nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

so this is getting complicated

nipper

So I was asked to write a perl script that would go though a bunch of DCRs and convert http://www.foo.com/ to http::/www.bar.com/

OK, easy enough.

Monkey wrench, from left field.

We have a related links section (replicant) of Name and URL.

format:

<item name="Related Links">
<value>
<item name="Name">
<value>M68HC08 Family</value>
</item>
<item name="URL">
<value>http://e-www.foo.com/blah.jsp?nodeId=12345</value>
</item>
</value>
</item>

Now they have asked to skip these.

Since my simple regex was not parsing the DCR I am not certain if this is possible.

If so it will be a hideously ugly regex.

I suspect I could do this if this was not a replicant, but when there can be 4 sets of URL/name, I do not know how to procede. Anyone care to point me in the correct direction ?

Andy

Find more posts tagged with

Comments

Migrateduser

I'm not clear on your current requirement. You want to replace some occurrences of "http://www.foo.com/", but not all. Which ones do you (or don't you) want to replace?

As has been mentioned many times in these Forums, simple regex patterns can sometimes be used when dealing with DCRs and XML files. But it is generally cleaner and safer to use modules that actually parse the documents.

Brinko Kobrin
Interwoven Staff Engineer

Migrateduser

Rewriting DCRs with the modules can be much more difficult than simply search and replace with regex - if you can get away with a regex, use a regex. I think in this case we just need the rule to be more clearly defined.

pflory

Andy,

Although it might be a tad bit more work, it's safer, more rewarding, and ultimately will be
easier to use xml parsing modules to do what you want. Actually, you'll want to use both
xml parsing AND regexes.

Example snippet. More in the online TS help on PTs.


use TeamSite::XMLparser;
use TeamSite::XMLnode;
        ### Parse the XML of the DCT... catch any problems in the eval block.
        eval {
            $rootnode = TeamSite::XMLnode->new($xml);
        };

        if ($@){
         ...  # error
        }
        else{
            ### Read the XML object elements.
            my $reviewDate = $rootnode->value('contributor.RevDate');
            ...
          }

See http://TSserver/iw/help/tst/pt/ on your TS server for additional examples.

Paul Flory

nipper

>I'm not clear on your current requirement. You want to replace some occurrences of "http://www.foo.com/", but not
>all. Which ones do you (or don't you) want to replace?

Yes that is the requirement. We have URLs in text areas and other fields. We also have URLs in an area called Related Links, which is a replicant of NAME/URL. I do not want to change those inside Related Links.

I think this just got easier since I found out that we do not use the field URL anywhere expect in Related Links. I was afraid I would have to look for
the Related Links tag and then find the close </item> (while matching 1 to n Name/URL pairs). Do not have to do that.

So after slurping this DCR into $txt:

<item name="Related Links">
<value>
<item name="Name">
<value>M68HC08 Family</value>
</item>
<item name="URL">
<value>http://e-www.foo.com/blah.jsp?nodeId=12345</value>
</item>
</value>
</item>

I want to strip out the URL item:

I was thinking a regex like this:

$txt = s|<item name="URL">(.*)</item>||g;

but I think I will get hit from regex being greedy and going to the last </item>

Tips/Pointers/RTFMs appreciated.

Andy

JonathonG

Add one char to your regex:

$txt = s|<item name="URL">(.*)</item>||g;

becomes

$txt = s|<item name="URL">(.*?)</item>||g;

The ? specifies minimal matching, reversing the default "greediness" of .*

Jonathon
Independent Interwoven Contractor