RegExp Doubts

[rotten]

Greetings,

I'm trying to write a grep command that prints all lines of a word file (represented by $WORDSFILE var) that begins with "something" and has precisely X (say 11) characters length.

I've managed with awk:

Code: Select all

 grep "^something" $WORDSFILE | awk '{ if (length() == 11) print $0}' 

But all my attempts using only grep seems to fail:

Code: Select all

--> grep -E "^something\{11\}" $WORDSFILE
--> grep -E "^something*\{11\}" $WORDSFILE
--> grep -E '^[something]\{11\}$' $WORDSFILE
--> grep '^something\{11\}$' $WORDSFILE

Could you kindly clarify the required regex syntax to accomplish this one using grep?

Thanks a lot for your support!
Best Regards,
rOttEna

[drgrussell]

grep -E "^something" | grep -E '^.{11}$'

otherwise you have to subtract the length of "something" from 11, and do "^something.{remainder}$" where remainder is the result of the calculation...

[rotten]

Thank you again, Mr. Gordon! That worked!

I'm having hard times constructing the regexp inside grep. For instance: when i want to match distinct character group, or subsets repetitions like the word "Berserker" which has two 'distinct' characters "er" repeating 3 times.. i'm using:

Code: Select all

 grep -i '\(\w\)\(\w\)[[:alnum:]]\+\1\2[[:alnum:]]\+\1\2' $DICTIONARY_FILE 

Which gives me a nice output when i run against the words file. 42 lines total, omitting some with ......).

• Christchurch
  Ouagadougou
  Panamanian
  Panamanian's
  Panamanians
  Tanzanian
  Tanzanian's
  Tanzanians
  ancestresses
  confrontation
  confrontation's
  .............
  fingerprinting
  incinerating
  incriminating
  indoctrinating
  infringing
  inseminating
  insinuating
  intercontinental
  interlinking
  intermingling
  pinpointing
  possessiveness
  possessiveness's
  priestesses
  restlessness
  .............

But i'm under the impression i'm still missing something..

I appreciate your attention and i'm looking forward for you answer.

Best regards,
rOttEn