#pentagram@irc.freenode.net logs for 13 Apr 2003 (GMT)

Archive Today Yesterday Tomorrow
Pentagram homepage


[01:10:07] --> Kirben has joined #pentagram
[01:10:07] --- ChanServ gives channel operator status to Kirben
[02:36:14] --> Cashman has joined #pentagram
[03:11:44] --> CashmanZzz has joined #pentagram
[03:12:11] <CashmanZzz> heh! my connection was left on the server! and now I'm back with CashmanZzz and Cashman
[03:12:12] <CashmanZzz> eh
[03:12:25] --- CashmanZzz is now known as Cash
[03:13:04] <Cash> I just tried to compile pentagram again! - havnt compiled it for ages and I got problems
[03:14:47] <Cash> ucprocess.h - I got parse errors
[03:15:07] <Cash> just currently trying to compile object.cpp and application.cpp from kernel files
[03:20:36] <-- Cashman has left IRC (Read error: 60 (Operation timed out))
[03:20:57] --- Cash is now known as Cashman
[03:29:42] <Cashman> Kirben - are you around! does pentagram current cvs compile fine as I'm compiling with dev c++ and I have problems with ucprocess.h
[03:37:54] <Cashman> hmm I guess I'll wait and try and catch wjp or darke sometime..
[03:37:59] <-- Cashman has left IRC ()
[04:13:23] --> Cashman has joined #pentagram
[04:13:31] <-- Cashman has left #pentagram ()
[04:22:06] --> Cashman has joined #pentagram
[04:22:14] --- DarkeZzz is now known as Darke
[04:22:19] <Cashman> I fixed the prob!
[04:22:20] --- ChanServ gives channel operator status to Darke
[04:22:39] <Darke> Cool. What was the problem? *grin*
[04:22:57] <Cashman> compiler didnt tell me about process.h that was required!! and my lack of c/c++ knowledge screwed me up with that
[04:24:03] <Darke> Ahh.
[04:24:05] <-- Cashman has left IRC (Read error: 54 (Connection reset by peer))
[04:24:29] --> Cashman has joined #pentagram
[04:24:31] <Cashman> I've recreated the project for dev c++ - so yeah I have to change 1 or 2 things around! oh and that cvs checkout works great now!!
[04:24:47] <Cashman> have to go for now darke! talk l8er
[04:24:47] <Darke> Yay!
[04:24:50] <Darke> Bye!
[06:08:53] --> Cash has joined #pentagram
[06:09:28] <Cash> eeh has Cashman been in here long
[06:09:41] <Cash> because thats me - another screw up
[06:10:10] <Cash> hurry up and timeout you idiot
[06:10:24] <Darke> *grin* Have you registered that nick?
[06:10:35] <Cash> no!
[06:10:58] <Cash> I disconnect from pentagram well I disconnect from the internet actually on my dial-up connection
[06:11:10] <Darke> Register it with nickserv and you can ghost kill it to get it back.
[06:11:29] <Cash> aye!
[06:12:15] <Cash> how do I reg with nickserv?!
[06:13:03] <-- Cash has left #pentagram ()
[06:13:06] --> Cash has joined #pentagram
[06:13:20] <Darke> `/msg nickserv help` will tell you.
[06:13:51] <Darke> I think it's a case of changing to the nick you want (you'll have to let cashman timeout first...), then typing something like, `/msg nickserv register <password>`.
[06:14:59] <Cash> ok and does that last even if I disconnect! -will it be avaliable everytime I connect to freenode - is that what its for
[06:15:30] <Cash> hmm can you possibly kick cashman for me
[06:16:10] <Darke> In part. You can actually set it up so that someone else can't use your nick, but the default isn't that.
[06:16:39] <Darke> Won't help. He has to time out on the server, not on the channel. The only thing it'll do is kick him off the channel, and he'll still be lurking around the server until he times out.
[06:17:10] <Cash> hmm ok
[06:17:25] <-- Cashman has left IRC (Read error: 60 (Operation timed out))
[06:17:31] <Cash> timed out quick last time it happened
[06:17:34] <Cash> coolz
[06:18:06] <Darke> Sometimes it does, sometimes it doesn't.
[06:18:15] <Cash> -nickserv register
[06:18:18] <Cash> -nickserv register cashman
[06:18:28] <-- Cash has left #pentagram ()
[06:18:45] --> Cash has joined #pentagram
[06:19:52] --- Cash is now known as Cashman
[06:22:24] <Cashman> I got everything to compile except the filesystem.cpp!
[06:22:57] <Cashman> 333 C:\Nick\Pentagram\pentagram\project\filesys\FileSystem.cpp
[06:22:58] <Cashman> no matching function for call to `stat::stat(const
[06:27:38] <Darke> Veird. What compiler?
[06:27:58] <Cashman> dev using mingw
[06:28:59] <Cashman> I'll try and isolate the prob and get back to you
[06:29:01] <Darke> What's the rest of that error line? It's cut off at 'stat(const' here.
[06:29:27] <Cashman> 333 C:\Nick\Pentagram\pentagram\project\filesys\FileSystem.cpp
[06:29:28] <Cashman> no matching function for call to `stat::stat(const
[06:29:30] <Cashman> 333 C:\Nick\Pentagram\pentagram\project\filesys\FileSystem.cpp
[06:29:30] <Cashman> no matching function for call to `stat::stat(const
[06:29:57] <Cashman> what tha
[06:29:59] <Cashman> bugger
[06:30:04] <Cashman> I'll get back to you
[06:31:48] <Cashman> 333 C:\Nick\Pentagram\pentagram\project\filesys\FileSystem.cpp
[06:31:48] <Cashman> no matching function for call to `stat::stat(const
[06:31:59] <Cashman> ---------------------------------------------------------------------------
[06:32:00] <Cashman> filesys/FileSystem.cpp: In member function `bool FileSystem::IsDir(const
[06:32:01] <Cashman> std::string&)':
[06:32:01] <Cashman> filesys/FileSystem.cpp:335: no matching function for call to `stat::stat(const
[06:32:01] <Cashman> char*, stat*)'
[06:32:31] <Cashman> ---------------------------------------------------------------------------
[06:40:36] <Darke> Weird. Ok, I just grabbed the latest cvs and it works here. Time to do some more checking.
[06:41:12] <Darke> This, of course, is under linux, so that's probably the reason. *grin*
[06:43:02] <Kirben> worked fine under mingw too when I tried last night.
[06:46:16] <Kirben> Have you defined HAVE_SYS_STAT_H ?
[06:46:25] <Darke> The only thing I can think of, is that's it's not finding stat.h properly.
[06:46:50] <Kirben> mingw build should use -DHAVE_SYS_STAT_H -DHAVE_SYS_TYPES_H -DHAVE_UNISTD_H
[06:49:07] <-- Cashman has left IRC (Read error: 60 (Operation timed out))
[08:05:49] --> Cashman has joined #pentagram
[08:06:11] <Cashman> isp went down the drain for a while - they were doing maintenance
[08:06:35] <Darke> Ick.
[08:06:41] <Cashman> yeah strange but true
[08:07:11] <Cashman> anyway are you there Kirben?!
[08:07:29] <Cashman> hmmm 1hr 20mins 31secs
[08:07:36] <Cashman> idle
[08:10:07] <Cashman> filesys/FileSystem.cpp: In member function `bool FileSystem::IsDir(const
[08:10:08] <Cashman> std::string&)':
[08:10:08] <Cashman> filesys/FileSystem.cpp:335: no matching function for call to `stat::stat(const
[08:10:08] <Cashman> char*, stat*)'
[08:10:08] <Cashman> C:/Dev-Cpp/include/wchar.h:188: candidates are: stat::stat()
[08:10:08] <Cashman> C:/Dev-Cpp/include/wchar.h:188: stat::stat(const stat&)
[08:10:10] <Cashman> filesys/FileSystem.cpp:337: `S_ISDIR' undeclared (first use this function)
[08:10:12] <Cashman> filesys/FileSystem.cpp:337: (Each undeclared identifier is reported only once
[08:10:14] <Cashman> for each function it appears in.)
[08:10:16] <Cashman> make.exe: *** [OBJ/FileSystem.o] Error 1
[08:10:18] <Cashman> Execution terminated
[08:11:06] <Cashman> ek! icky! this is still that filesytem part - either I have somthing missing or its windows, who compiles under win32 besides me?
[08:12:14] <Cashman> hmm bb just reconnecting
[08:13:29] --> Cash has joined #pentagram
[08:14:11] <Darke> Did you see what Kirben said just before you disconnected lasttime?
[08:14:43] <Cash> no! just looking up the logs
[08:16:04] <Cash> ok if I have to define HAVE_SYS_STAT_H which I see in source! where do I add the switches!
[08:16:26] <Cash> aye! yeah I looked at Kirbens makefile
[08:16:44] <-- Cashman has left IRC (Read error: 54 (Connection reset by peer))
[08:17:43] <Darke> Just on the command line for make, there's probably an option in your dev-c++ project gui to set them.
[08:20:29] <Cash> bingo!
[08:20:34] <Cash> thanks a mil
[08:21:07] <Cash> now I'll just have to make myself a reference file for the future
[08:24:36] <Darke> Yay!
[08:26:15] <Cash> complete compile..... umm hopefully! lets see
[08:27:29] <Cash> I see there is still a bit of source! mdl loader which isnt being use yet right!! that screw's you up if you include it
[08:28:23] <Darke> Umm... it might. *grin*
[08:29:16] <Cash> 59 C:\Dev-Cpp\include\c++\backward\Map.h
[08:29:17] <Cash> In file included from C:/Dev-Cpp/include/c++/backward/Map.h
[08:29:22] <Cash> 41 C:\Nick\Pentagram\pentagram\project\kernel\Application.cpp
[08:29:22] <Cash> from kernel/Application.cpp
[08:29:24] <Cash> 2 C:\Dev-Cpp\include\c++\backward\backward_warning.h:32
[08:29:29] <Cash> #warning This file includes at least one deprecated or antiquated header. Please consider using one of the 32 headers found in section 17.4.1.2 of the C++ standard. Examples include substituting the <X> header for the <X.h> header for C++ includes, or <sstream> instead of the deprecated header <strstream.h>. To disable this warning use -Wno-deprecated.
[08:29:38] <Cash> !!! Now the project compiles - is this to be ignored this warning!!!
[08:33:46] <Cash> plus any special paths required now or just the normal u8 dirs and have fixedfont
[08:35:05] <Cash> hmm e.g. root\data for fixedfont.tga now! and root\static and root\usecode!?
[08:35:36] <Darke> Yes, you're free to ignore that warning. *grin* We probably should remove that header though at some point in time.
[08:35:57] <Cash> ok thanks
[08:36:36] <Darke> I can't remember them myself, I think there's a quick readme in there to tell you what you need.
[08:44:45] <Cash> Init SDL
[08:44:57] <Cash> Create Kernel
[08:44:57] <Cash> Create UCMachine
[08:44:57] <Cash> Create FileSystem
[08:44:57] <Cash> Create Configuration
[08:44:57] <Cash> Loading configuration files:
[08:44:58] <Cash> @data/pentagram.cfg... Failed
[08:45:00] <Cash> @home/pentagram.cfg... Ok
[08:45:02] <Cash> Reading "config/paths/data" config key.
[08:45:04] <Cash> Data Path: C:/Nick/Pentagram/pentagram/project/BIN/data
[08:45:06] <Cash> Reading "config/paths/u8" config key.
[08:45:08] <Cash> U8 Path: C:/Nick/Pentagram/pentagram/project/BIN
[08:45:10] <Cash> Set Video Mode
[08:45:12] <Cash> Resize Console
[08:45:14] <Cash> Load Palette
[08:45:16] <Cash> Load Shapes
[08:45:18] <Cash> Testing Pentagram...
[08:45:20] <Cash> Testing Ultima8 SKF...
[08:45:22] <Cash> Testing Ultima8...
[08:45:24] <Cash> Initialize World
[08:45:26] <Cash> Load Confont
[08:46:00] <Cash> eh! Fatal signal: Segmentation Fault (SDL Parachute Deployed)
[08:47:11] <Darke> At least it's working, eh? *grin*
[08:47:15] <Cash> got it working
[08:47:22] <Darke> Cool!
[08:47:46] <Cash> yeah so you just see some usecode stuff and the avatar in the middle?!
[08:47:55] <Cash> ok nice!
[08:48:26] <Cash> is this correct?
[08:50:46] <Darke> Yep.
[08:50:50] <Darke> Nothing much else happens.
[08:51:26] <Cash> just heaps of cant creat item etc..
[08:51:57] <Cash> ok yeah! - glad I got it compiled imbertween other things heh
[08:53:24] <Darke> That's pretty much all, yeah.
[08:54:01] <Cash> @data/pentagram.cfg... Failed - is this old! or a future plan?!
[08:55:59] <Darke> Hmm... future plan I think. I don't think we're loading the pentagram.cfg 'correctly' yet.
[08:56:58] <Cash> ok
[08:57:07] --- Cash is now known as Cashman
[08:57:21] --- Cashman is now known as Cash
[08:57:46] --- Cash is now known as Cash-away
[08:59:03] <Cash-away> I dont know jack about xml - but is that why you have xml, to handle ur config files?
[08:59:12] --- Cash-away is now known as Cash
[09:00:25] <Darke> Umm... err... someone decided to do it? *grin*
[09:00:32] <Darke> It's as useful as any other I suppose.
[09:01:21] <Cash> yeah it is useful! a nice standard in layout thats for sure
[09:35:08] <Cash> how are you guys far away from getting a map to display again! with the new code
[09:35:41] <Cash> I think I remember maps being display once apon a time
[09:35:46] <Cash> heh
[09:36:19] <Darke> Yeah, if you grab the /old/ cvs module, you can find a few displayer programs for maps/shapes/gumps.
[09:37:57] <Kirben> Or just use binary at http://exult.sourceforge.net/download.php
[09:39:11] <Cash> aye binary downloads now!
[09:39:27] <Cash> oh yeah exult
[09:39:34] <Cash> nothing on pentagram page yet
[09:39:58] <Kirben> at end of page is snapshots of old pentagram module.
[09:41:19] <Kirben> Current pentagram snapshots will be at http://pentagram.sourceforge.net/snapshots/ when i get go ahead.
[09:41:36] <Cash> I remember that actually yeah
[09:41:55] <Cash> ok look forward to new shots! when we get somthing to look at
[10:30:50] --> wjp has joined #pentagram
[10:30:50] --- ChanServ gives channel operator status to wjp
[10:47:58] <Cash> hey how are you wjp, I got pentagram compiled under dev again, are we possibly going to see a bit of action in the way of map display soon
[10:48:06] <Cash> I was looking back at the logs and I saw that it might happen
[10:48:19] <wjp> hi
[10:48:31] <wjp> well, it would probably be one of the next steps, yes
[11:01:10] --- Cash is now known as Cash-Away
[12:50:01] --- Cash-Away is now known as CashZzz
[13:08:25] <-- CashZzz has left IRC ()
[13:40:43] --> Colourless has joined #Pentagram
[13:40:43] --- ChanServ gives channel operator status to Colourless
[14:03:48] <wjp> bah... globs are going to be messy whichever way we implement them
[14:04:07] <wjp> even if we fully expand them into the object list, changing globs will be quite messy
[14:05:21] <Colourless> aye
[14:09:39] <wjp> still, that's probably the easiest/best way to handle them from what I can see now
[14:09:50] <wjp> a GlobEgg would have to keep a list of its contents
[14:17:19] <Colourless> use container functionality was my original idea, but it would need some 'hacking'
[14:17:51] <wjp> I think that would too easily cause unexpected side-effects
[14:18:01] <Colourless> aye
[14:18:07] <Colourless> something 'different' is required
[14:18:20] <Colourless> which isn't really a problem since we will not be saving in their format anyway
[14:18:45] <wjp> well, we do need to save them to preserve objIDs
[14:19:51] <Colourless> an 'idea'
[14:20:30] <Colourless> have the glob egg object. it then has a private field that says how many objects it has spawned. then the next 'n' objects after the glob egg in the object list are the globs spawned objects
[14:21:05] <wjp> that does assume we'll always be able to find n consecutive empty places in the list
[14:21:25] <Colourless> assumption: globs only created at map load :-)
[14:21:40] <Colourless> how about this then
[14:21:49] <Colourless> create a 'linked' list structure
[14:22:03] <Colourless> each object owned by the glob points to the next one
[14:22:25] <wjp> didn't you say earlier that sometimes globs were changed?
[14:22:47] <wjp> linked list would probably work
[14:22:48] <Colourless> moved = yes, changed not sure
[14:23:05] <wjp> hm, if they're only moved it would make things considerably easier
[14:23:43] <Colourless> i only know about u8 though remember
[14:23:44] <wjp> although this idea doesn't really prevent changing them
[14:23:56] <Colourless> assumptions about u8 may not relate to crusader
[14:24:29] <wjp> weren't crusader's maps more static?
[14:24:31] <Colourless> "void Item::setGlobNum(uint)", // Set Q (GlobEgg family) - Unused
[14:24:52] <wjp> interesting
[14:25:23] <wjp> so they did (or were planning to) support it, but never used it
[14:25:39] <wjp> we don't have any intrinsic names for crusader, right?
[14:25:44] <Colourless> no we dont
[14:26:09] <Colourless> probably decided it was too icky to handle safely
[14:27:53] <Colourless> for the linked list, it is probably going to be wanted to either link to the previous objects, or the 'owner' glob egg
[14:28:09] <Colourless> if an item gets deleted, we really don't want to screw up the linked list :-)
[14:28:27] <wjp> yes :-)
[14:28:50] <wjp> or we could have an in-glob flag that doesn't allow items to be deleted
[14:29:11] <Colourless> that would work
[14:29:26] <Colourless> just send them to the void if they get erased
[14:31:04] --> Cless has joined #Pentagram
[14:31:05] <-- Colourless has left IRC (Read error: 54 (Connection reset by peer))
[14:31:32] --- Cless is now known as Colourless
[14:31:36] --- ChanServ gives channel operator status to Colourless
[14:32:02] <wjp> do we want to use the same linked list structure for container contents when we have the field anyway?
[14:32:32] <wjp> hm, no, nvm... would cause the same issue with deleting :-)
[14:32:51] <wjp> (was just thinking of the original's use of the 'next' field)
[14:33:35] <Colourless> yes i know thats what I thought of it
[14:33:44] <Colourless> s/what/why/
[14:36:48] --> Cless has joined #Pentagram
[14:36:49] <-- Colourless has left IRC (Read error: 104 (Connection reset by peer))
[14:37:16] --- Cless is now known as Colourless
[14:37:20] --- ChanServ gives channel operator status to Colourless
[14:43:23] --> Cless has joined #Pentagram
[14:43:23] <-- Colourless has left IRC (Read error: 104 (Connection reset by peer))
[14:44:37] --- Cless is now known as Colourless
[14:44:38] --- ChanServ gives channel operator status to Colourless
[15:23:20] --> Cless has joined #Pentagram
[15:23:21] <-- Colourless has left IRC (Read error: 104 (Connection reset by peer))
[15:23:40] --- Cless is now known as Colourless
[15:23:42] --- ChanServ gives channel operator status to Colourless
[15:28:07] <-- Kirben has left IRC (Read error: 54 (Connection reset by peer))
[16:22:49] <wjp> hm, those lists have to consist of Item*'s, I guess (to prevent them taking up objIDs when not in the current map)
[16:22:56] <wjp> so we need to be careful with saving them
[16:23:24] <Colourless> what?
[16:23:27] <wjp> OTOH, we could just forget about them completely when they're not in the current map
[16:23:49] <wjp> and just re-expand the globegg when the map becomes active again
[16:24:09] <Colourless> as i said, discard the items on map changes
[16:25:04] <wjp> will be slightly tricky to get entirely right, but should be doable :-)
[16:26:10] * wjp adds EXT_GLOB flag
[16:26:32] <wjp> make that EXT_INGLOB
[16:34:50] <wjp> bbl, dinner
[16:46:19] --> Mat|lda has joined #pentagram
[16:47:13] <-- Mat|lda has left IRC ("changing servers")
[18:22:01] <wjp> back
[18:54:11] <Colourless> gone
[18:54:12] <-- Colourless has left IRC ("casts invisibility")
[19:59:32] --> Dark-Star has joined #pentagram
[21:03:13] --- Darke is now known as DarkeZzz
[21:03:22] <wjp> night
[21:03:39] * DarkeZzz goes to sleep after awakening, haveing forgotten to go to sleep before doing so. *grin*
[21:03:40] <wjp> or should I say morning? :-)
[21:03:58] <DarkeZzz> The latter is more technically correct. *grin*
[21:04:00] <wjp> you're not going to be sleeping at work, are you? :-)
[21:04:43] <DarkeZzz> Unfortunately, what with my work being so... active, no. *grin*
[21:06:17] <wjp> item lists aren't fun... *sigh*
[21:07:27] <DarkeZzz> Ah well. Goodluck!
[21:07:57] <wjp> I think the logic is correct atm... now I just need to implement it :-)
[21:15:30] * DarkeZzz snickers.
[22:59:02] <-- wjp has left IRC ("Zzzz...")
[23:44:46] <-- Dark-Star has left IRC ()