Re: [GMG-Devel] PDF support

[Christopher Allan Webber]

Note, this is another one of the very old posts :)

We actually did very recently land PDF support, but from a different
branch!  (Thanks alonl!)

Sebastian Spaeth writes:

> Hi all,
> here is my first and rough cut at a PDF supporting media handler.
>
> This is commit 566fdd79e3bf44cf92e2cada4bf987fee10b4fde in my branch 
> media_handler_pdf at git://gitorious.org/mediagoblin/mediagoblin.git. 
> Patch attached.
>
> Some comments on the state of things:
>
> Uploading of PDF works. So far, downloading does not work, as I simply 
> use the "image preview" template and that attempts to stick the pdf in a 
> <img> tag, but that's only a matter of creating a PDF specific template.
>
> What remains to be done:
>   - the gallery object template somewhere prefixes "/mgoblin_media" to
>     the thumbnail URL, even if we want the default thumbnail starting
>     with a "/mgoblin_static" URL. THis prevents the default thumbnail
>     from being displayed at the moment.
>   - Create a HTML template that suits PDF file better
>   - Actually extract some metadata from the file (author, description,
>     created on, created with, ....)
>   - Display the real first page as a thumbnail, needs converter
>   - The thumbnail icon is hardcoded for now. This needs to get the value
>     from ./__init__.py:MEDIA_MANGER['default_thumb'] but I don't know
>     how to access that value from processing.py.
>
> Longterm: Use pdf.js to actually display the PDF file?!
>
> I am giving up for now, but I think merging in this code post-release 
> would be a good foundation to get something better going. My biggest 
> roadblock is the auto-prefixing of "/mgoblin_media" somewhere and I need 
> to track that down.
>
> Sebastian
> _______________________________________________
> devel mailing list
> address@hidden
> http://lists.mediagoblin.org/listinfo/devel