Extracting an XML element

[ddrillich]

Good Day,

We have a daily cron job that captures an XML document for statistics purposes. This large XML file ends up being in two lines.

I would like to extract the value of first occurrence of the documents element that looks like -

<documents>1615453</documents>

Any ideas?

Regards,
Dan

 

[Annihilannic]

Does grep -o '<documents>[0-9]+</documents>' filename.xml produce what you need, or does it match multiple items?

Annihilannic
[small]tgmlify - code syntax highlighting for your tek-tips posts[/small]

 

[ddrillich]

Thank you Annihilannic!

[autonomy@sbkj2ksrchdahs01 statistics]$ grep -o '<documents>[0-9]*</documents>'  stats_docs_11-01-2011 | head -1
<documents>1617752</documents>

One more thing - how can I get the number itself?

Regards,
Dan

 

[ddrillich]

Not much luck with -

[autonomy@sbkj2ksrchdahs01 statistics]$ grep -o '<documents>[0-9]*</documents>'  stats_docs_11-01-2011 | head -1 | grep -o '[0-9]*'
[autonomy@sbkj2ksrchdahs01 statistics]$

 

[feherke]

Hi

If you want only the value, it is simple with XmlStarlet:

xml sel -t -v '//documents' stats_docs_11-01-2011

By the way, your code at 2 Nov 11 11:20 works for me.

Feherke.

 

[ddrillich]

Feherke, weird as it doesn't work for me. Right, I should have such an app that maybe supports XPath...

Regards,
Dan