jsin@seashell.seas.ucla.edu (Just Another John) (07/17/90)
In article <3448@tuminfo1.lan.informatik.tu-muenchen.dbp.de> rommel@lan.informatik.tu-muenchen.dbp.de (Kai-Uwe Rommel) writes: >I also ported GNU diff 1.14 and patch 2.0 to MS-DOS and OS/2. They work >well, I used patch for applying patches 1-9 to nethack 3.0, for example, >that's several megabytes of diffs and it worked and I am using diff also >regularly, but it fails on files with more than about 2000 lines but I >can live with this. Few month ago, I had ported RCS 3.? to DOS and got it to work ok with MKS Toolkit's diff.exe. It was pretty easy, once you decide how to deal with DOS's filename limitations. I hadn't really tested it thoroughly, so I don't know if my port was any good. If GNU diff supports -n option, perhaps someone can port the latest version of RCS to DOS. Any takers? I had planned on doing it when I have more spare time... -- John (Jonghoon) Sin (Above opinions are my own etc, etc, etc...) UCLA SEASnet Facilities InterNet: jsin@seas.ucla.edu 2567 Boelter Hall UUCP: ...!(uunet,ucbvax,rutgers)!seas.ucla.edu!jsin Los Angeles, CA. 90024 Phone: (213) 825-3556
alexande@drivax.UUCP (Mark Alexander) (07/19/90)
In article <886@lee.SEAS.UCLA.EDU> jsin@seashell.seas.ucla.edu (Just Another John) writes: >Few month ago, I had ported RCS 3.? to DOS and got it to work ok with >MKS Toolkit's diff.exe. It was pretty easy, once you decide how to deal >with DOS's filename limitations. I hadn't really tested it thoroughly, so >I don't know if my port was any good. If GNU diff supports -n option, >perhaps someone can port the latest version of RCS to DOS. > >Any takers? I had planned on doing it when I have more spare time... I recently completed a port of GNU RCS to MS-DOS, using Turbo C 2.0. I didn't use GNU diff, because I was worried about memory usage (it reads both files into memory). Instead, I used a public domain diff that keeps only the line hash tables in memory. It works well, but it is a bit slower. I haven't thoroughly tested all the features yet (such as three-way merges), but the basic commands seem pretty solid. If anyone is interested in the package, please reply by email. Maybe by then I'll have decided how to send out the source (email or ZOO file on floppy). -- Mark Alexander (uunet!drivax!alexande)
NU013809@NDSUVM1.BITNET (Greg Wettstein) (07/20/90)
I don't know about the rest of the group but I would think that everyone would appreciate at least the posting of the executable to this group. On a related note. I packaged up my port of Version 1.7 of GNU DIFF and shipped it off to Mr. Davidsen yesterday. Hopefully if all is in order it will be processed into the newsgroup in due time. As I mentioned in the accompanying documentation Version 1.7 is a little older than the 1.14 which was posted elsewhere but this one is completely clean as far as I have been able to test it. I use it extensively and it has no problems either with diff'ing two directories or with handling files larger than 65Kbytes. As to the memory usage required by diff this can be a problem although I have diff'ed two files of around 200,000 bytes each with minimal difficulties. Problems usually begin to arise when two directories are diff'ed against one another. Ever since I have been looking at GNU DIFF there has been an option to use disk rather than main memory to hold the files being differenced. Probably in this day and age of large virtual memories there has not been enough incentive for anyone to work out the code modifications necessary to do this. If someone was really interested in a spare time project (which we all have so much of) adding this as an extension to the currently active ports would enhance the usefulness of the utility greatly. After GNU DIFF clears the newsgroup I will package up the port of GNU Make for MS-DOS and ship it to the moderator. I will follow that with GNU sed, GNU tar and if there is interest GNU grep. Hopefully everyone will find them as useful as I have. As always, Dr. G.W. Wettstein `The truest mark of a man's wisdom is his ability to listen to other men expound their wisdom.'