[pmwiki-users] TextExtract (Search recipe) update

Hans design5 at softflow.co.uk
Mon Sep 21 07:20:57 CDT 2009

New TextExtract version 2009-09-21 now released.

If you are upgrading you will have to change some of the
markup used!

This release normalises the input syntax to correspond with standard
pagelist syntax.
Most importantly:
All single arguments are treated as search terms, not just the first one.
PageName arguments need to be set with name= or group= or page= parameters.
(:extractresults:) is not used any longer, but (:searchresults:) instead.

This release makes use of Pmwiki's MakePageList() function to
generate the list of source pages.
(:pagelist:) and (:searchbox:) directives can be used with a
fmt=extract parameter, and all other pagelist and textextract
parameters, as can markup expression {(extract)} and earch form
directive (:extract:).

I kept the (:extract:) directive to generate a search form with
separate fields for search terms and page names as before, for sake
of conveniance. A similar or more customised search from can be built
with Input markup, using action=search and fmt=extract.

In addition to standard search terms as handled in PmWiki Searchbox
and pagelist regular expressions can be used for search input (set
regex=1 for this).

I added also a general way to handle anchor sections given with
PageNames with wild cards, so only text from the specified anchor
section will be used from the generated list of source pages.

This release integrates TextExtract with pagelist and standard search
most comprehensively, and hopefully adopting a similar markup syntax
will make it easier to use. It did make the code a little simpler.
And the performance seems to be fine.

Apologies to all users for having to change some existing markup!
Your feedback and comments are apppreciated and most welcome as


More information about the pmwiki-users mailing list