extract figures from pdf #1119
extract figures from pdf #1119
Conversation
|
LGTM 🚀 with a caveat. The analysis:
No functional safety/security issues or breaking changes that would require fixing in dependent codebase. One potential concern: If components expect Otherwise LGTM.
|
|
Enhanced PDF parsing with folder-based caching and image extraction.
Removed unused "salt" in hashing and replaced "-" with "_" in filenames
Added a note on using the scale parameter to improve image rendering.
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
| @@ -36,14 +36,21 @@ pages.slice(0, 2).forEach((page, i) => { | |||
| }) | |||
| ``` | |||
|
|
|||
| ## Rendering to images | |||
| ## Images and figures | |||
There was a problem hiding this comment.
Header changed from "Rendering to images" to "Images and figures". Consider keeping the original header for consistency.
AI-generated content by pr-docs-review-commit
header_changemay be incorrect
| const { data } = await parsers.PDF(env.files[0]) | ||
| ``` | ||
|
|
||
| ## Rendering pages to images |
There was a problem hiding this comment.
Code block is missing after the new section title.
AI-generated content by pr-docs-review-commit
missing_code_blockmay be incorrect
| const { data } = await parsers.PDF(env.files[0]) | ||
| ``` | ||
|
|
||
| ## Rendering pages to images | ||
|
|
||
| Add the `renderAsImage` option to also reach each page to a PNG image (as a buffer). This buffer can be used with a vision model to perform | ||
| an OCR operation. | ||
|
|
||
| ```js wrap |
There was a problem hiding this comment.
The code snippet is not properly formatted. Ensure consistent use of backticks or triple backticks for code blocks.
AI-generated content by pr-docs-review-commit
code_formattingmay be incorrect
| ``` | ||
|
|
||
| You can control the quality of the rendered image using the `scale` parameter (default is 3). | ||
|
|
||
| ## PDFs are messy |
There was a problem hiding this comment.
The code block for controlling image quality is missing. It should be included under the "Rendering pages to images" section.
AI-generated content by pr-docs-review-commit
missing_code_blockmay be incorrect
PDF Parsing Enhancements
Utility Import Expansion
isUint8ArrayandisUint8ClampedArrayto better handle numeric data in PDFs.Advanced Filtering Capabilities
Image Processing Support
parseFigureandparsenotated-ocr) for rendering PDF figures as images and extracting OCR annotations directly in prompt templates.