perl-docs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Per Einar Ellefsen <>
Subject Re: searching for $^T
Date Wed, 28 Aug 2002 07:26:46 GMT
At 05:09 28.08.2002, Stas Bekman wrote:
>Bill Moseley wrote:
>>On Tue, 27 Aug 2002, Stas Bekman wrote:
>>>Any idea how to make our search work for things like "$^T", if I 
>>>remember correctly that's not possible because the way it's indexed, right?
>>Hi Stas,
>>There's two ways to do that.  One would be to add "$" and "^" to swish-e's
>>WordCharacter list.  My guess is that would end up indexing too many
>>things we don't want indexed.
>>The second option is to use the BuzzWords feature and add the perl special
>>variables.  Buzzwords are accepted before words are split on
>Thanks Bill! I prefer the second option.
>>So if BuzzWords are used you will need to define Ignore[First|Last]Char
>>settings.  Probably something like
>>BuzzWords $^T
>>IgnoreFirstChar \"'
>>IgnoreLastChar .,;"'
>>Swish is case-insensitive, so that will get indexed as $^t.
>>The best thing is to setup a test document and run swish-e with -T
>>indexed_words trace option and see what swish decides are "words".
>well, I've tested it on the whole thing and guess what... it worked. 
>Though it doesn't get highlighted, but that's probably OK.
>Please tell me if the following patch seems to be OK. Mostly the 
>IgnoreFirstChar and IgnoreLastChar that I'm not sure about.
>What other buzzwords should we add?

What about the whole perlvar? Like that we're done with it. Oh yeah, and 
$r, $s and $c.

Per Einar Ellefsen

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message