Subler still crashes when saving VobSub tracks

Issue #81 closed
Former user created an issue

Greetings,

Subler 0.18 was able to export a given MKV file containing an English VobSub file and OCR it into a file. That was the last version before subsequent ones crashed hard doing so.

I have kept 0.18 on my system to run for exactly this purpose, but as of 10.10.3, it will no longer run properly for whatever reason. (It runs, but doesn't create a new window or open a file.) Perhaps it stopped running after the 10.10 upgrade.

So, now I come to you, the dear developer, and plead to you: Please fix this bug :-) I no longer have a faithful way of OCR'ing VobSub tracks automagically without resorting to either Windows or complex command-line solutions.

I can provide any supporting data you need. The specific use case is:

  1. ripping a DVD to an MKV file
  2. creating a new Subler MP4 container with tracks from the MKV
  3. Removing all but the desired VobSub track
  4. Saving to a file (.m4v in this case)

Step 4 causes a crash immediately.

I'll attach the output from the crash reporter.

Comments (4)

  1. Jubal Kessler

    And here's the eng.traineddata file I have been using:

    % ls -ld /Users/user/Library/Application\ Support/Subler/tessdata/eng.traineddata

    -rw-r--r--@ 1 user staff 1926792 Dec 3 2010 /Users/jubal/Library/Application Support/Subler/tessdata/eng.traineddata

  2. Jubal Kessler

    Looks like an updated eng.traineddata file from Oct 17, 2012 downloaded from here works, and solves the problem:

    https://code.google.com/p/tesseract-ocr/downloads/detail?name=tesseract-ocr-3.02.eng.tar.gz

    -rw-r--r--@ 1 user staff 21876572 Oct 17 2012 eng.traineddata

    Two requests:

    1. Please "trap" this type of crash, if you can, and return something useful to the end user, e.g. "This seems to be a bad tesseract data file"

    2. Please document the OCR feature with more detail, if you wouldn't mind.

  3. Log in to comment