https://github.com/ansemjo/ocrallmypdf

container image to run ocr processing on all files in inotify-watched directory
https://github.com/ansemjo/ocrallmypdf

Last synced: 6 months ago
JSON representation

container image to run ocr processing on all files in inotify-watched directory

• Host: GitHub
• URL: https://github.com/ansemjo/ocrallmypdf
• Owner: ansemjo
• License: mit
• Created: 2019-08-05T09:31:16.000Z (almost 7 years ago)
• Default Branch: main
• Last Pushed: 2024-03-25T04:38:30.000Z (over 2 years ago)
• Last Synced: 2024-03-25T05:33:07.889Z (over 2 years ago)
• Language: Shell
• Size: 43.9 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ocrallmypdf

This is a minimal wrapper around the `jbarlow83/ocrmypdf` container which

adds an inotify-based directory watcher and automatically runs OCR processing

on all incoming files in the input directory.

Think of this as a simplified version of [cmccambridge/ocrmypdf-auto](https://github.com/cmccambridge/ocrmypdf-auto),

where `ocrallmypdf` is using the existing tools `incron` and `task-spooler`

to achieve a similar effect.

## USAGE

Use your container runtime of choice and run:

    docker run -d \

      -v /path/to/input:/input \

      -v /path/to/output:/output \

      ghcr.io/ansemjo/ocrallmypdf

Existing files in the input directory are not processed. The processing only

triggers on `CLOSE_WRITE` and `MOVED_TO` events.

The container can be customized with a number of environment variables:

| env | description | default |

| --- | ----------- | ------- |

| `INPUT` | watched input directory inside of the container | `/input` |

| `OUTPUT` | output directory for processed file in the container, *be careful not to create an inotify loop by using the same directories!* | `/output` |

| `OCR_OPTIONS` | options passed to `ocrmypdf` command | `--clean --deskew --output-type pdfa --skip-text` |

| `OCR_LANGUAGE` | language used with tesseract (`-l` flag of `ocrmypdf`), missing languages will be installed on startup | `deu+eng` |

| `REMOVE_ORIGINAL` | truthy value whether to remove original input files | `yes` |

| `JOBS` | number of concurrent jobs in task spooler, *using too many at once may lead to ocr timeouts, since `ocrmypdf` uses all cores per process by default* | `1` |