headers
P T Withington | 2 Jul 15:22
Picon
Favicon
Gravatar

For Review: Change 20070702-ptw-D Summary: Preserve Debug.printLength when overriding

Change 20070702-ptw-D by ptw <at> dueling-banjos.local on 2007-07-02  
09:05:57 EDT
     in /Users/ptw/OpenLaszlo/legals
     for http://svn.openlaszlo.org/openlaszlo/branches/legals

Summary: Preserve Debug.printLength when overriding

Bugs Fixed:
LPP-4234 'DHTML Debug.inspect clobbers printLength'

Technical Reviewer: ben (pending)
QA Reviewer: frisco (pending)

Details:
     Guard with a try/catch.  Also, move length computation out of
     object test so length will be displayed for strings.

Tests:
     ant lztest, smokecheck

     lzx> Debug.printLength = 24
     24
     lzx> "A long string that will be abbreviated"
     «string(38)| 'A l...»
     lzx> Debug.inspect('A long string that will be abbreviated')
     'A long string that will be abbreviated'
     «string(38)| 'A l...»
     lzx> Debug.printLength
     24
     lzx>
(Continue reading)

Permalink | Reply | 2 comments |
headers
Benjamin Shine | 2 Jul 18:56
Favicon

Re: For Review: Change 20070702-ptw-D Summary: Preserve Debug.printLength when overriding

Approved... Just one little thing: the indentation on the whole try/ 
catch block down there is off-by-two -- imho it would be nice to  
indent it to match the new block structure.

-ben

On Jul 2, 2007, at 6:22 AM, P T Withington wrote:

> Change 20070702-ptw-D by ptw <at> dueling-banjos.local on 2007-07-02  
> 09:05:57 EDT
>     in /Users/ptw/OpenLaszlo/legals
>     for http://svn.openlaszlo.org/openlaszlo/branches/legals
>
> Summary: Preserve Debug.printLength when overriding
>
> Bugs Fixed:
> LPP-4234 'DHTML Debug.inspect clobbers printLength'
>
> Technical Reviewer: ben (pending)
> QA Reviewer: frisco (pending)
>
> Details:
>     Guard with a try/catch.  Also, move length computation out of
>     object test so length will be displayed for strings.
>
> Tests:
>     ant lztest, smokecheck
>
>     lzx> Debug.printLength = 24
>     24
(Continue reading)

Permalink | Reply | 1 comment |
headers
Benjamin Shine | 2 Jul 19:05
Favicon

Re: [Laszlo-checkins] r5578 - tools/trunk/svn

This looks reasonable -- two issues...
1) I didn't get a For Review mail -- did you send one?
2) I don't really, ahem, *know perl* so I can't really say whether  
"use Encode qw(encode);" makes any sense.

On Jul 2, 2007, at 6:24 AM, ptw <at> openlaszlo.org wrote:

> Author: ptw
> Date: 2007-07-02 06:24:19 -0700 (Mon, 02 Jul 2007)
> New Revision: 5578
>
> Modified:
>    tools/trunk/svn/review-mail.pl
> Log:
> Change 20070702-ptw-I by ptw <at> dueling-banjos.local on 2007-07-02  
> 09:22:46 EDT
>     in /Users/ptw/OpenLaszlo/tools/trunk
>     for http://svn.openlaszlo.org/tools/trunk
>
> Summary: svn tools tweaks
>
> Technical Reviewer: ben (pending)
> QA Reviewer: jgrandy (pending)
>
> Details:
>     encode mail body before escaping
>
> Tests:
>     IWFM
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Benjamin Shine | 2 Jul 19:20
Favicon

running Open Laszlo in jdk 1.6 and tomcat 5.5


A user has reported this tidbit to me, regarding his experience  
running Open Laszlo 3.4 with jdk 1.6 and tomcat 5.5.23:

To use the "deploy" target in build.xml, you'll need to undo a hack I  
had to put in to work with tomcat 5.0.30. In old tomcat, it was  
necessary to append a space to the localWar attribute of the deploy  
task:

tomcat 5.0.30
>   <deploy url="${tom.url}"
>              username="${tom.username}"
>              password="${tom.password}"
>              path="/${build.branch}"
>              localWar="file://${LPS_HOME} " />
                                            ^ this space right here  
is a hack

In tomcat 5.5.23, that space is no longer necessary, and in fact, it  
causes problems. Therefore, eliminate the space, like this:
>          <deploy url="${tom.url}"
>              username="${tom.username}"
>              password="${tom.password}"
>              path="/${build.branch}"
>              localWar="file://${LPS_HOME}"/>

This issue is reported as http://jira.openlaszlo.org/jira/browse/ 
LPP-4235, but I wanted to get this information out to the community  
before I get a chance to address the issue in code.
(Continue reading)

Permalink | Reply | 501 comments |
headers
P T Withington | 2 Jul 19:26
Picon
Favicon
Gravatar

Re: For Review: Change 20070702-ptw-D Summary: Preserve Debug.printLength when overriding

I did that to make it easy to see the diff.  Now that you have  
approved it, I will re-indent and check in.

On 2007-07-02, at 12:56 EDT, Benjamin Shine wrote:

> Approved... Just one little thing: the indentation on the whole try/ 
> catch block down there is off-by-two -- imho it would be nice to  
> indent it to match the new block structure.
>
> -ben
>
> On Jul 2, 2007, at 6:22 AM, P T Withington wrote:
>
>> Change 20070702-ptw-D by ptw <at> dueling-banjos.local on 2007-07-02  
>> 09:05:57 EDT
>>     in /Users/ptw/OpenLaszlo/legals
>>     for http://svn.openlaszlo.org/openlaszlo/branches/legals
>>
>> Summary: Preserve Debug.printLength when overriding
>>
>> Bugs Fixed:
>> LPP-4234 'DHTML Debug.inspect clobbers printLength'
>>
>> Technical Reviewer: ben (pending)
>> QA Reviewer: frisco (pending)
>>
>> Details:
>>     Guard with a try/catch.  Also, move length computation out of
>>     object test so length will be displayed for strings.
>>
(Continue reading)

Permalink | Reply | 0 comments |
headers
P T Withington | 2 Jul 19:30
Favicon

Re: [Laszlo-checkins] r5578 - tools/trunk/svn

I did not send review mail -- just checked in because I needed it to  
send the Debugger review!

I don't know perl either.  I copied the change from a google answer.  :P

Without this change, I couldn't send the review for the debugger  
change.  With this change, it worked.  QED.

On 2007-07-02, at 13:05 EDT, Benjamin Shine wrote:

> This looks reasonable -- two issues...
> 1) I didn't get a For Review mail -- did you send one?
> 2) I don't really, ahem, *know perl* so I can't really say whether  
> "use Encode qw(encode);" makes any sense.
>
> On Jul 2, 2007, at 6:24 AM, ptw <at> openlaszlo.org wrote:
>
>> Author: ptw
>> Date: 2007-07-02 06:24:19 -0700 (Mon, 02 Jul 2007)
>> New Revision: 5578
>>
>> Modified:
>>    tools/trunk/svn/review-mail.pl
>> Log:
>> Change 20070702-ptw-I by ptw <at> dueling-banjos.local on 2007-07-02  
>> 09:22:46 EDT
>>     in /Users/ptw/OpenLaszlo/tools/trunk
>>     for http://svn.openlaszlo.org/tools/trunk
>>
>> Summary: svn tools tweaks
(Continue reading)

Permalink | Reply | 0 comments |
headers
John Sundman | 2 Jul 20:08
Favicon

Re: running Open Laszlo in jdk 1.6 and tomcat 5.5

Thanks, noted.

jrs

On Jul 2, 2007, at 1:20 PM, Benjamin Shine wrote:

>
> A user has reported this tidbit to me, regarding his experience  
> running Open Laszlo 3.4 with jdk 1.6 and tomcat 5.5.23:
>
> To use the "deploy" target in build.xml, you'll need to undo a hack  
> I had to put in to work with tomcat 5.0.30. In old tomcat, it was  
> necessary to append a space to the localWar attribute of the deploy  
> task:
>
> tomcat 5.0.30
>>   <deploy url="${tom.url}"
>>              username="${tom.username}"
>>              password="${tom.password}"
>>              path="/${build.branch}"
>>              localWar="file://${LPS_HOME} " />
>                                            ^ this space right here  
> is a hack
>
> In tomcat 5.5.23, that space is no longer necessary, and in fact,  
> it causes problems. Therefore, eliminate the space, like this:
>>          <deploy url="${tom.url}"
>>              username="${tom.username}"
>>              password="${tom.password}"
>>              path="/${build.branch}"
(Continue reading)

Permalink | Reply | 501 comments |
headers
Benjamin Shine | 3 Jul 00:11
Favicon

For Review: Change 20070702-ben-8 Summary: improvements to runlztest to run more tests, more quietly

Change 20070702-ben-8 by ben <at> bens-g5.local on 2007-07-02 14:47:12 PDT
     in /Users/ben/src/svn/openlaszlo/branches/legals-clean
     for http://svn.openlaszlo.org/openlaszlo/branches/legals

Summary: improvements to runlztest to run more tests, more quietly

New Features:

Bugs Fixed: LPP-4236 runlztest should run to completion, not bail on  
first failure

Technical Reviewer: hqm (pending)
QA Reviewer: mkratt (pending)
Doc Reviewer: (pending)

Documentation:
This change tweaks build-tools/runlztest.sh to be less verbose about
standard operations, and to run an entire batch of tests. Before this
change, runlztest stopped after a single test failure. Now it runs
all tests, then reports a list of all tests that failed.

Also removed some sleeps so that skipped tests take approx no time,
instead of 2 seconds per skipped test.

Release Notes:

Details:

Tests:
When there are tests in the smokecheck which fail in lztest:
(Continue reading)

Permalink | Reply | 0 comments |
headers
P T Withington | 3 Jul 14:24
Picon
Favicon
Gravatar

[Reminder] Bug Scrub TUESDAY 2007-07-03T20:00Z

Bug Scrub TUESDAY

  1:00pm Pacific
  4:00pm Eastern
20:00Z

Date: 2007-07-03T20:00Z
Location: irc://irc.freenode.org/#laszlo
Ground rules:  1 hour, no more than 5 minutes/bug.

Agenda this week:

1. Scrub any new bugs (Bug Scrub: New filter)
Permalink | Reply | 0 comments |
headers
P T Withington | 3 Jul 15:47
Favicon

Re: For Review: Change 20070621-ben-x Summary:

I find the smokecheck very hard to view now.  The progress bar and  
summary status are overlaid on all sorts of test graphics.  Is this  
what you intended?  If so, how about adding an opaque background  
behind the test summary so it is legible?

On 2007-06-28, at 16:12 EDT, Benjamin Shine wrote:

> I am papering over a difference in behavior between the swf and  
> dhtml runtimes, but checking in this change would make it more  
> possible to report and debug errors in the dhtml runtime, so please  
> can I check it in without actually understanding the problem and  
> why this is the solution?
>
> Normally that's poor behavior but criminy, our smokechecks are  
> failing left and right, and I'd like to know why. This would help.
>
> On Jun 22, 2007, at 3:26 AM, P T Withington wrote:
>
>> It is not clear to me from your comments whether you are papering  
>> over a difference in behavior between SWF and DHTML runtimes, or  
>> you are saying the same broken behavior is observed in both  
>> runtimes and this is the fix for it.
>>
>> If the latter, approved.
>>
>> On 2007-06-21, at 18:40 EDT, Benjamin Shine wrote:
>>
>>> Change 20070621-ben-x by ben <at> cooper.local on 2007-06-21 15:34:48 PDT
>>>     in /Users/ben/src/svn/openlaszlo/branches/legals
>>>     for http://svn.openlaszlo.org/openlaszlo/branches/legals
(Continue reading)

Permalink | Reply | 5 comments |

Gmane