DR800/DR1000 Website archive browser (website in .ZIP file)

[luite]

As you all probably know, the DR800, and the DR1000 with 2.0 firmware, have a fairly decent HTML browser based on WebKit. Unfortunately, storing whole websites on the reader is not really feasible, since they usually contain a large number of files, making the indexing phase of the boot process much slower.

As an alternative, I have built a modified browser, based on the irex one, that allows you to browse within a .ZIP file. The idea is that you first download the site (with wget or your favourite downloading tool), make a ZIP archive with the contents and copy that file to the reader.

When you open the ZIP file from the file browser, it will open the modified browser, which looks for a /index.htm or /index.html in the ZIP file. If such a file is found, it is shown like a regular website. Images, css files and all other pages are also loaded from the same ZIP file.

I currently have this working on the DR800 emulator, but I want to wait for the 2.0 sources to be released, to release the final version (if they don't release the sources soon, I'm a bit worried about the chapter 11 situation, I'll probably do a release based on the old sources).

I now have a few questions for you guys:
- Do you think this is useful?
- What extension would you want to be associated with the modified browser? using .ZIP means of course that all .ZIP files will be opened by this program
- Any other feature requests or suggestions?

[luite]

For the people who want to know how it works:
I have modified the browser to start an embedded web server. This web server is based on libmicrohttpd ( http://www.gnu.org/software/libmicrohttpd/ ), and automatically extracts each requested file from the zip file and delivers it to the browser. The unzip part uses minizip ( http://www.winimage.com/zLibDll/minizip.html ).

The URLs look like this: http://127.0.0.1:7766/media/mmcblk0p...LES/index.html

It would be fairly easy to modify the browser to show files from other kinds of archives (it requires modifying only one function). I've seen a few requests for an offline Wikipedia browser. Does anyone know a good archive format for Wikipedia?

[owl123]

That's definitely a useful piece of software. Thanks!

[Gertjan]

Its always nice to have choices, but DR800 (1.0/2.0) and DR1000 (2.0) already support MHT (MHTML) web archives natively.

[Dellaster]

Originally Posted by luite

I now have a few questions for you guys:
- Do you think this is useful?

I can see uses for it. There are quite a few online manuals that do not have any print or ebook versions. I wouldn't mind the ability to grab some of those with Backstreet Browser for nice e-ink reading or reference.

Originally Posted by luite

- What extension would you want to be associated with the modified browser? using .ZIP means of course that all .ZIP files will be opened by this program

Would .WEB conflict with anything? It would be hard to forget what kind of file that was.

Originally Posted by luite

- Any other feature requests or suggestions?

Resizing with resampling for image archives would be nice. Something like how Kolenka's CBZ Document Plugin does it. (Much better looking than the built-in shrink-to-fit software.)

[Dellaster]

Originally Posted by Gertjan

Its always nice to have choices, but DR800 (1.0/2.0) and DR1000 (2.0) already support MHT (MHTML) web archives natively.

I had no idea. Hm, interesting. Thanks for letting us know. And best wishes to you, Gertjan, during these trying times.

[owl123]

There are a lot of of ebooks and ezines that aren't in .mht but consist of hundreds of individual .html files.

Thanks, luite!

[repods]

Originally Posted by luite

- Do you think this is useful?
- What extension would you want to be associated with the modified browser? using .ZIP means of course that all .ZIP files will be opened by this program
- Any other feature requests or suggestions?

well the main use which I think of is with chm ebooks..
as you know those books can not be read by dr800 or dr1000, but they can easily converted in html pages, they are an html pages archive essentially, so this can be an use of your program..

btw why don't you try to do directly a chm reader ?

[ThR]

An other possibilty is to save the "website" or the "html-ebook" as mht-file, e.g. with http://maf.mozdev.org/

[luite]

Originally Posted by Gertjan

Its always nice to have choices, but DR800 (1.0/2.0) and DR1000 (2.0) already support MHT (MHTML) web archives natively.

Ah, I have to admit that I didn't know that, but I was under the impression that MHT files are used for a single page, while I intend to use this for complete sites (for example Doxygen or Javadoc documentation, which often consist of hundreds of html files).

[luite]

Originally Posted by repods

btw why don't you try to do directly a chm reader ?

I have to admit that it was only a very small modification of the erbrowser code, zip was just the easiest format to support. CHM is a good suggestion, and could be done with chm lib ( http://www.jedrea.com/chmlib/ ), but I'm not sure how much work it is to properly support the table of contents an index features of CHM files.

Originally Posted by ThR

An other possibilty is to save the "website" or the "html-ebook" as mht-file, e.g. with http://maf.mozdev.org/

Thanks for the suggestion. MAFF seems to be a ZIP archive, and it seems very easy to add support (the directory structure is slightly different).

[Iñigo]

Originally Posted by luite

I have to admit that it was only a very small modification of the erbrowser code, zip was just the easiest format to support. CHM is a good suggestion, and could be done with chm lib ( http://www.jedrea.com/chmlib/ ), but I'm not sure how much work it is to properly support the table of contents an index features of CHM files.

You can use FBReader to read .chm files. It works ok if the files are not too big.

[luite]

Originally Posted by Iñigo

You can use FBReader to read .chm files. It works ok if the files are not too big.

Ok, I don't think CHM support will be necessary then.

And as promised, now that the 2.0 source code is available, here's a preview of the zip browser. It should work on both the DR800S(G) and DR1000, with the 2.0 firmware. Installing associates .web and .zip files with the program. Keep in mind that zipbrowser looks for a file called index.html or index.htm in the root directory of the zip file. You can zip complete websites with multiple html files, images and other resources.

MAFF/maf files are not supported yet, but I plan to add support in the future (it requires parsing some rdf files).

Installation:
- Enable "Folder based navigation" in Advanced Settings
- Connect your device via USB
- Extract the zip file to the root of your device
- Disconnect USB
- Go to SD Card / Programs / Install, click on AssociateZipbrowser.desktop
- Wait for the device to reboot

[luite]

Oops, I found a bug. The installation script incorrectly associated the djvu file format with zipbrowser if you had installed the preview of the djvu plugin for the new firmware ( https://www.mobileread.com/forums/sho...t=73790&page=2 ) before you installed zipbrowser. If you experience this problem, just reinstall the djvu plugin.

Here's an updated zipbrowser that fixes this.

[jsway]

Originally Posted by luite

Oops, I found a bug. The installation script incorrectly associated the djvu file format with zipbrowser if you had installed the preview of the djvu plugin for the new firmware ( https://www.mobileread.com/forums/sho...t=73790&page=2 ) before you installed zipbrowser. If you experience this problem, just reinstall the djvu plugin.

Here's an updated zipbrowser that fixes this.

I just downloaded it. Thanks a lot!