< mari
a
a
a
a
a
a
chi >
[ Page 3 of 6 ]
From: Nicholas Clark
Date: 21:45 on 07 May 2006
Subject: gcc
Dear gcc,
Please be noting in the language known as "C", this source code is not valid:
void foo ()
{
}
void bar ()
{
return foo();
}
and that a conformant "C" compiler *must* return a diagnostic.
C
You keep using that word. I do not think it means what you think that it means.
Nicholas Clark
From: Nicholas Clark
Date: 20:12 on 14 Feb 2006
Subject: The Internet
$ scp -p saigo.etla.org:bin/assimilate ~/bin
ssh: connect to host saigo.etla.org port 22: No route to host
What do you mean "No route to host"?
I'm only fucking logged in here from saigo.etla.org.
There must be a route for that error message to have reached my terminal.
And attempting to route from 1 machine (in France) to another (also in France)
via Seattle (not in France, not even in Europe) considered, well, brain dead.
Please just be working. Thanks in advance.
Nicholas Clark
From: Nicholas Clark Date: 12:23 on 18 Nov 2005 Subject: emacs In a text editor, which action I do more often? A: Set the font for the text? B: Go to a specific line? Place bets now! Correct answer was B. If you bet A, you were wrong. So why the fuck is the default action for Meta-G in emacs "set face"? RMS need to be hit around the head with a Larry Wall(*) until he grasps the principles of Huffman coding. Common actions should be shorter. Nicholas Clark *: Not the real Larry Wall. A 98% replica. Or how however the spam puts it. Heck, even a cardboard cutout would do for this job.
From: Nicholas Clark Date: 21:18 on 31 Oct 2005 Subject: Preview $ file ~/tmp/Printout /Users/nick/tmp/Printout: PostScript document text conforming at level 2.0 I don't care what its fucking name is. It's postscript. Damn well open it. Don't sit there with your pathetic dialogue box showing it greyed out because it doesn't conform to your numskull blinkered idea of how files should be named. I am the user and I am right dammit. Not Steve fucking Jobs. Nicholas Clark
From: Nicholas Clark
Date: 23:31 on 16 Sep 2005
Subject: perforce file type metadata
$ p4 sync ...@24049
//depot/perl/opnames.h#16 - updating /home/nick/p4perl/perl/opnames.h
Can't clobber writable file /home/nick/p4perl/perl/opnames.h
//depot/perl/reentr.c#17 - updating /home/nick/p4perl/perl/reentr.c
Can't clobber writable file /home/nick/p4perl/perl/reentr.c
//depot/perl/reentr.h#19 - updating /home/nick/p4perl/perl/reentr.h
Can't clobber writable file /home/nick/p4perl/perl/reentr.h
$ p4 edit opnames.h reentr.c reentr.h
//depot/perl/opnames.h#19 - opened for edit
//depot/perl/reentr.c#23 - opened for edit
//depot/perl/reentr.h#25 - opened for edit
$ p4 revert opnames.h reentr.c reentr.h
//depot/perl/opnames.h#19 - was edit, reverted
//depot/perl/reentr.c#23 - was edit, reverted
//depot/perl/reentr.h#25 - was edit, reverted
$ ll opnames.h reentr.c reentr.h
-rw-r--r-- 1 nick nick 9378 Jun 9 10:43 opnames.h
-rw-r--r-- 1 nick nick 16323 Jul 8 17:22 reentr.c
-rw-r--r-- 1 nick nick 78393 Jul 5 19:21 reentr.h
$ p4 sync ...@24049
//depot/perl/opnames.h#16 - updating /home/nick/p4perl/perl/opnames.h
Can't clobber writable file /home/nick/p4perl/perl/opnames.h
//depot/perl/reentr.c#17 - updating /home/nick/p4perl/perl/reentr.c
Can't clobber writable file /home/nick/p4perl/perl/reentr.c
//depot/perl/reentr.h#19 - updating /home/nick/p4perl/perl/reentr.h
Can't clobber writable file /home/nick/p4perl/perl/reentr.h
Hello. You mean you can't cope with VERSION CONTROLLING YOUR METADATA?
No. No. No. That would be non-hateful.
So instead, if I attempt to sync back before this one:
$ p4 describe 24775
Change 24775 by nicholas@ship-in-a-bottle on 2005/06/09 09:22:18
Change perforce filetype from text to text+w (so regen.pl is happy)
Affected files ...
... //depot/perl/opnames.h#19 edit
... //depot/perl/reentr.c#22 edit
... //depot/perl/reentr.h#24 edit
Differences ...
==== //depot/perl/opnames.h#19 (text+w/text) ====
==== //depot/perl/reentr.c#22 (text+w/text) ====
==== //depot/perl/reentr.h#24 (text+w/text) ====
you bleat. FUCKING SHUT UP and JUST DAMN WELL WORK.
Please?
Is that too much to ask?
Nicholas Clark
From: Nicholas Clark Date: 11:43 on 06 Sep 2005 Subject: svn diff So, I have a subversion checkout with some local changes. I think that they are just whitespace. Obviously the regular svn diff shows me lots of noise, but, hey, diff has an option to ignore whitespace, so surely svn diff can too... Its help says ... Valid options: ... -x [--extensions] arg : pass ARG as bundled options to GNU diff And gnu diff says that -b is the option I want so, let's see: $ svn diff -x -b svn: '-b' is not supported $ svn diff -x b svn: 'b' is not supported $ svn diff -x --ignore-space-change svn: '--ignore-space-change' is not supported $ svn diff -x ignore-space-change svn: 'ignore-space-change' is not supported $ svn diff -x=-b svn: '=-b' is not supported $ svn diff -x -- -b svn: invalid option character: b Type 'svn help' for usage. ggr. hateful dysfunctional thing. What's the point of telling me about something that doesn't work. Even perforce can manage this whitespace trick. Finally, a solution. Abuse one of the other hateful side effects of a svn checkout: $ diff -b lib/Vx/SOAP/DataStore/Storage.pm lib/Vx/SOAP/DataStore/.svn/text-base/Storage.pm.svn-base $ Yes. Just whitespace. Dear svn, why can't you tell me that? Nicholas Clark
From: Nicholas Clark Date: 17:36 on 30 Aug 2005 Subject: Hotel room keys So I'm attending a conference, and I'm staying in a room in a hotel (as you do) and I'm sharing a twin room with a friend (as is not unknown) and we both know a lot of other people attending and might actually have two separate lives (shock horror). The room has a swipe card lock. Moreover, you need the swipe card to make the lights work inside. So with one key you generally have to stick together... Us: Can we have a second key for room [XXX] please? Receptionist: Sure. But when I make the second key the first will stop working. Us: Oh. But we haven't lost the first key. Receptionist: But you can only have one key for the room. Us: Oh, why? Receptionist: Security, you see We politely say that we understand and thank him and decline the new key. Hateful hotel software. Nasty evil thing. Security be bollocks. Given that we needed no proof that we were the legitimate residents of room XXX, it's because the sub-muppet software is too damn stoooopid to write two keys to have the same magnetic strip. Or if the keys have fixed stripes, then it's the firmware in the door is tooooo damn stupid to be programmable to accept more than one guest-issue key. Either way, it's software. And it's hateful. Nicholas Clark
From: Nicholas Clark Date: 16:18 on 23 Aug 2005 Subject: gdb on Linux (gdb) b aio_proc Function "aio_proc" not defined. Make breakpoint pending on future shared library load? (y or [n]) y Breakpoint 1 (aio_proc) pending. (gdb) r Starting program: /home/nick/Reference/5.8.7-g/bin/perl5.8.7 -Mblib t/03_errors.t Breakpoint 2 at 0xb741b88f: file AIO.xs, line 279. Pending breakpoint "aio_proc" resolved Program terminated with signal SIGKILL, Killed. The program no longer exists. (gdb) where No stack. WTF? Oi, gdb, where's the fucking program? It's not small - it can't just disappear. Have you checked behind the sofa? Did it disappear up your own arse? With "tools" like these, who needs bugs? Nicholas Clark
From: Nicholas Clark Date: 19:20 on 21 May 2005 Subject: AIX patch Patching file ./cv.h using Plan A... patch: 3016-007 Cannot open file ./cv.h Hunk #1 ignored at 12. 1 out of 1 hunks ignored--saving rejects to ./av.h.rej errr? yes? Why did you put it in av.h.rej? Hateful thing. Can't cope with unified diffs either. Get New Utilities. Nicholas Clark
< mari
a
a
a
a
a
a
chi >
[ Page 3 of 6 ]
Generated at 10:26 on 16 Apr 2008 by mariachi