24

I am looking for a good tool to perform grep-like operations on XML - for example, extract certain attributes only.

Grep itself can't handle it - any DFA-equivalent tool can handle only non recursive matches, and mine may be recursive.

I have tried xgrep, but it is quite unstable, and I want a stable and reliable tool.

Any recommendations?

EDIT: I prefer open source tools that work well under Linux.

5 Answers 5

22

XMLStarlet (Wikipedia) is a command line tool which comes close to grep.  It is open source software (MIT license) and works well on Linux and Windows.

The XMLStarlet website describes it as follows.

XMLStarlet is a set of command line utilities (tools) which can be used to transform, query, validate, and edit XML documents and files using simple set of shell commands in similar way it is done for plain text files using UNIX grep, sed, awk, diff, patch, join, etc commands.

The Debian/Ubuntu package is named xmlstarlet. But beware: Contrary to what the manpage says, the binary is named xmlstarlet in Debian/Ubuntu and not xml.

There are also Windows binaries on SourceForge.

For a nice little introduction, see IBM's Start working with XMLStarlet (now via archive.org).

4
  • Remove the trailing slash from the first link.
    – Bkkbrad
    Aug 23, 2009 at 20:27
  • I can't get it to work... It never matches on any xpath except '/' (the whole document), which is pretty worthless :( Jan 13, 2012 at 16:29
  • @HendyIrawan - Are you sure it's not how you're trying to use xpath? (Like your XML has a default namespace that you're not accounting for?) Mar 2, 2012 at 16:15
  • I tried to clone it, but it seems that the repository is broken. May 4, 2018 at 15:35
5

The XPath syntax in various languages is best for finding things in xml. In fact one of the tools recommended by the makers of xgrep is basically a Perl XML parser that accepts XPath input.

5

A tool that works under Linux is xml_grep. It fully understands XML and is not a line-by-line tool.

xml_grep is included as a stand-alone tool in the XML::Twig package. The grepping functionality is quite powerful as it supports XPath specifications.

Sample command-line (extracting posts edited after the middle of February from the triology Data Dump):

xml_grep -p --cond="row[@LastEditDate>'2010-02-14']"  posts.xml  > lateEditedPosts.xml

Installation is easy. Either

  • sudo cpan -i "XML::Twig", as described in the xml_grep cookbook referenced below.

or


More information:

The best introduction I have found for xml_grep is xml_grep cookbook, about two pages. Other:

1
  • I have fixed a broken link, but the triology Data Dump link is also broken. I will see what I can do. Dec 14, 2016 at 20:52
0

I would advise NOT to use a grep-like tool on XML, but use a library to parse XML in stead.
What exactly do you need it for? Any programming language? I think the .NET built-in XML parser would fit the job easily if you're willing to write a program for it though.

Update: for Linux, a well known XML parser library is libxml2.

-1

XMLSpy is an amazing tool, if a bit spendy.

Not the answer you're looking for? Browse other questions tagged .