[pmwiki-users] TextExtract (Search recipe) update

Hans design5 at softflow.co.uk
Wed Sep 9 11:44:47 CDT 2009

Wednesday, September 9, 2009, 4:55:12 PM, Oliver Betz wrote:

> Excuse my ignorance, I don't know the problem with "highlighting two
> words of an AND condition" compared to "highlighting two words of an
> OR condition".

At present TextExtract search term input is passed as
a regular expression, more or less.
So input ABC|XYZ will look for ABC OR XYZ, and each term will  be
separately highlighted.
But input ABC XYZ will look for one string "ABC XYZ", not look to see
if both are present anywhere in the page, or paragraph, or line.
A search term entered as ABC XYZ would need to be split into two:
ABC and XYZ, and two regex searches need to be done.
And entering "ABC XYZ" with quotes will be treated as one string.
There is no regex pattern like ABC&&XYZ. So it gets complicated
to program a ABC XYZ input as "ABC" AND "XYZ" need to be present in
the text (page, paragraph, row) to count as a match.


More information about the pmwiki-users mailing list