AGAIN XML::Twig tag conversion

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

Dear Monks

I have already asked same type of the doubt earlier. I have asked for converting the text splitted with tab space to elements.

Now im facing 2 problems. Even though i some what managed the first one, im unable to solve the second.

First one is:

Im having the input file separated by tabspaces

Input:
<thead>
abc   bcd   def
abc   bcd   def
abc   bcd   def
</thead>
OUTPUT:
<thead>
<row><col>abc</col><col>bcd</col><col>def</col></row>
<row><col>abc</col><col>bcd</col><col>def</col></row>
<row><col>abc</col><col>bcd</col><col>def</col></row>
</thead>
[download]

I have done it by using subs_text(qr{(.+?)\n}m,"row") and then for each row again doing subs_text(qr{(.+?)\t}m,"col"). the problem im getting is first two columns are get converted but not the last column because of absence of tab at last. I have managed to convert last column with "col" by getting last child of the row which is text and then wrap it with "col" value. Am I doing right. I think this will be done more easier than what i have done. If so How?

My second question is:

If i already have some elements inside the same content given above. For example,

Input:
<thead>
abc   <ex>bcd</ex>   <lp>def</lp>
abc   <ex>bcd</ex>   def
abc   bcd   def
</thead>
OUTPUT:
<thead>
<row><col>abc</col><col><ex>bcd</ex></col><col><lp>def</lp></col></row
+>
<row><col>abc</col><col><ex>bcd</ex></col><col>def</col></row>
<row><col>abc</col><col>bcd</col><col>def</col></row>
</thead>
[download]

For the above code each "\n" is for the row separation but i m not able to convert the input into the required output. I dont know how to select the content upto tab space. I face problem only when a tab space is between two elements.

Please give me suggestions for the above questions.

Thanks in advance

--Murugesan.

Comment on AGAIN XML::Twig tag conversion Select or Download Code

Replies are listed 'Best First'.
Re: AGAIN XML::Twig tag conversion by mirod (Canon) on May 14, 2004 at 08:34 UTC
I think you are tackling this problem the wrong way. You are using an XML tools to work on non-XML data. XML::Twig can help there, but as much as it costs me to admit it, it can't perform miracles ;--( Where does the data come from? The proper time to add tags to the `abc bcd def` lines is _before_ you create the XML, or before these lines get into the XML. You don't have necessarily to create the whole `row`/`col` structure, but at least tag each cell. This will be trivial in a non-XML context, and this way you won't have to fight with the data. If this is really the content you have to work with, then just add a first pass, where you add tags to those columns, by just taking the content of your `thead` element, using regexps to split it, and just `print` to create the proper XML: you will have to deal with escaping markup characters but that will be simpler than what you are trying to do at the moment.	[reply]
Re: AGAIN XML::Twig tag conversion by chanio (Priest) on May 15, 2004 at 05:28 UTC
Try something like this: `perl -e'$tx="aaa \t bbb \t ccc";@let=split(/\t/,$tx);print "<row>";s/\ +s//g,print "<col>$_</col>" foreach (@let);print "</row>\n"'` [download] After getting each row. Split and join deal with things in the middle of lists, ok? .{\('v')/} _`(___)' __________________________	[reply] [d/l]