henry@utzoo.uucp (Henry Spencer) (09/17/89)
Sigh. Sometimes life would be a whole lot simpler if patch didn't try
to be "smart", or at least if there were a way to tell it to stop thinking
and follow orders because we know what we're doing dammit... The part of
the 14-Sep-1989 patch that meant to turn the empty relay/regress/out/stderr
into a one-line file ends up creating cnpatch/tmp.file instead. Here is
a patch to the patch (!) to fix this. A corrected complete patch will be
next. The only adverse effects of applying the original 14-Sep-1989 patch
are the creation of a useless directory cnpatch and a useless file in it,
and the fact that the relaynews regression test will fail unnecessarily.
If you want a fix that doesn't involve re-applying patches, just do this:
cp cnpatch/tmp.file relay/regress/out/stderr
rm -rf cnpatch
*** 14-Sep-1989.orig Sat Sep 16 18:49:35 1989
--- 14-Sep-1989 Sat Sep 16 18:49:38 1989
***************
*** 1566,1572 ****
(readcnt=fread(block, 1, (int)min(art->a_unread, COPYSIZE), in)) > 0;
art->a_unread -= readcnt, art->a_charswritten += readcnt)
! *** cnpatch/tmp.file Thu Sep 14 15:54:41 1989
--- relay/regress/out/stderr Wed Sep 13 20:35:17 1989
***************
*** 0 ****
--- 1566,1572 ----
(readcnt=fread(block, 1, (int)min(art->a_unread, COPYSIZE), in)) > 0;
art->a_unread -= readcnt, art->a_charswritten += readcnt)
! *** cnpatch/tmp.nonexistent.file.with.very.long.name.for.stupid.patch Thu Sep 14 15:54:41 1989
--- relay/regress/out/stderr Wed Sep 13 20:35:17 1989
***************
*** 0 ****
--
V7 /bin/mail source: 554 lines.| Henry Spencer at U of Toronto Zoology
1989 X.400 specs: 2200+ pages. | uunet!attcan!utzoo!henry henry@zoo.toronto.eduamos@taux01.UUCP (Amos Shapir) (09/17/89)
In article <1989Sep16.225749.23376@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes: >If you want a fix that doesn't involve re-applying patches, just do this: > > cp cnpatch/tmp.file relay/regress/out/stderr >## rm -rf cnpatch ### NOT SO FAST!! ### Since the empty-file misfeature of patch causes it to ignore the next patch too, and since the 'old file' line specifies cnpatch/tmp.file, there's still the unapplied patches to take care of, in cnpatch/tmp.file.rej and cnpatch/tmp.file.rej.orig (fortunately, there were only two of these as far as I can see, or else we'd have to extract them out of the Patch-89-09-14 file). Anyway, do patch relay/sh/anne.jones < cnpatch/tmp.file.rej patch relay/relaynews.c < cnpatch/tmp.file.rej.orig If you already rm'ed cnpatch, the rejected patches are right after a line containing '* 0 *' which is not preceded by a very-long-name (in the old patch file, not the patched patch file...). -- Amos Shapir amos@taux01.nsc.com or amos@nsc.nsc.com National Semiconductor (Israel) P.O.B. 3007, Herzlia 46104, Israel Tel. +972 52 522261 TWX: 33691, fax: +972-52-558322 34 48 E / 32 10 N (My other cpu is a NS32532)
jeff@quark.WV.TEK.COM (Jeff Beadles) (09/18/89)
In article <1989Sep16.225749.23376@utzoo.uucp> henry@utzoo.uucp (Henry Spencer) writes: >Sigh. Sometimes life would be a whole lot simpler if patch didn't try >to be "smart", or at least if there were a way to tell it to stop thinking >and follow orders because we know what we're doing dammit... Well Henry, if you don't like it, write something better. I happen to LIKE patch, and am quite happy that Larry Wall has let us mere mortals have access to it. There is nothing "forcing" you to use patch. Try and develop a better way however, that most everyone can use. 1/2 :-) -Jeff -- Jeff Beadles jeff@quark.WV.TEK.COM
pst@anise.acc.com (Paul Traina) (09/18/89)
So Henry,
Is this new patch 16-Sep-1989, or is it 14-Sep-1989, or what is the
date of the patch to the patch?
Do we pretend that 14-Sep-1989 never occured, or the first one never
occured, or the second one is 16-Sep-1989, or the patch to the patch
is 16-Sep-1989?
By the way... :-) (nudge, nudge)
Cheers!
--
"I just forgot about how Americans are. They're always grinning. You can
always spot Americans here. They're the ones with this big sh*t-eating
grin on their faces!" [ the difference between Swedes and Yanks as observed
by David Walden's friend as reported in soc.culture.nordic ]henry@utzoo.uucp (Henry Spencer) (09/18/89)
In article <4575@orca.WV.TEK.COM> jeff@quark.WV.TEK.COM (Jeff Beadles) writes: >>Sigh. Sometimes life would be a whole lot simpler if patch didn't try >>to be "smart", or at least if there were a way to tell it to stop thinking >>and follow orders because we know what we're doing dammit... > >Well Henry, if you don't like it, write something better... "Patch: love it or leave it."? Sorry, I don't buy that argument. It's a very useful program that could be still more useful if there were a way to override some of its sometimes-excessively-smart features. -- "Where is D.D. Harriman now, | Henry Spencer at U of Toronto Zoology when we really *need* him?" | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
henry@utzoo.uucp (Henry Spencer) (09/19/89)
In article <2596@taux01.UUCP> amos@taux01.UUCP (Amos Shapir) writes: >### NOT SO FAST!! ### > >Since the empty-file misfeature of patch causes it to ignore the next >patch too, and since the 'old file' line specifies cnpatch/tmp.file, >there's still the unapplied patches to take care of... Groan. My apologies again; Amos is correct. -- "Where is D.D. Harriman now, | Henry Spencer at U of Toronto Zoology when we really *need* him?" | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
henry@utzoo.uucp (Henry Spencer) (09/19/89)
In article <1989Sep17.191213.18030@anise.acc.com> pst@anise.acc.com (Paul Traina) writes: > Is this new patch 16-Sep-1989, or is it 14-Sep-1989, or what is the > date of the patch to the patch? > Do we pretend that 14-Sep-1989 never occured, or the first one never > occured, or the second one is 16-Sep-1989, or the patch to the patch > is 16-Sep-1989? > By the way... :-) (nudge, nudge) The new one is the correct version of 14-Sep-1989. Pretend the old one was hit by line noise or something like that. :-) :-) -- "Where is D.D. Harriman now, | Henry Spencer at U of Toronto Zoology when we really *need* him?" | uunet!attcan!utzoo!henry henry@zoo.toronto.edu