Need Help With Java Regex

[DartmanX]

Hi, I'm trying to write a regex to match text in the following format:

Bld Rm
(several lines)
Page ## of ##

However, the following code fragment doesn't work:

fullDocument = fullDocument.replaceAll("Bld (\\w)+Page [0-1000] of
[0-1000]","");

(fullDocument is of type String)

I'm absolutely horrible with regex's, so any help would be appreciated.

Jason

 

[hiwa]

Hi, I'm trying to write a regex to match text in the following format:

Bld Rm
(several lines)
Page ## of ##

However, the following code fragment doesn't work:

fullDocument = fullDocument.replaceAll("Bld (\\w)+Page [0-1000] of
[0-1000]","");

(fullDocument is of type String)

I'm absolutely horrible with regex's, so any help would be appreciated.

Jason

public class DotAll{
public static void main(String[] args){
String fulldoc = "Never Let Me Go\n" +
"Small Crimes\n" +
"Bld Rm\n" +
"The Faceless System\n" +
"Market For The Death\n" +
"Page 201 of 1380\n" +
"Age of Abundance\n" +
"Medici Money\n";

String regex = "(?s:Bld.+Page \\d+ of \\d+\\n)";

System.out.println(fulldoc.replaceAll(regex, ""));
}
}

 

[Dale King]

Hi, I'm trying to write a regex to match text in the following format:

Bld Rm
(several lines)
Page ## of ##

However, the following code fragment doesn't work:

fullDocument = fullDocument.replaceAll("Bld (\\w)+Page [0-1000] of
[0-1000]","");

The [0-1000] in a regex does not mean any number between 0 and 1000. The
[] expression is a class of characters where you specify the characters
you want to search for and can use ranges. For example, [A-Za-z] will
specify any letter A-Z in either case.

In your case the expression [0-1000] equates to the characters from 0 to
1 or 0 or 0 or 0. So it only succeeds for the characters 0 or 1.

What you probably want is something like: Page \\d+ of \\d+

\d+ says to find one or more digits 0-9

 

[Harald]

Hi, I'm trying to write a regex to match text in the following format:

Bld Rm
(several lines)
Page ## of ##

However, the following code fragment doesn't work:

fullDocument = fullDocument.replaceAll("Bld (\\w)+Page [0-1000] of
[0-1000]","");

Several things need to be changed:
1) With "[0-1000]" you seem to indent to match integers in the given
range, but this is not what "[]" does. Rather, "[]" describes a set of
characters to be matched. What you want is "[0-9]+" instead to match a
sequence of digits.

2) (\\w)+ matches a word, but you want a repetition of pairs (space
word). This should be covered by "(\\s+\\w+)+" . In particular the \s
should match over end of line.

3) This does not account yet for the last space in front of "Page", so
you add another \\s: "(\\s+\\w+)+\\s+Page"

4) This still does not what you want, because always the longest match
is sought. If you have many records of the described structure in your
text, the match will be from the start of the first to the end of the
last of those. You may have luck with non-greedy matching:
"(\\s+\\w+)+?\\s+Page

And don't forget to set the flag "MULTILINE" when calling compile().

And don't believe a single word I wrote because normally I prefer my
own regex-package (see sig

Harald.