Beta testing libprs500's new GUI

[kovidgoyal]

As for multiple databases, the chances are abysmal unless youre willing to write the code

What do you mean by a tag tree?

[macr0t0r]

Finally took the time to install version 0.4.33 of libprs500 on my iMac. Man, what an improvement over the 0.3.x days! I find it to be an excellent tool for organizing my eBooks due to the search filter and the ability to store multiple formats of each book. The USB interface to the Sony Reader is VERY nice, so I'm now putting my books into the Main memory so that I can do collections. It also does a nearly PERFECT job of converting my purchased (and DRM-stripped) MS Lit files.

Not everything is perfect, though (of course). There are many issues with conversion:
1. Not a good tool for converting Plain Text files to LRF. It removes any indentations and tends to gobble up most carriage returns. I've fiddled with the settings, but it still tends to create one giant, flat paragraph. I find it's better to convert Plain Text to RTF with the OS X "textutil" and then convert from there.

2. Very crash-prone when converting files. It's best to just convert one file at a time.

3. It either needs a hotkey to display MetaData, or there needs to be an option to have ALL MetaData shown in the column display. Numbering books in a collection is a pain. (Select file, edit medata, type book number, click save, select next file, etc...)

4. Available formats should be in Column display so I can see at a glance what I need to convert.


But, for organizing and sending eBooks, this thing is far superior to Sony's interface. Thanks for all your hard work on this!

- Jim

[kovidgoyal]

1. The plain text conversion is really meant to define a plain text ebook format using lightweight markup (specifically markdown). So that you can create plain txt files with links and headers and formatting and lists and even tables. It is not meant to "do the right thing" with every text file you throw at it. That's not going to change as I prefer to run dedicated tools on a txt file from a given source (e.g. gutenmark) and then convert the HTML.

2. Are the multifile conversion still causing crashes? I though I'd squashed that bug a while ago.

3. Making the displayed columns customizable is on my TODO list

4. In version 0.3.34 you'll be able to search for the term !lrf to get a list of books that dont have an LRF format (this is not a 100% safe since the search will be over all metadata so some book that has the string "lrf" in its metadata wont show up, but this is the best you'll get until i implement an advanced search dialog.

[Shadowcat]

I just got my Sony 505 reader and was very happy to find the libprs500 program. So kovidgoyal, excellent job!!

Now on to the meat of the post:

- On the multiple conversion crashes, I've ran across this and I'm pretty sure it is system specific. i.e. Something on that specific PC is causing the crash. I've run the libprs500 program on my desktop and my laptop, where the desktop exhibits the problem but the laptop has been completely crash free.

Library request: I know there have been a few other posts requesting the importing of save library feature, and I'd like to add a slight twist to that. Is there any way to add the feature of importing the metadata from a previously saved (saved by the libprs500 to disk) ebook to the current library? This would be a big help in a few areas:
- The "series" or "collection" information could be kept with the save disk file. This would make it much easier if you have multiple readers and want to have customized libraries on each without having to redue all of the work creating the series.
- The search feature is nice, but when you have large collections and just want to browse for a book it's a pain to walk through the whole collection. Instead if all of your books are in a nice directory structure (which the libprs500 does right now when you save to disk), it's easy to browse and grab the ones you want with the libprs500 and then upload them into the ebook. (you can do this right now, but since the metadata isn't also loaded if you add say 6 books from a series there is no "collection" info being passed to the Sony Reader)
- Also if you want to share books that you have spent a lot of time customizing the metadata, the only way to do this is to share the libprs500 database. This isn't really the optimal solution since each persons reading preferences are different. Do I really want to read my girlfriends romance novels? <grin>)

Once again, excellent program and much better than the sony software. Keep up the great work!

[kovidgoyal]

Do the desktop and laptop have the same OS?

Right now the save to disk feature saves the metadata as an OPF file, in the same directory as the ebook files. When I implement recursive ebook import, it will import the metadata back into the database as well.

The equivalent of browsing through a folder structure in libprs500 is to use the search, so for example if you want to browse books by Asimov you type author:Asimov in the search field, then if you want to browse books tagged with "foundation" you type "author:Asimov tag:foundation" and so on...I find this a much more powerful browsing system than using folders. After all folders are basically hierarchical tags. But general tags, like libprs500 has, can mimic the functionality of hierarchical tags and provided more features as well. For instance, if you want to browse through books by two different authors, say Dostoevsky and Russell (don't ask me why those two), there's no convenient way to do it using folders, but with a search interface you just type [authorostoevskt author:Russell] and then you can sort the list on title or date or rating and browse through your collection easily.

EDIT: I hope your girlfriend isn't reading this ;-)

[Shadowcat]

Hi Kovidgoyal!

Yes, both the desktop and the laptop are running Windows XP SP2 with all of the latest Windows patches. So the only two areas that are different are the software packages I've got installed (the PC has many more pieces of software installed), and the specific hardware drivers. I'll see if I can get a good error message to generate on the desktop, and if so I'll do a screen capture and post it.

I'm happy to hear that you've already got the recursive ebook import on your feature request list. For now I'll just leave the opc files in the saved file folders and look forward to this feature sometime in the future.

I use the search function now and it works quite well, but not really designed for random browsing. Of course if I have the files saved to disk I could always browse through there (which means i could organize them any way I wanted), and then just do a search in the libprs500 and load it to the ebook.

And no, my girlfriend isn't reading this. But of course I had to go and tell her my comments about the romance novels, which elicited a smack in the arm <grin>.

[lovebeta]

Hi Kovidgoyal,

Thanks for the great software. I have a PRS-500. If I understand correctly, I cannot use libprs500 together with ebook library due to the driver issue, right. Now my question is, can I use libprs500 on one computer and use ebook library on another computer? I really like your software, but I have to use the sony one to purchase books.

Thanks!

[kovidgoyal]

No you shouldn't use both to manage the books on your reader. I would suggest you simply save the books/news in the libprs500 GUI to a folder that you then set as the auto-sync folder in the Connect software. That way, you can use the Connect software to buy books and using the libprs500 GUI is almost as convenient as if you were using it to transfer books.

[Chris Rastyk]

I had seen this on windows, probably about six months ago. I haven't used the sony reader for awhile, since the windows tools were crashing on me, it was taking too long to update the reader with news articles because of it. Today, I checked out the linux release of libprs500 (using ubuntu gutsy 7.10). This is a well-done application, I was able to upload a number of technical documents and this is very, very handy!

[Xenophon]

I'm getting lots of errors like this one:

NoneType: Could not run job: any2lrf ([u'libprs500', u'--minimum-indent', u'0.0', u'--base-font-size', u'9.0', u'--author-sort', u'Saberhagen, Fred', u'--author', u'Fred Saberhagen', u'--force-page-break-before-attr', u'$,,$', u'--wordspace', u'2.5', u'--chapter-regex', u'chapter|book|appendix', u'--right-margin', u'10', u'--title', u'Rogue Berserker', u'--page-break-before-tag', u'h[12]', u'--left-margin', u'10', u'--top-margin', u'5', u'--publisher', u'Baen Books', u'--force-page-break-before-tag', u'$', u'--headerformat', u'%t by %a', u'--bottom-margin', u'0', u'--baen', u'--profile', u'prs500', u'--cover', u'/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_Ivm7PS.jpeg', '-o', '/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_ib-3m1.lrf', u'/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_FtRHGI.zip', '--verbose'],)
Failed to perform job: Convert book:Rogue Berserker
Detailed traceback:
Traceback (most recent call last):
  File "libprs500/gui2/jobs.pyo", line 105, in run
  File "libprs500/parallel.pyo", line 123, in run
RuntimeError: Could not run job: any2lrf ([u'libprs500', u'--minimum-indent', u'0.0', u'--base-font-size', u'9.0', u'--author-sort', u'Saberhagen, Fred', u'--author', u'Fred Saberhagen', u'--force-page-break-before-attr', u'$,,$', u'--wordspace', u'2.5', u'--chapter-regex', u'chapter|book|appendix', u'--right-margin', u'10', u'--title', u'Rogue Berserker', u'--page-break-before-tag', u'h[12]', u'--left-margin', u'10', u'--top-margin', u'5', u'--publisher', u'Baen Books', u'--force-page-break-before-tag', u'$', u'--headerformat', u'%t by %a', u'--bottom-margin', u'0', u'--baen', u'--profile', u'prs500', u'--cover', u'/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_Ivm7PS.jpeg', '-o', '/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_ib-3m1.lrf', u'/var/folders/vi/visJUvLg2RaGnE+8ZKwa9+++59k/-Tmp-/libprs500_0.4.42_FtRHGI.zip', '--verbose'],)
Log:

Using libprs500 v0.4.42. Any hints?

Xenophon

[Xenophon]

And, even weirder, repeated attempts to convert the same book sometimes lead to success! Perhaps a more detailed error message might help with diagnosis.

I have notice, by the way, that there are a number of points during the conversion process where I see neither CPU usage nor disk I/O. It seems odd to me that the whole process sometimes sits there for a minute (or two, or five) seeming to do absolutely nothing.

All this on a Mac Pro (early 2008) running Mac OS X 10.5.2, with the latest libprs500.

Xenophon

[kovidgoyal]

Hmm sounds like OS X doesn't like the new child process architecture. I've uploaded a new 0.4.42 dmg can you try with that, the error message may be more helpful

[iacchi]

Kovid, I get this error giving libprs500_postinstall:

Iacopo:/home/iacopo# libprs500_postinstall
Trying to setup udev rules...
/bin/sh: /etc/init.d/hald: No such file or directory
Restarting Hardware abstraction layer: hald.
Setting up bash completion... done
Setting up desktop integration...
Installing MAN pages...
Traceback (most recent call last):
  File "/usr/bin/libprs500_postinstall", line 8, in <module>
    load_entry_point('libprs500==0.4.42', 'console_scripts', 'libprs500_postinstall')()
  File "build/bdist.linux-i686/egg/libprs500/linux.py", line 348, in post_install
  File "build/bdist.linux-i686/egg/libprs500/linux.py", line 324, in install_man_pages
  File "/usr/lib/python2.5/subprocess.py", line 594, in __init__
    errread, errwrite)
  File "/usr/lib/python2.5/subprocess.py", line 1147, in _execute_child
    raise child_exception
OSError: [Errno 2] No such file or directory

[kovidgoyal]

You need to install the program help2man

[BobWhite]

Kovid,

What is done with the series and book number within series information? Is that information passed over to the PRS-500?

I read a lot of books in various series. When using libprs500 to look up the book information and cover, if I have the book within series info in the title, the lookup can't find it. A lot of times, that information is already in the filename, so to look up the cover, I have to take it out, then put it back. With 300+ books on my reader, it's hard to keep track of the order in which to read some of them!

Would it be possible to have that info optionally prepended to the title when the book is moved to the Sony?

For example, take David Weber's Starfire series: (1) Insurrection, (2) Crusade, (3) In Death Ground, and so forth (there are 5 total).

I'd like to specify the following metadata in the library: title - "Insurrection", author - "David Weber", series - "Starfire", book - "1" for the first book, and so forth for the rest of them.

What I'd like to see on the PRS-500 would be title: "Starfire 01 - Insurrection", author - "David Weber". The second book would have the title "Starfire 02 - Crusade", the third book "Starfire 03 - In Death Ground", and so forth. This would keep books in a series grouped together without having to use a collection (which would be limited to what you could cram into main memory).

Thanks,

Bob