Search Criteria
Package Details: docspell-joex 0.25.1-1
| Git Clone URL: | https://aur.archlinux.org/docspell.git (read-only, click to copy) |
|---|---|
| Package Base: | docspell |
| Description: | Assists in organizing your piles of documents, resulting from scanners, e-mails and other sources with miminal effort. (Job executer) |
| Upstream URL: | https://docspell.org/ |
| Licenses: | |
| Groups: | |
| Submitter: | Lucki |
| Maintainer: | Lucki |
| Last Packager: | Lucki |
| Votes: | 1 |
| Popularity: | 0.039235 |
| First Submitted: | 2020-10-29 11:57 |
| Last Updated: | 2021-07-30 12:28 |
Dependencies (8)
- ghostscript
- java-runtime-headless (jdk5, jre7-openjdk-headless-infinality, java-8-openjdk-shenandoah, jre8-openjdk-headless-infinality, jdk9, jre9, jdk-devel, jre10, jdk6, jdk12-openj9-bin, zulu-12-bin, jre6, jre12, jdk8-openjdk-dcevm, server-jre, jdk13-openj9-bin, jdk9-openj9-bin, jdk10, jdk11-openjdk-dcevm, jre8-openjdk-jetbrains-headless, jre8-openjdk-shenandoah-headless, jdk10-openj9-bin, jdk14-openj9-bin, jdk7, jre7, jre14, jdk8-jetbrains, zulu-14-bin, zulu-13-bin, jdk-bcl, jre-bcl, amazon-corretto-15, jdk11-openj9-bin, jdk11-openjdk-dcevm-conflicts-fixed, zulu-embedded-jdk8, sapmachine-jdk-bin, jdk15-adoptopenjdk, jdk-arm, liberica-jdk-8-full-bin, liberica-jdk-11-bin, liberica-jdk-11-full-bin, liberica-jdk-11-lite-bin, liberica-jdk-full-bin, liberica-jre-11-bin, liberica-jre-11-full-bin, jdk-openj9-bin, amazon-corretto-8, amazon-corretto-11, jdk11-msopenjdk-bin, liberica-jre-8-full-bin, zulu-embedded-jdk11, jre8-adoptopenjdk-headless, jdk-adoptopenjdk, jdk11-adoptopenjdk, zulu-16-bin, jre8, jdk8, jre11, jdk7-j9-bin, jdk7r1-j9-bin, jdk8-arm, jdk8-openj9-bin, jdk8-j9-bin, jdk11-j9-bin, jre11-jetbrains-headless-imfix, jdk-jetbrains, java-openjdk-bin, zulu-11-bin, zulu-8-bin, zulu-15-bin, jre, amazon-corretto-16, jre-jetbrains, java-openjdk-ea-bin, jre-openj9-headless, java-openjdk-loom-ea-bin, jre-openjdk-headless, jre11-openjdk-headless, jre7-openjdk-headless, jre8-openjdk-headless)
- tesseract (tesseract-git, tesseract-ocr-git)
- unoconv
- wkhtmltopdf (wkhtmltopdf-static)
- python (python-dbg) (make)
- ocrmypdf (optional) – adds an OCR layer to scanned PDF files to make them searchable
- unpaper (optional) – pre-processes images to yield better results when doing ocr
Latest Comments
Lucki commented on 2021-03-06 00:38
https://docspell.org/docs/joex/file-processing/#natural-language-processing
The recommendation is to have 1.4G or more if you're intending to use NLP in mode "full". If you don't have that much you either have to adapt your config or - as you've done - limit javas memory access and hope for the best.
chrido commented on 2021-03-05 21:27
docspell-joex runs out of memory with the default java configuration in docspell-joex.sh when nlp.mode is set to full. I use zulu-15.
To resolve this, I added /usr/bin/java -Xmx1400M to docspell-joex.sh. 1400M is the recommendation when full is set according to the docspell configuration.