Re: Filter writing becomes even easier




D Bera writes:
 > Writing beagle filters became even easier. Joe checked in FilterExternal with 
 > which the output of any program can be indexed. (As of now only text output 
 > can be indexed, metadata/properties can not be indexed).
 > 
 > For e.g. to index the text content of tex files, find any tex2ascii or tex2txt 
 > program which outputs the txt content to stdout and create an entry in 
 > external-filters.xml (its in the prefix/etc/beagle) like this:
 > 
 > <filter>
 >   <mimetype>text/x-tex</mimetype>
 >   <extension>.tex</extension>
 >   <command>tex2ascii</command>
 >   <arguments>%s</arguments>
 > </filter>
 > 
 > The next time beagle crawler finds .tex files, it will run tex2ascii on the 
 > file and index the text of the output. For more information, see the file 
 > beagle/Filters/external-filters.xml .
 > 
 > You dont even need to know C# ... some command-line tools, maybe a little bit 
 > scripting and you can index any file on your computer (beagle already covers 
 > the major formats ... whatever rest isnt yet covered can be covered via 
 > this). Isnt it cool ?

Yep... really cool....

Would it be possible to define  in the XML  language simple regexps to
recognize basic metadata such as authors, title, date,... ?

-dav




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]