Posts: 79
	Threads: 23
	Joined: Jan 2019
	
Reputation: 
0
	 
	
	
		Looks like OCRFeeder is using python.  I know nothing about tmpfs, private /tmp, and systemd.  Can someone help me get this program to work?
	
	
	
	
	
 
 
	
	
	
		
	Posts: 541
	Threads: 20
	Joined: May 2017
	
Reputation: 
0
	 
	
	
		How did you install OCRFeeder?  Using Lite Software usually solves "dependencies" which are missing pieces of other software that the other needs.  Forgive me if I have assumed too much.  If OCRFeeder is missing dependencies, you can open Synaptoc Package Manager (Menu -> System _> Add/Remove Software), find OCRFeeder and highlight it, then selct "Install Recommended."  Let us know!
	
	
	
	
	
 
 
	
	
	
		
	Posts: 79
	Threads: 23
	Joined: Jan 2019
	
Reputation: 
0
	 
	
	
		When I installed OCRFeeder from synaptics I did allow all dependencies to install at the same time.  It was a standard install from Synaptics.
I'm not sure why it can't write to /tmp.  I think it has something to do with systemd which I don't know how to adjust.  I hope someone with more knowledge than myself can offer some help.
	
	
	
	
	
 
 
	
	
	
		
	Posts: 79
	Threads: 23
	Joined: Jan 2019
	
Reputation: 
0
	 
	
	
		OCRFeeder is a front-end for OCR.  It calls tesseract by default to do the actual OCR.  This is when the /tmp file errors are encountered. 
It seems tesseract creates tmpfiles somewhere else than the OCRFeeder frontend or can't write to the OCRFeeder tmp directory.
When OCRFeeder, starts, it does create a directory in /tmp so that is ok.  But when tesseract runs the OCR, it doesn't write the files in the OCRFeeder tmp directory.
Any suggestions?
	
	
	
	
	
 
 
	
	
	
		
	Posts: 1,849
	Threads: 21
	Joined: Sep 2015
	
Reputation: 
0
	 
	
	
		Sorry I don't use OCR programs but perhaps some of the following links may point in a direction, if you haven't...
Quick Reads:  DPI settings can make a difference, Command being in the PATH, mics configuration tweaks..??...
https://ubuntuforums.org/showthread.php?t=2158633
https://askubuntu.com/questions/644217/o...t-anything
https://medium.freecodecamp.org/getting-...6a6b1cf75e
https://stackoverflow.com/questions/3560...g-pytesser
	
LL4.8 UEFI 64 bit ASUS E402W - AMD E2 (Quad) 1.5Ghz  - 4GB - AMD Mullins Radeon R2
LL5.8 UEFI 64 bit Test UEFI Kangaroo (Mobile Desktop) - Atom X5-Z8500 1.44Ghz - 2GB - Intel HD Graphics
LL4.8 64 bit HP 6005- AMD Phenom II X2 - 8GB - AMD/ATI RS880 (HD4200)
LL3.8 32 bit Dell Inspiron Mini - Atom N270 1.6Ghz - 1GB - Intel Mobile 945GSE Express  -- Shelved 
BACK LL5.8 64 bit Dell Optiplex 160 (Thin) - Atom 230 1.6Ghz - 4GB-SiS 771/671 PCIE VGA - Print Server
Running Linux Lite since LL2.2 
	
	
 
 
	
	
	
		
	Posts: 79
	Threads: 23
	Joined: Jan 2019
	
Reputation: 
0
	 
	
	
		OMG it was tesseract all the time.  Days of trying to get this work and the problem was the OCRFeeder synaptics package installs the tesseract ocr program as a dependency but no language packs except for English.  I was trying to OCR a German file.  And tesseract had no output to indicate a missing language file.  Crazy.
Just on a hunch I went to synaptics and searched on tesseract there are dozens of language files, but only English was installed.  I installed the German file and now it works fine.
There is a tesseract-ocr-all package but OCRFeeder doesn't install that, it just installs tesseract-ocr.
Wow, couple days wasted, but oh well at least it works now.
   :o