Menu
▾
▴
uglycode-coders — UM1 Coders
You can subscribe to this list here.
| 2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(22) |
Dec
(53) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
(56) |
Feb
(12) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Peter C. <pet...@ne...> - 2001-11-28 12:14:24
|
> From: ReaperMan [mailto:re...@re...] > I'm working on a few, interface.c is one, can't really remember > the others. Oh yes, db.h, objects.h and c. Might be some more. Ah. The core of the system, then. > I would like to request that I am the next code change to > go in. I should have it finished soon once I ask Grim > what a socket is. It's what we plug you into if your fixes don't work. - Peter |
|
From: ReaperMan <re...@re...> - 2001-11-28 12:13:52
|
Adrian St. John wrote:
> Can I point out that the game now records the last 10 commands typed
> before a crash in an attempt to be able to recreate what happened,
> assuming the core dump is useless.
If you are looking for interesting things to fix then I saved
a few interesting stderrs in the log directory. Most of them are
probably null pointers but you may want to have a quick look.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-28 12:10:41
|
Adrian St. John wrote:
> Of course this is going to take a while to implement, and I will be
> touching
> every file, so if anyone is working on anything big, please tell me and
> I'll
> avoid touching the necessary files until they've been checked in.
I'm working on a few, interface.c is one, can't really remember
the others. Oh yes, db.h, objects.h and c. Might be some more.
As suggested in the code process thing, I would suggest that
you do not check anything in to the CVS at all until you
are almost ready but that is what I think you mean anyway.
I would like to request that I am the next code change to
go in. I should have it finished soon once I ask Grim
what a socket is.
Reaps.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: Peter C. <pet...@ne...> - 2001-11-28 12:01:12
|
> From: Adrian St. John [mailto:AS...@pa...] > Well, another day, another NULL pointer. Surprise. > As this is a reoccuring nightmare (I feel I'm typing value_or_empty in > my sleep!), I propose to overhaul the game to use a custom > string class > (not std::string because it is overkill) that handles NULL pointers > gracefully (ie it returns an empty string). I've always shied away from this beause it requires large, not-quite-systematic changes; but it's needed doing for about the last eight years. If you can do it, that would be marvellous! > Of course this is going to take a while to implement, and I will be > touching > every file, so if anyone is working on anything big, please > tell me and I'll > avoid touching the necessary files until they've been checked in. I'm working on the Network Inference classifier. Please don't touch it... :-) - Peter |
|
From: Adrian S. J. <AS...@pa...> - 2001-11-28 09:23:21
|
Well, another day, another NULL pointer. For your information, latest release (and 'running' version) is um1_004, which consists entirely of NULL pointer fixes since the last release. As this is a reoccuring nightmare (I feel I'm typing value_or_empty in my sleep!), I propose to overhaul the game to use a custom string class (not std::string because it is overkill) that handles NULL pointers gracefully (ie it returns an empty string). Of course this is going to take a while to implement, and I will be touching every file, so if anyone is working on anything big, please tell me and I'll avoid touching the necessary files until they've been checked in. Adrian. -- Adrian St. John Software Engineer Pan Security International Ltd. T 0121 780 0619 as...@pa... =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=20 Pan Security International Limited, Vienna House, International Square, Birmingham International Park, Bickenhill Lane, Solihull B37 7GN United Kingdom. Registered company number 3965174=20 This message may contain information which is confidential and subject to legal privilege. If you are not the intended recipient, you may not peruse, use, disseminate, distribute or copy this message. If you have received this message in error, please notify the sender immediately by e-mail, facsimile or telephone and return and/or destroy the original message. Pan Security International Limited accept no responsibility for any changes made to this message after it has been sent by the original author. |
|
From: Peter C. <pet...@ne...> - 2001-11-27 09:59:19
|
> From: Adrian St. John [mailto:AS...@pa...] [...] > To try and > investigate the bug I reset the @csucc and @cfail values to see were > they > necessary to avoid the crash... Resetting them also executed the > commands they had previously been linked to, which was odd. [...] > The second part, I suspect, is a side-effect of the fuse being on > Humblehope, > and of course they will trigger as you do things. But, since > it is part > of > the bug report, has anyone got any other suggestions as to why this is > true? If a fuse is STICKY, it executes after the command itself is run; if not, it executes before. I suspect this fuse is not STICKY. - Peter |
|
From: Adrian S. J. <AS...@pa...> - 2001-11-27 09:53:08
|
> From: Adrian St. John=20 >=20 > > From: Peter Crowther [mailto:pet...@ne...] > >=20 > > (Blink). What problem? Just so Peter gets the entire thing... BUG REPORT I'm sending this internally as I did not want to announce game crashing bugs over the um...@jm... list. Recently, our only worry was a dodgy array that only wizards could examine, so not much threat, this is different. A bug has recently appeared in @fuse. If you create a fuse without setting the ticks (@desc) or reset value (@drop) then examining the fuse crashes us. Both the ticks and reset value MUST be set before examining it. Other bugs were highlighted in trying to determine the nature of this bug. I created a fuse, set the ticks, reset value, and @csucc, @cfail. To try and investigate the bug I reset the @csucc and @cfail values to see were they necessary to avoid the crash... Resetting them also executed the commands they had previously been linked to, which was odd. So, @desc and @drop new fuses, and be careful of executing csuccs/cfails in resetting them. Humblehope. Now, the first part of the problem (description & drop NULL pointers) has been fixed. The second part, I suspect, is a side-effect of the fuse being on Humblehope, and of course they will trigger as you do things. But, since it is part of the bug report, has anyone got any other suggestions as to why this is true? Adrian. |
|
From: Adrian S. J. <AS...@pa...> - 2001-11-26 12:13:35
|
> From: Peter Crowther [mailto:pet...@ne...] >=20 > > From: Adrian St. John [mailto:AS...@pa...] > > Ok, since nobody has mailed round publicly what the problem is, > > I shall just say that I am fixing it now. Once done, I'll > > say exactly what I did, and how people can fix this in the future > > without me... >=20 > (Blink). What problem? The fuse, Solaris and number 0 problem. It is a common theme with all our recent crashes that the Solaris C library doesn't like NULL pointers as strings. > - Peter, who has obviously missed something I missed it until today because it was mud-mailed to avoid hitting any public email address. I don't agree with the policy, as we have suitable addresses that *should* be private to the admin, but it was somebody else... Adrian, typer of value_or_empty() |
|
From: Peter C. <pet...@ne...> - 2001-11-26 11:22:12
|
> From: Adrian St. John [mailto:AS...@pa...] > Ok, since nobody has mailed round publicly what the problem is, > I shall just say that I am fixing it now. Once done, I'll > say exactly what I did, and how people can fix this in the future > without me... (Blink). What problem? - Peter, who has obviously missed something |
|
From: Adrian S. J. <AS...@pa...> - 2001-11-26 11:18:58
|
Ok, since nobody has mailed round publicly what the problem is, I shall just say that I am fixing it now. Once done, I'll say exactly what I did, and how people can fix this in the future without me... Adrian. -- Adrian St. John Software Engineer Pan Security International Ltd. T 0121 780 0619 as...@pa... =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=20 Pan Security International Limited, Vienna House, International Square, Birmingham International Park, Bickenhill Lane, Solihull B37 7GN United Kingdom. Registered company number 3965174=20 This message may contain information which is confidential and subject to legal privilege. If you are not the intended recipient, you may not peruse, use, disseminate, distribute or copy this message. If you have received this message in error, please notify the sender immediately by e-mail, facsimile or telephone and return and/or destroy the original message. Pan Security International Limited accept no responsibility for any changes made to this message after it has been sent by the original author. |
|
From: J.P. K. <jp...@he...> - 2001-11-21 16:24:39
|
> Can you spot what's wrong with this picture? Time went backwards. The machine Ugly is running on is my router, and ntp is broken on it. An admin reset the clock by hand. The problem will disappear when something is back up and running. Julian |
|
From: John A. <jar...@in...> - 2001-11-21 16:19:27
|
Can you spot what's wrong with this picture?
Current Players: Idle
14s shah mat splange 1 e4 a6 2 d4 b5 (hysterical laughter) (W) 00s
-954s Frijj Magnutt Bored Operator From Hell... (B) 00s
57m 09s SalmonDays Disco watched the skies and fell asleep (BF) 8m 31s
1h 18m Charmarl (A) 8m 39s
1h 20m Parking Ticket Tarka - current ulcer count is 5 (Bw) 2m 41s
1h 22m Rincewind; the man, the myth, the legend... (Ww)12m 08s
1h 35m fantastic fifi knows sb loves her really (Bw) 12s
1h 44m Refresher visible(boss)?work(hard):spod(a,bit); (B) -1347s
Users: 8 (Peak 11) Up: 1 hour, 45 minutes.
--
"I broke my arm trying to fold a bed. It wasn't the kind that folds."
- Steven Wright
_____________________________________________________________________
I prefer encrypted mail (see headers for PGP key)
Why encrypt? http://www.heureka.clara.net/sunrise/pgpwhy.htm
_____________________________________________________________________
|
|
From: ReaperMan <re...@re...> - 2001-11-21 02:22:27
|
I will tag the current version in CVS and then
I am adding code to do an @recall function
which allows the user to recall the last n lines
of output.
Also this will allow people to reconnect to the
game at the position in the who list where they
left. They can then use @recall to see what they
missed.
Primary use is for those of us on 2 hour connection
limits.
Reaps.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-20 00:51:52
|
This could get complex.
I suggest we don't let it.
The CVS manual always insists that communication is the key
to the whole thing which is why I created the coders alias.
Suggested Rules and Procedures:
Rule 1 - Game only runs on tagged code.
Fix required to @version so that it shows the tag
Rule 2 - netmud.running sym link to netmud-release_00x
Keep previous netmud binary (minimum).
Procedure for coding:
We could either do it with tagged branches so that we
can fix branches while coders work on the head. Ick
for the moment I think. Suggest that tested fixes
and extensions are added to the head.
You have a bug/feature that needs fixing/adding.
1. Email the coders list with what you are going to do.
2. Get the latest code out.
3. Make your changes. Test changes locally or on
one of the Ugly servers. Do not check code in
to the CVS while you are developing. Take
local daily copies if you need backups.
If anyone else runs the whole cycle while you
are developing then you need to cvs update,
check your code and test it with their fixes
in as well before you move on.
4. Once you are ready for your change to go live
(or just be added to the repository) then email
the coder alias to indicate that you will be updating.
At this stage the code should compile, run and
have been tested.
5. If this is to be a code release then add that information
to the tag_list file (in the repository). If it
is not then just add the broad changes to the tag
list, ready for next version.
6. Check in your code (you may need to update first).
7. If a release tag head code with release number.
8. Email the coder alias that you have finished and
with details of your changes.
9. If a release then use the uglymug account to
checkout the code. Build on game machine. Move
binary to run directory using netmud-release_00x
and remake sym link. If urgent restart the game
with new binary. Alter motd.txt
4-8 should really be an 'atomic' operation on the CVS.
Until someone has said that they have completed
their submission then no-one else should be playing
with the CVS.
Things not to do:
- Don't check in code while you are developing. This
would prevent someone making an urgent fix before
you are ready.
- Don't not have backups of your code since it
is not in CVS until it is complete.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-19 01:34:43
|
Chisel Wright wrote:
> On Mon, Nov 19, 2001 at 01:06:54AM +0000, ReaperMan wrote:
>
>>And I have added Refresher since he does stuff too.
>>
> OK
>>Freddy_C has also started doing some coding.
>>Any objections to adding him to source forge and
>>this list?
>>
> None - reaps, I gave you permissions to screw up the project accordingly on SF didn't I?
You did. I was just being polite...
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-19 01:08:01
|
And I have added Refresher since he does stuff too.
Freddy_C has also started doing some coding.
Any objections to adding him to source forge and
this list?
Reaps.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-19 01:05:02
|
Dear Coders,
I have created this list to discuss UM1 code issues and
processes. It runs through sourceforge for the moment
but please can you use the uglymug address since that
will be it's permanent entry point.
This list contains only the four people currently
in the SF list and also Ozzard.
Reaps.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
|
From: ReaperMan <re...@re...> - 2001-11-19 00:52:34
|
If you get this message please email me,
Thanks,
Keith.
--
----------------------------------------------------------------------
re...@re... www.reaperman.org +44 7976 696 407
----------------------------------------------------------------------
|
