egghelp/eggheads community

HELP! Regexp/Regsub

2012-12-05T15:57:27-04:00

Thanks for all the help. I got it working with some modifications to the code to not use a pubm and used a regular bind pub and still used the split text that Nimos provided.

I decided against the regexp method because of overlapping binds using pubm unless there is something i missed to stop this

Statistics: Posted by maphex — Wed Dec 05, 2012 3:57 pm

HELP! Regexp/Regsub

2012-12-05T12:00:57-04:00

Code:

   if {[regexp -nocase -- {^!.* .*$} $text]} {            set trigger [lindex [split $text] 0]      set search [lrange [split $text] 1 end]         regsub -all -- {\*} $search % search             set searchvar [list]       foreach word $search {          lappend searchvar db.`searchcolum` like '%[mysqlescape $word]%'"       }           do sql stuff like getting results         where [join $searchvar " AND " ]    AND SearchChannel LIKE '#%$searchchan%'    }

Statistics: Posted by Nimos — Wed Dec 05, 2012 12:00 pm

HELP! Regexp/Regsub

2012-12-05T02:22:23-04:00

Could you provide me with an example? I understand a little better when its laid out and can see how it works? Thanks for the fast response!

Statistics: Posted by maphex — Wed Dec 05, 2012 2:22 am

HELP! Regexp/Regsub

2012-12-04T22:15:11-04:00

Why not simply use [split $text] on the text.
Then [lindex $text 0] would be the first word and [lrange $text 1 end] would be all the other words, still as a list.

Statistics: Posted by SpiKe^^ — Tue Dec 04, 2012 10:15 pm

HELP! Regexp/Regsub

2012-12-04T21:22:33-04:00

So what im trying to do here is make a trigger that is based on a SQL entry.
So the first !(.*) will be the trigger.
Then I want the next (.*) to be the search term(s).

Everything works fine if the search term is one word but if the term is broken by a space like !searchchan search words here the script fails.

So I need to make it take the second (.*) and extend to the end of the line and save it as a var. Then the regsub will then take it and break it into usable content the rest of the script can use.

Code:

    bind pubm - * bind_search    proc bind_search {nick uhost handle channel text} {       filter_mirc_codesset regexp {^!(.*) (.*) *$}if {[regexp -nocase -- $regexp $text -> searchchan search]} {  regsub -all -- {\*} $search % searchset searchvar [list]foreach word [split $search] {lappend searchvar db.`searchcolum` like '%[mysqlescape $word]%'"}do sql stuff like getting results         where [join $searchvar " AND " ]AND SearchChannel LIKE '#%$searchchan%'}}

Added this just in case someone need to know theproc for the filter_mirc_codes

Code:

    proc filter_mirc_codes {} {      upvar text text      regsub -all -- "(\002|\017|\026|\037|\003(\[0-9\]\[0-9\]?(,\[0-9\]\[0-9\]?)?)?)"  $text "" text}

Big thanks in advance! I am open to other options that maybe fast or cleaner to the way I am doing it.

Statistics: Posted by maphex — Tue Dec 04, 2012 9:22 pm