replacing continuesly occuring pattern

Ananda has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: replacing continuesly occuring pattern by CountZero (Bishop) on Apr 16, 2005 at 09:34 UTC
Several options are open: Painstakingly go through your xml-file with a regex on a line-by-line basis, selecting everyting between `<name></name>`-tags and pushing this into an array and finally outputting the array between a `<names></names>` tag. This may work, but is considered prone to breaking if your xml-file spans the tags over multiple lines, or your tags will include other tags, or ... . Much better is it to use a module like XML::Parser which will parse your xml-file and allow you to extract whatever you want from the xml-file Or you could go the XSLT-way and transform your xml-file into another xml-file. Excellent XSLT-modules exist, such as Xml::Sablotron or XML-LibXSLT or use XML::XSLT::Wrapper which provides a consistent interface to various XSLT-engines. CountZero "If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law	[reply] [d/l] [select]
Re: replacing continuesly occuring pattern by hv (Prior) on Apr 16, 2005 at 15:39 UTC
It's generally not considered a good idea to manipulate XML with regular expressions - dedicated XML parsing modules tend to do a better job. That's said, here's one way to do it: `1 while $text =~ s{ <names?> ([^<]) </names?> <name> ([^<]) </name> }{<names>$1,$2</names>}gx;` [download] This assumes that the 'name' tags cannot contain additional XML entities. It also isn't particularly efficient, but should be fine if your strings aren't longer than a few KB or so. Hugo	[reply] [d/l]
Re: replacing continuesly occuring pattern by Tanktalus (Canon) on Apr 16, 2005 at 16:48 UTC
My favourite XML method ... XML::Twig! ;-) Read more... (2 kB) There is probably an easier way to do this with XML::Twig using the twig handlers, but those confuse the heck out of me, so I try to avoid them ;-) I find this method, even if a bit longer, and probably slower and more memory-consuming, to be easier to understand. To me, it has a much higher degree of WUD (thanks, friedo, for that term!) :-)	[reply] [d/l] [select]
Re: replacing continuesly occuring pattern by friedo (Prior) on Apr 16, 2005 at 09:34 UTC
Something like this ought to do the trick: `$xml =~ s/<(.*?)name>/<$1names>/g;` [download] [id://CountZero] is right, I misread the original question. I thought the OP simply wanted to replace the "name" tags with "names." Oops.	[reply] [d/l]
Re^2: replacing continuesly occuring pattern by CountZero (Bishop) on Apr 16, 2005 at 09:43 UTC
Are you sure that works? When I tried it I get as a result: `<names>abc</names><names>def</names>` which is not what was requested. CountZero "If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law	[reply] [d/l]