Alex Rousskov wrote:
On Sun, 2008-10-05 at 15:23 +1300, Amos Jeffries wrote:
All expected operational features are now in or about to go into trunk.
I'm going to start the 3.1 branch sequence shortly.
I thought we planned to do the source layout changes _before_ branching.
Did I misunderstood or did the plans change? It would be much easier to
shovel the code between trunk and a fresh branch if the code has the
same basic layout...
We did plan that initially. Though IIRC I was the holdout/veto on the
initial import. I've since been converted.
The only catch is that we all need to follow the maintenance cycle. To
do the merges back in order as they were added to HEAD in the first place.
As a part of this action, the major part of source code re-format will
take place. Which is expected to cause patch clashes. If you are aware
of bug patches needing to go into trunk please commit them.
Only existing Code is affected.
I'm aware of the following enhancement bugs with patches listed.
But need a confirmation that they are actually ready to go in.
2038 check reply_body_max_size before ICAP
2042 Propagate TCP_RESET to the client
2055 Honor Max-Connections
I can re-review these this week, but will need more than 24 hours your
other message seems to allow.
Okay. The branch happened slightly earlier than intended because Henrik
and I managed to get a rare few hours synchronized.
The re-format is still pending. So can wait for these.
Meanwhile I'm still distracted tracking down the new bugs.
I've got all the configure ones now. We are back at that "class
std::exception" issue you can't seem to reproduce.
Amos
--
Please use Squid 2.7.STABLE4 or 3.0.STABLE9
