Re5: Estimating Vocabulary

On a linux 2.4.9 box running perl 5.6.0 perl /tmp/a /usr/share/dict/words 1 yields such things as:

databases
fritter
Saracens
stammerer
when
Whitmanize
willing
writes
Wuhan
youthfully
zigzag
Zoroaster
Zulu
Zulus
Zurich
wc -l could have told you this is 45424 words
[download]

shrivel
topologies
wetter
Wilkins
wristwatch
Yeager
yellowed
zoom
zooms
Zoroastrian
Zulu
Zulus
Zurich
wc -l could have told you this is 45424 words
[download]

valuably
wins
wriggles
Zennist
zoning
Zoroaster
Zulu
Zulus
Zurich
wc -l could have told you this is 45424 words
[download]

knave
requisitioning
seismology
sentimentally
tail
Telnet
Welles
Whipple
winner
workbooks
workmen
Yates
yeas
Yokohama
zodiac
zonally
zone
Zoroaster
Zoroastrian
Zulu
Zulus
Zurich
wc -l could have told you this is 45424 words
[download]

-- perl -pe "s/\b;([st])/'\1/mg"

Comment on Re5: Estimating Vocabulary Select or Download Code

Replies are listed 'Best First'.
Re: Re5: Estimating Vocabulary by I0 (Priest) on Mar 28, 2002 at 02:12 UTC
I can't duplicate your problem. It should be impossible since `$ARGV[0]-@lines` would be negative while `rand($line-$.)` is positive.	[reply] [d/l] [select]
Re7: Estimating Vocabulary by belg4mit (Prior) on Mar 28, 2002 at 02:16 UTC
heh heh heh, woops. just re-rechecked and it seems I somehow managed to merge the top of yours with the bottom (while) of mine. Doh! `-- perl -pe "s/\b;([st])/'\1/mg"`	[reply]