no-reply@allegro.cc (TeaRDoWN) — Mon, 03 Feb 2020 16:44:37 +0000

EDIT: Solved!

Years back I made a function that check for which mods are available in subfolders under a subfolder called "Mods" in my game's project directory. When finding a mod the name is stored and used for loading different assets.
However now some of the file system routines that previously worked fine are no longer doing the same thing. I don't know if it is a combination of Windows 10 + Allegro 5.0 or something else.
Anyone else know if new versions of Allegro have had changes made to adapt to file system changes in newer versions of Windows?

Here's a simplified code snippet of what I am doing and the results I get:

#SelectExpand
  1ALLEGRO_PATH *ap;
  2char *projectPath;
  3char *projectName;
  4...
  5while(true)
  6{
  7   ap = createAnAllegroPathToEachSubfolderUnderTheModsFolder();
  8   ...
  9   sprintf(modPath, "%s\\file.txt", al_path_cstr(ap, ALLEGRO_NATIVE_PATH_SEP));
 10   ...

projectPath will contain the correct full path "C:\\Projects\\MyProject\\Mods\\Zelda\\file.txt".

However when using al_get_path_tail() I do not get the last component in the path but the second to last:

#SelectExpand
 10   ...
 11   sprintf(projectName, al_get_path_tail(ap));
 12   ...
 13}

projectName will contain "Mods" instead of "Zelda".

Same result will happen when using al_get_path_component() with -1 (first from right) as component id:

#SelectExpand
 10   ...
 11   sprintf(projectName, al_get_path_component(ap, -1));
 12   ...
 13}

projectName will again contain "Mods" instead of "Zelda".

It's like the last component of the path is ignored but back in 2003 when I used the code it worked. The only thing I can think of that has changed is my Windows version.

Trying to find a newer version of Allegro (5.2.5) to try that but no luck finding binaries of this (yet).

---

EDIT: I just noticed that my ALLEGRO_PATH *get_next_subdirectory_in_directory() function used:
return al_create_path((char*)al_get_fs_entry_name(file));
Instead of:
return al_create_path_for_directory((char*)al_get_fs_entry_name(file));

So a bug due to me. However I don't think I've touched this code since 2013...aaand I think it worked back then. But who remembers. Oh well, problem solved!

File system routines - al_get_path_tail() returning wrong path component