On 18 May 2005, at 14:25, Basile Starynkevitch [news] wrote:

[snip the script]

> A typical output of the above script is
>
>   ab      "12"
>   cd      "45"
>
> where the file starts at the column a, and where keys and data are
> separated by a tabulation.
>
> My goal was to code the corresponding loading script gdbmload; the
> overall motivation for these 2 scripts is to manage under version
> control (Subversion) a GDBM file (by dumping it to & reloading it from
> a textual format, and by having a "canonical" dump format of it by
> sorting the keys; I don"t want to version control the GDBM binary file
> -because it is machine dependent- but a dump format of it, obtained
> thru this gdbmdump script and the opposite gdbmload script to be
> written)

Ah-ha!

You want YAML.  It gives you a great plain-text representation that  
you can load/dump with ease and works great with revision control.

-- 
Eric Hodel - drbrain / segment7.net - http://segment7.net
FEC2 57F1 D465 EB15 5D6E  7C11 332A 551C 796C 9F04