finding regex patterns in text file

[jessie]

hi,

i have a text file which contains information in this format..

[ -LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP ] [ said/VBD [ it/PRP ] [ and/CC [ Belgium/NNP ] [ 's/POS IMEC/NNP ] [ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ] [ to/TO develop/VB [ advanced/JJ metallization/NN process/NN technology/NN ./.In ] [ a/DT statement/NN ] [ late/RB on/IN [ Monday/NNP ]

i need to list the words that are in between the square brackets only...i mean the output should be something like

-LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP
said/VBD
it/PRP
Belgium/NNP and so on...

is it possible to do it with regex patterns? i tried but nothing worked... pls help..

thnks
jessie

[eRaaaa]

have you made a mistake?

[ said/VBD [ it/PRP ] : here you wrote that you want to find out said/VBD and
it/PRP

[ and/CC [ Belgium/NNP ] : here you only want the second one?

are your results wrong or your posted format?

[jessie]

hi,

now its corrected..sorry i posted the wrong one....

i have a text file which contains information in this format..

[ -LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP ] said/VBD [ it/PRP ] [ and/CC Belgium/NNP ] [ 's/POS IMEC/NNP ] [ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ] to/TO develop/VB [ advanced/JJ metallization/NN process/NN technology/NN ./.In ] [ a/DT statement/NN ] late/RB on/IN [ Monday/NNP ]

i need to list the words that are in between the square brackets only...i mean the output should be something like

-LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP
it/PRP
and/CC Belgium/NNP and so on...

is it possible to do it with regex patterns? i tried but nothing worked... pls help..

thnks
jessie

[eRaaaa]

ok summary: you have a long text with some text in brackets which do you want to extract? no nested brackets ? sorry but i ask that again, because you posted it again:

"[ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ]"

perhaps, if you have a pattern like [...] ... [...] ...[..] [..]... (no nested)
you could use the following:

		Pattern p = Pattern.compile("\\[(.+?)\\]");
		Matcher m = p.matcher(s);
		while (m.find()) {
			System.out.println(m.group(1).trim());
		}

s is your posted string

the output:

-LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP
it/PRP
and/CC Belgium/NNP
's/POS IMEC/NNP
a/DT three-year/JJ collaboration/NN agreement/NN
advanced/JJ metallization/NN process/NN technology/NN ./.In
a/DT statement/NN
Monday/NNP

[StormyWaters]

What's the output you would be expecting?

Here's basically what you currently have:

[ -LRB-/-LRB- Applied/NNP Materials/NNPS Inc/NNP ]
said/VBD
[ it/PRP ]
[ and/CC Belgium/NNP ]
[ 's/POS IMEC/NNP ]
[ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ]
to/TO develop/VB
[ advanced/JJ metallization/NN process/NN technology/NN ./.In ]
[ a/DT statement/NN ] late/RB on/IN [ Monday/NNP ]

How would you handle the following?

[ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ]

Would you just return the "a/DT three-year/JJ collaboration/NN agreement/NN"?

If there was a matching end bracket for the "had/VBD entered/VBN into/IN", so you had...

[ had/VBD entered/VBN into/IN [ a/DT three-year/JJ collaboration/NN agreement/NN ]]

...how would you want to handle that?