gm@romeo.cs.duke.edu (Greg McGary) (01/27/89)
I recently grabbed the GNU e?grep and compiled it with Turbo-C. After
clearing up some initial include-file problems, changing a couple of
int's to size_t's, and putting in tests which prevent calling
malloc(0), it ran O.K. It still fails to match any of the lines in the
khadafi test, and it crashes the machine on some of the Spencer-tests
(at least 72--75) It does succeed on Tests 1--72, tho I haven't gotten
around to testing beyond that...
Anyway, It does work pretty well for most of the kinds of things I want
to search for and it runs 3 to 4 *times* faster than the MKS egrep! I
would like to have a fully functional version, however. If no one has
done this, I may find some time over the next month or two to fix the
remaining porting-bugs...
-- Greg McGary
-- 4201 University Drive #102, Durham, NC 27707 voice: (919) 490-6037
-- {decvax,hplabs,seismo,mcnc}!duke!gm data: (919) 493-5953
-- gm@cs.duke.eduloafman@convexs.uucp (Kenneth W. Loafman) (01/29/89)
In article <13283@duke.cs.duke.edu> gm@duke.UUCP () writes: >... It still fails to match any of the lines in the >khadafi test, and it crashes the machine on some of the Spencer-tests >(at least 72--75) It does succeed on Tests 1--72, tho I haven't gotten >around to testing beyond that... What are the khadafi and Spencer tests? Are these Public Domain or part of a 'commercial' testing suite? Any information on how to acquire these, and perhaps a brief rundown of what they are would be appreciated. ----- Kenneth W. Loafman @ Convex Computer Corp, Dallas, Texas email: {allegra,uiucdcs,ctvax}!convex!loafman phone: (214) 952-0829
toma@tekgvs.GVS.TEK.COM (Tom Almy) (01/30/89)
In article <13283@duke.cs.duke.edu> gm@duke.UUCP () writes: >I recently grabbed the GNU e?grep and compiled it with Turbo-C. [...] >It still fails to match any of the lines in the >khadafi test, and it crashes the machine on some of the Spencer-tests >(at least 72--75) [...] > >[...] it runs 3 to 4 *times* faster than the MKS egrep! Gee, I can write a version that is even *FASTER*, lets say 100 times! (Providing it doesn't have to work!) Tom Almy toma@tekgvs.tek.com Standard Disclaimers (including :-) ) Apply
nelson@sun.soe.clarkson.edu (Russ Nelson) (01/31/89)
Yes. I have it fully working. I will put it up on grape.ecs.clarkson.edu in the uploads area. BTW, it's much faster if you use _read() instead of read(), but that breaks $ because $ is checking for \n, and _read will return \r at end of line. I didn't see any obvious way to convince $ to check for \r instead of \n. -- --russ (nelson@clutx [.bitnet | .clarkson.edu]) "I saved the whales!" - Rebecca L. Nelson, 3.5 years old, on receiving her Christmas present of a whale "adoption" certificate. Bless her liberal heart.