On Sun, 7 Apr 2013 16:57:23 -0700, Sam Siegel wrote: >On Sun, Apr 7, 2013 at 4:16 PM, Steve Thompson wrote: >> > >> http://pubs.opengroup.org/onlinepubs/9699919799/utilities/patch.html >> >> <snip> >> >> I read this rapidly. I'm trying to understand certain headaches "we" have >> with certain products where it seems that you have to do a re-install to >> put on fixes [and I do not necessarily mean here at IBM]. >> >> I did notice that the "Patch" document refers specifically to a POSIX >> environment and "C" protocols as one would (at least me) expect to find in >> a *nix world. >> The POSIX environment has been available on MVS since 5.2.2 It's a couple levels down from that in the link: "diff -u" isn't available and you must rely on "diff -c". Similary effective; the patches are slightly more verbose.
The "C" involvement is that with correct options, "patch" will update a "C" source file by inserting "#ifdef" (conditional compilation) instructions. IEBUPDTE provides no corresponding facility to generate AIF, AGO, and ANOP instructions for HLASM. A delightful feature is that "patch" detects previously applied patches and prompts you for whether you want to ignore them or undo them. Most formats of "diff" output are revertible patches. >Patch is the unix version of SUPERC that shows the delta between source >files. > Slight correction. The UNIX "version of SUPERC" is "diff". (I suspect they use similar algorithms.) "patch" is the UNIX "version of" IEBUPDTE. >diff3 allows a 3 way merge between the original source and 2 different >updates to the original source. see: http://en.wikipedia.org/wiki/Diff3 > >I believe the point is that source maintenance can be done without the line >numbers. > Yup. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
