[GUI Plugin] Count Pages

[stevenlebeau]

Yeah, I just tried looking at the source. Way too complicated to figure out what's going on there. I think I'm just going to convert all my epubs to .txt and then make a command-line script in Python to do the unique word counts.

[BetterRed]

@stevenlebeau - you could shovel the results from parsing the text files into a custom column in the metadata database via the Import List PI - I don't think the calibredb ebook-meta command allows updating of custom columns.

BR

[stevenlebeau]

That could work! I'll give it a shot.

[davidfor]

Originally Posted by stevenlebeau

Yeah, I just tried looking at the source. Way too complicated to figure out what's going on there. I think I'm just going to convert all my epubs to .txt and then make a command-line script in Python to do the unique word counts.

I had a look when you first asked, and it wouldn't be that hard. If you follow the count words code (in statistics.py), that extracts the text from all the contained files. The lazy coder in me would convert that to a list and then a set and get the size of the set. That would give a crude count. And as a hack, returning it as the word count, lets you see and store the number.

There's probably a more efficient way to do it than that, plus maybe you want to filter some words. Should "a" and "I" be counted? Or "the", "and" etc. Do plurals get counted separately? What about proper nouns? Numbers?

Adding it as an option and using a separate custom column is, to me anyway, the hard bit. But, it is mainly copying similar sections and changing the names to protect the innocent. Getting the GUI all looking good with decent field descriptions is probably the hardest bit.

And I'm regretting typing this. I'm giving myself a "If it's so easy, why don't you do it" feeling

[stevenlebeau]

Based on what you wrote, I actually changed the lines in statistics.py to make this happen. Only thing is, calibre won't let me install it. I re-zipped and tried to install the plug-in manually, and it complained about not having a __init__.py file (which it DOES have, so I'm not sure what to do next).

I suppose I could simply change the already-installed statistics.py to do the work for me, but I haven't been able to figure out where calibre stores its' plug-ins in OS X.

[BetterRed]

Originally Posted by stevenlebeau

Based on what you wrote, I actually changed the lines in statistics.py to make this happen. Only thing is, calibre won't let me install it. I re-zipped and tried to install the plug-in manually, and it complained about not having a __init__.py file (which it DOES have, so I'm not sure what to do next).

I suppose I could simply change the already-installed statistics.py to do the work for me, but I haven't been able to figure out where calibre stores its' plug-ins in OS X.

Maybe in the plugins subdirectory of the configuration folder, on Windows the config folder is wherever you land when you click the Preferences->Miscellaneous->Open calibre configuration button

BR

[davidfor]

Originally Posted by stevenlebeau

Based on what you wrote, I actually changed the lines in statistics.py to make this happen. Only thing is, calibre won't let me install it. I re-zipped and tried to install the plug-in manually, and it complained about not having a __init__.py file (which it DOES have, so I'm not sure what to do next).

Did you accidentally create a zip file with folder inside it which has all the files?

I suppose I could simply change the already-installed statistics.py to do the work for me, but I haven't been able to figure out where calibre stores its' plug-ins in OS X.

Which probably wouldn't help as the plugins are still zip files in that directory.

[stevenlebeau]

Okay, I figured it all out. I did archive the root directory and not just its contents. Once I fixed that, it worked like a charm.

So basically I replaced the word count with code that set the book text to lowercase, split it into a list, and then made a set of the list and counted the length of the set.

I am actually shocked at how many different words are even in the simplest of children's books! But I have a much better idea of where I'll be starting now.

Thanks to everyone for all their help!

[crumba]

how do you fix (No goodreads id)%s

getting it for 99% of my books. also getting false page numbers on the majority of my books using estimate.

[93terp]

I'm having same issue as crumba in post #699. Any help to troubleshoot is appreciated.

[davidfor]

Originally Posted by crumba

how do you fix (No goodreads id)%s

getting it for 99% of my books.

Is it wrong of me to think that "No goodreads id" tells you exactly what is wrong? The book needs to have a goodreads id in its metadata.

also getting false page numbers on the majority of my books using estimate.

What are you comparing the page count to? You can't really compare this to a paper book as each edition would have a different page count. I use the ADE algorithm and it is a reasonably close.

[davidfor]

Originally Posted by 93terp

I'm having same issue as crumba in post #699. Any help to troubleshoot is appreciated.

crumba reported two problems. Which one are you seeing?

[JSWolf]

I've found a bug. Count Pages treats words with an em dash between them as one word.

So word1—word2 is counted as one word. I'm betting it's that way for ... and en dash and so forth.

Can this be fixed? Thanks.

[sophieturgeon]

Hello, I'm new at this so please bear with me. I have installed the Count Pages plugins. I have created the custom columns. I have customized the preferences of the plugins so the columns have been selected.

Calibre is installed on the following path:
'K:\10 - Library' , which is an external hard drive.
My books are stored using the following path:
'K:\10 - Library\Calibre Library'

I run the plugins and it looks like it is working fine. I look up the Count log and I can see the appropriate and correct information. When running the plugins and clicking yes to update the columns, a '1' appears in each columns. In the count log the following information can be read (something tells me you might need to know that):
"InputFormatPlugin: EPUB Input running
on C:\Users\Turgeon\AppData\Local\Temp\calibre_onffru \a3vesk_count_pages\2759.epub
Found HTML cover titlepage.xhtml"

What am I doing wrong. Please keep in mind: I have absolutely no programming knowledge/understanding. I know just enough about computers to get myself in trouble. Your help would be greatly appreciated.

[theducks]

Originally Posted by sophieturgeon

Hello, I'm new at this so please bear with me. I have installed the Count Pages plugins. I have created the custom columns. I have customized the preferences of the plugins so the columns have been selected.

Calibre is installed on the following path:
'K:\10 - Library' , which is an external hard drive.
My books are stored using the following path:
'K:\10 - Library\Calibre Library'

I run the plugins and it looks like it is working fine. I look up the Count log and I can see the appropriate and correct information. When running the plugins and clicking yes to update the columns, a '1' appears in each columns. In the count log the following information can be read (something tells me you might need to know that):
"InputFormatPlugin: EPUB Input running
on C:\Users\Turgeon\AppData\Local\Temp\calibre_onffru \a3vesk_count_pages\2759.epub
Found HTML cover titlepage.xhtml"

What am I doing wrong. Please keep in mind: I have absolutely no programming knowledge/understanding. I know just enough about computers to get myself in trouble. Your help would be greatly appreciated.

There should be one more line: Page Count: nnn.n
What page count Algorithm are you using?