natural-pdf 0.1.19__tar.gz → 0.1.21__tar.gz

{natural_pdf-0.1.19/natural_pdf.egg-info → natural_pdf-0.1.21}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.1.19
+Version: 0.1.21
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT
@@ -34,13 +34,6 @@ Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Requires-Dist: pytest-xdist; extra == "test"
 Requires-Dist: setuptools; extra == "test"
-Provides-Extra: search
-Requires-Dist: lancedb; extra == "search"
-Requires-Dist: pyarrow; extra == "search"
-Provides-Extra: favorites
-Requires-Dist: natural-pdf[deskew]; extra == "favorites"
-Requires-Dist: natural-pdf[ocr-export]; extra == "favorites"
-Requires-Dist: natural-pdf[paddle]; extra == "favorites"
 Provides-Extra: dev
 Requires-Dist: black; extra == "dev"
 Requires-Dist: isort; extra == "dev"
@@ -58,25 +51,19 @@ Requires-Dist: nbclient; extra == "dev"
 Requires-Dist: ipykernel; extra == "dev"
 Requires-Dist: pre-commit; extra == "dev"
 Requires-Dist: setuptools; extra == "dev"
-Provides-Extra: deskew
-Requires-Dist: deskew>=1.5; extra == "deskew"
-Requires-Dist: img2pdf; extra == "deskew"
 Provides-Extra: all
 Requires-Dist: natural-pdf[ocr-export]; extra == "all"
 Requires-Dist: natural-pdf[deskew]; extra == "all"
 Requires-Dist: natural-pdf[test]; extra == "all"
 Requires-Dist: natural-pdf[search]; extra == "all"
-Requires-Dist: natural-pdf[extras]; extra == "all"
 Requires-Dist: natural-pdf[favorites]; extra == "all"
-Provides-Extra: paddle
-Requires-Dist: paddlepaddle>=3.0.0; extra == "paddle"
-Requires-Dist: paddleocr>=3.0.1; extra == "paddle"
-Requires-Dist: paddlex>=3.0.1; extra == "paddle"
-Provides-Extra: extras
-Requires-Dist: surya-ocr>=0.13.0; extra == "extras"
-Requires-Dist: doclayout_yolo; extra == "extras"
-Requires-Dist: easyocr; extra == "extras"
-Requires-Dist: natural-pdf[paddle]; extra == "extras"
+Requires-Dist: natural-pdf[export-extras]; extra == "all"
+Provides-Extra: deskew
+Requires-Dist: deskew>=1.5; extra == "deskew"
+Requires-Dist: img2pdf; extra == "deskew"
+Provides-Extra: search
+Requires-Dist: lancedb; extra == "search"
+Requires-Dist: pyarrow; extra == "search"
 Provides-Extra: ocr-export
 Requires-Dist: pikepdf; extra == "ocr-export"
 Provides-Extra: export-extras
@@ -101,14 +88,28 @@ Natural PDF lets you find and extract content from PDFs using simple code that m
 pip install natural-pdf
 ```
 
-For optional features like specific OCR engines, layout analysis models, or the interactive Jupyter widget, you can install one to two million different extras. If you just want the greatest hits:
+Need OCR engines, layout models, or other heavy add-ons? Install the **core** once, then use the helper CLI to pull in exactly what you need:
+
+```bash
+# add PaddleOCR (+paddlex) after the fact
+npdf install paddle
+
+# Surya OCR and the YOLO Doc-Layout detector in one go
+npdf install surya yolo
+
+# see what's already on your machine
+npdf list
+```
+
+Light-weight extras such as `deskew` or `search` can still be added with
+classic PEP-508 markers if you prefer:
 
 ```bash
-# deskewing, OCR (surya) + layout analysis (yolo), interactive browsing
-pip install natural-pdf[favorites]
+pip install "natural-pdf[deskew]"
+pip install "natural-pdf[search]"
 ```
 
-See the [installation guide](https://jsoma.github.io/natural-pdf/installation/) for more details on extras.
+More details in the [installation guide](https://jsoma.github.io/natural-pdf/installation/).
 
 ## Quick Start
 

{natural_pdf-0.1.19 → natural_pdf-0.1.21}/README.md

@@ -15,14 +15,28 @@ Natural PDF lets you find and extract content from PDFs using simple code that m
 pip install natural-pdf
 ```
 
-For optional features like specific OCR engines, layout analysis models, or the interactive Jupyter widget, you can install one to two million different extras. If you just want the greatest hits:
+Need OCR engines, layout models, or other heavy add-ons? Install the **core** once, then use the helper CLI to pull in exactly what you need:
 
 ```bash
-# deskewing, OCR (surya) + layout analysis (yolo), interactive browsing
-pip install natural-pdf[favorites]
+# add PaddleOCR (+paddlex) after the fact
+npdf install paddle
+
+# Surya OCR and the YOLO Doc-Layout detector in one go
+npdf install surya yolo
+
+# see what's already on your machine
+npdf list
+```
+
+Light-weight extras such as `deskew` or `search` can still be added with
+classic PEP-508 markers if you prefer:
+
+```bash
+pip install "natural-pdf[deskew]"
+pip install "natural-pdf[search]"
 ```
 
-See the [installation guide](https://jsoma.github.io/natural-pdf/installation/) for more details on extras.
+More details in the [installation guide](https://jsoma.github.io/natural-pdf/installation/).
 
 ## Quick Start
 

{natural_pdf-0.1.19 → natural_pdf-0.1.21}/docs/installation/index.md

@@ -30,6 +30,38 @@ pip install natural-pdf[favorites]
 
 Other OCR and layout analysis engines like `surya`, `easyocr`, `paddle`, `doctr`, and `docling` can be installed via `pip` as needed. The library will provide you with an error message and installation command if you try to use an engine that isn't installed.
 
+After the core install you have two ways to add **optional engines**:
+
+### 1 · Helper CLI (recommended)
+
+```bash
+# list optional groups and their install-status
+npdf list
+
+# install PaddleOCR stack
+npdf install paddle
+
+# install Surya OCR + YOLO layout detector
+npdf install surya yolo
+```
+
+The CLI runs each wheel in its own resolver pass, so it avoids strict
+version pins like `paddleocr → paddlex==3.0.1` while still upgrading to
+`paddlex 3.0.2`.
+
+### 2 · Classic extras (for the light stuff)
+
+```bash
+# Deskewing
+pip install "natural-pdf[deskew]"
+
+# Semantic search service
+pip install "natural-pdf[search]"
+```
+
+If you attempt to use an engine that is missing, the library will raise an
+error that tells you which `npdf install …` command to run.
+
 ## Your First PDF Extraction
 
 Here's a quick example to make sure everything is working: