comiclrf - Comics(CBZ) to LRF converter

[FangornUK]

what about:
comiclrf.pl c:\"test.cbr"

[phurren]

Thanks, just tried it, but came up with the same error.

[Ryan2065]

I am getting the same error as phurren. I downloaded a pack and the first 3 .cbr files went through comiclrf just fine but now it doesn't seem to work on any of the other .cbr files in the pack. I've used it to convert other comics with no problems and it is currently converting some other .cbr files I have... I'm not really sure what is going on with it. I've changed the file name of the cbr file to take out the spaces, deleted the comiclrf files and re-copied them, moved the cbr file from the d drive to c and even to the directory in c that holds comiclrf but nothing seems to work with this file. I've also tried using the *.cbr command and test.cbr but both still give me the error.

I don't have a cbr viewer but 7zip has no problems opening it and all the pictures display fine when it is unzipped.

You have made a great program though! Great job on it.


C:\utils>comiclrf.pl "c:\test.cbr"
Converting Comic "test"...
Extracting...
RAR failed! at C:\utils\comiclrf.pl line 138.

[Ryan2065]

After finding a few more comics with this same error I'm starting to think the problem is with the comic file and not the program. I'm seeing this error alot with comics that won't open in a comic viewer or comics that won't load one or two pages in the comic viewer. There still are one or two comics that open without error in the comic viewer program but get this error but I'd still say they are the files fault and not the program. I just extract the image files and re-make the comic and then all is good.

[FangornUK]

Perhaps the comic files have subdirectories?

[mimoviz]

Dear Farngorn: I seem unable to make your wonderful script to work. This is the output and it always happens whenever it reaches the fourth file of any cbr.
Output:
dbe495437:~ mimoviz$ perl comiclrf.pl Akira01.cbr tes.lrf
Converting Comic "Akira01"...
Extracting...
Processing 1 of 182 "Akira - I - 001.jpg"
Processing 2 of 182 "Akira - I - 002.jpg"
Processing 3 of 182 "Akira - I - 003.jpg"
Processing 4 of 182 "Akira - I - 004.jpg"
convert: geometry does not contain image `/tmp/comiclrf/Akira - I - 004.jpg'.
ImageMagick failed, path set wrong? at comiclrf.pl line 218.

The path is Ok. In fact I have tried the convert command by hand using the same options as per comiclrf in line 218 and it works flawlessly. So am still confused.
Could you help me?

Thanks

BTW: I run an OSX Tiger machine. I downloaded and installed imagemagick myself in /usr/local/bin and it is accesible from any terminal.

[FangornUK]

Not sure what's wrong. You're calling comiclrf.pl incorrectly but I don't see how that would cause it to fail. Use: 'perl comiclrf.pl Akira01.cbr'

[mimoviz]

Originally Posted by FangornUK

Not sure what's wrong. You're calling comiclrf.pl incorrectly but I don't see how that would cause it to fail. Use: 'perl comiclrf.pl Akira01.cbr'

Same thing happens.... I dont know why. If there was something wrong in the path then the individual commands wouldn't work right?
See... this is what happens:

dbe495437:~ mimoviz$ perl comiclrf.pl Akira01.cbr
Converting Comic "Akira01"...
Extracting...
Processing 1 of 182 "Akira - I - 001.jpg"
Processing 2 of 182 "Akira - I - 002.jpg"
Processing 3 of 182 "Akira - I - 003.jpg"
Processing 4 of 182 "Akira - I - 004.jpg"
convert: geometry does not contain image `/tmp/comiclrf/Akira - I - 004.jpg'.
ImageMagick failed, path set wrong? at comiclrf.pl line 218.

I have checked both convert and identify binaries...they work well... perhaps there are some library dependencies that are not working? I wonder how can I tell?
Do you happen to know how to uninstall perl completely from terminal in a mac? It is not the Activestate version, just a plain simple port for darwin.


Thanks

[mimoviz]

Ok. I installed ActivePerl and its fully functional. But the problem remains... My current guess it must have something to do with imagemagick, but I am still at loss.
perhaps the name of the files with spaces between the numbers?
perhaps something else?

[mimoviz]

I did it! I just had to update a library within /sw/lib and everything was smooth. Thanks... The problem was in Imagemagick after all...
Thanks man
ç

[mimoviz]

Ahh... come on... this is just awful. when I thought everything was OK I have a new issue here: could you look at the code and surmise what is happening?

Processing 182 of 182 "Akira - I - 182.jpg"
Converting LRS to LRF...
Traceback (most recent call last):
  File "/Applications/libprs500.app/Contents/Resources/lrs2lrf.py", line 8, in <module>
    from libprs500.ebooks.lrf.lrs.convert_from import main
  File "libprs500/__init__.pyo", line 30, in <module>
  File "PyQt4/QtCore.pyo", line 18, in <module>
  File "PyQt4/QtCore.pyo", line 11, in __load
ImportError: dlopen(/Applications/libprs500.app/Contents/Resources/lib/python2.5/lib-dynload/PyQt4/QtCore.so, 2): Symbol not found: __cg_TIFFSetErrorHandler
  Referenced from: /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/Versions/A/ImageIO
  Expected in: /opt/local/lib/libTIFF.dylib

LRS2LRF failed! at comiclrf.pl line 247.

[x5nder]

I have a feature request. Looking in your README.TXT I see the following 'feature':

Added auto image rotation for landscape images.

What I would really like, as these rotated images usually turn out too small to read, would be for your program to automatically split landscape images in half and encode them as two seperate pages. Would that make sense? I, for one, would love it!

[Xombie]

Originally Posted by mimoviz

Ahh... come on... this is just awful. when I thought everything was OK I have a new issue here: could you look at the code and surmise what is happening?

Processing 182 of 182 "Akira - I - 182.jpg"
Converting LRS to LRF...
Traceback (most recent call last):
  File "/Applications/libprs500.app/Contents/Resources/lrs2lrf.py", line 8, in <module>
    from libprs500.ebooks.lrf.lrs.convert_from import main
  File "libprs500/__init__.pyo", line 30, in <module>
  File "PyQt4/QtCore.pyo", line 18, in <module>
  File "PyQt4/QtCore.pyo", line 11, in __load
ImportError: dlopen(/Applications/libprs500.app/Contents/Resources/lib/python2.5/lib-dynload/PyQt4/QtCore.so, 2): Symbol not found: __cg_TIFFSetErrorHandler
  Referenced from: /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/Versions/A/ImageIO
  Expected in: /opt/local/lib/libTIFF.dylib

LRS2LRF failed! at comiclrf.pl line 247.

Hi. Try this. Open up the comiclrf.pl file and go to line 218. Change...

system("\"$imagemagick\" -size ".$scrwidth."x".$scrheight." \"$fromfile\" $rotate -fuzz 40% -trim +repage $im_param -type grayscale -colors $colours \"$tofile\"") == 0 || die "ImageMagick failed, path set wrong?";

...to...

system("\"$imagemagick\" -size ".$scrwidth."x".$scrheight." \"$fromfile\" $rotate -fuzz 40% $im_param -type grayscale -colors $colours \"$tofile\"") == 0 || die "ImageMagick failed, path set wrong?";

That gets rid of the -trim function so it's not a permanent fix. However, most of the people that put out manga already trim their pages. Well, the decent/good groups, I suppose.

I had the same initial problem as you (geometry does not contain image) and removing trim fixed it without the need for installing any other software like ActivePerl, etc... I'm on Ubuntu 7.10 if it helps.

I may spend more time investigating if nobody else figures it out in the meantime.

Good luck!

[nrapallo]

Sorry for butting in here, but those who want to try a new Comic (CBZ/CBR) to lrf converter may want to try PDFRead 1.8.1 which was just released in the Content forum. Input support for .CBZ/.CBR was recently added in addition to pdf/djvu/tiff/imglist. Output Profiles have been provided for the Sony PRS-500 and PRS-505 based on past discussions here and elsewhere.

PDFRead 1.8.1 can be found here.

While it may require the (free) eBook Publisher software installed, it allows for Cropping, rotating, dilating (thickening the text) and edge enhancing with colors/grayscale from 2 to 256! It also outputs in various landscape and portrait modes to suit your needs.

I'd be interested in some feedback if the default settings are to your liking. Please respond in the PDFRead 1.8.1 thread so as to lessen the impact here.

Thanks in advance!

[mimoviz]

Xombie: Thanks for your input. I finally solved the problem. It was not on the code or the software. In fact, the issue has to do with some idiosyncrasies of OSX... whenever it finds more than 1 library, even they are the same version, the execution stops.
I had to read a LOT to get here, and it was so simple after all...

Thanks to all of you guys for your support.