75 lines
2.7 KiB
Markdown
75 lines
2.7 KiB
Markdown
|
Looks like this when fetched:
|
||
|
![[Pasted image 20221212171637.png]]
|
||
|
# Obsidian Extract PDF Annotations Plugin
|
||
|
|
||
|
This is a plugin for [Obsidian](https://obsidian.md). It extracts Annotations from PDF Files.
|
||
|
|
||
|
## Usage
|
||
|
|
||
|
This Plugin visits all PDF files in a given directory and extracts comments and highlights from the PDF files. It treats the first line of every comment as *Topic* for grouping the comments.
|
||
|
|
||
|
Assume we have in a folder in our Vault containing PDF files, e.g:
|
||
|
|
||
|

|
||
|
|
||
|
and we have highlighted the Julia Hello World Programm with a note 'Hello World':
|
||
|
|
||
|

|
||
|
|
||
|
In the editor (e.g. \_Extract) we run the plugin's command `Extract PDF Annotations` (Hotkey Ctrl-P for all Commands). This will fetch all annotations in the PDF files in the current folder and sort them by *Topic*:
|
||
|
|
||
|

|
||
|
|
||
|
As such, you can relate comments for your topics (here 'Hello World') from several PDF files.
|
||
|
|
||
|
### Commands
|
||
|
* `Extract PDF Annotations` Works when editing a markdown note. Searches all PDF files in current Folder for annotations, and inserts them at the current position of the open note.
|
||
|
* `Extract PDF Annotations on single file` Works while displaying a PDF file. Extracts annotations from this file and writes them to the note `Annotations for <filename>`
|
||
|
|
||
|
### Plugin Settings:
|
||
|
|
||
|
* Use the first line of the comment as 'Topic' (and sort accordingly), or not
|
||
|
* Use folder name or PDF-Filename for sorting
|
||
|
|
||
|
## Versions
|
||
|
|
||
|
1.0.4 clean up hyphenation https://github.com/munach/obsidian-extract-pdf-annotations/issues/5
|
||
|
1.0.3 updated highlight fetching to use QuadPoints instead of Rectangles
|
||
|
|
||
|
## Installation / Build
|
||
|
|
||
|
Fetch repository:
|
||
|
```bash
|
||
|
$ git clone https://github.com/munach/obsidian-extract-pdf-annotations.git
|
||
|
$ cd obsidian-extract-pdf-annotations
|
||
|
```
|
||
|
Install dependencies:
|
||
|
```
|
||
|
$ npm i
|
||
|
```
|
||
|
|
||
|
Transpile `main.ts`:
|
||
|
```
|
||
|
$ npm run build
|
||
|
```
|
||
|
|
||
|
Then create the plugin directory and copy the files `main.js` and `manifest.json`, e.g.;
|
||
|
```
|
||
|
$ mkdir ~/MyVault/.obsidian/plugins/obsidian-extract-pdf-annotations
|
||
|
$ cp main.js manifest.json ~/MyVault/.obsidian/plugins/obsidian-extract-pdf-annotations/
|
||
|
```
|
||
|
|
||
|
Enable the plugin in Obsidan's setting.
|
||
|
|
||
|
## Issues / Bugs
|
||
|
|
||
|
[] works only on left-to-right highlights
|
||
|
|
||
|
## Credits
|
||
|
|
||
|
This plugin builds on ideas from Alexis Rondeaus Plugin https://github.com/akaalias/obsidian-extract-pdf-highlights, but uses obsidians build-in pdf.js library.
|
||
|
|
||
|
## Author
|
||
|
|
||
|
Franz Achermann
|