Michael Meeks wrote:
On Fri, 2006-08-25 at 11:23 +0200, Stephan Bergmann wrote:
You got me wrong. I did not mean to suggest to automatically intern
each string upon construction. ("Intern" being the act of replacing a
string instance with a unique instance with identical content from the
cache.) I still wanted to add intern calls in all the places where
necessary, and merely have only one (singleton) implementation of intern().
Ah - nice; like the glib GQuark feature - of course, I love the idea; I
believe I've suggested it before, but I forget the consensus - clearly
chunks of code do this with custom code already (xmloff eg.).
Either way - someone greater than I [wrt. effecting real movement in
the infrastructure] needs to make the decision to push this, in the
meantime I'll keep tweaking.
As discussed, the single global solution has some drawbacks compared to
any local solutions (which can potentially get away with never removing
from the cache, or not being thread-safe), so a list of places that
already use or would most probably benefit from using this technique
would be interesting when making a decision.
-Stephan
[...]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
