RE: [SMARTY-DEV] Trouble compiling tpl from resource plugin

Blagh. that should be "ignores double byte characters." >-----Original Message----- >From: Gary Loescher [mailto:gary@loescher.org] >Sent: Tuesday, July 16, 2002 11:44 AM >To: Monte Ohrt >Cc: Jason Sweat; smarty-dev@lists.php.net >Subject: RE: [SMARTY-DEV] Trouble compiling tpl from >resource plugin > > >Well, I vote for urlencode. Basename only looks for >slashes (\ and /) and double byte characters, >according to the code. >Gary > >>-----Original Message----- >>From: Monte Ohrt [mailto:monte@ispi.net] >>Sent: Tuesday, July 16, 2002 11:12 AM >>To: gary@loescher.org >>Cc: Jason Sweat; smarty-dev@lists.php.net >>Subject: Re: [SMARTY-DEV] Trouble compiling tpl from >>resource plugin >> >> >>Using strpos, We are assuming that ":" is the only >>character we're >>looking for. Are there any other characters that >>basename wouldn't >>catch? In any event, urlencode is much faster and >>covers everything. >> >>Gary Loescher wrote: >>> Heh. I was just writing an email to suggest this >>> same code. Great minds, Jason :) >>> >>> Looks like we're stuck with munged filenames >>> again, or a performance hit. Personally, the >>> filename doesn't bother me. I actually like >>> being able to easily identify the resource >>> used to get the template. In the "file:c:/foo.tpl" >>> case, though, I guess that the price for >>> using an absolute path would be funny-looking >>> compiled template names. ;) Seems to me >>> using an absolute path is a bad practice, anyway. >>> >>> Gary >>> >>> >>>>-----Original Message----- >>>>From: Jason Sweat [mailto:jsweat_php@yahoo.com] >>>>Sent: Tuesday, July 16, 2002 10:36 AM >>>>To: smarty-dev@lists.php.net; gary@loescher.org; >>monte@ispi.net >>>>Subject: Re: [SMARTY-DEV] Trouble compiling tpl from >>>>resource plugin >>>> >>>> >>>>What about this ? >>>>if ($offset=strpos($string,':')) { >>>> $offset++; >>>>} >>>>$match = substr($string,$offset); >>>> >>>> >>>> >>>>Well we may have to use urlencode instead of trying to >>>>get the filename. >>>>preg_match and preg_split are get what we want but are >>>>much slower. >>>> >>>>repeated 100,000 times: >>>> >>>>urlencode(basename($string)); --> 2.920 >>>>secs >>>> >>>>preg_match('![^:|/\\\\$]!',$string,$match); --> 7.627 >>>>secs >>>>explode(':',basename($string)); --> 5.436 secs >>>> >>>>The second two get it right, but they take much >>>>longer. Unless there are >>>>any other ideas? >>>> >>>>Monte >>>> >>>>__________________________________________________ >>>>Do You Yahoo!? >>>>Yahoo! Autos - Get free new car price quotes >>>>http://autos.yahoo.com >>>> >>>>-- >>>>Smarty Development Mailing List (http://smarty.php.net/) >>>>To unsubscribe, visit: http://www.php.net/unsub.php >>>> >>> >>> >> >> >> >>-- >>Monte Ohrt <monte@ispi.net> >>Director of Technology, ispi Inc. >> > >-- >Smarty Development Mailing List (http://smarty.php.net/) >To unsubscribe, visit: http://www.php.net/unsub.php >