headers
Remy Maucherat | 1 Sep 02:08
Picon
Favicon
Gravatar

Re: svn commit: r571578 - in /tomcat/trunk/java/org/apache: catalina/CometEvent.java catalina/connector/CometEventImpl.java catalina/connector/Request.java coyote/http11/Http11NioProcessor.java tomcat/util/net/PollerInterest.java

fhanik <at> apache.org wrote:
> Author: fhanik
> Date: Fri Aug 31 13:59:00 2007
> New Revision: 571578
> 
> URL: http://svn.apache.org/viewvc?rev=571578&view=rev
> Log:
> Move back to original idea, use primitive int instead of enum, extremely inefficient using enums and
arrays to hold them. 

Good. Two of the biggest gropes I had in this portion of the API and 
impl are now gone (after arguing they were great ideas, and ignoring my 
complaints about them for months :( ). How about adopting my interface ? 
It's clearly the next step in the evolution.

The main problem continues, though: API changes are being made without 
being discussed in any way beforehand. I think this needs to stop, and I 
will restart the vote for moving this branch to the sandbox.

Rémy

PS: The fix for the close problem is horrible, but good job on finding 
the bug.
Permalink | Reply | 2 comments |
headers
Remy Maucherat | 1 Sep 02:09
Picon
Favicon
Gravatar

[VOTE] Move trunk to sandbox

Hi,

Due to continuing independent development in trunk by Filip, I am 
calling for a second vote to move it to the sandbox, where Filip could 
continue developing his ideas in a non disruptive way. I think the whole 
thing has been debated pretty extensively in the previous vote.

Following this, debate should take place to determine if a new main 
trunk is needed, and if commit procedure changes would be good 
(personally, I do like Jean-Frédéric's ideas).

<ballot>
[ ] +1
[ ] 0
[ ] -1
</ballot>

Rémy
Permalink | Reply | 9 comments |
headers
Filip Hanik - Dev Lists | 1 Sep 04:56

Re: svn commit: r571578 - in /tomcat/trunk/java/org/apache: catalina/CometEvent.java catalina/connector/CometEventImpl.java catalina/connector/Request.java coyote/http11/Http11NioProcessor.java tomcat/util/net/PollerInterest.java

Remy Maucherat wrote:
> fhanik <at> apache.org wrote:
>> Author: fhanik
>> Date: Fri Aug 31 13:59:00 2007
>> New Revision: 571578
>>
>> URL: http://svn.apache.org/viewvc?rev=571578&view=rev
>> Log:
>> Move back to original idea, use primitive int instead of enum, 
>> extremely inefficient using enums and arrays to hold them. 
>
> Good. Two of the biggest gropes I had in this portion of the API and 
> impl are now gone (after arguing they were great ideas, and ignoring 
> my complaints about them for months :( ). 
pardon me, I always wanted the int, you were arguing the enums, so I 
compromised. then I realized how bad it was, so I took it out.
> How about adopting my interface ? It's clearly the next step in the 
> evolution.
don't like that interface, but a simple vote will resolve the dispute.
>
> The main problem continues, though: API changes are being made without 
> being discussed in any way beforehand. I think this needs to stop, and 
> I will restart the vote for moving this branch to the sandbox.
you slandered several people in the last vote thread without really 
answering any of their questions, don't see the point of another one 
vote, since you bowed out of the real questions in the last one.
The vote should be, and should have always been, about the API.

Filip
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Bill Barker | 1 Sep 06:11
Favicon

Re: [VOTE] Move trunk to sandbox


"Remy Maucherat" <remm <at> apache.org> wrote in message 
news:46D8ADAB.6000703 <at> apache.org...
> Hi,
>
> Due to continuing independent development in trunk by Filip, I am calling 
> for a second vote to move it to the sandbox, where Filip could continue 
> developing his ideas in a non disruptive way. I think the whole thing has 
> been debated pretty extensively in the previous vote.
>
> Following this, debate should take place to determine if a new main trunk 
> is needed, and if commit procedure changes would be good (personally, I do 
> like Jean-Frédéric's ideas).
>
> <ballot>
> [X] +1
> [ ] 0
> [ ] -1
> </ballot>
>

I agree that the competing Comet implementations should have their own 
sandbox entries for now, and allow code-Darwanism determine the one for 
6.next or 7.0 (whichever comes first).  At the moment it is clear that the 
current trunk (aka Filip's sandbox) can't be released due to lack of 
community support, so there is no reason to call it "trunk".  I'd be infavor 
of having a "trunk" that contains the Geronimo injection changes, but as 
Remy said that is a separate [VOTE].  (And for the httpd lurkers on the 
list, yes, we used to be a Jakarta project, so we are still very vote-happy 
:).
(Continue reading)

Permalink | Reply | 0 comments |
headers
Remy Maucherat | 1 Sep 09:20
Picon
Favicon
Gravatar

Re: svn commit: r571578 - in /tomcat/trunk/java/org/apache: catalina/CometEvent.java catalina/connector/CometEventImpl.java catalina/connector/Request.java coyote/http11/Http11NioProcessor.java tomcat/util/net/PollerInterest.java

Filip Hanik - Dev Lists wrote:
> Remy Maucherat wrote:
>> fhanik <at> apache.org wrote:
>>> Author: fhanik
>>> Date: Fri Aug 31 13:59:00 2007
>>> New Revision: 571578
>>>
>>> URL: http://svn.apache.org/viewvc?rev=571578&view=rev
>>> Log:
>>> Move back to original idea, use primitive int instead of enum, 
>>> extremely inefficient using enums and arrays to hold them. 
>>
>> Good. Two of the biggest gropes I had in this portion of the API and 
>> impl are now gone (after arguing they were great ideas, and ignoring 
>> my complaints about them for months :( ). 
> pardon me, I always wanted the int, you were arguing the enums, so I 
> compromised. then I realized how bad it was, so I took it out.

I vetoed these API changes almost immediately, and, despite starting 
with similar calls, my own proposal never included these new enums (it 
used boolean flags), or any of the related complexity. Seriously :(

My position has been fairly consistent so far: significant and possibly 
controversial changes like API changes should be discussed and agreed 
upon *in full details* before hitting the commit button. This applies 
especially to you, since it's about as impossible as it gets to have you 
modify anything afterwards :(

>> How about adopting my interface ? It's clearly the next step in the 
>> evolution.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Peter Rossbach | 1 Sep 09:38
Picon
Favicon
Gravatar

Re: [VOTE] Move trunk to sandbox

+1

move 6.0 trunk at a sandbox. We must made a check that we not lose  
some non comet changes :-(

Am 01.09.2007 um 02:09 schrieb Remy Maucherat:

> Hi,
>
> Due to continuing independent development in trunk by Filip, I am  
> calling for a second vote to move it to the sandbox, where Filip  
> could continue developing his ideas in a non disruptive way. I  
> think the whole thing has been debated pretty extensively in the  
> previous vote.
>
> Following this, debate should take place to determine if a new main  
> trunk is needed, and if commit procedure changes would be good  
> (personally, I do like Jean-Frédéric's ideas).
>
> <ballot>
> [ ] +1
> [ ] 0
> [ ] -1
> </ballot>
>
> Rémy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe <at> tomcat.apache.org
> For additional commands, e-mail: dev-help <at> tomcat.apache.org
(Continue reading)

Permalink | Reply | 3 comments |
headers
pero | 1 Sep 09:52
Picon
Favicon

svn commit: r571739 - /tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java

Author: pero
Date: Sat Sep  1 00:52:52 2007
New Revision: 571739

URL: http://svn.apache.org/viewvc?rev=571739&view=rev
Log:
reset parent attribute to made useless recovery thread not useable.

Modified:
    tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java

Modified: tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java
URL: http://svn.apache.org/viewvc/tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java?rev=571739&r1=571738&r2=571739&view=diff
==============================================================================
---
tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java (original)
+++
tomcat/container/tc5.5.x/modules/cluster/src/share/org/apache/catalina/cluster/mcast/McastServiceImpl.java
Sat Sep  1 00:52:52 2007
@@ -396,7 +396,7 @@

         public RecoveryThread(McastServiceImpl parent) {
             this.parent = parent;
-            if (!init(this)) parent = null;
+            if (!init(this)) this.parent = null;
         }

         public static synchronized boolean init(RecoveryThread t) {
Permalink | Reply | 0 comments |
headers
Henri Gomez | 1 Sep 10:17
Picon
Gravatar

Re: [VOTE] Move trunk to sandbox

+1

We need a stable TC 6.x, experimental code should go to some sort of
incubation land
Permalink | Reply | 0 comments |
headers
Jean-frederic Clere | 1 Sep 11:21
Picon
Gravatar

Re: [VOTE] Move trunk to sandbox

Remy Maucherat wrote:
> Hi,
>
> Due to continuing independent development in trunk by Filip, I am 
> calling for a second vote to move it to the sandbox, where Filip could 
> continue developing his ideas in a non disruptive way. I think the 
> whole thing has been debated pretty extensively in the previous vote.
>
> Following this, debate should take place to determine if a new main 
> trunk is needed, and if commit procedure changes would be good 
> (personally, I do like Jean-Frédéric's ideas).
>
> <ballot>
> [ X] +1

Cheers

Jean-Frederic

> [ ] 0
> [ ] -1
> </ballot>
>
> Rémy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe <at> tomcat.apache.org
> For additional commands, e-mail: dev-help <at> tomcat.apache.org
>
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
jim | 1 Sep 14:34
Picon
Favicon

svn commit: r571787 - /tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c

Author: jim
Date: Sat Sep  1 05:34:35 2007
New Revision: 571787

URL: http://svn.apache.org/viewvc?rev=571787&view=rev
Log:
&a[0] == a

Modified:
    tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c

Modified: tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c
URL: http://svn.apache.org/viewvc/tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c?rev=571787&r1=571786&r2=571787&view=diff
==============================================================================
--- tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c (original)
+++ tomcat/connectors/trunk/jk/native/common/jk_lb_worker.c Sat Sep  1 05:34:35 2007
@@ -584,7 +584,7 @@
     for (i = 0; i < p->num_of_workers; i++) {
         if (strcmp(p->lb_workers[i].s->route, name) == 0) {
             rc = &p->lb_workers[i];
-            rc->r = &(rc->s->route[0]);
+            rc->r = rc->s->route;
             break;
         }
     }
@@ -622,7 +622,7 @@
     }

     if (candidate) {
-        candidate->r = &(candidate->s->domain[0]);
(Continue reading)

Permalink | Reply | 0 comments |

Gmane