Indeed.

Though, when using the syntax validator, it reports that **/*.meta and **/.meta are similar; that they’ll still grab the .meta files, and that’s essentially what I need.

**/*.meta (in this case vehicles.meta) recursively searches for the string .meta within any given subdirectories; essentially locating all existing files that are literally named .meta. Specifying **/*vehicles.meta would perform a recursive search for all files that are literally named vehicles.meta; not all .meta files, even without the specifying asterisk denoting all instances of the given string as shown.

All .meta files of each subdirectory would be located if only **/*.meta was specified, as shown.

As shown, **/.meta or in this case **/vehicles.meta, would recursively search for all files which are literally named .meta (or any given filename; vehicles.meta) within any given subdirectories, and so I wasn’t entirely sure whether it was actually necessary to retain the specifying asterisk denoting all instances of a given file here.

I’ll try to explain as best as I can.

Since __resource.lua requires each vehicle add-on’s .meta files to be defined in this part of the __resource.lua file, I figured I should still do that. I figured the same applied for the lower part of the file for each data_file entry. Where DATA/**/vehicles.meta, is written, for example, the __resource.lua file would perform a recursive search within the subdirectories of the /DATA/ folder for any files that are literally named vehicles.meta; the same for DATA/**/carvariations.meta, and so forth.

I’ve only partially experimented with this last night; specifying only DATA/**/*.meta produced the same result in my Windows server, and it successfully parsed the models without any issues. To avoid any arbitration while experimenting, I initially avoided specifying the asterisk for the entries in this section of the file. Adding the asterisk to the entries in this section while retaining DATA/**/ in every entry (i.e. DATA/**/*vehicles, carvariations.meta and so forth) yielded the same results. You may be right about the actual matter of difference in the entries’ possession of asterisks, but I’ll have to experiment more with this on the Linux server to verify that there even is one. I just know that the syntax itself could be problematic there.

:pushpin: Edit: Another user just confirmed that they experienced this very same issue on their own Linux server as well.
:pushpin: Edit: A fellow Linux enthusiast stressed to me that case-sensitivity was an important matter to consider when operating anything in its proprietary environment, and that I should have tried uncapitalizing everything within __resource.lua along with any relevant folders. Doing this resolved the entire issue.

So, for anyone who is viewing this thread and is experiencing this very specific issue—uncapitalize all relevant or pertinent data that is intended to be parsed within a Linux environment; this includes folders, files and corresponding data written within any files, due to Linux’s case-sensitivity.

:white_check_mark: Solution:
Your resource file, when using the aforementioned method I’ve discussed, should appear as such.

:pushpin: Edited 8/31 for referential accuracy.

1 Like