New Buggins Firmware - 20100716

[aegeus]

New test firmware from Buggins, sorry if it was mentioned before:

- Fixed rendering inline images
- Added support for title attribute for images
- Support for semitransparent PNG

V5: v5setup-20100716.zip (86.2 MB)

http://files.mail.ru/S7JL4S

Regards.

[Dave_S]

Originally Posted by aegeus

New test firmware from Buggins, sorry if it was mentioned before:

- Fixed rendering inline images
- Added support for title attribute for images
- Support for semitransparent PNG

V5: v5setup-20100716.zip (86.2 MB)

http://files.mail.ru/S7JL4S

Regards.

I am running both the 5" and 6" versions of that release and do not see any new problems, but I have not tried to test the new features mentioned above. The discussions concerning that release on the-ebook.org also provided information on making your own script to clear the CoolReader cache without needing to connect to a PC. If you want to try that, here is one that I made to clear both the internal and SD cache simultaneously. If you edit the script, it needs to be done with a Linux compatible editor since Windows line endings will cause the script to fail.

[Blossom]

Thanks Dave! Where do I install the file to and how do I use it?

[Dave_S]

Originally Posted by Blossom

Thanks Dave! Where do I install the file to and how do I use it?

After you unzip it, you should be able to put it anywhere that you like because it appears to the BookShelf as a DJVU book. When you select the "book" to read, it will be executed by the Linux Bash shell instead of opening to read. As it is written, it will entirely clear the cache on both the internal book memory AND the SD card book memory.

I do not think that DJVU is a format that is supported in the Astak firmware, so this little script will probably only work on the Lbook firmware. But then, as it is written, it is only useful to clear the cache on the newest versions of CoolReader. Buggins provided it as an example, so the only thing that I did was copy it and paste it into the Notepad++ text editor so that I could save it in the Linux text format. The other change that I made was to use the DJVU extension as suggested by Jotas instead of the FB2 extension that Buggins suggested. That causes the script to be passed to the shell by the DJVU viewer instead of by CoolReader. The process just works a little faster that way.

[jusmee]

Interesting. Thanks for the script Dave.

I worry a little about this mechanism for executing a script though. Looks like an open invitation for malware. I can imagine downloading a 'book' that really is a malicious script.

[Dave_S]

Originally Posted by jusmee

Interesting. Thanks for the script Dave.

I worry a little about this mechanism for executing a script though. Looks like an open invitation for malware. I can imagine downloading a 'book' that really is a malicious script.

The nice thing about scripts is that they are human readable, so just open the script in a text editor to see what it does before loading and executing it on the reader.

BTW, Buggins has already updated the firmware again on July 19th.

6" V3 version: http://files.mail.ru/CBMZB8
5" V5 version: http://files.mail.ru/69G2PM
6" V3+ version: http://files.mail.ru/3YHBN5

New test firmware 

- Fixed translucent PNG to V3 
- Fixed built-in document styles (FB2 stylesheet) 
- Added support for lists (CSS: display: list-item; list-style-type, list-style-position), tagging ul / ol / li 

For ul / ol / li - need to update fb2.css, epub.css, htm.css, chm.css 

In fb2 to maintain compatibility to be elements of the list of the usual sections, using styles style = "display: list-item"

[jusmee]

Originally Posted by Dave_S

The nice thing about scripts is that they are human readable, so just open the script in a text editor to see what it does before loading and executing it on the reader.

Oh for sure, but if it is named as if it is a book, and you think you have downloaded a book, you're likely to just open it to read it - and blam! I certainly don't check the contents of book files before I open them usually.

[Dave_S]

Originally Posted by jusmee

Oh for sure, but if it is named as if it is a book, and you think you have downloaded a book, you're likely to just open it to read it - and blam! I certainly don't check the contents of book files before I open them usually.

That is a possibility, but a very very small one given that book readers are no where near as prevalent as Windoze PCs. The good news is that now you are aware that scripts can be run on your reader, and probably on all Linux based readers. In any case, I do not worry about malware scripts sneaking onto my reader, since I always know the source of books that I put on my reader.

[Dave_S]

Originally Posted by Dave_S

The other change that I made was to use the DJVU extension as suggested by Jotas instead of the FB2 extension that Buggins suggested. That causes the script to be passed to the shell by the DJVU viewer instead of by CoolReader. The process just works a little faster that way.

First, I need to apologize that I only tested this script on my 6" EZREader and not on my PocketPRO. This morning, I just noticed that the DJVU extension does not work on my PocketPRO, and the I had to change the extension to FB2 to get it to work. Either extension works on my 6" EZReader, but in my case ONLY the FB2 extension works on my PocketPRO.

Again, I sincerely apologize for my oversight!!!!!!!!!

[booknut]

I ran it from my SD card on my V5 where it showed up as cacheclear.exe and a djvu file and it worked fine down to the last wedge on the hourglass where it is now stuck. Does this mean it's now working or does it take a long time? O well, that's what I get for not believeing Dave_S... ;-)

[Dave_S]

Originally Posted by booknut

I ran it from my SD card on my V5 where it showed up as cacheclear.exe and a djvu file and it worked fine down to the last wedge on the hourglass where it is now stuck. Does this mean it's now working or does it take a long time? O well, that's what I get for not believeing Dave_S... ;-)

That is what I saw too, and that is what made me apologize for my oversight. After I pushed the reset button to get the PocketPRO going again, I found that the cache had not been cleared. I think that the PocketPRO version of the DJVU viewer actually tried to open the file as a DJVU document. If you change the last file extension from DJVU to FB2 I think that you will find that it takes about 5 seconds or so to clear the cache, with most of that time taken just for Linux to load CoolReader. BTW, you can name the file anything that you like, but it MUST end in.exe.fb2. If you leave out the first extension (EXE), then CoolReader will just open and display the file as text. I am assuming that the reason Buggins is requiring filename.exe.fb2 is so that there can be no confusion about the fact that the file is an executable and not a normal fb2 book?

[booknut]

It's actually not clearing anything for me regardless of the extension, tried it in SD and HD, as cacheclear.exe.fb2 and cacheclear.exe.djvu. I am running today's firmware from Buggins...

Is there something I need to specify in the file itself, some sort of path, since the file is easily opened in Notepad?

[Dave_S]

Originally Posted by booknut

It's actually not clearing anything for me regardless of the extension, tried it in SD and HD, as cacheclear.exe.fb2 and cacheclear.exe.djvu. I am running today's firmware from Buggins...

I am running Buggins' current release too, and it still works OK. You can put the script anywhere that you want to because the argument to the rm command is a full path specification to the location of the cache. Buggins puts the cache in the same places on both the 6" and 5" readers.

Originally Posted by booknut

Is there something I need to specify in the file itself, some sort of path, since the file is easily opened in Notepad?

Editing the file on a Windows PC can be very tricky and Notepad definitely will cause the file to no longer work. That is because the file must be a Linux text file and not a Windows or Mac text file. I initially had a lot of the same trouble that you are seeing because I just cut and pasted the suggested commands from Buggins into my normal text editor. I spent hours of failed attempts trying to get it to work. That is why I uploaded a ZIP file of my working script instead of just posting those few lines of code. It was really frustrating trying to get the script to work because there is no way to view the error messages from the Bash shell when trying to run the script. I ended up writing my working script in Notepad++, where I could ensure that it was a UNIX text file with only ANSI characters in it.
I can only say that I did get it working, and it works well on both my 6" and 5" readers with the current Buggins firmware. If you need to edit the script yourself, I hope that my explanation made some sense?

[booknut]

Thanks for the info, it all makes perfect sense - it just does not work for me...maybe I need to start with a fresh copy....

Mm, noticed a file size discrepancy, after loading the unzipped file onto my PP, it is always just 1 K, I wonder if that's the problem.

[Spiffy]

Originally Posted by Dave_S

I am running Buggins' current release too, and it still works OK. You can put the script anywhere that you want to because the argument to the rm command is a full path specification to the location of the cache. Buggins puts the cache in the same places on both the 6" and 5" readers.


Editing the file on a Windows PC can be very tricky and Notepad definitely will cause the file to no longer work. That is because the file must be a Linux text file and not a Windows or Mac text file. I initially had a lot of the same trouble that you are seeing because I just cut and pasted the suggested commands from Buggins into my normal text editor. I spent hours of failed attempts trying to get it to work. That is why I uploaded a ZIP file of my working script instead of just posting those few lines of code. It was really frustrating trying to get the script to work because there is no way to view the error messages from the Bash shell when trying to run the script. I ended up writing my working script in Notepad++, where I could ensure that it was a UNIX text file with only ANSI characters in it.
I can only say that I did get it working, and it works well on both my 6" and 5" readers with the current Buggins firmware. If you need to edit the script yourself, I hope that my explanation made some sense?

Its a Unicode issue, right? Whatever text editor you use has to be able to save that way.