Parse multiple XML tags using regex

Question

I need to parse a few tags (their value) from am XML. This must be done by regex (don't ask me why :) )

For example:

 <name>AAA</name>
 <id>1234</id>
 <gender>M</gender>

I know the pattern needed for each tag using regex

string name =  "(?<=<name>).+?(?=</name>)";
string id = "(?<=<id>).+?(?=</id>)";
string gender = "(?<=<gender>).+?(?=</gender>)";

I just don't know how to init the Regex object to handle all of them.

I can do:

private static readonly Regex rgx1 = new Regex(name);
private static readonly Regex rgx2 = new Regex(id);
private static readonly Regex rgx3r = new Regex(gender);

but I'm guessing that's a terrible waste....

So my question is: how to init a single Regex to handle multiple patterns?

And once I did it, how to extract the values from it?

p.s: I'm programming in C# if anyone need to know....

10x alot !

Why don't you use Xml parser instead?

ysrb
– ysrb

2011-11-13 07:42:24 +00:00
Commented Nov 13, 2011 at 7:42 — ysrb
– ysrb, Commented Nov 13, 2011 at 7:42

misha · Accepted Answer · 2011-11-13 07:56:46Z

3

You can try this:

  string input = @" <name>AAA</name>
                                <id>1234</id>
                                <gender>M</gender>";
          string pattern = @"<name>(?<name>.+)</name>
                                <id>(?<id>.+)</id>
                                <gender>(?<gender>.+)</gender>";
          Match m = Regex.Match(input, pattern);
          Console.WriteLine(m.Groups["name"]);
          Console.WriteLine(m.Groups["id"]);
          Console.WriteLine(m.Groups["gender"]);

answered Nov 13, 2011 at 7:56

misha

2,8793 gold badges30 silver badges30 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Dominic Cronin · Accepted Answer · 2011-11-13 08:38:19Z

2

You say "don't ask me why", but I'm afraid I'm going to invoke programmer's prerogative and ask you why. If nothing else, because the solution will vary based on what the actual problem is. So for example, even using regexes, if you take misha's example (assuming it's fixed up to process the whitespace between the elements properly), it will only work on exactly the XML you posted.

In other words, with XML like this:

<name>AAA</name>
<id>1234</id>

the match would fail.

The purpose of XML is to allow for generic processing of this kind of data. Now sure, you can fix up the regex to make sure it deals with a missing gender tag, but if your real-world case is even a little more complex than your example, you will end up with a very complex regex indeed, and the responsibility for ensuring it performs well will fall on you. (Good quality modern XML parsers are highly tuned for good performance.)

So there you have it: to answer your question properly, we need to know the actual problem, and in this context, a constraint such as "you must use regexes" is quite interesting.

Say for example, that the XML in question isn't actually well-formed XML, so an XML parser would fall at the first hurdle. Knowing this would allow us to question whether the problem could be broken down into simpler parts, such as first extracting a well-formed XML fragment.

There could be other reasons, but whatever the reason is, it is crucial to the solution. Please share.

answered Nov 13, 2011 at 8:38

Dominic Cronin

6,2002 gold badges26 silver badges57 bronze badges

3 Comments

Dardar Over a year ago

in a nutshell the reason is performance issues. i need a very fast code and that is why i was instructed to use regex

Kirk Broadhurst Over a year ago

@Dardar might that be premature optimization? Is it proven that using an XML library will be too slow? Has it even been tested?

Dominic Cronin Over a year ago

I agree with Kirk. In practice, the only way to get high-performance code is to measure the performance of one or more implementations, and change your design as necessary depending on the results. While regexes can be very fast, it is also quite easy to write a regex that will perform very, very badly. This is expecially so when the logic of the match has to be general rather than specific, and you get a lot of backtracking. On the other hand, XML parsers can perform very well.

Michael Kay · Accepted Answer · 2011-11-13 14:41:02Z

2

You can't expect the kind of person who answers questions on this list to accept "don't ask me why" as a constraint. No self-respecting software engineer would accept a demand to use the wrong design for the task without first asking why.

answered Nov 13, 2011 at 14:41

Michael Kay

165k11 gold badges97 silver badges173 bronze badges

1 Comment

Mark Good Over a year ago

This is a comment, not an answer.

Koder101 · Accepted Answer · 2015-03-01 15:11:29Z

2

A more Generic Solution where you don't even have to know the Tags of the XML in advance :

static void Main(string[] args)
    {
        string path = @"C:\TestFile.xml";
        string input = File.ReadAllText(path);

        string pattern = @"<(.*)>(.*)</\1>";

        foreach (Match m in Regex.Matches(input, pattern))
        {
            System.Console.WriteLine(m.Groups[2].Value);
            System.Console.WriteLine("\n");
        }
    }

Use (m.Groups[1].Value) to get the name of XML Tag.

answered Mar 1, 2015 at 15:11

Koder101

9121 gold badge15 silver badges31 bronze badges

Collectives™ on Stack Overflow

Parse multiple XML tags using regex

4 Answers 4

Comments

3 Comments

1 Comment

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

3 Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related