Java Regex Multiline issue

Question

I have a String read from a file via apache commons FileUtils.readFileToString, which has the following format:

<!--LOGHEADER[START]/-->
<!--HELP[Manual modification of the header may cause parsing problem!]/-->
<!--LOGGINGVERSION[2.0.7.1006]/-->
<!--NAME[./log/defaultTrace_00.trc]/-->
<!--PATTERN[defaultTrace_00.trc]/-->
<!--FORMATTER[com.sap.tc.logging.ListFormatter]/-->
<!--ENCODING[UTF8]/-->
<!--FILESET[0, 20, 10485760]/-->
<!--PREVIOUSFILE[defaultTrace_00.19.trc]/-->
<!--NEXTFILE[defaultTrace_00.1.trc]/-->
<!--ENGINEVERSION[7.31.3301.368426.20141205114648]/-->
<!--LOGHEADER[END]/-->
#2.0#2015 03 04 11:04:19:687#+0100#Debug#...(few lines to follow)

I am trying to filter out everything between the LOGHEADER[START] and LOGHEADER[END] line. Therefore I created a java regex:

String fileContent = FileUtils.readFileToString(file);
String logheader = "LOGHEADER\\[START\\].*LOGHEADER\\[END\\]";
Pattern p = Pattern.compile(logheader, Pattern.DOTALL);
Matcher m = p.matcher(fileContent);
System.out.println(m.matches());

(Dotall since it is a Multiline pattern and i want to cover linebreaks as well) However this pattern does not match the String. If I try to remove the LOGHEADER\[END\] part of the regex I get a match, that contains the whole String. I don't get why it is not matching for the original RegEx.

Any help is appreciated - thanks a lot!

Regexes are probably not the best tool for the job here. Since you know the frontiers, just use a BufferedReader and use .readLine(). — fge
– fge, Commented Mar 17, 2015 at 11:48
Or if you want to rely on RegEx, try your RegEx with one of this tools — Zorian
– Zorian, Commented Mar 17, 2015 at 11:55
Did you read what matches() does? Are you sure you want to use this method? Maybe you are looking for find(). Also .* should probably be reluctant. — Pshemo
– Pshemo, Commented Mar 17, 2015 at 11:56

Wiktor Stribiżew · Accepted Answer · 2015-03-17 12:31:34Z

1

The important thing to remember about this Java matches() method is that your regular expression must match the entire line.

So, you have to use find() this way to capture all in-between  and n<!--LOGHEADER[END]/--:

String logheader = "(?<=LOGHEADER\\[START\\]/-->).*(?=<!--LOGHEADER\\[END\\])";
        Pattern p = Pattern.compile(logheader, Pattern.DOTALL);
        Matcher m = p.matcher(fileContent);
        while(m.find()) {
         System.out.println(m.group());
       }

Or, to follow the logics you suggest (just using matches), we need to add ^.* and .*$:

String logheader = "^.*LOGHEADER\\[START\\].*LOGHEADER\\[END\\].*$";
Pattern p = Pattern.compile(logheader, Pattern.DOTALL);
Matcher m = p.matcher(fileContent);
System.out.println(m.matches());

edited Mar 17, 2015 at 12:31

answered Mar 17, 2015 at 12:07

Wiktor Stribiżew

631k41 gold badges502 silver badges632 bronze badges

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

Comments

Avinash Raj · Accepted Answer · 2015-03-17 12:09:35Z

You actually need to use Pattern and Matcher classes along with find method. The below regex will fetch all the lines which exists between LOGHEADER[START] and LOGHEADER[END].

String s = "<!--LOGHEADER[START]/-->\n" + 
        "<!--HELP[Manual modification of the header may cause parsing problem!]/-->\n" + 
        "<!--LOGGINGVERSION[2.0.7.1006]/-->\n" + 
        "<!--NAME[./log/defaultTrace_00.trc]/-->\n" + 
        "<!--PATTERN[defaultTrace_00.trc]/-->\n" + 
        "<!--FORMATTER[com.sap.tc.logging.ListFormatter]/-->\n" + 
        "<!--ENCODING[UTF8]/-->\n" + 
        "<!--FILESET[0, 20, 10485760]/-->\n" + 
        "<!--PREVIOUSFILE[defaultTrace_00.19.trc]/-->\n" + 
        "<!--NEXTFILE[defaultTrace_00.1.trc]/-->\n" + 
        "<!--ENGINEVERSION[7.31.3301.368426.20141205114648]/-->\n" + 
        "<!--LOGHEADER[END]/-->\n" + 
        "#2.0#2015 03 04 11:04:19:687#+0100#Debug#...(few lines to follow)";
Matcher m = Pattern.compile("(?s)\\bLOGHEADER\\[START\\][^\\n]*\\n(.*?)\\n[^\\n]*\\bLOGHEADER\\[END\\]").matcher(s);
while(m.find())
{

System.out.println(m.group(1));

}

Output:

<!--HELP[Manual modification of the header may cause parsing problem!]/-->
<!--LOGGINGVERSION[2.0.7.1006]/-->
<!--NAME[./log/defaultTrace_00.trc]/-->
<!--PATTERN[defaultTrace_00.trc]/-->
<!--FORMATTER[com.sap.tc.logging.ListFormatter]/-->
<!--ENCODING[UTF8]/-->
<!--FILESET[0, 20, 10485760]/-->
<!--PREVIOUSFILE[defaultTrace_00.19.trc]/-->
<!--NEXTFILE[defaultTrace_00.1.trc]/-->
<!--ENGINEVERSION[7.31.3301.368426.20141205114648]/-->

If you do want to match also the LOGHEADER lines, then a capturing group would be an unnecessary one.

Matcher m = Pattern.compile("(?s)[^\\n]*\\bLOGHEADER\\[START\\].*?\\bLOGHEADER\\[END\\][^\\n]*").matcher(s);
while(m.find())
{

System.out.println(m.group());

}

Collectives™ on Stack Overflow

Java Regex Multiline issue

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related