Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

seamonkey-2.7.2.en-US.os2.10.4esr.zip

18 views
Skip to first unread message

Dave Yeo

unread,
May 4, 2012, 10:18:12 AM5/4/12
to
I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
improvement is focus should work. Same cross-platform fixes as Firefox
10.04ESR. Requires libc065.dll
Dave

Lewis Rosenthal

unread,
May 4, 2012, 10:59:51 AM5/4/12
to
On 05/04/12 10:18 am, Dave Yeo thus wrote :
Focus: thank you, thank you, thank you, thank you....!!!!

--
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC www.2rosenthals.com
Need a managed Wi-Fi hotspot? www.hautspot.com
visit my IT blog www.2rosenthals.net/wordpress
-------------------------------------------------------------

Lewis Rosenthal

unread,
May 4, 2012, 12:34:37 PM5/4/12
to
On 05/04/12 10:59 am, Lewis Rosenthal thus wrote :
> On 05/04/12 10:18 am, Dave Yeo thus wrote :
>> I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
>> improvement is focus should work. Same cross-platform fixes as Firefox
>> 10.04ESR. Requires libc065.dll
>> Dave
>
> Focus: thank you, thank you, thank you, thank you....!!!!
>

Unfortunately, while the focus issue is better, it's still present, as
Dave (Saville) noted in FF, in the "test build 3" thread.

I haven't tested the RMB stuff, yet.

Thanks for the progress, though, Dave! It is surely a big step in the
right direction.

Lewis Rosenthal

unread,
May 4, 2012, 6:23:12 PM5/4/12
to
On 05/04/12 12:34 pm, Lewis Rosenthal thus wrote :
> On 05/04/12 10:59 am, Lewis Rosenthal thus wrote :
>> On 05/04/12 10:18 am, Dave Yeo thus wrote :
>>> I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
>>> improvement is focus should work. Same cross-platform fixes as Firefox
>>> 10.04ESR. Requires libc065.dll
>>> Dave
>>
>> Focus: thank you, thank you, thank you, thank you....!!!!
>>
>
> Unfortunately, while the focus issue is better, it's still present, as
> Dave (Saville) noted in FF, in the "test build 3" thread.
>
> I haven't tested the RMB stuff, yet.
>
> Thanks for the progress, though, Dave! It is surely a big step in the
> right direction.
>

Still the RMB crash (though I don't think as consistently as before),
and unlike what Dave mentioned about the immediate exit (now) under FF,
I still get the same window pattern on the way out. But, as I say, I
don't think it's happening as often (I *know* I've seen my context menu
on a page and was surprised that I hadn't crashed it).

So, again, big step in the right direction, and thanks so, so much for
all of your hard work on these latest updates, Dave!!

Dave Yeo

unread,
May 4, 2012, 11:52:32 PM5/4/12
to
Lewis Rosenthal wrote:
> On 05/04/12 10:59 am, Lewis Rosenthal thus wrote :
>> On 05/04/12 10:18 am, Dave Yeo thus wrote :
>>> I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
>>> improvement is focus should work. Same cross-platform fixes as Firefox
>>> 10.04ESR. Requires libc065.dll
>>> Dave
>>
>> Focus: thank you, thank you, thank you, thank you....!!!!
>>
>
> Unfortunately, while the focus issue is better, it's still present, as
> Dave (Saville) noted in FF, in the "test build 3" thread.

I haven't seen any focus problems yet, where are they showing up?

>
> I haven't tested the RMB stuff, yet.

I never did end up doing anything about the RMB (actually one of the
test builds disabled some likely looking code which didn't help) yet
here it recovered quite well. Strange but when things get carried away
recursing who knows.

>
> Thanks for the progress, though, Dave! It is surely a big step in the
> right direction.
>

Mostly I've just rebased Rich and Walters patches though some are mine.
Dave

Lewis Rosenthal

unread,
May 5, 2012, 11:29:49 AM5/5/12
to
On 05/04/12 11:52 pm, Dave Yeo thus wrote :
> Lewis Rosenthal wrote:
>> On 05/04/12 10:59 am, Lewis Rosenthal thus wrote :
>>> On 05/04/12 10:18 am, Dave Yeo thus wrote :
>>>> I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
>>>> improvement is focus should work. Same cross-platform fixes as Firefox
>>>> 10.04ESR. Requires libc065.dll
>>>> Dave
>>>
>>> Focus: thank you, thank you, thank you, thank you....!!!!
>>>
>>
>> Unfortunately, while the focus issue is better, it's still present, as
>> Dave (Saville) noted in FF, in the "test build 3" thread.
>
> I haven't seen any focus problems yet, where are they showing up?
>
They take a little time, but if I'm working in another app and come back
directly to a browser window to, say, select text or into a message
window to go back to typing, I lose focus. If I remember to click in the
url bar or the subject header first, I'm good to click in the main
window. This is essentially how it was previously (I think), but it
doesn't happen immediately. This could be a resource issue, perhaps. I'm
typically running at least one or two other libc apps (OOo and/or
JavaRDP - the latter running under OpenJDK 1.6), and my shared arena
gets pretty fragmented. Typically, after a few hours, a shutdown will
yield the XWP warning that my ini files could not be saved and will
likely be corrupt on the next boot. None of that started happening
before I started using JavaRDP in January (to access a remote XP VM). I
mention the resource issue because that's the only thing I can think
might cause the focus issue to show up so late into the running of the
app. I have not run Theseus to check for leaks, though.


>>
>> I haven't tested the RMB stuff, yet.
>
> I never did end up doing anything about the RMB (actually one of the
> test builds disabled some likely looking code which didn't help) yet
> here it recovered quite well. Strange but when things get carried away
> recursing who knows.
>

Indeed. In that case, all bets are off. We don't get anything written to
the logs, just the "screensaver" effect.

>>
>> Thanks for the progress, though, Dave! It is surely a big step in the
>> right direction.
>>
>
> Mostly I've just rebased Rich and Walters patches though some are mine.
>

But you've taken the time to do that, compile, and package it for
distribution, which is so much appreciated. Your modesty is misplaced,
my friend. Your work - hard or not - shows *every* day we use these (now
vital) apps. I wish there was more that we could do to help facilitate
your efforts. I know that you are bandwidth constrained. Is broadband
available at all where you are?

Dave Yeo

unread,
May 5, 2012, 1:20:36 PM5/5/12
to
Strange that I haven't seen it. I haven't bothered with Java yet but do
often have FF, SM, and TB open at the same time and compiling can use
quite a few resources. I've been using this build for close to a week
and before that an almost identical one that reported itself as ver 2.7.
>
>>>
>>> I haven't tested the RMB stuff, yet.
>>
>> I never did end up doing anything about the RMB (actually one of the
>> test builds disabled some likely looking code which didn't help) yet
>> here it recovered quite well. Strange but when things get carried away
>> recursing who knows.
>>
>
> Indeed. In that case, all bets are off. We don't get anything written to
> the logs, just the "screensaver" effect.

I always got various things written to popup.log though due to running
out of stack they don't convey much info.

>
>>>
>>> Thanks for the progress, though, Dave! It is surely a big step in the
>>> right direction.
>>>
>>
>> Mostly I've just rebased Rich and Walters patches though some are mine.
>>
>
> But you've taken the time to do that, compile, and package it for
> distribution, which is so much appreciated. Your modesty is misplaced,
> my friend. Your work - hard or not - shows *every* day we use these (now
> vital) apps. I wish there was more that we could do to help facilitate
> your efforts. I know that you are bandwidth constrained. Is broadband
> available at all where you are?
>

No broadband currently, even the satellites aren't visible enough
especially as I'm in a rain forest so surrounded by 100+ ft trees.
There is a local ISP that was supposed to be putting in wireless last
Sept. It sounded pretty positive but depended on getting a government
grant and access to a local mountain top to receive a 3.2GHz signal and
resend it at 800 Mhz. Hopefully when the snow goes away it'll happen.
There's also rumours of a cell tower going in, probably to support smart
meters, but the cell companies in Canada make the American ones seem good.
Dave

Felix Miata

unread,
May 6, 2012, 10:59:59 AM5/6/12
to
On 2012/05/04 70:18 (GMT-0700) Dave Yeo composed:

> I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
> improvement is focus should work. Same cross-platform fixes as Firefox
> 10.04ESR. Requires libc065.dll

Something's still not right with focus. At random times, moving mouse across
email window, focus shifts to my cmd window containing FC/2, where the contents
of an unopened SM error console are dumped into the FC/2 window as if each
error entry was a command entered at a prompt. e.g. "J:\>(firefox-bin:3295):
GLib-GObject-CRITICAL **: g_object_ref: assertion `G_IS_OBJECT (object)`
failed; SYS1079: : was unexpected at this time". At last instance, 6 similar
lines pollute FC/2 cmd history. I don't recall having opened Firefox since last
boot either. Though it's possible I may have, there is no instance in the
Ctrl-ESC window list.

Bug 686187 appeared again on first open only.
--
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata *** http://fm.no-ip.com/

Dave Yeo

unread,
May 6, 2012, 1:21:52 PM5/6/12
to
Felix Miata wrote:
> On 2012/05/04 70:18 (GMT-0700) Dave Yeo composed:
>
>> > I've uploaded seamonkey-2.7.2.en-US.os2.10.4esr.zip to netlabs. Main
>> > improvement is focus should work. Same cross-platform fixes as Firefox
>> > 10.04ESR. Requires libc065.dll
> Something's still not right with focus. At random times, moving mouse across
> email window, focus shifts to my cmd window containing FC/2, where the contents
> of an unopened SM error console are dumped into the FC/2 window as if each
> error entry was a command entered at a prompt. e.g. "J:\>(firefox-bin:3295):
> GLib-GObject-CRITICAL **: g_object_ref: assertion `G_IS_OBJECT (object)`
> failed; SYS1079: : was unexpected at this time". At last instance, 6 similar
> lines pollute FC/2 cmd history. I don't recall having opened Firefox since last
> boot either. Though it's possible I may have, there is no instance in the
> Ctrl-ESC window list.

Where would references to glib be coming from? While the build system
used to use glib and it is still a dependency, it was only used to build
a program that processed IDL files into headers and has been replaced by
a python script. Not only that but I'm pretty sure gobject is glib2
thing and I've only got glib-1.2.1 (actually glib2 is installed in
\usr\local but it is not on any path). Firefox-bin is also IIRC the Unix
name for Firefox which does use the GTK toolkit and glib2.
Almost like you have pollution from a Linux version but I can't imagine
how that would happen.
>
> Bug 686187 appeared again on first open only.

Strange, I have yet to see a focus problem with this build.
Dave

Lewis Rosenthal

unread,
May 6, 2012, 1:50:53 PM5/6/12
to
On 05/06/12 01:21 pm, Dave Yeo thus wrote :
Now that Felix mentions it, I thought it was unrelated, but I've seen a
couple instances where my cursor has "flown" to the upper left of the
screen. I thought I was causing it myself, but the occurrence in
conjunction to Felix's report is too much for coincidence.

I keep my Xcenter at the top of the display, and my Xbutton gets hit
(either as a result of me clicking and not knowing that the mouse is
focused there, or automatically selected). Felix, is it possible that
you have a hot corner displayed or something which might cause what
you're seeing or perhaps our z-ordering happens to hit on those two apps
when the problem shows up?

Dave Yeo

unread,
May 6, 2012, 2:12:57 PM5/6/12
to
Dave Yeo wrote:
>> Bug 686187 appeared again on first open only.
>
> Strange, I have yet to see a focus problem with this build.

Got mixed up on the bug#. This release /should/ set the preferences
correctly to stop that crash but it is possible that it didn't at first.
You can go to about:support to see your changed preferences and verify
that gfx.downloadable_fonts.enabled is not on the list.
Dave

Steve Wendt

unread,
May 6, 2012, 4:01:36 PM5/6/12
to
You are still mixed up. :-) Bug 686187 is about the missing bitmaps on
scrollbars (which I still see regularly).

Felix Miata

unread,
May 6, 2012, 11:08:38 PM5/6/12
to
On 2012/05/06 13:50 (GMT-0400) Lewis Rosenthal composed:

> Dave Yeo wrote :

>> Felix Miata wrote:

>>> Something's still not right with focus. At random times, moving mouse across
>>> email window, focus shifts to my cmd window containing FC/2, where the contents
>>> of an unopened SM error console are dumped into the FC/2 window as if each
>>> error entry was a command entered at a prompt. e.g.
>>> "J:\>(firefox-bin:3295):
>>> GLib-GObject-CRITICAL **: g_object_ref: assertion `G_IS_OBJECT (object)`
>>> failed; SYS1079: : was unexpected at this time". At last instance, 6 similar
>>> lines pollute FC/2 cmd history. I don't recall having opened Firefox since last
>>> boot either. Though it's possible I may have, there is no instance in the
>>> Ctrl-ESC window list.

>> Where would references to glib be coming from? While the build system
>> used to use glib and it is still a dependency, it was only used to build
>> a program that processed IDL files into headers and has been replaced by
>> a python script. Not only that but I'm pretty sure gobject is glib2
>> thing and I've only got glib-1.2.1 (actually glib2 is installed in
>> \usr\local but it is not on any path). Firefox-bin is also IIRC the Unix
>> name for Firefox which does use the GTK toolkit and glib2.
>> Almost like you have pollution from a Linux version but I can't imagine
>> how that would happen.

After some thought it became obvious this was clipboard content. Source was
text file "out" on L:, which is a cifs connection from my eCS to this Linux,
from which I used FC/2 to copy 6 lines of content later pasted into a Bugzilla
comment in a 2.7.2 tab on eCS. So what's new here is focus trouble escaping to
the clipboard, then moving its content to another app's window somehow.

> Now that Felix mentions it, I thought it was unrelated, but I've seen a
> couple instances where my cursor has "flown" to the upper left of the
> screen. I thought I was causing it myself, but the occurrence in
> conjunction to Felix's report is too much for coincidence.

Multiple times daily my mouse pointer disappears into the ether only to be
rediscovered by me after moving it around enough to grab my attention. It's
been happening far too long to recall if it ever didn't happen.

> I keep my Xcenter at the top of the display, and my Xbutton gets hit
> (either as a result of me clicking and not knowing that the mouse is
> focused there, or automatically selected). Felix, is it possible that
> you have a hot corner displayed or something which might cause what
> you're seeing or perhaps our z-ordering happens to hit on those two apps
> when the problem shows up?

Don't know what you mean by "hot corner". Typically the browser window fills
the entire height from 1" from left edge to 1/4" from right edge. Email goes
beyond right edge and about 1/4" left of left of browser window, with top of
browser window down to urlbar peeking above mail window. Underneath it all full
width but from bottom of screen up about 70% is the FC/2 window. Peeking out at
upper left is a piece of desktop that includes "Local System" and the pager. So
the paste escape must happen when the mouse somehow skips left to the peeking
left end of the FC/2 window.

SET AUTOSTART=PROGRAMS,TASKLIST,FOLDERS,WARPCENTER

WarpCenter is at top of screen, SM status bar is usually at bottom.

Current uptime is about 92 hours.

I get the feeling something somehow is triggering mouse gestures, which I would
never enable intentionally.

Who's to say that the many months ago manual replacement of Estyler 1.0 DLL
with 1.1's has nothing to do with this? Maybe I should get around to installing
eCS 2.0 to replace this ancient 1.14 before long. I downloaded it over a year
ago. :-p

Dave Yeo

unread,
May 7, 2012, 12:40:29 AM5/7/12
to
OK, it was from a Linux version. While I have had my mouse pointer fly
to the upper left of the screen which I blamed on where I'm currently
using my optical mouse and it going over the edge or the desk. Seems
also I've only seen this on eCS 2.1, not Warp 4.5.
I have yet had any clipboard contents accidentally paste into anything
including fc/2 which is usually open.

>
>> > Now that Felix mentions it, I thought it was unrelated, but I've seen a
>> > couple instances where my cursor has "flown" to the upper left of the
>> > screen. I thought I was causing it myself, but the occurrence in
>> > conjunction to Felix's report is too much for coincidence.
> Multiple times daily my mouse pointer disappears into the ether only to be
> rediscovered by me after moving it around enough to grab my attention. It's
> been happening far too long to recall if it ever didn't happen.
>

I've had the disappearing mouse cursor occasionally for a long time,
sometimes it completely vanishes and I have to grope around to launch
e.exe to bring it back.
Whether browser related I don't know.
Dave

Steve Wendt

unread,
May 7, 2012, 12:48:29 AM5/7/12
to
On 05/06/12 08:08 pm, Felix Miata wrote:

>> Now that Felix mentions it, I thought it was unrelated, but I've seen a
>> couple instances where my cursor has "flown" to the upper left of the
>> screen. I thought I was causing it myself, but the occurrence in
>> conjunction to Felix's report is too much for coincidence.
>
> Multiple times daily my mouse pointer disappears into the ether only to be
> rediscovered by me after moving it around enough to grab my attention. It's
> been happening far too long to recall if it ever didn't happen.

I believe I've seen this with the Windows version (more so than on OS/2,
at least).

Felix Miata

unread,
May 8, 2012, 2:04:33 AM5/8/12
to
On 2012/05/06 10:59 (GMT-0400) Felix Miata composed:

> Something's still not right with focus.

Too hard to use. I'm back on 2.6.whatever from December 21 until something
newer sounds fixed, or I can find 2-3 days to install eCS 2.0 to see if the OS
installation or configuration is the problem.

Lewis Rosenthal

unread,
May 9, 2012, 12:02:12 AM5/9/12
to
Hi, Felix... Apologies for the late follow-up.

On 05/06/12 11:08 pm, Felix Miata thus wrote :
> On 2012/05/06 13:50 (GMT-0400) Lewis Rosenthal composed:
>
>> Dave Yeo wrote :
>
>>> Felix Miata wrote:
>
>>>> Something's still not right with focus.

<snip>
>
> After some thought it became obvious this was clipboard content. Source was
> text file "out" on L:, which is a cifs connection from my eCS to this Linux,
> from which I used FC/2 to copy 6 lines of content later pasted into a Bugzilla
> comment in a 2.7.2 tab on eCS. So what's new here is focus trouble escaping to
> the clipboard, then moving its content to another app's window somehow.
>
>> Now that Felix mentions it, I thought it was unrelated, but I've seen a
>> couple instances where my cursor has "flown" to the upper left of the
>> screen. I thought I was causing it myself, but the occurrence in
>> conjunction to Felix's report is too much for coincidence.
>
> Multiple times daily my mouse pointer disappears into the ether only to be
> rediscovered by me after moving it around enough to grab my attention. It's
> been happening far too long to recall if it ever didn't happen.
>
>> I keep my Xcenter at the top of the display, and my Xbutton gets hit
>> (either as a result of me clicking and not knowing that the mouse is
>> focused there, or automatically selected). Felix, is it possible that
>> you have a hot corner displayed or something which might cause what
>> you're seeing or perhaps our z-ordering happens to hit on those two apps
>> when the problem shows up?
>
> Don't know what you mean by "hot corner".

I was referring to the Screen settings object, Screen borders tab. This
assumes that you have XWP or eWP installed, of course.

<snip>

> I get the feeling something somehow is triggering mouse gestures, which I would
> never enable intentionally.
>
> Who's to say that the many months ago manual replacement of Estyler 1.0 DLL
> with 1.1's has nothing to do with this? Maybe I should get around to installing
> eCS 2.0 to replace this ancient 1.14 before long. I downloaded it over a year
> ago. :-p

Well, we were both in on that discussion, too. ;-) On an up-note, I just
did two upgrades for a client, one from 1.2 and another form 2.0 RC6,
and both went off without a hitch. In your case, though, I suspect
you'll want to do a clean install and not a migration, though what's
been holding me off on my own upgrades has been the thought of all of
that *stuff* either having to be re-installed manually or the risk of a
migration chewing it up. Still, our builds are getting to an age where
it's hard to tell what antiquated/patched/hacked version of what could
be causing odd behavior!

I do see that you rolled back again. I'm sticking with this build for
now, as overall, it seems stable for me (other than the few points we've
discussed here).

Cheers/2

Dave Yeo

unread,
May 9, 2012, 12:28:43 AM5/9/12
to
Lewis Rosenthal wrote:
> Well, we were both in on that discussion, too. ;-) On an up-note, I just
> did two upgrades for a client, one from 1.2 and another form 2.0 RC6,
> and both went off without a hitch. In your case, though, I suspect
> you'll want to do a clean install and not a migration, though what's
> been holding me off on my own upgrades has been the thought of all of
> that *stuff* either having to be re-installed manually or the risk of a
> migration chewing it up. Still, our builds are getting to an age where
> it's hard to tell what antiquated/patched/hacked version of what could
> be causing odd behavior!

My other install is Warp v4 from '96 with lots of updates etc. It is
getting awfully crufty, crashes when shutting down lately, get
complaints when starting up, SMP causes a trap when doing anything
including looking at the mouse :) yet it is more stable then eCS 2.1
with the latest ACPI, certain things work that I haven't gotten to work
on eCS and is very familiar after using for 15+ years.
SeaMonkey and other Mozilla apps run just the same on the old crusty
system as the new install. Perhaps even better as they stay up for
longer. An old install doesn't need to equal bad.
Dave

Dave Yeo

unread,
May 9, 2012, 12:32:34 AM5/9/12
to
Dave Yeo wrote:
> No broadband currently, even the satellites aren't visible enough
> especially as I'm in a rain forest so surrounded by 100+ ft trees.
> There is a local ISP that was supposed to be putting in wireless last
> Sept. It sounded pretty positive but depended on getting a government
> grant and access to a local mountain top to receive a 3.2GHz signal and
> resend it at 800 Mhz. Hopefully when the snow goes away it'll happen.

Good news (I hope). The local ISP is trying to get hold of me (Yah!) and
with any luck it is to set a date to install the dish (pointed at a
local mountain) and router. God, I would like to get of of dial-up.
Dave

Elmer Fudd

unread,
May 9, 2012, 7:49:23 AM5/9/12
to
Where are you - Vancouver Island? Not that many places in North America with
rainforest.

On Tue, 08 May 2012 21:32:34 -0700, Dave Yeo wrote:

Dave Yeo

unread,
May 9, 2012, 11:15:53 AM5/9/12
to
Elmer Fudd wrote:
> Where are you - Vancouver Island? Not that many places in North America with
> rainforest.

In the mountains east of Vancouver.
Dave

Elmer Fudd

unread,
May 9, 2012, 12:26:11 PM5/9/12
to
Ah, OK.

On Wed, 09 May 2012 08:15:53 -0700, Dave Yeo wrote:

Brendan McCullough

unread,
May 10, 2012, 3:15:25 PM5/10/12
to
Thanks very much for this. It seems a little quicker, two more sites
load consistently, focus works and no abends when sending email.

Dave Yeo

unread,
May 18, 2012, 11:12:08 PM5/18/12
to
Has anyone else noticed the bug where restoring a session with more then
one window, the lower windows have their content at the same point as
the top window, so no menu and perhaps a cm of blank at the bottom?
Resizing it is the fix.
Dave

Steve Wendt

unread,
May 18, 2012, 11:21:34 PM5/18/12
to
On 5/18/2012 8:12 PM, Dave Yeo wrote:

> Has anyone else noticed the bug where restoring a session with more then
> one window, the lower windows have their content at the same point as
> the top window, so no menu and perhaps a cm of blank at the bottom?
> Resizing it is the fix.

That one has been discussed here before, so it isn't new.

Dave Yeo

unread,
May 19, 2012, 1:38:10 AM5/19/12
to dev-po...@lists.mozilla.org
I'm obviously overtired <gr>
Dave

Dave Yeo

unread,
May 19, 2012, 3:24:56 AM5/19/12
to
Firefox 11 really outdoes this. No scroll bars and the UI scrolls up.
Tabs scroll over the buttons then even the menus scroll off of the top
of the page, bloody depressing. Extensions are also getting more and
more broken, things can't connect, popups are empty besides the yes and
no buttons and trying to install a persona resulted in a browser that
didn't display forcing me to resort to a backup.
Dave

Alfredo Fernández Díaz

unread,
May 19, 2012, 11:52:14 AM5/19/12
to
Dave Yeo escribi�:
> Steve Wendt wrote:
>> On 5/18/2012 8:12 PM, Dave Yeo wrote:
>>
>>> Has anyone else noticed the bug where restoring a session with more then
...
> Firefox 11 really outdoes this. No scroll bars and the UI scrolls up. Tabs
> scroll over the buttons then even the menus scroll off of the top of the page,
> bloody depressing. Extensions are also getting more and more broken, things
> can't connect, popups are empty besides the yes and no buttons and trying to
> install a persona resulted in a browser that didn't display forcing me to
> resort to a backup.

Is resizing still a fix?


Dave Yeo

unread,
May 19, 2012, 12:21:32 PM5/19/12
to
Alfredo Fernández Díaz wrote:
> Dave Yeo escribió:
No, I'll have to try to find the commit that broke it.
Dave

Dave Yeo

unread,
May 19, 2012, 1:23:35 PM5/19/12
to
Dave Yeo wrote:
> Alfredo Fern�ndez D�az wrote:
>> Dave Yeo escribi�:
Or rather the typo that broke it. Amazing what a stray ! can do <gr>
Dave

Alfredo Fernández Díaz

unread,
May 19, 2012, 1:50:40 PM5/19/12
to
Dave Yeo escribi�:
> Dave Yeo wrote:
>> Alfredo Fern�ndez D�az wrote:
>>> Dave Yeo escribi�:
>>>> Steve Wendt wrote:
>>>>> On 5/18/2012 8:12 PM, Dave Yeo wrote:
>>>>>
>>>>>> Has anyone else noticed the bug where restoring a session with more
>>>>>> then
>>> ...
>>>> Firefox 11 really outdoes this. No scroll bars and the UI scrolls up.
>>>>...
>>>
>>> Is resizing still a fix?
>>
>> No, I'll have to try to find the commit that broke it.
>
> Or rather the typo that broke it. Amazing what a stray ! can do <gr>

Not a lot if you don't commit it! ;)

Steve Wendt

unread,
May 19, 2012, 9:04:12 PM5/19/12
to
On 05/19/12 10:23 am, Dave Yeo wrote:

> Or rather the typo that broke it. Amazing what a stray ! can do <gr>

Those single character fixes can take an extraordinary amount of time to
track down. :-p

Dave Yeo

unread,
May 20, 2012, 12:58:10 AM5/20/12
to
In a way it was more of a senior moment :) and after some thought it was
like, that ! in nsFrameManager.h was probably wrong and it was.
Still left with a few subtle bugs. Installing a persona and restarting
resulted in a hang, worked around by killing Firefox and starting in
safemode. Re-enabling all extensions and restarting worked.
Now I notice that clicking the download button for a preference or
persona doesn't work. Worked around by RMB and opening in new tab.
Now I can't scroll the tabs right and left is way too quick. Also that
add-on still displays a blank popup with working buttons. And this is 5
month old 11a. All I can do is try to trace down the commits that broke
this and that. I need to test on trunk tip and file bugs from there. And
it took me 4 days to fix the crash so how long will it take to trace
down these subtle bugs.
I had hopes for a 13 or 14 release but now if I spend a lot of time we
might see a 17esr release. I do want an up to date browser so'll
probably keep working on this but really it is probably a full time job
for someone, hopefully with a faster computer then I have. (1.25 hrs to
compile Firefox here, much better then the 8 hours it used to take).
Also can't believe how fast keeping all these trees around eats up disk
space.
Dave, who wants to exercise his brain to strive of the Alzheimers that
runs in the family :)

Dave Yeo

unread,
May 26, 2012, 2:33:13 AM5/26/12
to
Dave Yeo wrote:
> Now I can't scroll the tabs right and left is way too quick. Also that
> add-on still displays a blank popup with working buttons. And this is 5
> month old 11a. All I can do is try to trace down the commits that broke
> this and that. I need to test on trunk tip and file bugs from there. And
> it took me 4 days to fix the crash so how long will it take to trace
> down these subtle bugs.

The tab problem was caused by part 2 of bug#704171 and fixed in #706323.
Wasted 5 evenings chasing cross platform bug down :(
Oh well, onward.
Dave

Lewis Rosenthal

unread,
May 26, 2012, 3:49:06 PM5/26/12
to
On 05/26/12 02:33 am, Dave Yeo thus wrote :
At this point, Dave, I'm wondering whether you might want to try a build
of one of the newer SM's, as someof this stuff may have already been
addressed between 2.7.2 and where the code base is now. Seems a shame to
spend so much time on these things.

That said, your efforts are so greatly appreciated.

Dave Yeo

unread,
May 26, 2012, 6:43:41 PM5/26/12
to dev-po...@lists.mozilla.org
On 05/26/12 12:49 pm, Lewis Rosenthal wrote:
>> The tab problem was caused by part 2 of bug#704171 and fixed in #706323.
>> > Wasted 5 evenings chasing cross platform bug down:(
>> > Oh well, onward.
> At this point, Dave, I'm wondering whether you might want to try a build
> of one of the newer SM's, as someof this stuff may have already been
> addressed between 2.7.2 and where the code base is now. Seems a shame to
> spend so much time on these things.

SeaMonkey is basically a wrapper around (the internals of) Firefox so
generally if Firefox won't build then neither will SeaMonkey. Though the
tab problem wouldn't have affected SeaMonkey, the crash that I
previously fixed would have. So the procedure is build a working
Firefox, then SeaMonkey and Thunderbird which usually just work at that
point.
Firefox is also faster to build, though still taking over an hour using
both cores (close to 2 hours with one core, make -j2 generally means
compiling 2 object files at once).
Dave
ps If anyone has an old Core2Duo E6600 or E6700 CPU laying around, that
would be a useful donation here. Currently using a E6300

Lewis Rosenthal

unread,
May 26, 2012, 9:15:01 PM5/26/12
to
On 05/26/12 06:43 pm, Dave Yeo thus wrote :
> On 05/26/12 12:49 pm, Lewis Rosenthal wrote:
>>> The tab problem was caused by part 2 of bug#704171 and fixed in #706323.
>>> > Wasted 5 evenings chasing cross platform bug down:(
>>> > Oh well, onward.
>> At this point, Dave, I'm wondering whether you might want to try a build
>> of one of the newer SM's, as someof this stuff may have already been
>> addressed between 2.7.2 and where the code base is now. Seems a shame to
>> spend so much time on these things.
>
> SeaMonkey is basically a wrapper around (the internals of) Firefox so
> generally if Firefox won't build then neither will SeaMonkey. Though the
> tab problem wouldn't have affected SeaMonkey, the crash that I
> previously fixed would have. So the procedure is build a working
> Firefox, then SeaMonkey and Thunderbird which usually just work at that
> point.

Ah... Thanks or that.

> Firefox is also faster to build, though still taking over an hour using
> both cores (close to 2 hours with one core, make -j2 generally means
> compiling 2 object files at once).

Geez...

> ps If anyone has an old Core2Duo E6600 or E6700 CPU laying around, that
> would be a useful donation here. Currently using a E6300

Dave what system board do you have? Seems to me, if we can get you into
a quad core (Q8400 or Q9400 or better), that would be a better move from
an E6300. Alternatively, a newer system with an i5 or i7 might be a good
upgrade. I would donate time & materials to this for you.

Dave Yeo

unread,
May 26, 2012, 10:09:14 PM5/26/12
to
Lewis Rosenthal wrote:
>> ps If anyone has an old Core2Duo E6600 or E6700 CPU laying around, that
>> would be a useful donation here. Currently using a E6300
>
> Dave what system board do you have? Seems to me, if we can get you into
> a quad core (Q8400 or Q9400 or better), that would be a better move from
> an E6300. Alternatively, a newer system with an i5 or i7 might be a good
> upgrade. I would donate time & materials to this for you.

It's an old ASUS P5PE-VM, originally had a Pentium 4D, the maximum
update for it is an E6700.
Dave

kbr_n...@guzzi.demon.nospam.nl

unread,
May 27, 2012, 9:17:44 AM5/27/12
to
In <J6mdnYqJ3OtQEVzS...@mozilla.org>, on 05/26/12
at 07:09 PM, Dave Yeo <dave....@gmail.com> said:



>It's an old ASUS P5PE-VM, originally had a Pentium 4D, the maximum
>update for it is an E6700.

I just upgraded to a quad core. Have the old E7500 duo lying around.

Cheers, Bjorn.
--
-----------------------------------------------------------
k...@guzzi.demon.nl
-----------------------------------------------------------

Lewis Rosenthal

unread,
May 29, 2012, 9:00:17 PM5/29/12
to
On 05/26/12 10:09 pm, Dave Yeo thus wrote :
Dave, it looks like that board can take a Core 2 Extreme X6800 2.93GHz,
which I *think* I may have on the shelf in New York (good pull, not
brand new). I'll check when I get back home next week and will let you
know. If I have it, it's yours.

Cheers/2

Trevor Hemsley

unread,
Jun 3, 2012, 6:23:47 PM6/3/12
to
On Wed, 30 May 2012 01:00:17 UTC in mozilla.dev.ports.os2, Lewis Rosenthal
<lgros...@2-de-sp-am-2rosenthals.com> wrote:

> Dave, it looks like that board can take a Core 2 Extreme X6800 2.93GHz,
> which I *think* I may have on the shelf in New York (good pull, not
> brand new). I'll check when I get back home next week and will let you
> know. If I have it, it's yours.

Neat, at the time of release that would've been one of Intel's US$999 processors
:-)

--
Trevor Hemsley, Brighton, UK
Trevor dot Hemsley at ntlworld dot com
0 new messages