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

Hi, I am fairly new to Perl and got a task where I need some help. I want to parse a xml file which has its own xml structure into an array and manipulate the data in the xml. Depending on other child information within the xml file I want to create another parent/child in it. I tried the package XML::SIMPLE and I can read out an XML but when I try to read my data by just selection the info I need it wont return a value. This is my dtd: (snipe) 
<!ELEMENT DEFJOB (JOB+)>

<!ELEMENT JOB (TAG_NAMES*,(TAG_NAMES | INCOND | OUTCOND | AUTOEDIT | Q
+UANTITATIVE | CONTROL | SHOUT | STEP_RANGE | ON)*)>
<!ELEMENT INCOND EMPTY>

<!ELEMENT OUTCOND EMPTY>

<!ELEMENT AUTOEDIT EMPTY>

<!ELEMENT QUANTITATIVE EMPTY>

<!ELEMENT CONTROL EMPTY>

<!ELEMENT SHOUT EMPTY>

<!ELEMENT STEP_RANGE EMPTY>

<!ELEMENT ON (DO|DOAUTOEDIT|DOSHOUT|DOFORCEJOB|DOCTBRULE|DOSYSOUT|DOIF
+RERUN|DOCOND|DOMAIL)*>
        <!ELEMENT DO EMPTY>
        <!ELEMENT DOSHOUT EMPTY>
        <!ELEMENT DOCOND EMPTY>
        <!ELEMENT DOAUTOEDIT EMPTY>
        <!ELEMENT DOFORCEJOB EMPTY>
        <!ELEMENT DOSYSOUT EMPTY>
        <!ELEMENT DOIFRERUN EMPTY>
        <!ELEMENT DOMAIL EMPTY>
        <!ELEMENT DOCTBRULE EMPTY>

<!ELEMENT TAG_NAMES EMPTY>
<!ATTLIST TAG_NAMES
        TAG_NAME CDATA #IMPLIED
                 CDATA           #IMPLIED>

[download]
Below is a xml example: Within the data-set I find the GROUP / NODEID from witch I need to create the new child "QUANTITATIVE NAME". 
<?xml version='1.0' encoding='ISO-8859-1' ?>
<!DOCTYPE DEFJOB SYSTEM "/aa/em/ddd/e62xstf3/resource/defjob.dtd">
 <DEFJOB >
  <JOB
   GROUP="xp_maint"
   NODEID="localhost"
   USE_INSTREAM_JCL="0"
>
   <QUANTITATIVE NAME="QR1-INIT" QUANT="1" />
   <QUANTITATIVE NAME="QR2-NODEID" QUANT="1" />
   <QUANTITATIVE NAME="QR3-GROUP" QUANT="1" />
/>
   </ON>
  </JOB>

[download]
My Question: 1. What would be the best package to use? XML::SIMPLE? I was thinking to read the data into an array, then to see if any "QUANTITATIVE NAME" is used, if not I would create new ones with the GROUP / NODEID. Thanks in advanced for any help

Replies are listed 'Best First'.
Re: Reading and Manipulation XML Files
by Anonymous Monk on Mar 26, 2010 at 11:02 UTC
    tried the package XML::SIMPLE and I can read out an XML but when I try to read my data by just selection the info I need it wont return a value.

    What? Also, your XML isn't valid.

    Read more... (1053 Bytes)
    Read more... (2 kB)

    1. What would be the best package to use? XML::SIMPLE?

    It depends. You can't go wrong with XML::Twig, XML::Rules, XML::LibXML...

    I was thinking to read the data into an array, then to see if any "QUANTITATIVE NAME" is used, if not I would create new ones with the GROUP / NODEID

    Um, try it and you'll see why its a bad idea, but if you can make it work, or like suffering .... to each his own :)

[reply]
[d/l]
[select]
      Thanks for your answer, much appreciated. I didn't want to spam the page so I didn't add the entire xml/dtd just a snipe of it. As it still feels somewhat hard to describe my problem technical and on the dot I tried not to over-spam. Now you did say that it is rather a bad idea, what would you suggest how I can solve this sleeker? The aim of this task is to prove a migration script that would read an extract (out.xml) and look for this QuantResources,and complete them depending on there NodeID/Group. Then the System would be able to control the components better. This is for a tool called ControlM. thanks again
[reply]
        ..so I didn't add the entire xml/dtd just a snipe of it.

        Thats fine, but you could have made it a valid snip, unless the purpose of your post was to troubleshoot your xml (which would be off topic , How (Not) To Ask A Question)

        . Now you did say that it is rather a bad idea, what would you suggest how I can solve this sleeker?

        Like I already said and showed. Using proper parser always beats hand parsing xml.

[reply]
Re: Reading and Manipulation XML Files
by Anonymous Monk on Mar 26, 2010 at 11:08 UTC
[reply]

Back to Seekers of Perl Wisdom