in reply to index for large text file

Typeglob filehandles make my eyes bleed. I know they're out there in the wild. But if you can avoid creating new code that uses them you'll be happier in the longrun.

If you could post what your expected output is, as opposed to what you're getting, that would be helpful. Also, when you say all blocks are identical, do you mean identical? If that's the case, you have 30 gigs of millions of identically repeated four-line blocks? That doesn't make any sense. I re-read the question a few times and just couldn't come up with a concept of what you mean to say. ...probably my fault. But could you clarify what your dataset looks like, what output you're getting, and what you're expecting?


Dave

Replies are listed 'Best First'.
Re^2: index for large text file
by cafeblue (Novice) on Mar 28, 2011 at 07:11 UTC

    thank you, maybe I should not use the word "block", I should say pattern.

    like the text file I have given in the first post, the first line starts with a symbol "@", and the third line starts with "+",the other part are identical, but not identical in other four-lines.

    the even lines are of the same length in the whole text file.

    the whole file follow this pattern.

    I am a newbie. sorry for my careless.

[reply]

In Section Seekers of Perl Wisdom