How is Berkley gdbm typically used?

Question

Gdbm is a simple key – value storage. I suspect that its use is based on various tricks. For example, from end-user perspective, I would like to have a record named user/project, that should have associated the following:

• protocol, git:// or http://
• domain like: github.com, bitbucket.com, etc. – forming url like github://user/project
• enabled / disabled
• method of loading – here 3 different methods like "native", "binary", "normal"
• main file – "main.c", or other
• ignore files – say "&" separated string with file paths to ignore
• depth – number of commits to download

I think this is enough. What I mean is that real world uses are sophisticated and Gdbm doesn't seem to defend itself. Is there a way to apply its use for above scenario?

Answer 1

Will this work? GDBM uses hashes, and if you use MLDBM you can store hashes of hashes as references.

%bobs_game_prj = { proto => git
  domain => blah.com
  ignore => \%ignore_files = { file1 => path, file2 => path}
  etc.}

I do something like this with perl, but I imagine it works in any programming language.