Finally!! Lightning 0.5 & Sunbird 0.5 release is out
Simon Paquet
Finally, the 0.5 release has been completed!
About two months after our initial scheduled release date, Lightning 0.5
and Sunbird 0.5 bring us a huge step closer to our 1.0 release, which is
scheduled for the first half of 2008.
Notable improvements of this release are:
* Much more polished user interface in the calendar views
* Event invitations can be sent and received via iMIP/iTIP
(Lightning only)
* Working hours are back
* Automatic migration of data in Sunbird 0.2, iCal.app, and Evolution
* Much improved printing functionality
* Better integration of Lightning into Mozilla Thunderbird (Printing,
Undo/redo, Copy/paste)
* Support for Google Calendar (via the Google Calendar Provider
extension[1])
* and much more. A full list of changes that have gone into this release
can be found on The Rumbling Edge weblog[2].
Lightning 0.5 and Sunbird 0.5 are available for Windows, Mac OS X
(universal builds) and Linux in 22 different languages including English.
Please use the following links to download the release:
* Lightning 0.5 (fully localized for 22 languages)
http://www.mozilla.org/projects/calendar/lightning/download.html
* Sunbird 0.5 in English
http://www.mozilla.org/projects/calendar/sunbird/download.html
* Sunbird 0.5 in other languages
http://www.mozilla.org/projects/calendar/sunbird/l10n_download.html
Please read the release notes for Lightning 0.5[3] and Sunbird 0.5[4]
before downloading.
Thank you again to all our developers, contributors, localizers, testers,
and supporters. We would not be able to do this without your assistance!
Cya
Simon
[1] https://addons.mozilla.org/en-US/thunderbird/addon/4631
[2]
http://weblogs.mozillazine.org/rumblingedge/archives/2006/12/sb_0-5.html
[3] http://www.mozilla.org/projects/calendar/releases/lightning0.5.html
[4] http://www.mozilla.org/projects/calendar/releases/sunbird0.5.html
--
Calendar l10n coordinator
Calendar Website Maintainer: http://www.mozilla.org/projects/calendar
Calendar developer blog: http://weblogs.mozillazine.org/calendar
Ricardo Palomares Martinez
> Lightning 0.5 and Sunbird 0.5 are available for Windows, Mac OS X
> (universal builds) and Linux in 22 different languages including English.
>
> Please use the following links to download the release:
>
> * Lightning 0.5 (fully localized for 22 languages)
> http://www.mozilla.org/projects/calendar/lightning/download.html
> * Sunbird 0.5 in English
> http://www.mozilla.org/projects/calendar/sunbird/download.html
> * Sunbird 0.5 in other languages
> http://www.mozilla.org/projects/calendar/sunbird/l10n_download.html
Please, be aware of these glitches when downloading:
It is somewhat difficult to download the Linux Lightning version from
Windows and viceversa. Even if your provides links to Windows, Linux
and Mac versions from your Lightning download page, AMO is /too/ smart
and just shows the one appropiate for the OS in which the browser is
running. You have to follow the link "Complete version history" to
access other systems.
With Sunbird localized, it is even worse. The links (at least, the
es-ES links) redirect to http://www.mozilla.org/ and no download seems
to appear. From main's mozilla.org, I've followed links thru Sunbird
0.5 -> Download Sunbird -> Other Systems & Languages -> (find my way
through the FTP site). It looks like the releases.mozilla.org
redirection is not working properly.
I was looking to provide the links from our l10n website, but I don't
think you want us to provide direct links to ftp.mozilla.org, so I
would be grateful if you let us know when these issues are fixed (if
they can be fixed).
BTW, congratulations! :-)
--
If it's true that we are here to help others,
then what exactly are the OTHERS here for?
Jim S
I cannot get it to downoad. It just takes me back and forth between two
pages!
> * Sunbird 0.5 in other languages
> http://www.mozilla.org/projects/calendar/sunbird/l10n_download.html
>
> Please read the release notes for Lightning 0.5[3] and Sunbird 0.5[4]
> before downloading.
>
> Thank you again to all our developers, contributors, localizers, testers,
> and supporters. We would not be able to do this without your assistance!
>
> Cya
> Simon
>
> [1] https://addons.mozilla.org/en-US/thunderbird/addon/4631
> [2]
> http://weblogs.mozillazine.org/rumblingedge/archives/2006/12/sb_0-5.html
> [3] http://www.mozilla.org/projects/calendar/releases/lightning0.5.html
> [4] http://www.mozilla.org/projects/calendar/releases/sunbird0.5.html
--
Jim S
Tyneside UK
http://www.jimscott.co.uk
Stefan Sitter
the builds need to show up on all mirrors and the links are
currently being fixed too (Bug 386093). So hopefully the links work
within the next hours.
/Stefan
Simon Paquet
>With Sunbird localized, it is even worse. The links (at least, the
>es-ES links) redirect to http://www.mozilla.org/ and no download seems
>to appear. From main's mozilla.org, I've followed links thru Sunbird
>0.5 -> Download Sunbird -> Other Systems & Languages -> (find my way
>through the FTP site). It looks like the releases.mozilla.org
>redirection is not working properly.
>
>I was looking to provide the links from our l10n website, but I don't
>think you want us to provide direct links to ftp.mozilla.org, so I
>would be grateful if you let us know when these issues are fixed (if
>they can be fixed).
They have been fixed temporarily as I have changed the website and added
the direct links to the ftp-server there.
Once the redirects are working across the whole mozilla-mirror-network, I
will switch back to the pointing to downloads.mozilla.org
Simon
Lionel Valero
Denis Bitouzé
Simon Paquet <si...@gmx.de> a écrit :
> Finally, the 0.5 release has been completed!
Many thanks!
--
Denis
Damian
most changes behind the scenes in the code? All of the things that still
seem undeveloped and beta'ish to me are exactly as they were.
Sorry to be talking sown something I've been waiting anxiously for, just
seemed to be a bit of an anti climax. It seems I still shouldn't take
the ageing 'Schedule+' off of my machine because it just does some thing
so much better.
Damian
Simon Paquet
>This is great news, but much less has changed than I was hoping for. Are
>most changes behind the scenes in the code? All of the things that still
> seem undeveloped and beta'ish to me are exactly as they were.
That may very well be, depending on your requirements. But quite a lot
has changed since 0.3.^1 as can be seen from the list of changes:
http://weblogs.mozillazine.org/rumblingedge/archives/2006/12/sb_0-5.html
Simon
Peter Lairo
> Finally, the 0.5 release has been completed!
Is there a way to get Lightning 0.5 to work with the Tb trunk builds?
I've tried editing the "maxversion" in install.rdf with no success.
--
Regards,
Peter Lairo
The browser you can trust: www.GetFirefox.com
Reclaim Your Inbox: www.GetThunderbird.com
Israel - Myths & Facts: http://www.jewishvirtuallibrary.org/
Dangers of Islam (German): http://www.PoliticallyIncorrect.de/
Church of the Flying Spaghetti Monster: http://www.venganza.org/
Stefan Sitter
> Is there a way to get Lightning 0.5 to work with the Tb trunk builds?
No. For Thunderbird trunk builds you need to use Lightning trunk
builds. Lightning trunk builds are only available as nightly builds
for testing.
/Stefan
Peter Lairo
OK. Thanks. I thought Lightning had only the trunk (no branches :-\ )
Andreas Tscharner
> [Followup-To mozilla.dev.apps.calendar set]
>
> Finally, the 0.5 release has been completed!
>
As I have mentioned, these releases don't work on the Debian build of
Thunderbird (called Icedove). It turned out, that I'm not the only
Debian user having these problems.
Anyway, there is a compiled XPI for Icedove here:
http://jean-christophe.dubacq.fr/index.php?post/2007/07/03/Lightning-05-for-icedove
Best regards
Andreas
--
Andreas Tscharner andreas....@metromec.ch
------------------------------------------------------------------------
And so at last the beast fell and the unbelievers rejoiced. But all was
not lost, for from the ash rose a great bird. The bird gazed down upon
the unbelievers and cast fire and thunder upon them. For the beast had
been reborn with its strength renewed, and the followers of Mammon
cowered in horror. -- The Book of Mozilla, 7:15
Arthur Norman
old postings about speed I suspect I HAVE to assume that there is
something truly stupid or terribly unusual that I a doing. I am trying
sunbird and find that it can be several MINUTES for it to start up and
several second for it to even page on by a week. It at present seems
beyond belief that this can be happening to others and not causing lots
of traffic here, so what is my mistake?
I have tried a 0.5 install on both Windows XP and on Vista Home Premium.
When those went badly I fetched sources and built from the latest
sources I could find on Fedora 7. I got my data by using FinchSync to
fetch from my ipaq to an .ics file which is only 1.6 Mbytes, 73K lines
and has the text BEGIN:VEVENT in it 6500 times. It has 562 lines that
start RRULE. My ipaq and MS Outlook can process that much data with no
significant pause - for ME I find that sunbird just gives the impression
of stalling, using all of one of the CPUs on my AMDx64x2 and up to
several hundred megabytes of memory.
I recompiled the Fedora version with "gcc -pg" to try to use gprof to
see where the time went, but even after installing a wrapper round
pthread_create to get clock interrupts into all threads (I hope!) gprof
faile dto record any of the time spent - so I had a really frustrating
time trying to and failing to install even that basic instrumentation to
let me understand for myself where the bottleneck was!
I also tried to import the .ics database into what I hope was a sunbird
native one. The import tooke AGES, and when it had happened the calendar
was slow on either database.
So I have tried selecting all but the last few months of date and going
"delete selected events". That is NOT what I want to do at all because
many annual recurring items get lost that way. But having started it
that has now been using CPU for quite a few hours.
My experience is not of the style of "10 seconds to do something
reasonably complicated": it is of much longer times to do things I
expect to be simple on a machine with a fast dual CPU and 3G of memory
that is not overloaded with other things.
I am doing all of this on one computer with no attempt to set up any
multi-computer scheme for sharing appointments and with all files on an
ordinarily fast local SATA HD. Well OK, the Fedora experiment was under
vmware, but that does not normally slow things down a terrible amount.
Advice please and feedback as to whether 1.6 Mbytes of text-file is
thought of as a "large" amount of datya these days...
Arthur
Arthur Norman
The Venkman tool has the ability to
> do some javascript profiling, and if you'd like to run it for us on your
> set of events, we'd be really appreciative.
Very many thanks - I have fetched that but will probably need a day or
so to work out exactly how to drive it, but if I get any info I will
pass it back. I will also perhaps expurgate my .ics file by putting
random text in in place of all my actual info in case that will make a
sample event-list towards the large size for others to play with.
>> Advice please and feedback as to whether 1.6 Mbytes of text-file is
>> thought of as a "large" amount of datya these days...
> This file has 5600 events in it. 500 of them are recurring. This is a
> very large file. At my last company we did quite a few informal surveys
> of our users and our executives and the largest calendars we found were
> those that had been used constantly for five years and contained about
> 4000 events.
That reassures me somewhat - I had not dreamed for many moments that my
case might be in any way *really* unusual. My data started on a Palm
IIIx in May 1999, moved to an iPaq a few years ago and is full of
teaching timetables and the like that REALLY help me keep track of where
I should be during term-time. I have never felt any pressure to purge
old appointments and indeed I just looked back through the years to the
"purchased a Palm IIIx" entry where I can be reminded of how much it
cost me. But knowing I can improve my experience by chucking away old
events may help.
OK, I have tried one tiny run with the venkman "profile" option, it was
to click "on a week" 3 or 4 times, and I get a load of stuff like the
following - if you would like me to email that in I will since the full
lot looks ugly and bulky for general posting... so if you say that
stuff like this will help I will send you more of it!!!! Thanks again.
Arthur
15 <file:/D:/Program%20Files%20(x86)/Mozilla%20Sunbird/js/calEvent.js>
calEvent.js: 2500 - 1000000 milliseconds
Function Name: anonymous (Lines 90 - 99)
Total Calls: 125962 (max recurse 0)
Total Time: 7391.14 (min/max/avg 0/10.74/0.06)
-------------------------------------------------------------------------------
calEvent.js: 750 - 2500 milliseconds
Function Name: anonymous (Lines 294 - 296)
Total Calls: 164663 (max recurse 0)
Total Time: 1840.7 (min/max/avg 0/6.84/0.01)
-------------------------------------------------------------------------------
calEvent.js: 75 - 250 milliseconds
Function Name: anonymous (Lines 142 - 144)
Total Calls: 3495 (max recurse 0)
Total Time: 146.48 (min/max/avg 0/0.98/0.04)
-------------------------------------------------------------------------------
calEvent.js: 10 - 50 milliseconds
Function Name: anonymous (Lines 133 - 135)
Total Calls: 85 (max recurse 0)
Total Time: 41.99 (min/max/avg 0/0.98/0.49)
Function Name: anonymous (Lines 146 - 148)
Total Calls: 3410 (max recurse 0)
Total Time: 29.29 (min/max/avg 0/0.98/0.01)
Michiel van Leeuwen
> OK, I have tried one tiny run with the venkman "profile" option, it was
> to click "on a week" 3 or 4 times, and I get a load of stuff like the
> following - if you would like me to email that in I will since the full
> lot looks ugly and bulky for general posting... so if you say that
> stuff like this will help I will send you more of it!!!! Thanks again.
> Arthur
The profile you show below looks exactly what we need to try to find out
what is taking so much time.
Could you post the file on some public place, or else send it to me?
Thanks,
Michiel
Dan Mosedale
Give that all the functions are "anonymous", going in and giving the
functions in those files all names would probably make that data
significantly more useful.
Dan
Michiel van Leeuwen
> The profile you show below looks exactly what we need to try to find out
> what is taking so much time.
> Could you post the file on some public place, or else send it to me?
I've taken a quick look at the profile, and one part I noticed was this:
calItemBase.js: 2500 - 1000000 milliseconds
Function Name: anonymous (Lines 128 - 163)
Total Calls: 6539 (max recurse 0)
Total Time: 65796.53 (min/max/avg 0/1048.76/10.06)
Function Name: anonymous (Lines 354 - 366)
Total Calls: 751092 (max recurse 1)
Total Time: 18160.99 (min/max/avg 0/4081.77/0.02)
Function Name: anonymous (Lines 750 - 755)
Total Calls: 400898 (max recurse 0)
Total Time: 8426.26 (min/max/avg 0/4081.77/0.02)
Function Name: anonymous (Lines 541 - 629)
Total Calls: 6514 (max recurse 0)
Total Time: 6211.5 (min/max/avg 0/671.83/0.95)
the first is making the item immutable. That takes quite a bit of time,
and shouldn't be all that necessary. After all, we are creating the
event. It should be possible to just set the immutable bit, instead of a
lot of copying around of data. This is something that needs some attention.
The next two are for getting properties of an item. That should not take
this long. It should be really fast, because it's simple. This might be
related to bug 362762. A possible workaround is in bug 382121. I think
we should get that workaround in.
The last is for creating the items. That isn't even the bulk of the
time. But I can imagine that taking some time, because a lot of work is
done.
There might be more in the profile, but it's a but hard to read, since
you can't get the time spend inside a function, and you can't sort in
any way. But that's not something you can do anything about. Thanks for
profiling this!
Michiel