uglycode-coders — UM1 Coders

Re: [UM1-coders] trying to run netmud...

[J.P. K. <jp...@he...>]

> export LD_LIBRARY_PATH=/usr/local/lib

Or just ask.

I have added it to the default ld.so.1 path, and I have removed
the entry from uglymug's .profile.

> Adrian.

Julian

RE: [UM1-coders] trying to run netmud...

[Adrian S. J. <AS...@pa...>]

> From: Chisel Wright [mailto:ch...@he...]
>=20
> bash-2.03$ ./netmud -s start.db sanity.filler 8888
> ld.so.1: ./netmud: fatal: libstdc++.so.2.10.0: open failed:=20
> No such file or directory
> Killed
> bash-2.03$=20
>=20
>=20
> on something.
> any ideas?

export LD_LIBRARY_PATH=3D/usr/local/lib

Adrian.

[UM1-coders] trying to run netmud...

[Chisel W. <ch...@he...>]

bash-2.03$ ./netmud -s start.db sanity.filler 8888
ld.so.1: ./netmud: fatal: libstdc++.so.2.10.0: open failed: No such file or directory
Killed
bash-2.03$ 


on something.
any ideas?

bash-2.03$ find /usr/local/lib -name libstdc++.so.2\*
/usr/local/lib/libstdc++.so.2.10.0
bash-2.03$ 

Does solaris have an equivalent of ld.so.conf? Should/could /usr/local/lib be added to it?

-- 
Chisel Wright (ch...@he...)

I'm so unlucky, if there is such a thing as reincarnation,
I'll probably come back as myself.

[UM1-coders] cvs lock

[Chisel W. <ch...@he...>]

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/uglycode/UglyCODE/

The lock seems to have gone.
It's late though.

In the morning I will try checking out mudstring.h from um, then into UglyCODE.
Then we''l see if there's a problem at grim's end, sf's end, or if there are
no problems at all now.

Chiz
-- 
Chisel Wright (ch...@he...)

I'm so unlucky, if there is such a thing as reincarnation,
I'll probably come back as myself.

[UM1-coders] cgitelnet

[J.P. K. <jp...@he...>]

Ok, I believe that I have added cgitelnet as a subdirectory of the 
UglyCODE directory.

I appear to have finally committed enough to get it working, and up
to date.

Please feel free to test it, and make sure that it works for you.

Currently I have only tested it on Solaris, but I'll work on making
sure that it all works correctly on our other target platforms.  Just
not right at the moment.

Comments, etc, welcome.

Julian

Re: [UM1-coders] sticky?

[Chisel W. <ch...@he...>]

On Mon, Feb 04, 2002 at 10:12:28AM +0000, Chisel Wright wrote:
> cvs server: sticky tag `um1_029' for file `interface.c' is not a branch
> cvs server: sticky tag `um1_029' for file `tag_list' is not a branch
> cvs [server aborted]: correct above errors first!
> cvs commit: saving log message in /tmp/cvsBAA8LaGFd

My fault.

cvs up -A on my working copy fixed this.


-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

Re: [UM1-coders] . up

[J.P. K. <jp...@he...>]

> Odd, the error says -l is an illegal option, then lists it in the options....
> 
> bash-2.03$ . up 
> bash: ulimit: illegal option: -l
> ulimit: usage: ulimit [-SHacdflmnpstuv] [limit]
> bash: ulimit: illegal option: -m
> ulimit: usage: ulimit [-SHacdflmnpstuv] [limit]
> bash-2.03$ 

Yeah, this is my sloppy script :-)

I suspect that Solaris doesn't allow non-root users to set -l 
(The maximum size that may be locked into memory) and -m
(The maximum resident set size)

It worked on FreeBSD, it still works on Solaris, but spews errors.
I could fix it, but it is too low down my priority list at the moment.

If anyone fancies doing it then I recommend doing stuff like
OS=`uname`

switch "$OS"
...



Julian

[UM1-coders] . up

[Chisel W. <ch...@he...>]

Odd, the error says -l is an illegal option, then lists it in the options....

bash-2.03$ . up 
bash: ulimit: illegal option: -l
ulimit: usage: ulimit [-SHacdflmnpstuv] [limit]
bash: ulimit: illegal option: -m
ulimit: usage: ulimit [-SHacdflmnpstuv] [limit]
bash-2.03$ 

-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

[UM1-coders] sticky?

[Chisel W. <ch...@he...>]

cvs server: sticky tag `um1_029' for file `interface.c' is not a branch
cvs server: sticky tag `um1_029' for file `tag_list' is not a branch
cvs [server aborted]: correct above errors first!
cvs commit: saving log message in /tmp/cvsBAA8LaGFd



Clues?
-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

[UM1-coders] HALFQUIT

[Chisel W. <ch...@he...>]

I'm getting the following on something with the HALFQUIT code:

Program received signal SIGSEGV, Segmentation fault.
0x94d90 in shutdownsock__15descriptor_data (this=0x18dc68) at interface.c:2107
(gdb) list 2090,2110
2090                    if (next)
2091                            next->prev = prev;
2092    
2093                    delete (this);
2094                    ndescriptors--;
2095            }
2096            else
2097            {
2098                    Trace( "HALFQUIT descriptor |%d|%d| player |%s|%d| at |%02d/%02d/%02d %02d:%02d\n",
2099                    CHANNEL(),
2100                    channel,
2101                    getname (get_player()),
2102                    get_player(),
2103                    now->tm_year,
2104                    now->tm_mon+1,
2105                    now->tm_mday,
2106                    now->tm_hour,
2107                    now->tm_min);
2108            }
2109    }
2110    

(gdb) display now 
3: now = (tm *) 0x0

Does this mean 'now' is empty?

-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

Re: [UM1-coders] releasing

[Chisel W. <ch...@he...>]

On Mon, Jan 28, 2002 at 01:55:04PM -0000, Adrian St. John wrote:
> Forgot a bit. I'm quoting the entire lot again so that we've got
> a single email with the entire release procedure.

It's online now:
https://sourceforge.net/docman/display_doc.php?docid=9300&group_id=33517


-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

[UM1-coders] SF Package Releases (not game-tag, compile and run)

[Chisel W. <ch...@he...>]

I've partially engaged my brain; all releases now go into the 'uglycode'
package, and the release is based on the tag.
I've hidden all the other packages. For some reason there doesn't seem to be
a Delete Package option.

I think this allows people to monitor 'uglycode' and find out when there are
new releases.

I've just about finished package.sh, which

1. does connect to sourceforge, retrieve, and tar the source for a tag
2. echoes FTP instructions for transferring the file, then connects to the remote FTP server
3. gives URLs and an outline of how to release the file

I'd like to further automate the FTP (but not sure how is bash)

I've attached the script so far.


Chiz
-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

Re: [UM1-coders] Developer request

[J.P. K. <jp...@he...>]

> Keithy has given me another temp password, as you were
> heavily idle (ahem) at the time on Ugly.  I'll find you

Fair enough.  Shame, I think you'd have appreciated
the password I set for you. ;-)

> [Julian] there tomorrow to ask more questions.  (I was
Ok.  Although perhaps we should generate a FAQ for
new admin.

[So you've been banged up]
[So you've gone and been caught again]

etc.

> forced to wimp out of work early today owing to the fact
> that the Novocaine that the nice dentist lady gave me [when
> she was drilling half my face off] wore off and I couldn't
> think from the throbbing stabbing pain.  Huzzah for modern
> dentistry).
Boo to modern dentist receptionists who don't ring back and
don't tell you when your bloomin' appointment is.

> inw

Julian

Re: [UM1-coders] Developer request

[Ian W. <ia...@wh...>]

>>I will take 'inw', please, in order to be thoroughly
>>uncontroversial.  And to preserve what little remains of my
>>sanity.
>Ok. I have a (temporary) password for you. Page me on Ugly
>and I'll give it to you, or failing that sent me a crypted
>string, and I'll bung it in the password file.

Keithy has given me another temp password, as you were
heavily idle (ahem) at the time on Ugly.  I'll find you
[Julian] there tomorrow to ask more questions.  (I was
forced to wimp out of work early today owing to the fact
that the Novocaine that the nice dentist lady gave me [when
she was drilling half my face off] wore off and I couldn't
think from the throbbing stabbing pain.  Huzzah for modern
dentistry).


>P.S. Are you on the um1coders list? I keep CCing you in
>case you aren't.

I don't believe that I am, as I never get two copies.
Keithy, can you copy the inw...@wh... address
into the list?

inw

-- 
Ian Whalley                 <first name> @ <last name> . org

Re: [UM1-coders] Developer request

[J.P. K. <jp...@he...>]

> I will take 'inw', please, in order to be thoroughly
> uncontroversial.  And to preserve what little remains of my
> sanity.

Ok.  I have a (temporary) password for you.  Page me on Ugly
and I'll give it to you, or failing that sent me a crypted
string, and I'll bung it in the password file.

> inw

Julian

P.S. Are you on the um1coders list? I keep CCing you in case you aren't.

Re: [UM1-coders] Developer request

[Ian W. <ia...@wh...>]

>>Or alternatively, as Julian has specified that I am on the
>>list, I propose that FIan has an account on something.
>Ok.  FIan, what name would you like, we have the following
>models available
>inw
>fian
>bursar
>keithweloveyou

I will take 'inw', please, in order to be thoroughly
uncontroversial.  And to preserve what little remains of my
sanity.

inw

-- 
Ian Whalley                 <first name> @ <last name> . org

Re: [UM1-coders] Developer request

[J.P. K. <jp...@he...>]

> Or alternatively, as Julian has specified that I am on the list,
> I propose that FIan has an account on something.

Ok.  FIan, what name would you like, we have the following
models available

inw
fian
bursar
keithweloveyou

Other models are available on request.

> Adrian.

Julian

RE: [UM1-coders] Developer request

[Adrian S. J. <AS...@pa...>]

> From: Chisel Wright [mailto:ch...@he...]
>=20
> On Tue, Jan 29, 2002 at 11:08:13AM -0500, Ian Whalley wrote:
> > I am 'inw' on sourceforge (remarkably), and I would like to
> > be given the opportunity to, once again, modify Keith's
> > code.
>=20
> You're now added to the sf-project as a developer, with some=20
> admin rights,
> for poking around.
>=20
> I have a feeling I'm on the list of having some say in what=20
> happens on 'something'.
> Since I can't remember, I won't say.

Or alternatively, as Julian has specified that I am on the list,
I propose that FIan has an account on something.

And then it can be like the good-old-days, where we chatted with
write/wall when the game was down, discussing what Keith had
screwed up this time, and who was going to put it back up.

> Chisel Wright (ch...@he...)

Adrian.

Re: [UM1-coders] Developer request

[Chisel W. <ch...@he...>]

On Tue, Jan 29, 2002 at 11:08:13AM -0500, Ian Whalley wrote:
> I am 'inw' on sourceforge (remarkably), and I would like to
> be given the opportunity to, once again, modify Keith's
> code.

You're now added to the sf-project as a developer, with some admin rights,
for poking around.

I have a feeling I'm on the list of having some say in what happens on 'something'.
Since I can't remember, I won't say.

I wonder if I still have any relevant email(s)...

-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

[UM1-coders] Developer request

[Ian W. <ia...@wh...>]

Good afternoon;

I am 'inw' on sourceforge (remarkably), and I would like to
be given the opportunity to, once again, modify Keith's
code.

Huzzah!

inw

-- 
Ian Whalley                 <first name> @ <last name> . org

Re: [UM1-coders] releasing

[Chisel W. <ch...@he...>]

On Mon, Jan 28, 2002 at 01:55:04PM -0000, Adrian St. John wrote:
> 16. (If someone can explain exactly how to do it)
>     Make an official SourceForge release.

Wasn't this covered in my other thread?


-- 
Chisel Wright (ch...@he...)

Typer of code; tearer of hair.

RE: [UM1-coders] releasing

[Adrian S. J. <AS...@pa...>]

> From: Adrian St. John=20
>=20

Forgot a bit. I'm quoting the entire lot again so that we've got
a single email with the entire release procedure.

> In your development CVS directory (whichever machine you're working
> on, its all the same...):
>=20
> Before committing any code:
>=20
> 1. cvs update (make sure you've got the latest changes)
> 1a. If no changes then go to step 2.
> 1b. Go through the build & testing phase again, to make sure nothing
>     breaks.
>=20
> 2. Update 'tag_list' with a summary of what you've done.
>    If you're making a formal release, put the release information
>    in as well.
>=20
> 3. cvs commit
>=20
> 4. cvs tag um1_XXX
>=20
> To build a release for uglymug:
>=20
> 5. ssh something.uglymug.org.uk
> 6. su uglymug
> 7. cd ~uglymug/ugly/CVS/UglyCODE
> 8. cvs update -r um1_XXX
> 9. make clean (the dependancies aren't correct yet!)
> 10. make
> 11. mv netmud ../../netmud_XXX
> 12. cd ../..
> 13. rm netmud
> 14. ln -s netmud_XXX netmud

15. Email ug...@ug... with the following:
    a. Reason for release (important bug fix, I felt like it, etc.)
    b. The details from tag_list

16. (If someone can explain exactly how to do it)
    Make an official SourceForge release.

Adrian.

Re: [UM1-coders] Erm, it doesn't compile

[ReaperMan <re...@re...>]

Chisel Wright wrote:

> On Mon, Jan 28, 2002 at 06:49:23AM +0000, ReaperMan wrote:
> 
>>It could be that I am on gcc 2 and everyone else is on 3
>>
> AFAIK I'm using gcc2.
> 
> 
>>I checked in my changes (oh god) with this as is.
>>
> It wouldn't compile so you checked it in?
No I checked it out like that. I didn't modify

it and since you hadn't mentioned it I saw no
reason to change...



-- 
-------------------------------------------------------------------------------
                re...@re...         +44 7976 696 407
-------------------------------------------------------------------------------

RE: [UM1-coders] Erm, it doesn't compile

[Adrian S. J. <AS...@pa...>]

> From: ReaperMan [mailto:re...@re...]
>=20
> Well I am sure it compiles for everyone else otherwise it
> would be checked in and I know someone said that this
> doesn't compile on their system too:
>=20
>   terminal_type.clear();
>=20
> g++ -c -g -Wall -Wcast-qual -Wparentheses -Wwrite-strings=20
> -Wconversion=20
> interface.c
> interface.c: In method `enum Command_status=20
> descriptor_data::terminal_set_termtype(const CString &, int)':
> interface.c:4271: no matching function for call to `String::wclear=20
> (WINDOW *&)'
> *** Error code 1
> make: Fatal error: Command failed for target `interface.o'

The problem is curses. It defines a stupid macro called 'clear()'
that expands to 'wclear(SOME_DEFAULT)'.

The latest code doesn't use clear() anymore, or at least it shouldn't.

And please don't commit code that doesn't compile. If you can't get
it to work then ask someone as they might know why...

Adrian.

RE: [UM1-coders] releasing

[Adrian S. J. <AS...@pa...>]

> From: Chisel Wright [mailto:ch...@he...]
>=20
> I'm not sure if this is right; one bit I'm sure is wrong:
>=20

Lots of it is wrong; I'll explain.

In your development CVS directory (whichever machine you're working
on, its all the same...):

Before committing any code:

1. cvs update (make sure you've got the latest changes)
1a. If no changes then go to step 2.
1b. Go through the build & testing phase again, to make sure nothing
    breaks.

2. Update 'tag_list' with a summary of what you've done.
   If you're making a formal release, put the release information
   in as well.

3. cvs commit

4. cvs tag um1_XXX

To build a release for uglymug:

5. ssh something.uglymug.org.uk
6. su uglymug
7. cd ~uglymug/ugly/CVS/UglyCODE
8. cvs update -r um1_XXX
9. make clean (the dependancies aren't correct yet!)
10. make
11. mv netmud ../../netmud_XXX
12. cd ../..
13. rm netmud
14. ln -s netmud_XXX netmud

Adrian.