Re: Thoughts about FilterChm

["Miguel Cabrera" <mfcabrera gmail com>]

On 4/24/06, Joe Shaw <joeshaw novell com> wrote:

Hi,

On Mon, 2006-04-24 at 17:59 +0200, @4u wrote:
> I think it makes sense for beagle to link each HTML file based on its
> content. If you are searching for one PHP function, it would be great if
> beagle links the file directly, so that a viewer program will not open
> the index CHM file but the HTML file containing the selected function.

>
> Say you are searching for mysql_error ():
>
> If you handle a CHM file like an archive, beagle will link it the
> following way:
> CHM file -> subsection whereever -> mysql_error.html

Well, even if I said before that would be nice to have a filter process for every html inside a chm file , not all chm files has such beutiful names inside them (for example, some chm html names are composed by number rather than name: 123331-ADF4.html) so I dont think that is a good idea to show Chm File -> "htmlfilename.html".
   

Ok, makes sense.  How are CHM files opened today on Linux?  What viewer
handles them?  What should the URI or file scheme passed into the
program be?

I know a couple of them, X-CHM and Gnochm, I think (if it exist at all) there isn't  a standart way to tell  the viewer program to go directly to a section inside the chm file.
 

Was Beagle treating them as just one big document before?

Yes,  for procesor usage issues it only was indexing the default page and the topics file, and as metadata it extracted the chm title. The idea is that, due to the improvements made to the html filter,  in the next version of the Filter, it indexes all the contents of the chm file.
 

Well, I does not fit inside the "child indexable problem", but it would be nice being able to use the Html filter from the Chm Filter, there is a way to do this? any clues? If i can't do that,  I will have to rewrite the code and include (copy and paste) some code from the html filter.


Miguel.