[ILUG] text flow tools

[Brian Foster]

  | Date: Wed, 27 Jul 2005 21:19:24 +0000
  | From: Paul Biggar <paul.biggar at gmail.com>
  | 
  | [ ... ]  Does anybody know a way of making this return a case where
  | word1 is at the end of one line, and word2 is at the start of the
  | next.

 just as I was crawling into bed, what may be a
 better, albeit longer (and rather more cryptic),
 solution occurred to me:

    sed -n -e '{:b;/word1[[:space:]]*$/{N;bb;}}' -e '/word1[[:space:]\n]\+word2/p'

 unlike the previous gawk(1) solution, that prints
 just the matching line(s), and does not care how
 yer paragraphs are written.  actually, it prints
 one line too many in the obscure case:

    line one ends with word1
    word2 starts line two also ends word1
    line three but no 2nd word

 in any case, it should work with any modern sed(1);
 i.e., any `sed' that supports EREs.

cheers!
	-blf-
-- 
Experienced (20+ yrs) kernel/software Eng: | Brian Foster   Montpellier,
 • Unix, embedded, &tc;  • Linux;  • doc;  | blf at utvinternet.ie   FRANCE
 • IDL, automated testing, process, &tc.   |  Stop E$$o (ExxonMobile)!
Résumé (CV) http://www.blf.utvinternet.ie  |     http://www.stopesso.com