Bindery

[ibr4him]

Windows, yes. Mac, no. I don't have admin access to a Mac, but I can give instructions on how to make a .app from the sources to someone who has one.

I can provide remote (admin) access to a mac, if you want? Anyhow, windows version is fine for me.

[Anonymous1]

It's kind of like beating a dead horse. Python is already cross platform to begin with (which is why I chose to work with it), so making binaries seems a bit counter productive. If you absolutely need them, I can make a Windows binary today (I can write up how to make a Mac binary).

Here's an updated screenshot of the program. I added more progress markers (it highlights the current item that is being processed in progressively darker shades of green after each pass. It first analyzes, the does OCR, and then binds, so there are three shades of green), a default icon for each page (as a placeholder while it's thumbnailing), and some menu options. Only the "File Previews" option works for now, and it's a bit buggy:

Screenshot-Bindery.png

Screenshot-Bindery-1.png

Most of the menu and other options are not functional (except the DjVu Encoding options. They all work.), but I'm still adding PDF support. I'll make a Windows binary when I have access to Windows.

Happy binding

[Anonymous1]

Here's the first binary build (for Ubuntu):

[Anonymous1]

And, for those Windows users, I can't make it "compile" for Windows (wine). It works fine when you run it from the command line, but I just can't make it work. Sorry

[strider1551]

@strider1551, are there any more settings that I should know about in djvubind?

Sorry for not getting back to you sooner. I just returned from an 8-day silent retreat, and I have a small mountain of things to get back to speed on.

No, djvubind doesn't have any settings that aren't documented (at least intentionally). There will be some new options in the next release pertaining to issue 25.

The screenshots look really good. Hopefully I'll get some time tonight to give it a try, maybe give some feedback

[Anonymous1]

Wow, that's amazing! If PDF doesn't have this sort of option, I can finally convince (some) people to use DjVu. Custom page names is definitely something I will use, as I've scanned some books which neglected to paginate blank pages and pages with pictures. It was interesting, as there were ~ 30 pages of each

After working with PyQt4 for a while, you get really used to it. It took me four days of StackOverflow-ing to figure out how to make it thumbnail, but after that, the rest was pretty simple. I'll have to accommodate for your changes in djvubind, as I included progress hooks within your encoding module to make the progressbar move along with everything else.

Just as a side question, what seminary do you attend?

[strider1551]

St. Mary Seminary up in Cleveland.

What version of PyQT4 are you developing with? And python3 I assume? I just cloned your repository and main.py tells me:

Code: Select all

Traceback (most recent call last):
  File "./main.py", line 7, in <module>
    import functions
  File "/home/strider1551/workbench/bindery/Bindery/functions.py", line 3, in <module>
    from gui import *
  File "/home/strider1551/workbench/bindery/Bindery/gui.py", line 352, in <module>
    from BookListWidget import BookListWidget
ImportError: No module named BookListWidget


Probably something stupid on my part, which is why my first thought is whether I even have the right environment setup.

[Anonymous1]

I'm working with Python 2.6.6. It was my original install, and all of my modules were working fine with it. I had to downgrade your code a little to work with Python2, but the changes were completely minor.

As for the error, I just cloned my repo and got it too. I just pushed a working branch (I forgot to add some files to be monitored by git), and it should be working.

[Anonymous1]

If you're committing, it won't work... I have to push a few more changes, and it's taking a long time. Maybe it will finish in 5 minutes...?

[Anonymous1]

Well, the git repo is working now, but I'm getting funky results with djvubind's OCR module. It keeps spitting out:

Code: Select all

wrn: tesseract produced a significant mismatch between textual data and character position data on "51.tif".  This may result in partial ocr content for this page.

What does this mean? The djvu file does have partial OCR, but why? I'm playing with UTF-8, but not getting anywhere...