Summary: Batch OCR for PDFs with heading restoration and visual content integration
 Home-page: https://github.com/franckalbinet/mistocr
 Author: Solveit
@@ -38,7 +38,7 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# mistocr
+# Mistocr
 
 
 <!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
@@ -69,12 +69,12 @@ fundamental challenges that raw OCR output leaves unsolved:
   markdown. This makes visual information searchable and accessible for
   downstream applications.
 
-- **Cost-efficient batch processing**: By exclusively using Mistral’s
-  batch API, mistocr cuts costs by 50% (\$0.50 vs \$1.00 per 1000 pages)
-  while eliminating the boilerplate code typically required.
+- **Cost-efficient batch processing**: The OCR step exclusively uses
+  Mistral’s batch API, cutting costs by 50% (\$0.50 vs \$1.00 per 1000
+  pages) while eliminating the boilerplate code typically required.
 
-**In short**: Production-ready batch OCR with intelligent postprocessing
-that ensures your documents are actually usable for AI systems.
+**In short**: Complete PDF OCR with heading hierarchy fixes and image
+descriptions for RAG and LLM pipelines.
 
 ## Get Started
 
@@ -94,7 +94,10 @@ os.environ['ANTHROPIC_API_KEY'] = 'your-key-here'  # for refine features (see Ad
 
 ### Complete Pipeline
 
-Full pipeline with all features:
+#### Single File Processing
+
+Process a single PDF with OCR (using Mistral’s batch API for cost
+efficiency), heading fixes, and image descriptions:
 
 ``` python
 from mistocr.pipeline import pdf_to_md
@@ -171,12 +174,12 @@ instead.
 
 ### Advanced Usage
 
-**Batch process entire folders:**
+**Batch OCR for entire folders:**
 
 ``` python
 from mistocr.core import ocr_pdf
 
-# Process all PDFs in a folder
+# OCR all PDFs in a folder using Mistral's batch API
 output_dirs = ocr_pdf('path/to/pdf_folder', dst='output_folder')
 ```
 

{mistocr-0.2.1 → mistocr-0.2.2}/README.md

@@ -1,4 +1,4 @@
-# mistocr
+# Mistocr
 
 
 <!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
@@ -29,12 +29,12 @@ fundamental challenges that raw OCR output leaves unsolved:
   markdown. This makes visual information searchable and accessible for
   downstream applications.
 
-- **Cost-efficient batch processing**: By exclusively using Mistral’s
-  batch API, mistocr cuts costs by 50% (\$0.50 vs \$1.00 per 1000 pages)
-  while eliminating the boilerplate code typically required.
+- **Cost-efficient batch processing**: The OCR step exclusively uses
+  Mistral’s batch API, cutting costs by 50% (\$0.50 vs \$1.00 per 1000
+  pages) while eliminating the boilerplate code typically required.
 
-**In short**: Production-ready batch OCR with intelligent postprocessing
-that ensures your documents are actually usable for AI systems.
+**In short**: Complete PDF OCR with heading hierarchy fixes and image
+descriptions for RAG and LLM pipelines.
 
 ## Get Started
 
@@ -54,7 +54,10 @@ os.environ['ANTHROPIC_API_KEY'] = 'your-key-here'  # for refine features (see Ad
 
 ### Complete Pipeline
 
-Full pipeline with all features:
+#### Single File Processing
+
+Process a single PDF with OCR (using Mistral’s batch API for cost
+efficiency), heading fixes, and image descriptions:
 
 ``` python
 from mistocr.pipeline import pdf_to_md
@@ -131,12 +134,12 @@ instead.
 
 ### Advanced Usage
 
-**Batch process entire folders:**
+**Batch OCR for entire folders:**
 
 ``` python
 from mistocr.core import ocr_pdf
 
-# Process all PDFs in a folder
+# OCR all PDFs in a folder using Mistral's batch API
 output_dirs = ocr_pdf('path/to/pdf_folder', dst='output_folder')
 ```
 

mistocr-0.2.2/mistocr/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.2.2"

{mistocr-0.2.1 → mistocr-0.2.2}/mistocr/refine.py

@@ -197,8 +197,8 @@ async def describe_imgs(
     imgs: list[Path], # List of image file paths to describe
     model: str = 'claude-sonnet-4-5', # Model to use for image description
     prompt: str = describe_img_prompt, # Prompt template for description
-    semaphore: int = 2, # Max concurrent API requests
-    delay: float = 1 # Delay in seconds between requests
+    semaphore: int = 10, # Max concurrent API requests
+    delay: float = 0.1 # Delay in seconds between requests
 ) -> dict[str, dict]: # Dict mapping filename to parsed description
     "Describe multiple images in parallel with rate limiting"
     sem = Semaphore(semaphore)

{mistocr-0.2.1 → mistocr-0.2.2/mistocr.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mistocr
-Version: 0.2.1
+Version: 0.2.2
 Summary: Batch OCR for PDFs with heading restoration and visual content integration
 Home-page: https://github.com/franckalbinet/mistocr
 Author: Solveit
@@ -38,7 +38,7 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# mistocr
+# Mistocr
 
 
 <!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
@@ -69,12 +69,12 @@ fundamental challenges that raw OCR output leaves unsolved:
   markdown. This makes visual information searchable and accessible for
   downstream applications.
 
-- **Cost-efficient batch processing**: By exclusively using Mistral’s
-  batch API, mistocr cuts costs by 50% (\$0.50 vs \$1.00 per 1000 pages)
-  while eliminating the boilerplate code typically required.
+- **Cost-efficient batch processing**: The OCR step exclusively uses
+  Mistral’s batch API, cutting costs by 50% (\$0.50 vs \$1.00 per 1000
+  pages) while eliminating the boilerplate code typically required.
 
-**In short**: Production-ready batch OCR with intelligent postprocessing
-that ensures your documents are actually usable for AI systems.
+**In short**: Complete PDF OCR with heading hierarchy fixes and image
+descriptions for RAG and LLM pipelines.
 
 ## Get Started
 
@@ -94,7 +94,10 @@ os.environ['ANTHROPIC_API_KEY'] = 'your-key-here'  # for refine features (see Ad
 
 ### Complete Pipeline
 
-Full pipeline with all features:
+#### Single File Processing
+
+Process a single PDF with OCR (using Mistral’s batch API for cost
+efficiency), heading fixes, and image descriptions:
 
 ``` python
 from mistocr.pipeline import pdf_to_md
@@ -171,12 +174,12 @@ instead.
 
 ### Advanced Usage
 
-**Batch process entire folders:**
+**Batch OCR for entire folders:**
 
 ``` python
 from mistocr.core import ocr_pdf
 
-# Process all PDFs in a folder
+# OCR all PDFs in a folder using Mistral's batch API
 output_dirs = ocr_pdf('path/to/pdf_folder', dst='output_folder')
 ```
 

{mistocr-0.2.1 → mistocr-0.2.2}/settings.ini

@@ -1,7 +1,7 @@
 [DEFAULT]
 repo = mistocr
 lib_name = mistocr
-version = 0.2.1
+version = 0.2.2
 min_python = 3.9
 license = apache2
 black_formatting = False

mistocr-0.2.1/mistocr/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.2.1"