ToME 2.4.0-ah released

[AnonymousHero]

I'm starting to wonder if I shouldn't just provide a Docker container which the correct toolchain pre-setup just for compiling...

[Sky]

Broadly speaking ... what are the differences between TOME 1, 2, 3 and 4?

I played tome 1 and iirc it was similar to angband in style - although it had wilderness, colors, quests - many of the stats, commands, gameplay were similar.

But then i played tome 4 - Tales Of Copyright Infringment - that did away with the tolkien setting and changed pretty much everything.

When did the big change happen? What is the last angband-esque build?? (And where can i find it?)

[wobbly]

Tome 3 doesn't exist. It was Darkgod's 1st attempt at the new engine. He ended up abandoning it and writing T4. The 1st versions of T4 are still middle earth, though that's stretching it. The last version based on angband, you're probably in the thread for it? Not really an expert on Tome.

[Derakon]

ToME 2 is the last Angband-like ToME. This is the thread for the most recent release of that line.

ToME 1 was called PernAngband and got hit by a C&D from Anne McCaffrey's lawyers, so 2 stripped out the Pern references at the same time that it bumped the version number.

[Gwarl]

For apparently no reason at all I managed to build it this time, having done exactly the same thing I did last time.

However, I don't seem to be able to specify the user directory from the command line? I don't remember having this problem before.

[Gwarl]

grabbed the 2.4.0 release and I have the same problem.

looking at main.c I don't see anywhere that the user directory is parsed.

this makes no sense because I swear this worked before.

[AnonymousHero]

Originally Posted by Gwarl

grabbed the 2.4.0 release and I have the same problem.

looking at main.c I don't see anywhere that the user directory is parsed.

this makes no sense because I swear this worked before.

Are/were you using the TOME_PATH environment variable, or...?

AFAIR it should still work -- I certainly haven't done anything to remove it or anything like that.

[Gwarl]

In my archive of the old site, in the source folder I have, within a folder labelled tome2-2.3.6-ah a main.c file with the following function:

/*
* Handle a "-d<what>=<path>" option
*
* The "<what>" can be any string starting with the same letter as the
* name of a subdirectory of the "lib" folder (i.e. "i" or "info").
*
* The "<path>" can be any legal path for the given system, and should
* not end in any special path separator (i.e. "/tmp" or "~/.ang-info").
*/
static void change_path(cptr info)
{
cptr s;

/* Find equal sign */
s = strchr(info, '=');

/* Verify equal sign */
if (!s) quit_fmt("Try '-d<what>=<path>' not '-d%s'", info);

/* Analyze */
switch (tolower(info[0]))
{
case 'a':
{
string_free(ANGBAND_DIR_APEX);
ANGBAND_DIR_APEX = string_make(s + 1);
break;
}

case 'f':
{
string_free(ANGBAND_DIR_FILE);
ANGBAND_DIR_FILE = string_make(s + 1);
break;
}

case 'h':
{
string_free(ANGBAND_DIR_HELP);
ANGBAND_DIR_HELP = string_make(s + 1);
break;
}

case 'i':
{
string_free(ANGBAND_DIR_INFO);
ANGBAND_DIR_INFO = string_make(s + 1);
break;
}

case 'u':
{
string_free(ANGBAND_DIR_USER);
ANGBAND_DIR_USER = string_make(s + 1);
break;
}

case 'x':
{
string_free(ANGBAND_DIR_XTRA);
ANGBAND_DIR_XTRA = string_make(s + 1);
break;
}

#ifdef VERIFY_SAVEFILE

case 'b':
case 'd':
case 'e':
case 's':
{
quit_fmt("Restricted option '-d%s'", info);
}

#else /* VERIFY_SAVEFILE */

case 'b':
{
string_free(ANGBAND_DIR_BONE);
ANGBAND_DIR_BONE = string_make(s + 1);
break;
}

case 'd':
{
string_free(ANGBAND_DIR_DATA);
ANGBAND_DIR_DATA = string_make(s + 1);
break;
}

case 'e':
{
string_free(ANGBAND_DIR_EDIT);
ANGBAND_DIR_EDIT = string_make(s + 1);
break;
}

case 's':
{
string_free(ANGBAND_DIR_SAVE);
ANGBAND_DIR_SAVE = string_make(s + 1);
break;
}

#endif /* VERIFY_SAVEFILE */

default:
{
quit_fmt("Bad semantics in '-d%s'", info);
}
}
}

I need this to keep everyone's character dumps and pref files seperate from each other. The only I really need is "case u:", and some variants do indeed collapse the -d switch into just this case (NPPangband is the first example that springs to mind).

The usual purpose of the ANGBAND_PATH variable (still referred to as such in the main.c I apparently used last time) is specifying the lib folder and hence every subfolder not just the user folder. main_real in github's main.cc looks a lot like main in the main.c I have, but comes around 200 lines earlier.

Now I remember, I think I resorted to 2.3.6 before and then forgot that I had done so. That explains why kobold was complaining that there were still traps in the game.

I will need -duser=<path> or at least -d<path> command line switches to host it on live.

[Gwarl]

for reference, this is NPP's far more concise handling of only the switch I need:

/*
* Handle a "-d<what>=<path>" option
*
* The "<what>" can be any string starting with the same letter as the
* name of a subdirectory of the "lib" folder (i.e. "i" or "info").
*
* The "<path>" can be any legal path for the given system, and should
* not end in any special path separator (i.e. "/tmp" or "~/.ang-info").
*/
static void change_path(cptr info)
{
if (!info || !info[0])
quit_fmt("Try '-d<path>'.", info);

string_free(ANGBAND_DIR_USER);
ANGBAND_DIR_USER = string_make(info);
}

He didn't bother to update the comments

[AnonymousHero]

Oh, yes, I see I did actually remove that -- it shouldn't be too much effort to bring it back, though.

Just curious... are you using the GCU/curses frontend or something else to handle the frontend for angband.live?

(I'm doing a... thing to the T2 codebase and if at all possible, I'd like to be able to remove curses. However, if angband.live needs it, then that might convince me to try harder to keep it. Still might get removed, obviously, but...)