From: pasp <pa...@ll...> - 2007-05-30 01:31:59
|
On Tue, 29 May 2007 23:40:08 +0200 "Tom Szilagyi" <tom...@gm...> wrote: > On 5/29/07, pasp <pa...@ll...> wrote: > > > > Since Aqualung rely on GLib I think we should to convert all standard > > type definitions to GLib portable equivalents (int to gint, char to > > gchar, strdup to g_strdup, etc.). > > This is definitely NOT going to happen. > > GLib data types (or the lack of them) pose no barrier to the > functionality of the > program. We use them when the code interacts with GLib, but otherwise > the proposed activity is entirely pointless and does not gear the source > towards the milestones we are trying to reach. Fixes no bugs, adds no > functionality. Forget it. > > Of course if you have more specific proposals about why using GLib > function instead of std C equivalent in this-and-that place is > *critical* then that's a different question... The only thing I wanted to obtain with this proposal was source code unification. Many GTK+ based apps I looked in uses "g_" types in all places. In other words, such approach is compatible with GLib paradigm: "...glib's various facilities are intended to have a consistent interface; the coding style is semi-object-oriented, and identifiers are prefixed with "g" to create a kind of namespace..." So, I don't see the point where we need to use GLib data types *only* for interaction with GLib. In this case we can use standard int/char/float, etc types. I know it is not required at this moment, because many more important things are still to do. I wanted to know your point of view. Now I know and I'm happy with it. > BTW thanks for asking :) Don't worry. I'm not going to spoil Aqualung without blessing from you :) > What we would really need ATM is a working metadata solution, not the > crap that TagLib is. I wish I had a couple of weeks to fix it once and > for all... OK, I don't want to bother you with such nonsense ideas at this moment. Time to do something required/useful :) Tomek |