On 11-02-15 02:09 PM, Jeroen T. Vermeulen wrote:
> Diverged implies current. Barring low-impact bugs, there are no non-current diverged messages.
Thanks for the info. It's surprising that you wouldn't want to know
which messages were previously been associated with a given template.
> So "diverged" is more specific than "current."
I don't understand why we have makeDivergedTranslation, then.
makeCurrentTranslation is more powerful, and according to you, diverged
translations must be current, so there's no reason to have both.
> Optimizing _setTranslation would be a lot more risky. We prefer to be very conservative with it.
Then you need more unit tests, so that you feel comfortable making
whatever performance improvements are necessary.
Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 11-02-15 02:09 PM, Jeroen T. Vermeulen wrote:
> Diverged implies current. Barring low-impact bugs, there are no non-current diverged messages.
Thanks for the info. It's surprising that you wouldn't want to know
which messages were previously been associated with a given template.
> So "diverged" is more specific than "current."
I don't understand why we have makeDivergedTra nslation, then. slation is more powerful, and according to you, diverged
makeCurrentTran
translations must be current, so there's no reason to have both.
> Optimizing _setTranslation would be a lot more risky. We prefer to be very conservative with it.
Then you need more unit tests, so that you feel comfortable making
whatever performance improvements are necessary.
Aaron enigmail. mozdev. org/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://
iEYEARECAAYFAk1 a0osACgkQ0F+ nu1YWqI1OOACfWB Qee2SdICZHcBfxk C1EQ/WH yHNheNj6psG9UzT S0LX0Orh
AFAAn00x/
=Kupi
-----END PGP SIGNATURE-----