[Clipart] SVG::Metadata 0.11 Released

[Jonadab the Unsightly One]

Bryce Harrington <bryce at bryceharrington.com> writes:

> Okay, cool.  In time we should probably expand on this to allow
> selection from the set of existing keywords, like we've been
> discussing.

Oooh, a dropdown lists?  Nifty idea.  Later, though.

>> However, I intend to make a separate field in the upload form for
>> this soon, now that keywords are being written into the RDF, and
>> also offer a link to a page showing keywords currently in use, and
>> what they mean.  This latter should probably be on the wiki -- is
>> there anything there currently, or should I create a page for it?
>
> I think we discussed it only on the mailing lists; you can probably
> dig it out of the list archives.

Okay; I will in that case create a page on the wiki for it, so that
the upload script can link to that as explanation.

>> As far as a script to add keywords to an existing file, that will
>> probably be my next project (assuming nobody beats me to it) after
>> I get the keyword and .zip support in the upload script ironed out.
>
> Cool.  I have that on my todo list but unfortunately won't be able
> to get to it any time soon, so if you could take care of it, that'd
> be very cool.  

Once I get the zip upload stuff fixed up, that's next on my agenda.

> I can add the script to SVG::Metadata to keep it together with the
> others.

Okay.  I'll let you know when I've got it written.

> One script that would be really useful right now is something that
> navigates through a directory tree and extracts keywords from the
> path to the svg file, then updates the metadata for the svg file
> accordingly.

Yes, I was thinking that already.  It's a very minor adaptation to the
above script and can probably be a command-line argument.

> I.e., if it finds a file at collection/animal/mammal/dog/collie.svg
> it would insert keywords 'animal', 'mammal', and 'dog' into
> collie.svg.

Yep, that's what it would do alright.

> This script would let us leverage the work that's been done to
> quickly categorize the images in the last release.  Then we'd only
> need to work on categorizing the newly submitted images.

Indeed.  I might get to such a script quite soon, depending on how
long it takes me to debug the zip upload stuff.

-- 
$;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}}
split//,"ten.thgirb\@badanoj$/ --";$\=$ ;-> ();print$/