Nabble - Dojo (Old Archive)

Re: Extending dojox.grid.DataGrid - How To Set Formatter?

2009-10-21T21:16:14Z

Did you fix this? I had the same problem and fixed it thusly :

{... formatter: function () { return this.grid.formatExpdate(arguments); }}

hope that helps...

Context of this in the view is the cell you're operating on ...

Re: Sending JSON Data to a server

2009-10-17T08:51:17Z

Hi,

I am new to Dojo.
Do you have any example that will send JSON data to server and receive the JSON data from server using XHRPost, then populate into grid ?

Thanks,
Techdisc

Query regarding Dojo Type Ahead component(FilteringSelect)

2009-09-09T16:09:11Z

Hi,

Dojo's Type ahead component (FilteringSelect) defaults behavior is, as user typing in text it auto populates the very first item in the dropdown list into the text box. Is there any way to switch off this auto population by changing configuration on the FilteringSelect component?

Only way I could do is by modifying the dojo source code. Just wanted to know if there is any better solution available.

Thanks In Advance,
Lax

dojo-1.3.2 grid headers misaligned

2009-09-09T13:49:56Z

Hi,

I was creating a programmatic grid with column width = 'auto' and I see that the column headers and columns are misaligned. Looks like a 1.3 bug, since it use to work fine in 1.2.3.

Here's the code:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title>Test dojox.grid.DataGrid Programmatic Instantiation</title>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"></meta>
<style type="text/css">
@import "../../../dojo/resources/dojo.css";
@import "../resources/Grid.css";
@import "../resources/tundraGrid.css";
.heading {
font-weight: bold;
padding-bottom: 0.25em;
}

#grid {
border: 1px solid #333;
width: 50em;
height: 30em;
}
</style>
<script type="text/javascript" src="../../../dojo/dojo.js"
djConfig="isDebug:false, debugAtAllCosts: false, parseOnLoad: true"></script>
<script type="text/javascript">
dojo.require("dijit.dijit"); // optimize: load dijit layer
dojo.require("dojox.grid.DataGrid");
dojo.require("dojo.data.ItemFileWriteStore");
dojo.require('dijit.Menu');
dojo.require("dojo.parser");
</script>
<script type="text/javascript" src="support/test_data.js"></script>
<script type="text/javascript">
var grid = null;
dojo.addOnLoad(function(){
var layout = [[
{name: 'Column 1', field: 'col1', width: 'auto'},
{name: 'Column 2', field: 'col2', width: 'auto'},
{name: 'Column 3', field: 'col3', width: 'auto'}
]];

grid = new dojox.grid.DataGrid({
id: "grid",
store: test_store,
structure: layout
}, document.createElement('div'));
dojo.byId("gridContainer").appendChild(grid.domNode);
grid.startup();
});
</script>
</head>
<body class="tundra">
<div class="heading">dojox.grid.DataGrid Programmatic Instantiation Test</div>
<div id="gridContainer"></div>
</body>
</html>

Re: build from svn requires rhino?

2009-09-09T13:19:24Z

Doh. Nevermind. This only happened because I ran the build script
from outside its directory.

Rob

On Sep 9, 2:35 pm, Rob <rob.nikan...@...> wrote:

> It mentions Rhino in the prerequisites, but only as the reason for
> needing Java. So I thought maybe the Rhino jars were in the source
> tree somewhere. But I get this when trying to build...
>
> Exception in thread "main" java.lang.NoClassDefFoundError: org/
> mozilla/javascript/tools/shell/Main
>
> So do I need to download Rhino?
>
> thanks,
> Rob
>
> _______________________________________________
> FAQ:http://dojotoolkit.org/support/faq
> Book:http://docs.dojocampus.org
> Dojo-inter...@...://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

build from svn requires rhino?

2009-09-09T11:35:07Z

Hi,

I want to start using the lastest code from SVN, so I'm reading this
page here:

http://docs.dojocampus.org/build/index

It mentions Rhino in the prerequisites, but only as the reason for
needing Java. So I thought maybe the Rhino jars were in the source
tree somewhere. But I get this when trying to build...

Exception in thread "main" java.lang.NoClassDefFoundError: org/
mozilla/javascript/tools/shell/Main

So do I need to download Rhino?

thanks,
Rob

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: Drag a whole Form

2009-09-09T10:51:50Z

Just specify "skipForm: true" on your source, like that:

<div dojoType="dojo.dnd.Source" skipForm="true"
id="dragForm" class="source">
<div dojoType="dijit.form.Form"
id="form" class="dojoDndItem">
<input type="text" id="tbHost"
dojoType="dijit.form.TextBox" name="host"/>
</div>
</div>

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/09/2009 06:53 AM, John Bindby wrote:

> Harry,
>
> How did you solve it ?
>
> I have the same problem, but just in Firefox.
>
> /John
>
> Harry Brandenburg wrote:
>> Ok problem solved!
>> Thanks anyway.
>>
>>
>> Greetz
>> Sprigg
>>
>> 2009/9/9 Harry Brandenburg <harrybrandenburg@...
>> <mailto:harrybrandenburg@...>>
>>
>> Hi,
>> i have a Problem with dragging a whole dijit.form.Form. The
>> Dragging works well, but now i can't focus my textboxes of the
>> form anymore. I guess with a click on them i'm activating the
>> drag. Is there any option to set the priority of the textboxes higher?
>>
>> Here a bit of the Code:
>> <div dojoType="dojo.dnd.Source" id="dragForm" class="source">
>> <div dojoType="dijit.form.Form" id="form"
>> class="dojoDndItem">
>> <input type="text" id="tbHost"
>> dojoType="dijit.form.TextBox" name="host"/>
>> </div></div>
>>
>> Greetings
>>
>> Sprigg
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> FAQ: http://dojotoolkit.org/support/faq
>> Book: http://docs.dojocampus.org
>> Dojo-interest@...
>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>
>
>

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: DateTextBox - value is not commited

2009-09-09T05:36:27Z

Hi Bill,
thanks a lot. There was a mistake in the generated HTML-code, confused
the form with a second form on the same page.

The firebug was new to me, great help for finding such mistakes!

Greetings
Kerstin

Bill Keese schrieb:

> Hmm, it looks OK to me, not sure why it isn't working.
>
> You can do a "view source" in your browser to make sure your PHP code
> generated the HTML you expected, and then in firebug make sure that
> inside the widget there's a hidden <input> field with name=datum set on it.
>
> IIRC there was a problem with this in the past, where we set name via
> dojo.attr() rather than at creation time, and IE didn't work right, but
> we fixed that.
>
> On Wed, Sep 9, 2009 at 8:27 PM, Kerstin <kfinke@...
> <mailto:kfinke@...>> wrote:
>
> Hi,
>
> I'm new with dojo and try to implement a datetextbox in a form.
>
> It looks well but when I submit the form the value is not commited in
> the POST-Variable.
>
> When I replace the DateTextBox by a standard-Textbox with id="datum" it
> works fine.
>
> Any ideas?
>
> Thanks in advance
> Kerstin
>
>
> Here's my PHP-code:
>
> $out_cont .= '<form
> action="index.php?module=anwesenheiten&action=view&id='.$id.'"
> method="post" name="anwesenheiten">';
>
> $heute = date("Y-m-d");
> if(empty($datum)) {
> $datum = $heute;
> }
>
> $out_cont .= '<input type="hidden" name="lg_id"
> value="'.$id.'">';
> $out_cont .= '<input type="text" id="datum" name="datum"
> value="'.$heute.'"
> dojoType="dijit.form.DateTextBox"
> promptMessage="Bitte Datum eintragen"
> constraints="{max:\''.$heute.'\'}"
> invalidMessage="Eingabe ungueltig! Bitte geben
> Sie das
> Datum im Format TT.MM.JJJJ an."
> required="true" />';
> $out_cont .= '<input type="submit">';
> $out_cont .= '</form>';
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> <mailto:Dojo-interest@...>
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

DateTextBox - value is not commited

2009-09-09T04:24:05Z

Hi,

I'm new with dojo and try to implement a datetextbox in a form.

It looks well but when I submit the form the value is not commited in
the POST-Variable.

When I replace the DateTextBox by a standard-Textbox with id="datum" it
works fine.

Any ideas?

Thanks in advance
Kerstin

Here's my PHP-code:

$out_cont .= '<form
action="index.php?module=anwesenheiten&action=view&id='.$id.'"
method="post" name="anwesenheiten">';

$heute = date("Y-m-d");
if(empty($datum)) {
$datum = $heute;
}

$out_cont .= '<input type="hidden" name="lg_id" value="'.$id.'">';
$out_cont .= '<input type="text" id="datum" name="datum"
value="'.$heute.'"
dojoType="dijit.form.DateTextBox"
promptMessage="Bitte Datum eintragen"
constraints="{max:\''.$heute.'\'}"
invalidMessage="Eingabe ungueltig! Bitte geben Sie das
Datum im Format TT.MM.JJJJ an."
required="true" />';
$out_cont .= '<input type="submit">';
$out_cont .= '</form>';

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

change font on dijit form input etc.

2009-09-08T19:25:17Z

Hi All,

I am working on a project were we want to display new chinese style
characters which are in no other font then we created ourself.

I looked quite a bit in the css files of tundra where I can change the
font... But in the end the only way I found to have an inline style in
the <input> tag itself.

I have one css I load last - should be able to overwrite all other
definitions.

I would very appreciate your help a lot!

Greetings - Achim
_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Display all the nodes of a tree

2009-09-08T08:45:47Z

Hi all,

I have a tree built from an ItemFileReadStore with few items but nested. I would like it to be completely open at any time (showing all leafs and disabling collapse). I guess it's uber-easy but I can only find the openOnClick attribute. What am I missing?

Regards

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Multiple Dialog support in 1.4

2009-09-07T11:46:36Z

Hi all,

I'm having some issues (minor but annoying nonetheless) with my current implementation of multiple Dialogs. I've seen http://archive.dojotoolkit.org/nightly/dojotoolkit/dijit/tests/test_Dialog.html to check the new multidialog support in 1.4 and I'm a little bit unsure because I've noticed the second dialog completely hides the first (I guess intentionally) but that won't be true in many situations and the visual effect may be unclear. I also missed a test with a lightbox launched from a Dialog (a use case that I also need). I was considering reimplementing my code (it controls the number of dialogs currently being shown and hides one before opening another using setTimeouts.. that seriously affect speed..) by sliding out the current visible dialog and showing the next onEnd (and all the way back, of course) but I have doubts because I may very well be wasting time that I could invest in new features. What's your recommendation? Given that the multiple dialog support is one of the main features in the next release of the widget could you add some more tests that better highlight the functionality?

Regards

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Dojox ContentPane and addOnLoad

2009-09-07T05:57:26Z

Hi all,

I'm trying to load some page fragment (no html, head or body tags) inside a dojox.layout.ContentPane widget (inside a tab container and loaded with href). This fragment includes several widgets and scripts. Everyhting works but one thing. I have calls to dojo.connect wrapped inside dojo.addOnLoad calls to detect events happening on widgets (for example, on dialog show). They seem to fail when loaded using AJAX because the widget is not ready when the script is evaluated. My guess is that:

Scripts are executed before widgets are parsed
dojo.addOnLoad is executed directly (the page is loaded after all)

I would like to change this kind of behavior because otherwise I need to clutter the code with something like function whenDialogIsReady(function()...) in addition to the addOnLoad wrapper. How can I force the lazy evaluation of scripts in there? Or how could I change the addOnLoad behavior to wait till the widgets are ready?

Regards

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Having several dijit.form.editors sharing a toolbar

2009-09-07T04:48:49Z

Hi,

I've been trying to grok the editor code that deals with generating the toolbar (that hold the bold, italic, et.c. icons for the editor and editor plugins) a while back and I found that I pretty much didn't :)

What I wonder is if anyone could gauge how hard it would be to ;

a) Separate the element that wrap the toolbar for an editor and append it to a different place from the editor content area

b) Have two or more editors share a common place for their toolbar, (which switches depending on which editor get the context).

I think that if I could get a hint of (1) I could probably do (2).

Also, if nobody have time to answer this, I won't blog about it :) I'll just figure it out myself (in six months or so).

Cheers,
PS

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Mailing list Rocks!

2009-09-06T16:21:51Z

25th time answering questions about mailing lists --- a jubilee!

1) Nabble (web interface).

Dojo forum that reflects this mailing list:
http://www.nabble.com/Dojo-%282%29-f36462.html

You can read/write from Nabble.

2) Gmane (web interface)

Dojo forum that reflects this mailing list:
http://dir.gmane.org/gmane.comp.web.dojo.user

You have a choice:

a) threads:
http://news.gmane.org/gmane.comp.web.dojo.user

b) flat blog-like:
http://blog.gmane.org/gmane.comp.web.dojo.user

You can read/write from Gmane.

3) Gmane (NNTP)

Subscribe with your favorite news reader:
nntp://news.gmane.org/gmane.comp.web.dojo.user

You can read/write using NNTP.

This is my choice --- I use Mozilla Thunderbird to read NNTP. It can
show a news group in a variety of ways, and it has message filters to boot!

While anybody can read, in order to write with options 1-3 you have to
subscribe to the mailing list:
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

The trick is to go to the mailing list settings, and disable all emails
(provided you don't want them) --- this way it is a pure "pull" schema
--- you go and read the mailing list whenever you like without an
avalanche of emails.

Gmane is good enough to carry open-source mailing lists of public
interest. If it is not there, you can always ask to add --- I did it on
multiple occasions.

Of course there is one more option:

4) Getting email.

It is not that bad as it sounds to many. There are some mailing lists I
don't feel like pushing on Gmane, so I use this option. Obviously I
don't want to be overwhelmed by 1000s of emails per day, so I organize.

a) gmail allows to classify emails using simple criteria. It is in
settings. Usually I use a sender's email, or a prefix in subject to
classify my email. It is in settings --- check it out. When email fits
my criteria I label it (in the settings too) and archive it. Later on I
click on the label and see new emails.

Useful hint: if your email is abcdefg@..., you can use
absdefg+something@... too. For example, you can subscribe to dojo
mailing lists using absdefg+dojo@..., and process all emails that
came to this address differently, e.g., forward them, or label them.

b) Actually it is a variation of (a) --- each label is a separate
folder, if you access gmail using IMAP. Just use your favorite email
program.

c) Thunderbird (and many other email programs) can classify email too. I
created a folder for email related to dojo checkins and Thunderbird
moves all such email to this folder. I can see how many unread items I
have on this topic, I can scan them when I have time, and mark all of
them as read.

Hopefully you can find whatever work for you. The point is if you have a
forum, you are stuck with it. With mailing lists options are countless.
That's why I always supported the mailing list option.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/06/2009 05:27 PM, Micheil Smith wrote:

> Well put iRoo, I'm just wondering if we'd be better off moving the
> mailing lists
> over to something like google groups, so that that way there's the solid
> forum style layout, but it's also a mailing list?
>
> - Micheil
>
> Insane Kangaroo wrote:
>> I read to the fourth paragraph and seen he wanted to still use Dojo,
>> even before I read to the end. The guy is a regular textbook case.
>>
>> He wants to use Dojo on *his* terms. He is welcomed to run a forum by
>> himself. Maybe he will contribute the cost of VBulletin to Dojo since he
>> wants to run a forum?
>>
>> I don't care much for his attitude and twisting of words when it comes
>> to "why the ML sucks." He sent the "Mailing list sucks" email with
>> malicious intent to waste interrupt a public mailing list.
>>
>> iRoo
>>
>>
>> Shane O'Sullivan wrote:
>>
>>> I saw that too, and it's a bit over the top.
>>>
>>> I wrote a long and (I think) reasonable comment on that post
>>> yesterday, but Chris as not approved it, even though other comments
>>> have been approved since mine.
>>>
>>> This makes sense unfortunately, given Chris' tone in these emails.
>>> It's a bit sad really. Ah well, all we can do is make Dojo better for
>>> everyone else I guess.
>>>
>>> Thanks
>>>
>>> Shane
>>>
>>> 2009/9/6 Jose Noheda <jose.noheda@...>:
>>>
>>>> This is more than I can accept really
>>>> http://www.dzone.com/links/rss/what_is_the_future_of_dojo.html
>>>> You have to develop a better tool than they could, in your free time, for
>>>> free, provide them the bandwith and the documentation, examples and a
>>>> campus, provide a mailing list and an IRC channel and your valuable time to
>>>> try to help the issues that may arise, build a foundation and side projects.
>>>> But no, it's not enough. So yeah, and this is a bandwagon I was not going to
>>>> join, go away don't use dojo! Forget completely about us and use whatever
>>>> other framework you want. Pretty please.
>>>> Regards
>>>>
>>>> On Fri, Sep 4, 2009 at 11:55 PM, Jean-Rubin Leonard <jrleonard@...>
>>>> wrote:
>>>>
>>>>> I didnt want to add to this but I feel I have to give my 2 canadian cents.
>>>>> I considered myself a total noob at dojo not too long ago. I joined as the
>>>>> forums was agonizing. I must say that while there are some weaknesses in the
>>>>> doc there are tools out there to help you along the dojo path (I don't know
>>>>> about grid). Using the mailing list + joining #dojo on irc is a two pronged
>>>>> approach that has allowed me to move forwards in my use of dojo.
>>>>> Now I'm not such a newb anymore. This is open source people, you must give
>>>>> a little in order to receive.
>>>>> JR
>>>>>
>>>>> On Fri, Sep 4, 2009 at 5:23 PM, Nikolai Onken <nonken@...>
>>>>> wrote:
>>>>>
>>>>>> Hi Chris and everybody else,
>>>>>> I've been reading over the thread and just want to say a couple of things
>>>>>> which are seriously important:
>>>>>> - Lets keep the tone right, just because we are sending virtual messages
>>>>>> it doesn't mean we can talk like we know each other really well :) - As
>>>>>> James states very rightly. Noone is doing anyone good with writing emails
>>>>>> without actual content, so lets try not to do this again.
>>>>>> - Of course it would be a pity to lose you as a Dojo user, and email
>>>>>> threads shouldn't be affecting this (especially threads with namebashing)
>>>>>> If you have issues with how certain things are setup, join the discussion
>>>>>> (if it is really important to you hang out during the dojo meetings on
>>>>>> wednesdays), we are trying to be open for any type of suggestion - sometimes
>>>>>> it is not easy to do so, so as others said be patient and persistent.
>>>>>> The efforts of getting the new website, docs and community resources up
>>>>>> and running are very much on the way currently and it really is a matter
>>>>>> weeks until you should have a much more clearly structured website, so you
>>>>>> can find what you are looking for hopefully in a better way.
>>>>>> So, heads up, and maybe join the irc channel more often and keep dojoing.
>>>>>> We all should have a dojo.beer() ;)
>>>>>> Regards,
>>>>>> Nikolai
>>>>>>
>>>>>> On Sep 4, 2009, at 10:53 PM, Chris Hardin wrote:
>>>>>>
>>>>>> This "ignorant whiner" is jumping ship right now. I am leaving the
>>>>>> mailing list and relegating the posts to the spam folder. It's obvious some
>>>>>> of you are frogs in slowly boiling water and you're not interested in
>>>>>> accomodating more people to get involved. This entire thread has made up my
>>>>>> mind. My next project will be Ext or another framework instead of Dojo. Some
>>>>>> real change needs to occur in the ranks here.
>>>>>>
>>>>>> Captain Kangaroo is right about one thing, if you don't like it leave. If
>>>>>> enough people stop participating in this ridiculous list, it will change.
>>>>>> It's a shame what I considered a great and powerful framework has been
>>>>>> reduced to this.
>>>>>>
>>>>>> Adios
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Fri, Sep 4, 2009 at 3:43 PM, Stefan Vesterlund
>>>>>> <stefan.vesterlund@...> wrote:
>>>>>>
>>>>>>> I don't know if 'stupid' beats 'insane', or if it matters, but Insane
>>>>>>> Kangaroo make some points that should be considered.
>>>>>>> It is IMO a quite common attitude that "it's free, so shut up" and "if
>>>>>>> you don't like it the way it is, go elsewhere". I think it must be
>>>>>>> remembered that all the people that is working for within an open source
>>>>>>> community is investing a lot of time and effort into it. They want it to
>>>>>>> be a success. The more effort they put into it, the more they _need_ it
>>>>>>> to be a success. Nobody can live on work alone because you can not eat a
>>>>>>> code snippet and you can not build a house out of widgets. Success is
>>>>>>> mainly measured in how many people are using the product of their effort
>>>>>>> and how satisfied they are.
>>>>>>>
>>>>>>> It is not only the developers who is investing time and effort, it is
>>>>>>> also the end users. A common process when selecting a product is to
>>>>>>> first of all get an overview of what is available for you. Then you
>>>>>>> compare them, both functionality and cost. Then you check the support
>>>>>>> options. Then you try to get an idea of what the users think about the
>>>>>>> product. After that you make a choice. If it turns out that the product
>>>>>>> failed to live up to your expectations you either live with it or you
>>>>>>> abandon it. If you abandon it, you will most likely advice other people
>>>>>>> to use something else as well if they ask you.
>>>>>>>
>>>>>>> Since no one that has comitted themselves to a product really want to
>>>>>>> abandon it, the end users are very much in the same situation as the
>>>>>>> people more directly involved in the development - they want the product
>>>>>>> to be a success so that the time and effort (which equals to money in
>>>>>>> most cases) is not wasted. We all have a responsibility in improving the
>>>>>>> product. Complaints are annoying, but if there is a lot of complaints
>>>>>>> maybe it needs that something should be changed. It is always easier to
>>>>>>> complain than to respond to complaints though, and one-dimensional
>>>>>>> complaints are rarely leading to anything but unrest and irritation.
>>>>>>> Irritated responses to irritated complaints leads to more irritation.
>>>>>>>
>>>>>>> So let's behave.
>>>>>>>
>>>>>>> Thanks
>>>>>>> Stefan
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Insane Kangaroo schrieb:
>>>>>>>
>>>>>>>> Excuse me... all of you who are complaining.
>>>>>>>>
>>>>>>>> Piss off.
>>>>>>>>
>>>>>>>> You're using a free resource, which is the mailing list. Don't like
>>>>>>>> it?
>>>>>>>> Start your own forum.
>>>>>>>>
>>>>>>>> I've heard complaints of "Dojo mail will fill up my INBOX." How stupid
>>>>>>>> are people? USE A SEPARATE EMAIL ACCOUNT FOR DOJO. This is not rocket
>>>>>>>> science. Can't search the list? Learn how to google, you obviously
>>>>>>>> don't
>>>>>>>> know how. A forum will not be any easier.
>>>>>>>>
>>>>>>>> Those who don't like the mailing list, tough. Want immediate Dojo
>>>>>>>> support? PAY FOR IT! We, the community, are not your own personal
>>>>>>>> bitch,
>>>>>>>> stop treating us like we're supposed to bend over for every question.
>>>>>>>>
>>>>>>>> Stop starting and contributing to these meaningless threads. Spend
>>>>>>>> your
>>>>>>>> time on developing or performing some meaningful action instead of
>>>>>>>> wasting time.
>>>>>>>>
>>>>>>>> Thank you very much to the developers and contributors. Everyone else
>>>>>>>> can go screw themselves.
>>>>>>>>
>>>>>>>> Sincerely,
>>>>>>>> Insane Kangaroo
>>>>>>>>
>>>>>>>>
>>>>>>>> Terry Carnes wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>> Peter,
>>>>>>>>>
>>>>>>>>> I can definitely identify with Chris. I've tried getting answers and
>>>>>>>>> have
>>>>>>>>> found that if the answer may not be readily available, the question
>>>>>>>>> quickly
>>>>>>>>> gets buried under a mass of other emails. I would then try to break
>>>>>>>>> down my
>>>>>>>>> problem into different parts and try to ask related questions that
>>>>>>>>> might
>>>>>>>>> help me accomplish my original task. But did not have very good
>>>>>>>>> results.
>>>>>>>>>
>>>>>>>>> I went back over my email questions for the past month. I averaged a
>>>>>>>>> 25%
>>>>>>>>> response rate.
>>>>>>>>>
>>>>>>>>> Date Subject
>>>>>>>>> Response
>>>>>>>>>
>>>>>>>>> ============================================================================
>>>>>>>>> ==========================
>>>>>>>>> 8/4/09 dojo grid - How to do Master with expanding Detail
>>>>>>>>> 1 - link to
>>>>>>>>> online example
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/11/09 dojo grid - How to do Master with expanding Detail
>>>>>>>>> none
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/14/09 dojo grid - How to do Master with expanding Detail
>>>>>>>>> none
>>>>>>>>> (included code of what I'm trying)
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/15/09 dojo grid - How to do Master with expanding Detail
>>>>>>>>> 1-
>>>>>>>>> suggestion to use Tree grid and
>>>>>>>>> (update and further questions)
>>>>>>>>> offer of sample code
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/17/09 Dojo Grid - How to use inRowIndex to get value from
>>>>>>>>> none
>>>>>>>>> Column
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/18/09 Dojo Grid - Trying to create Detail Rows
>>>>>>>>> none
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 8/18/09 dojo grid - How to do Master with expanding Detail
>>>>>>>>> none
>>>>>>>>> (taking up earlier offer for sample code)
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 9/1/09 Where do I find documentation for Dojo DataGrids
>>>>>>>>> none
>>>>>>>>> and use of Layout, get and formatter
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 9/1/09 Dojo DataGrid - combining fields in Get
>>>>>>>>> 2 - very
>>>>>>>>> helpful replies!
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 9/2/09 Sorting on Grid with sub row fails - bug?
>>>>>>>>> None
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>> 9/3/09 Sorting DataGrid with colspan dies
>>>>>>>>> none
>>>>>>>>> so far
>>>>>>>>>
>>>>>>>>> ----------------------------------------------------------------------------
>>>>>>>>> --------------------------
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Terry Carnes
>>>>>>>>> www.peacehilltechnologies.com
>>>>>>>>> www.peacehillfarm.com
>>>>>>>>> www.anemergentagrarian.com
>>>>>>>>>
>>>>>>>>> -----Original Message-----
>>>>>>>>> From: dojo-interest-bounces@...
>>>>>>>>> [mailto:dojo-interest-bounces@...] On Behalf Of
>>>>>>>>> Peter E
>>>>>>>>> Higgins
>>>>>>>>> Sent: Friday, September 04, 2009 8:12 AM
>>>>>>>>> To: dojo-interest@...
>>>>>>>>> Subject: Re: [Dojo-interest] Mailing list Sucks
>>>>>>>>>
>>>>>>>>> Peter Higgins wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> Most of the threads in my inbox have responses or go on and on. Did
>>>>>>>>>> you
>>>>>>>>>> ask something specifically and it was missed?
>>>>>>>>>>
>>>>>>>>>> Regards,
>>>>>>>>>> Peter
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Peter E Higgins
>>>>>>>>> Dojo Project Lead : http://dojotoolkit.org
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> FAQ: http://dojotoolkit.org/support/faq
>>>>>>>>> Book: http://docs.dojocampus.org
>>>>>>>>> Dojo-interest@...
>>>>>>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>>>>>>
>>>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> FAQ: http://dojotoolkit.org/support/faq
>>>>>>> Book: http://docs.dojocampus.org
>>>>>>> Dojo-interest@...
>>>>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>>>>
>>>>>> _______________________________________________
>>>>>> FAQ: http://dojotoolkit.org/support/faq
>>>>>> Book: http://docs.dojocampus.org
>>>>>> Dojo-interest@...
>>>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> FAQ: http://dojotoolkit.org/support/faq
>>>>>> Book: http://docs.dojocampus.org
>>>>>> Dojo-interest@...
>>>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>>>
>>>>>>
>>>>> _______________________________________________
>>>>> FAQ: http://dojotoolkit.org/support/faq
>>>>> Book: http://docs.dojocampus.org
>>>>> Dojo-interest@...
>>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>>
>>>>>
>>>> _______________________________________________
>>>> FAQ: http://dojotoolkit.org/support/faq
>>>> Book: http://docs.dojocampus.org
>>>> Dojo-interest@...
>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>
>>>>
>>>>
>>> _______________________________________________
>>> FAQ: http://dojotoolkit.org/support/faq
>>> Book: http://docs.dojocampus.org
>>> Dojo-interest@...
>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> FAQ: http://dojotoolkit.org/support/faq
>> Book: http://docs.dojocampus.org
>> Dojo-interest@...
>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

[DO NOT REPLY] test email from google groups

2009-09-06T16:14:43Z

*flops on the mailing list*

*sits here laying staring at people*
_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: dojo.forEach scope

2009-09-04T20:25:48Z

First of all --- I am not bashing jQuery. And I am not in the business
of impressing others.

According to your numbers you had a 2x savings. If your payload consists
of something more than iterating through a single array, or you do it in
a tight loop, or in a fast event handling code (see the link I gave to
you earlier) this savings can be substantial.

Then again it all depends on tasks at hand --- if all you want is to add
two numbers it really doesn't matter how fast is the addition: 1ns or
100ms --- they both will be "instantaneous".

jQuery and Dojo target different markets, and I trust that they are both
successful in what they want to solve.

But I trust it is beyond the original point. If you want to re-formulate
it, please start a different thread.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/04/2009 10:01 PM, Les Szklanny wrote:

> Here's my example, only 100,000 entries :)
>
> var a = [];
> for(var i = 0; i < 100000; ++i) a.push(i);
>
> console.time('dojo');
> dojo.forEach(a, function(){ return true; });
> console.timeEnd('dojo');
>
> console.time('jQuery');
> jQuery.each(a, function(){ return true; });
> console.timeEnd('jQuery');
>
> Results:
> dojo: 91ms
> jQuery: 208ms
>
> So I saved 117ms by using dojo.forEach on an array that has 100,000 entries.
> This is not an impressive number.
>
> On Fri, Sep 4, 2009 at 9:40 PM, Eugene Lazutkin
> <eugene.lazutkin@... <mailto:eugene.lazutkin@...>> wrote:
>
> I didn't expect that you want me to test the obvious, but sure, I can
> spare 2 minutes:
>
> var a = [];
> for(var i = 0; i < 1000000; ++i) a.push(i);
> var dojoTime = new Date().getTime();
> dojo.forEach(a, function(){ return true; });
> dojoTime = new Date().getTime() - dojoTime;
> var jqTime = new Date().getTime();
> jQuery.each(a, function(){ return true; });
> jqTime = new Date().getTime() - jqTime;
> console.log("jQuery is ",
> (jqTime / dojoTime).toFixed(2),
> " slower than Dojo on iterating through arrays");
>
> On my computer (FF/Linux 3.5) it prints:
>
> jQuery is 2.93 slower than Dojo on iterating through arrays
>
> I ran it 10 times and the number changes from time to time but the
> result is > 2.7 on all attempts.
>
> Google Chrome for Linux:
>
> jQuery is 3.09 slower than Dojo on iterating through arrays
>
> Opera 10.0 for Linux:
>
> jQuery is 1.49 slower than Dojo on iterating through arrays
>
> I think if we reverted to the native implementation, numbers would be
> even more impressive. And dojox.lang.functional.array methods being more
> lax are generally faster than the implementation in the Dojo base.
>
> Thanks,
>
> Eugene Lazutkin
> Dojo Toolkit, Committer
> http://lazutkin.com/
>
> On 09/04/2009 09:10 PM, Les Szklanny wrote:
> > Can you come up with an example that would back up your statement that
> > there's any noticeable performance gain by not checking the return
> value
> > of the function?
> >
> > On Fri, Sep 4, 2009 at 8:48 PM, Eugene Lazutkin
> > <eugene.lazutkin@... <mailto:eugene.lazutkin@...>
> <mailto:eugene.lazutkin@...
> <mailto:eugene.lazutkin@...>>> wrote:
> >
> > 12ms?
> >
> http://lazutkin.com/blog/2008/mar/23/javascript-edp-and-0ms-timeouts/
> > --- to save on reading scroll down to the table.
> >
> > And allow me to disagree politely with your performance vs.
> convenience
> > assessment.
> >
> > Thanks,
> >
> > Eugene Lazutkin
> > Dojo Toolkit, Committer
> > http://lazutkin.com/
> >
> > On 09/04/2009 08:04 PM, Les Szklanny wrote:
> > >>>> There is a reason why forEach doesn't support early
> termination:
> > > performance. If it checks for an exit value, it will
> penalize code
> > that
> > > doesn't need to terminate early.
> > >
> > > Performance degradation is negligible - forEach is used for
> > > convenience, not for performance anyway.
> > >
> > > Run the code below at http://docs.dojocampus.org/dojo/NodeList
> > >
> > > var els = dojo.query('*');
> > > console.log(els.length)
> > > console.time('dojo');
> > > dojo.forEach(els, function(e){return true;});
> > > console.timeEnd('dojo');
> > > console.time('jQuery');
> > > jQuery.each(els, function(e){return true;});
> > > console.timeEnd('jQuery');
> > >
> > > http://docs.dojocampus.org/dojo/NodeList
> > >
> > > I see these results on my computer:
> > >
> > > elements: 1616
> > > dojo: 12ms
> > > jQuery: 12ms
> > >
> > > You will need to jQuerify the page:
> > >
> > > http://www.learningjquery.com/2006/12/jquerify-bookmarklet
> > >
> > >
> > >
> > > On Fri, Sep 4, 2009 at 6:43 PM, Eugene Lazutkin
> > > <eugene.lazutkin@...
> <mailto:eugene.lazutkin@...> <mailto:eugene.lazutkin@...
> <mailto:eugene.lazutkin@...>>
> > <mailto:eugene.lazutkin@...
> <mailto:eugene.lazutkin@...>
> > <mailto:eugene.lazutkin@...
> <mailto:eugene.lazutkin@...>>>> wrote:
> > >
> > > Array functions are patterned after the array methods in the
> > upcoming
> > > ECMA-262 5th edition. The idea was to provide shims for
> > methods of the
> > > same name, or delegate to them if they are available. We
> tried
> > to match
> > > the behavior down to the argument list, and how the callback
> > is called.
> > >
> > > (Sidenote: we need to provide the fallback to native
> methods.
> > Right now
> > > we don't do that because of the performance problems
> with Firefox'
> > > implementation.)
> > >
> > > We used the Mozilla spec at that moment because that was
> the only
> > > codified thing available. Now, when the release candidate is
> > published,
> > > we can revise our implementation to match it more closely.
> > Namely the
> > > standard changed the default scope from the global to the
> > undefined (it
> > > is true for other array methods). We need to update our
> > > implementation too.
> > >
> > > There is a reason why forEach doesn't support early
> termination:
> > > performance. If it checks for an exit value, it will
> penalize
> > code that
> > > doesn't need to terminate early. But I understand the
> need for a
> > > versatile iterating function, and I think it would be a good
> > candidate
> > > for dojox.lang.functional methods. we can call it each()
> just
> > like other
> > > libraries do.
> > >
> > > Thanks,
> > >
> > > Eugene Lazutkin
> > > Dojo Toolkit, Committer
> > > http://lazutkin.com/
> > >
> > > On 09/04/2009 09:12 AM, Les Szklanny wrote:
> > > > IMHO you rely too much on the Mozilla spec, which is not
> > that great.
> > > >
> > > > Take for example the ability to early terminate the
> forEach
> > loop.
> > > This
> > > > feature is not in the Mozilla spec, but all other
> libraries
> > (Ext,
> > > > JQuery, Prototype, MooTools) provide the ability to early
> > > terminate forEach.
> > > >
> > > > In Dojo, however, I have to use dojo.some or dojo.every to
> > simulate
> > > > breaking out of forEach, which is not intuitive.
> > > >
> > > > Les
> > > >
> > > > On Fri, Sep 4, 2009 at 8:44 AM, Peter E Higgins
> > > <dante@... <mailto:dante@...>
> <mailto:dante@... <mailto:dante@...>>
> > <mailto:dante@... <mailto:dante@...>
> <mailto:dante@... <mailto:dante@...>>>
> > > > <mailto:dante@...
> <mailto:dante@...> <mailto:dante@...
> <mailto:dante@...>>
> > <mailto:dante@... <mailto:dante@...>
> <mailto:dante@... <mailto:dante@...>>>>> wrote:
> > > >
> > > > From:
> > > >
> > >
> >
> https://developer.mozilla.org/En/Core_JavaScript_1.5_Reference:Objects:Array:forEach
> > > >
> > > > ""
> > > > If a thisObject parameter is provided to forEach,
> it will be
> > > used as the
> > > > this for each invocation of the callback. If it is not
> > > provided, or is
> > > > null, the global object associated with callback
> is used
> > instead.
> > > > ""
> > > >
> > > > You can access the original array as "arr" in your
> example,
> > > just as you
> > > > would propose be 'this'. In jQuery, 'this' is the
> item,
> > which
> > > imo is
> > > > completely wrong. Easier to stick to generally
> accepted
> > specs.
> > > >
> > > > Maybe I'm just missing the advantage to referring
> to your
> > > original array
> > > > as 'this' over the third arg?
> > > >
> > > > You code could also be written as:
> > > > var array = [1,2,3];
> > > > dojo.forEach(array, function(el, i){ }, array); // if
> > you must.
> > > >
> > > > Regards,
> > > > Peter
> > > > > Why dojo.forEach default scope is dojo.global
> and not e.g.
> > > the array
> > > > > or the element being processed?
> > > > >
> > > > > var array = [1,2,3];
> > > > > var scope = array;
> > > > > dojo.forEach(array, function(element, index, arr) {
> > > > > debugger;
> > > > > }, scope);
> > > > >
> > > > > If I don't provide scope, 'this' object defaults to
> > > dojo.global, which
> > > > > is not as useful.
> > > > >
> > > >
> > > > --
> > > > Peter E Higgins
> > > > Dojo Project Lead : http://dojotoolkit.org
> > > >
> > > > _______________________________________________
> > > > FAQ: http://dojotoolkit.org/support/faq
> > > > Book: http://docs.dojocampus.org
> > > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>>
> > > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>>>
> > > >
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> > > >
> > > >
> > > >
> > > >
> > >
> >
> ------------------------------------------------------------------------
> > > >
> > > > _______________________________________________
> > > > FAQ: http://dojotoolkit.org/support/faq
> > > > Book: http://docs.dojocampus.org
> > > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>>
> > > > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >
> > > _______________________________________________
> > > FAQ: http://dojotoolkit.org/support/faq
> > > Book: http://docs.dojocampus.org
> > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>>
> > > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >
> > >
> > >
> > >
> >
> ------------------------------------------------------------------------
> > >
> > > _______________________________________________
> > > FAQ: http://dojotoolkit.org/support/faq
> > > Book: http://docs.dojocampus.org
> > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> >
> >
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> <mailto:Dojo-interest@...>
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: dojo.forEach scope

2009-09-04T19:40:20Z

I didn't expect that you want me to test the obvious, but sure, I can
spare 2 minutes:

var a = [];
for(var i = 0; i < 1000000; ++i) a.push(i);
var dojoTime = new Date().getTime();
dojo.forEach(a, function(){ return true; });
dojoTime = new Date().getTime() - dojoTime;
var jqTime = new Date().getTime();
jQuery.each(a, function(){ return true; });
jqTime = new Date().getTime() - jqTime;
console.log("jQuery is ",
(jqTime / dojoTime).toFixed(2),
" slower than Dojo on iterating through arrays");

On my computer (FF/Linux 3.5) it prints:

jQuery is 2.93 slower than Dojo on iterating through arrays

I ran it 10 times and the number changes from time to time but the
result is > 2.7 on all attempts.

Google Chrome for Linux:

jQuery is 3.09 slower than Dojo on iterating through arrays

Opera 10.0 for Linux:

jQuery is 1.49 slower than Dojo on iterating through arrays

I think if we reverted to the native implementation, numbers would be
even more impressive. And dojox.lang.functional.array methods being more
lax are generally faster than the implementation in the Dojo base.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/04/2009 09:10 PM, Les Szklanny wrote:

> Can you come up with an example that would back up your statement that
> there's any noticeable performance gain by not checking the return value
> of the function?
>
> On Fri, Sep 4, 2009 at 8:48 PM, Eugene Lazutkin
> <eugene.lazutkin@... <mailto:eugene.lazutkin@...>> wrote:
>
> 12ms?
> http://lazutkin.com/blog/2008/mar/23/javascript-edp-and-0ms-timeouts/
> --- to save on reading scroll down to the table.
>
> And allow me to disagree politely with your performance vs. convenience
> assessment.
>
> Thanks,
>
> Eugene Lazutkin
> Dojo Toolkit, Committer
> http://lazutkin.com/
>
> On 09/04/2009 08:04 PM, Les Szklanny wrote:
> >>>> There is a reason why forEach doesn't support early termination:
> > performance. If it checks for an exit value, it will penalize code
> that
> > doesn't need to terminate early.
> >
> > Performance degradation is negligible - forEach is used for
> > convenience, not for performance anyway.
> >
> > Run the code below at http://docs.dojocampus.org/dojo/NodeList
> >
> > var els = dojo.query('*');
> > console.log(els.length)
> > console.time('dojo');
> > dojo.forEach(els, function(e){return true;});
> > console.timeEnd('dojo');
> > console.time('jQuery');
> > jQuery.each(els, function(e){return true;});
> > console.timeEnd('jQuery');
> >
> > http://docs.dojocampus.org/dojo/NodeList
> >
> > I see these results on my computer:
> >
> > elements: 1616
> > dojo: 12ms
> > jQuery: 12ms
> >
> > You will need to jQuerify the page:
> >
> > http://www.learningjquery.com/2006/12/jquerify-bookmarklet
> >
> >
> >
> > On Fri, Sep 4, 2009 at 6:43 PM, Eugene Lazutkin
> > <eugene.lazutkin@... <mailto:eugene.lazutkin@...>
> <mailto:eugene.lazutkin@...
> <mailto:eugene.lazutkin@...>>> wrote:
> >
> > Array functions are patterned after the array methods in the
> upcoming
> > ECMA-262 5th edition. The idea was to provide shims for
> methods of the
> > same name, or delegate to them if they are available. We tried
> to match
> > the behavior down to the argument list, and how the callback
> is called.
> >
> > (Sidenote: we need to provide the fallback to native methods.
> Right now
> > we don't do that because of the performance problems with Firefox'
> > implementation.)
> >
> > We used the Mozilla spec at that moment because that was the only
> > codified thing available. Now, when the release candidate is
> published,
> > we can revise our implementation to match it more closely.
> Namely the
> > standard changed the default scope from the global to the
> undefined (it
> > is true for other array methods). We need to update our
> > implementation too.
> >
> > There is a reason why forEach doesn't support early termination:
> > performance. If it checks for an exit value, it will penalize
> code that
> > doesn't need to terminate early. But I understand the need for a
> > versatile iterating function, and I think it would be a good
> candidate
> > for dojox.lang.functional methods. we can call it each() just
> like other
> > libraries do.
> >
> > Thanks,
> >
> > Eugene Lazutkin
> > Dojo Toolkit, Committer
> > http://lazutkin.com/
> >
> > On 09/04/2009 09:12 AM, Les Szklanny wrote:
> > > IMHO you rely too much on the Mozilla spec, which is not
> that great.
> > >
> > > Take for example the ability to early terminate the forEach
> loop.
> > This
> > > feature is not in the Mozilla spec, but all other libraries
> (Ext,
> > > JQuery, Prototype, MooTools) provide the ability to early
> > terminate forEach.
> > >
> > > In Dojo, however, I have to use dojo.some or dojo.every to
> simulate
> > > breaking out of forEach, which is not intuitive.
> > >
> > > Les
> > >
> > > On Fri, Sep 4, 2009 at 8:44 AM, Peter E Higgins
> > <dante@... <mailto:dante@...>
> <mailto:dante@... <mailto:dante@...>>
> > > <mailto:dante@... <mailto:dante@...>
> <mailto:dante@... <mailto:dante@...>>>> wrote:
> > >
> > > From:
> > >
> >
> https://developer.mozilla.org/En/Core_JavaScript_1.5_Reference:Objects:Array:forEach
> > >
> > > ""
> > > If a thisObject parameter is provided to forEach, it will be
> > used as the
> > > this for each invocation of the callback. If it is not
> > provided, or is
> > > null, the global object associated with callback is used
> instead.
> > > ""
> > >
> > > You can access the original array as "arr" in your example,
> > just as you
> > > would propose be 'this'. In jQuery, 'this' is the item,
> which
> > imo is
> > > completely wrong. Easier to stick to generally accepted
> specs.
> > >
> > > Maybe I'm just missing the advantage to referring to your
> > original array
> > > as 'this' over the third arg?
> > >
> > > You code could also be written as:
> > > var array = [1,2,3];
> > > dojo.forEach(array, function(el, i){ }, array); // if
> you must.
> > >
> > > Regards,
> > > Peter
> > > > Why dojo.forEach default scope is dojo.global and not e.g.
> > the array
> > > > or the element being processed?
> > > >
> > > > var array = [1,2,3];
> > > > var scope = array;
> > > > dojo.forEach(array, function(element, index, arr) {
> > > > debugger;
> > > > }, scope);
> > > >
> > > > If I don't provide scope, 'this' object defaults to
> > dojo.global, which
> > > > is not as useful.
> > > >
> > >
> > > --
> > > Peter E Higgins
> > > Dojo Project Lead : http://dojotoolkit.org
> > >
> > > _______________________________________________
> > > FAQ: http://dojotoolkit.org/support/faq
> > > Book: http://docs.dojocampus.org
> > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>>
> > > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >
> > >
> > >
> > >
> >
> ------------------------------------------------------------------------
> > >
> > > _______________________________________________
> > > FAQ: http://dojotoolkit.org/support/faq
> > > Book: http://docs.dojocampus.org
> > > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> >
> >
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> <mailto:Dojo-interest@...>
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: dojo.forEach scope

2009-09-04T18:48:47Z

12ms?
http://lazutkin.com/blog/2008/mar/23/javascript-edp-and-0ms-timeouts/
--- to save on reading scroll down to the table.

And allow me to disagree politely with your performance vs. convenience
assessment.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/04/2009 08:04 PM, Les Szklanny wrote:

>>>> There is a reason why forEach doesn't support early termination:
> performance. If it checks for an exit value, it will penalize code that
> doesn't need to terminate early.
>
> Performance degradation is negligible - forEach is used for
> convenience, not for performance anyway.
>
> Run the code below at http://docs.dojocampus.org/dojo/NodeList
>
> var els = dojo.query('*');
> console.log(els.length)
> console.time('dojo');
> dojo.forEach(els, function(e){return true;});
> console.timeEnd('dojo');
> console.time('jQuery');
> jQuery.each(els, function(e){return true;});
> console.timeEnd('jQuery');
>
> http://docs.dojocampus.org/dojo/NodeList
>
> I see these results on my computer:
>
> elements: 1616
> dojo: 12ms
> jQuery: 12ms
>
> You will need to jQuerify the page:
>
> http://www.learningjquery.com/2006/12/jquerify-bookmarklet
>
>
>
> On Fri, Sep 4, 2009 at 6:43 PM, Eugene Lazutkin
> <eugene.lazutkin@... <mailto:eugene.lazutkin@...>> wrote:
>
> Array functions are patterned after the array methods in the upcoming
> ECMA-262 5th edition. The idea was to provide shims for methods of the
> same name, or delegate to them if they are available. We tried to match
> the behavior down to the argument list, and how the callback is called.
>
> (Sidenote: we need to provide the fallback to native methods. Right now
> we don't do that because of the performance problems with Firefox'
> implementation.)
>
> We used the Mozilla spec at that moment because that was the only
> codified thing available. Now, when the release candidate is published,
> we can revise our implementation to match it more closely. Namely the
> standard changed the default scope from the global to the undefined (it
> is true for other array methods). We need to update our
> implementation too.
>
> There is a reason why forEach doesn't support early termination:
> performance. If it checks for an exit value, it will penalize code that
> doesn't need to terminate early. But I understand the need for a
> versatile iterating function, and I think it would be a good candidate
> for dojox.lang.functional methods. we can call it each() just like other
> libraries do.
>
> Thanks,
>
> Eugene Lazutkin
> Dojo Toolkit, Committer
> http://lazutkin.com/
>
> On 09/04/2009 09:12 AM, Les Szklanny wrote:
> > IMHO you rely too much on the Mozilla spec, which is not that great.
> >
> > Take for example the ability to early terminate the forEach loop.
> This
> > feature is not in the Mozilla spec, but all other libraries (Ext,
> > JQuery, Prototype, MooTools) provide the ability to early
> terminate forEach.
> >
> > In Dojo, however, I have to use dojo.some or dojo.every to simulate
> > breaking out of forEach, which is not intuitive.
> >
> > Les
> >
> > On Fri, Sep 4, 2009 at 8:44 AM, Peter E Higgins
> <dante@... <mailto:dante@...>
> > <mailto:dante@... <mailto:dante@...>>> wrote:
> >
> > From:
> >
> https://developer.mozilla.org/En/Core_JavaScript_1.5_Reference:Objects:Array:forEach
> >
> > ""
> > If a thisObject parameter is provided to forEach, it will be
> used as the
> > this for each invocation of the callback. If it is not
> provided, or is
> > null, the global object associated with callback is used instead.
> > ""
> >
> > You can access the original array as "arr" in your example,
> just as you
> > would propose be 'this'. In jQuery, 'this' is the item, which
> imo is
> > completely wrong. Easier to stick to generally accepted specs.
> >
> > Maybe I'm just missing the advantage to referring to your
> original array
> > as 'this' over the third arg?
> >
> > You code could also be written as:
> > var array = [1,2,3];
> > dojo.forEach(array, function(el, i){ }, array); // if you must.
> >
> > Regards,
> > Peter
> > > Why dojo.forEach default scope is dojo.global and not e.g.
> the array
> > > or the element being processed?
> > >
> > > var array = [1,2,3];
> > > var scope = array;
> > > dojo.forEach(array, function(element, index, arr) {
> > > debugger;
> > > }, scope);
> > >
> > > If I don't provide scope, 'this' object defaults to
> dojo.global, which
> > > is not as useful.
> > >
> >
> > --
> > Peter E Higgins
> > Dojo Project Lead : http://dojotoolkit.org
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > <mailto:Dojo-interest@...
> <mailto:Dojo-interest@...>>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> >
> >
> >
> ------------------------------------------------------------------------
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://docs.dojocampus.org
> > Dojo-interest@...
> <mailto:Dojo-interest@...>
> > http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> <mailto:Dojo-interest@...>
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: Mailing list Sucks or Thanks for all the fish revisited

2009-09-04T17:00:22Z

Inline

On 09/04/2009 11:36 AM, Chris Hardin wrote:
>
> Someone who is in charge of Dojo needs to deprecate this mailing list
> and go to a forum like now. I never had such frustrations when I worked
> on a project using Ext. I love Dojo...want to keep using it, a mailing
> list isn't appropriate for a project of this magnitude anyway. Whoever
> decided to go to a mailing list instead of a forum, needs to have their
> head examined.

I understood you r negative feelings towards mailing lists. But you have
no arguments against it. What is wrong with the mailing list and how it
is different from a forum in your opinion?

Thanks,

Eugene

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: Mailing list Sucks

2009-09-04T16:56:37Z

Do you really think that by changing medium you'll get more people to
participate? If the problem with the medium --- what is it exactly in
your opinion?

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/04/2009 06:43 AM, Chris Hardin wrote:

> Guys,
>
> I don't want to offend anyone, but this mailing list sucks. many, many
> questons from people and few few answers. Without a bettter medium for
> folks to post their problems like a forum, Dojo will likely be overtaken
> by another framework with better support. I want to keep using Dojo, but
> outdated documentation, hard to find answers and examples that don't
> work because they represent earlier versions of the codebase are really
> getting to me. No doubt this email will generate responses from everyone
> just like the guy who say "thanks for all the fish" a few weeks ago, but
> it's a shame that the true technical questions rarely get responses.
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: dojo.forEach scope

2009-09-04T16:43:38Z

Array functions are patterned after the array methods in the upcoming
ECMA-262 5th edition. The idea was to provide shims for methods of the
same name, or delegate to them if they are available. We tried to match
the behavior down to the argument list, and how the callback is called.

(Sidenote: we need to provide the fallback to native methods. Right now
we don't do that because of the performance problems with Firefox'
implementation.)

We used the Mozilla spec at that moment because that was the only
codified thing available. Now, when the release candidate is published,
we can revise our implementation to match it more closely. Namely the
standard changed the default scope from the global to the undefined (it
is true for other array methods). We need to update our implementation too.

There is a reason why forEach doesn't support early termination:
performance. If it checks for an exit value, it will penalize code that
doesn't need to terminate early. But I understand the need for a
versatile iterating function, and I think it would be a good candidate
for dojox.lang.functional methods. we can call it each() just like other
libraries do.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/04/2009 09:12 AM, Les Szklanny wrote:

> IMHO you rely too much on the Mozilla spec, which is not that great.
>
> Take for example the ability to early terminate the forEach loop. This
> feature is not in the Mozilla spec, but all other libraries (Ext,
> JQuery, Prototype, MooTools) provide the ability to early terminate forEach.
>
> In Dojo, however, I have to use dojo.some or dojo.every to simulate
> breaking out of forEach, which is not intuitive.
>
> Les
>
> On Fri, Sep 4, 2009 at 8:44 AM, Peter E Higgins <dante@...
> <mailto:dante@...>> wrote:
>
> From:
> https://developer.mozilla.org/En/Core_JavaScript_1.5_Reference:Objects:Array:forEach
>
> ""
> If a thisObject parameter is provided to forEach, it will be used as the
> this for each invocation of the callback. If it is not provided, or is
> null, the global object associated with callback is used instead.
> ""
>
> You can access the original array as "arr" in your example, just as you
> would propose be 'this'. In jQuery, 'this' is the item, which imo is
> completely wrong. Easier to stick to generally accepted specs.
>
> Maybe I'm just missing the advantage to referring to your original array
> as 'this' over the third arg?
>
> You code could also be written as:
> var array = [1,2,3];
> dojo.forEach(array, function(el, i){ }, array); // if you must.
>
> Regards,
> Peter
> > Why dojo.forEach default scope is dojo.global and not e.g. the array
> > or the element being processed?
> >
> > var array = [1,2,3];
> > var scope = array;
> > dojo.forEach(array, function(element, index, arr) {
> > debugger;
> > }, scope);
> >
> > If I don't provide scope, 'this' object defaults to dojo.global, which
> > is not as useful.
> >
>
> --
> Peter E Higgins
> Dojo Project Lead : http://dojotoolkit.org
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> <mailto:Dojo-interest@...>
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Posting multidimensional JSON

2009-09-04T13:27:42Z

I am using this function to post data to the server:

---------------------------------------
function ajaxCall(jsnParams) {
var jsnResult,
strUrl = 'ajax.php',
strHandle = 'json';

jsnResult = dojo.rawXhrPost({
url: strUrl,
sync: true,
handleAs: strHandle,
content: jsnParams,
load: function(data,ioArgs) {
return data;
},
error: function(data,ioArgs) { return data; }
});
return eval(jsnResult.results[0]);
}
---------------------------------------

When using the function, this works:

----------------------
var jsnResult = ajaxCall({"var1":"val1","var2":"val2"});
----------------------

At the server side, I get $_POST['var1'] and $_POST['var2'] as expected.

But now I need to send a multidimensional JSON:

----------------------
var jsnResult =
ajaxCall({"var1":"val1","var2":"val2","var3":{"var3a":"val3a"}});
----------------------

I would expect to get $_POST['var3']['var3a'] set to the 'val3a' value
but instead I just get "[object Object]". This is the output of the PHP
var_dump() function:

-----------------------
array (3) {
["var1"]=>
string(4) "val1"
["var2"]=>
string(4) "val2"
["var3"]=>
string(15) "[object Object]"
}
-----------------------

Can you tell me why that is?

/Jette

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

problem with dojo xmpp

2009-09-03T01:46:22Z

Hi

I have implemented a web-based client IM with dojo.xmpp, but i have
one great problem!
I'm using OpenFire for jabber server on subdomain e.g. http://jabber.example.com
My application is on the top level domain e.g. http://example.com
When i try to access on HTTP-BIND directory of jabber server, for
connect the web-based client, the xmpp session return "uncaught
exception: Access to restricted URI denied (NS_ERROR_DOM_BAD_URI)"
because, i suppose, i cant access on a page that reside on another
server.
In apache there are a mod_proxy and mod_rewrite module for bypass
this, but i have IIS 5 and I don't have find any suitable solution to
resolve this problem.
I tried to build a x-domain release of dojo, put this on subdomain and
i called all scripts from this, but i have the same problem

There is any solution?

Tnx in advance
Regards Daniele
--
Leorsini Daniele
http://www.linkedin.com/in/leorsini
_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

How to get the parent of a tree node?

2009-09-02T14:53:12Z

Is there a way to get the parent of an item in a tree? If it matters I am using a XmlStore as the data source for the Tree with a TreeStoreModel in between.

Some insight into what I am wanting to do…

My XML file has a series of children elements that each contain a heading number (think of it like an outline). When a tree node is clicked on, I generate some information and display it on a different part of the page. Part of that information is a combination of the headings from every parent of the node up the tree. For example…

Node1 Heading = 1

-> Node2 Heading = 2

-> -> Node3 Heading = 3

When Node3 is clicked on, I want to generate the string 1.2.3 to display on the page. But having the Node3 item I can only get Node3’s Heading number.

Thanks.

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Using XmlStore with Tree

2009-09-02T14:18:46Z

I haven’t seen any examples online of using XMLStore as the data backend for a Tree widget. With the help of dmachi I got it working this afternoon. He may get an example put together for dojocampus in the near future but until then I wanted to get something online for other people to reference if they are trying to do the same thing.

Things to note…

1. When you define the XmlStore, you set the label to whatever XML element that contains the text that you want to appear in the tree. In this example the label is set to “description”. So any text in the <description> element will appear in the tree. It can be retrieved with XmlStore.getLabel().

2. All other XML elements are retrieved using XmlStore.getValue().

XML FILE

<?xml version="1.0" encoding="utf-8"?>

<root>

<node1>

<description>Node 1 Description</description>

<node2>

<description>Node 2 Description</description>

<node3>

<description>Node 3 Description</description>

<node4>

<description>Node 4 Description</description>

</node4>

<node4>

<description>Node 4 Description</description>

</node4>

</node3>

</node2>

</node1>

</root>

JAVASCRIPT

var store = new dojox.data.XmlStore({url: "xml_file.xml", label: "description"});

var model = new dijit.tree.TreeStoreModel(

{

store: store,

rootId: "root",

rootLabel: "root",

childrenAttrs: ["node2", "node3", "node4"]

});

var tree = new dijit.Tree({model: model});

dojo.connect(tree, "onClick", function(item)

{

console.log(store.getValue(item, “heading”) + “ “ + store.getLabel(item));

});

dijit.byId("TreePane").attr("content", tree.domNode); // FYI: TreePane is the id for a ContentPane.

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: pie charts cropped at the edges

2009-09-02T10:16:29Z

Nick,

It is clearly a bug, please open a ticket for that with a small test. If
you can see this defect on existing tests, just include a reference to them.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 09/02/2009 10:17 AM, Nick Mitchell wrote:

> hi everyone!
>
> if i make a pie chart with an unspecified radius, so that it scales,
> the left and right ends will be cropped by a few pixels. inspecting
> the DOM, i see that dojo places an 11-pixel wide white box is placed
> on either side of the pie chart. i imagine that this is space for the
> axes of a chart, which doesn't seem to make a lot of sense for a pie
> chart.
>
> i have worked around the problem by specifying an alpha-0 background
> for the chart, but it seems like a strange default. am i missing
> something?
>
> thanks!
> nick
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: How do I hide this content pane?

2009-09-01T16:46:53Z

dojo.style(dijit.byId("applicationHeader").domNode, "display", "none");
dojo.style(dojo.query("#applicationHeader"), "display", "hidden");

none of those seems to work for the contentpane inside a tabcontainer.

here's my tab structrue:
<div dojoType="dijit.layout.TabContainer" id="tab_container" doLayout="false">
<div dojoType="dijit.layout.ContentPane" id="subtab0" jsId="subtab0"
title="Warnings" layoutAlign="client" style="position:relative;">
........
</div>

<div dojoType="dijit.layout.ContentPane" id="subtab1"
jsId="subtab1" title="Step 1: Make selections below" >
......
</div>

<div dojoType="dijit.layout.ContentPane" id="subtab2"
jsId="subtab2" href="systemtiming_inputs.jsp" title="Step 2: Input
values below" refreshOnShow="on">
</div>

<div dojoType="dijit.layout.ContentPane" id="subtab3" jsId="subtab3"
href="systemtiming_results.jsp" title="Step 3: Results"
refreshOnShow="true" >
</div>
</div>

I'd like to make 'subtab0' go away if there is no warning. I used:

dojo.style(dijit.byId("subtab0").domNode, "display", "none");

Is there anything wrong about the syntax....?

Thanks,
Sharon

On Fri, Aug 22, 2008 at 10:14 AM, <dave.alvarado@...> wrote:

>
> Yes, everything's good here. I noticed that whenever I had "layoutAlign='client'" within the TabContainer DIV, the element got this "position:absolute" attribute assigned to it. So what I did was add a 'style="position:relative;"' to the DIV and all was good.
>
> Granted maybe there's more I'm not telling you. I inherited code that wasn't my own and so I'm trying to figure out what's going on. We're using Dojo 0.9 if that's of any help.
>
> Anyway, thanks again, - Dave
>
>
> > -------Original Message-------
> > From: Peter E Higgins <dante@...>
> > Subject: Re: [Dojo-interest] How do I hide this content pane?
> > Sent: Aug 22 '08 16:54
> >
> > Hmm, that may be a Bug -- unless we're still supporting layoutAlign for
> > the deprecated LayoutContainer, which would in fact make the node
> > position:absolute to that it can be sized and managed by the Container.
> > Doesn't seem like you have one of those, so I'm going to guess it's safe
> > to remove ... are you all worked out then?
> >
> > Regards,
> > Peter Higgins
> >
> >
> > dave.alvarado@... wrote:
> > > Ah, you are wise.  When I inspected the TabContainer element it had position=absolute, inserted there by some magic having to do with "layoutAlign".
> > >
> > >  - Dave
> > >
> > >
> > >>  -------Original Message-------
> > >>  From: Peter E Higgins <dante@...>
> > >>  Subject: Re: [Dojo-interest] How do I hide this content pane?
> > >>  Sent: Aug 22 '08 16:20
> > >>
> > >>  How is the tabContainer positioned? If it is position:absolute in the
> > >>  layout, it won't move when the surrounding div's change size...
> > >>  typically, setting display:none on a node removes it from the layout,
> > >>  and it takes no space, so something else seems to be going on here.
> > >>  dijit.byId("applicationHeader").domNode is the node that should be
> > >>  taking the space ... is this preview-able anywhere that some remote
> > >>  firebug style inspection might shed some light on?
> > >>
> > >>  fwiw, because you have jsId="applicationHeader", the node is available
> > >>  globally as applicationHeader.domNode
> > >>
> > >>  Regards,
> > >>  Peter Higgins
> > >>
> > >>
> > >>  dave.alvarado@... wrote:
> > >>  > Thanks.  Your solution succeeds in making the DIV invisible, but it doesn't succeed in moving the DIVs beneath it up to fill the blank space:
> > >>  >
> > >>  > http://screencast.com/t/hFogoQATkTa
> > >>  >
> > >>  > This was the question I posed earlier, how to make everything  move up?
> > >>  >
> > >>  > Thanks for your help, - Dave
> > >>  >
> > >>  >
> > >>  >
> > >>  >
> > >>  >>  -------Original Message-------
> > >>  >>  From: Jared Jurkiewicz <jared.jurkiewicz@...>
> > >>  >>  Subject: Re: [Dojo-interest] How do I hide this content pane?
> > >>  >>  Sent: Aug 22 '08 16:00
> > >>  >>
> > >>  >>  Or simpler:
> > >>  >>
> > >>  >>  dojo.style(dijit.byId("applicationHeader").domNode, "display", "none");
> > >>  >>
> > >>  >>  -- Jared
> > >>  >>
> > >>  >>
> > >>  >>
> > >>  >>  On Fri, Aug 22, 2008 at 11:58 AM, Jared Jurkiewicz
> > >>  >>  <jared.jurkiewicz@...> wrote:
> > >>  >>  > Style the domNode of the contentPane to hidden.
> > >>  >>  >
> > >>  >>  > dojo.style(dojo.query("#applicationHeader"), "display", "hidden");   //Or none.
> > >>  >>  >
> > >>  >>  > -- Jared
> > >>  >>  >
> > >>  >>  >
> > >>  >>  > On Fri, Aug 22, 2008 at 11:27 AM,  <dave.alvarado@...> wrote:
> > >>  >>  >> Hi,
> > >>  >>  >>
> > >>  >>  >> In certain situations I want to hide the DIV with id="applicationHeader", at the top of my page ...
> > >>  >>  >>
> > >>  >>  >> <body>
> > >>  >>  >>                <div dojoType="dijit.layout.ContentPane" layoutAlign="top" id="applicationHeader" jsId="applicationHeader">
> > >>  >>  >>                        <div> ...</div>
> > >>  >>  >>                        <div class="header">
> > >>  >>  >>                                <div class="headerRt" id="divHeader">
> > >>  >>  >>                                        NPS Application Configuration
> > >>  >>  >>                                </div>
> > >>  >>  >>                        </div>
> > >>  >>  >>                </div>
> > >>  >>  >>                <div dojoType="dijit.layout.TabContainer" id="tabs" jsId="tabs"  layoutAlign="client">
> > >>  >>  >>                ...
> > >>  >>  >>                </div>
> > >>  >>  >>
> > >>  >>  >> Without it being hidden, it looks like this:
> > >>  >>  >>
> > >>  >>  >> http://screencast.com/t/OrEkqRdLe
> > >>  >>  >>
> > >>  >>  >> but when I do manage to hide the DIV using this code:
> > >>  >>  >>
> > >>  >>  >>                dojo.query("#applicationHeader").removeClass("dijitContentPane");
> > >>  >>  >>                dojo.query("#applicationHeader").addClass("hidden");       // Simply contains display:none
> > >>  >>  >>
> > >>  >>  >> the DIV is hidden but nothing moves up ...
> > >>  >>  >>
> > >>  >>  >> http://screencast.com/t/27jU5jq2ln
> > >>  >>  >>
> > >>  >>  >> What do I need to do to hide the DIV and also have everything move up?
> > >>  >>  >>
> > >>  >>  >> Thanks, - Dave
> > >>  >>  >> _______________________________________________
> > >>  >>  >> FAQ: http://dojotoolkit.org/support/faq
> > >>  >>  >> Book: http://dojotoolkit.org/docs/book
> > >>  >>  >> Forums: http://dojotoolkit.org/forum
> > >>  >>  >> Dojo-interest@...
> > >>  >>  >> http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >>  >>  >>
> > >>  >>  >
> > >>  >>  _______________________________________________
> > >>  >>  FAQ: http://dojotoolkit.org/support/faq
> > >>  >>  Book: http://dojotoolkit.org/docs/book
> > >>  >>  Forums: http://dojotoolkit.org/forum
> > >>  >>  Dojo-interest@...
> > >>  >>  http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >>  >>
> > >>  >>
> > >>  > _______________________________________________
> > >>  > FAQ: http://dojotoolkit.org/support/faq
> > >>  > Book: http://dojotoolkit.org/docs/book
> > >>  > Forums: http://dojotoolkit.org/forum
> > >>  > Dojo-interest@...
> > >>  > http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >>  >
> > >>  >
> > >>
> > >>  _______________________________________________
> > >>  FAQ: http://dojotoolkit.org/support/faq
> > >>  Book: http://dojotoolkit.org/docs/book
> > >>  Forums: http://dojotoolkit.org/forum
> > >>  Dojo-interest@...
> > >>  http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >>
> > >>
> > > _______________________________________________
> > > FAQ: http://dojotoolkit.org/support/faq
> > > Book: http://dojotoolkit.org/docs/book
> > > Forums: http://dojotoolkit.org/forum
> > > Dojo-interest@...
> > > http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> > >
> > >
> >
> > _______________________________________________
> > FAQ: http://dojotoolkit.org/support/faq
> > Book: http://dojotoolkit.org/docs/book
> > Forums: http://dojotoolkit.org/forum
> > Dojo-interest@...
> > http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest
> >
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://dojotoolkit.org/docs/book
> Forums: http://dojotoolkit.org/forum
> Dojo-interest@...
> http://turtle.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Grid html content does not get rendered in tests under IE7

2009-08-29T11:20:15Z

In dojo 1.3.2 some of the supplied grid tests like

dojox/grid/tests/test_change_structure.html
dojox/grid/tests/test_events.html
dojox/grid/tests/test_subgrid.html
dojox/grid/tests/test_yahoo_images.html

only show the raw HTML code in the cell,
when for example an image should show up via <img> tag.

I am testing with IE7.

This used to work.
Could it be that the latest dojo release introduced a bug?

BTW does anybody know why the tests are not working under Firefox 3.x ?

Thx
Horst

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: IO Pipeline Topics

2009-08-28T00:56:02Z

What I meant is to use "/dojo/io/error". It can be used as a default
fallback for errors. Obviously for it to work properly an application
should not set error handlers or ignore globally processed errors.
Something like that:

var d = dojo;

d.subscribe("/dojo/io/error", function(err, ioArgs){
switch(ioArgs.xhr.status){
case 403:
alert("Your session expired!");
break;
case 500:
alert("Server is down, please try again in 5 minutes.");
break;
...
}
});

...

// let topics handle all errors
d.xhrGet({
...
// no error() at all
});

...

// let's ignore common errors
d.xhrGet({
...
error: function(err, ioArgs){
if(ioArgs.xhr.status == 403){
// skip it => defer to the global processing
}else{
// other error processing
alert("Some bad thing happened!");
}
}
});

Or something like that. Topics look like the appropriate place for all
default error handling.

Another good use of topics is to show I/O indicator ("Loading...",
"Busy", or an animated GIF) with "/dojo/io/start" and "/dojo/io/stop".

The same topics can be used to warn end users against navigating from a
page when I/O is still active, but if we ignore all read-only I/O and
want to track a particular "write" I/O request, we can use
"/dojo/io/send" and "/dojo/io/done".

With properly designed I/O between clients and a server (e.g., a
"universal" structure of transferred data, so it can be inspected
without going into much specifics) the possibilities are endless.

Thanks,

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 08/27/2009 10:24 PM, James Burke wrote:

> The IO topics are after the fact, they notify that an IO action
> happened, not something that happens before.
>
> We might be able use a default mixin set of properties that are mixed
> in with all requests, but that would be something new.
>
> James
>
> On Thu, Aug 27, 2009 at 3:23 PM, Eugene
> Lazutkin<eugene.lazutkin@...> wrote:
>> Can we use existing I/O topics for that?
>>
>> Eugene Lazutkin
>> Dojo Toolkit, Committer
>> http://lazutkin.com/
>>
>> On 08/26/2009 01:22 PM, James Burke wrote:
>>> You could override dojo.xhr() to add that onError block to each set of
>>> args passed in. Warning, this has not been tested, but hopefully gives
>>> you enough to work out a proper solution:
>>>
>>> var oldXhr = dojo.xhr;
>>> dojo.xhr = function(method, args, hasBody){
>>> //Make sure to not mess with original args in case they are reused
>>> var oldError = args.error;
>>> args = dojo.delegate(args);
>>> args.error = function(err, ioArgs) {
>>> //Call the old error function.
>>> if(oldError){
>>> oldError.apply(args, arguments);
>>> }
>>>
>>> //Do general work here.
>>> if(request.xhr.status == 403){
>>> alert("expired");
>>> }
>>> }
>>> return oldXhr.call(dojo, method, args, hasBody);
>>> }
>>>
>>> James
>>>
>>> On Wed, Aug 26, 2009 at 8:53 AM, Zhaidarbek
>>> Ayazbayev<zhaidarbek@...> wrote:
>>>> Good morning everyone!
>>>> Resending my previous email:
>>>>
>>>> I need to handle session timeouts wiht ajax. If session is invalid my server
>>>> responds with 403 status code. In simplest case it is handled as:
>>>> dojo.xhrGet({
>>>> onError: function(errorData, request){
>>>> if(request.xhr.status == 403){
>>>> alert("Your session expired!");
>>>> }
>>>> }});
>>>>
>>>> But i have different dojo xhr requests (get, post, put, delete) throughout
>>>> the page, also data stores uses ajax to fetch data from server. So i need to
>>>> do it at one place. What is the best/right way to implement it? Seems like
>>>> dojo pipelines is the way to go, but i don't know how to use it, could you
>>>> please provide an example code.
>>>> Thanks in advance.
>>>> _______________________________________________
>>>> FAQ: http://dojotoolkit.org/support/faq
>>>> Book: http://docs.dojocampus.org
>>>> Dojo-interest@...
>>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>>
>>>>
>>> _______________________________________________
>>> FAQ: http://dojotoolkit.org/support/faq
>>> Book: http://docs.dojocampus.org
>>> Dojo-interest@...
>>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>>
>>
>> _______________________________________________
>> FAQ: http://dojotoolkit.org/support/faq
>> Book: http://docs.dojocampus.org
>> Dojo-interest@...
>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

TabContainer with Custom widgets as ContentPlanes Problem

2009-08-27T23:37:33Z

Hi

I looked around and spent already a day on this problem... I need some advice!

I have a custom TabContainer:

dojo.declare("t3dojo_basicapp.DynamicTabContentPlane",
[dijit.layout.TabContainer],

When my script opens a Tab with some template file as url in href inside the tab all the containing widgets are rendered correctly. (I guess it has to do that I select the child tab before the href is loaded and rendered)

now comes the problem: If I create a custom templated widget (derived from ContentPlane) with this code. They are initially not rendered correctly. After resizing the browser the widget is rendered OK.

        tempTab = new zhdojo.dbforms.MainCharForm({title: strTitle, closable: true});
        this.oDynabicTabController.addChild(tempTab);
        this.oDynabicTabController.selectChild(tempTab);

What tried a lot : calling resize() or layout() of the TabController etc. Nothing helped.

My guess is, that since the tempTab is not visible when it is added to the TapController it is not rendered. - But I don't find how to make my custom widget render correctly.

Any help is much apprechiated!

Greetings - Achim

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: IO Pipeline Topics

2009-08-27T15:23:50Z

Can we use existing I/O topics for that?

Eugene Lazutkin
Dojo Toolkit, Committer
http://lazutkin.com/

On 08/26/2009 01:22 PM, James Burke wrote:

> You could override dojo.xhr() to add that onError block to each set of
> args passed in. Warning, this has not been tested, but hopefully gives
> you enough to work out a proper solution:
>
> var oldXhr = dojo.xhr;
> dojo.xhr = function(method, args, hasBody){
> //Make sure to not mess with original args in case they are reused
> var oldError = args.error;
> args = dojo.delegate(args);
> args.error = function(err, ioArgs) {
> //Call the old error function.
> if(oldError){
> oldError.apply(args, arguments);
> }
>
> //Do general work here.
> if(request.xhr.status == 403){
> alert("expired");
> }
> }
> return oldXhr.call(dojo, method, args, hasBody);
> }
>
> James
>
> On Wed, Aug 26, 2009 at 8:53 AM, Zhaidarbek
> Ayazbayev<zhaidarbek@...> wrote:
>> Good morning everyone!
>> Resending my previous email:
>>
>> I need to handle session timeouts wiht ajax. If session is invalid my server
>> responds with 403 status code. In simplest case it is handled as:
>> dojo.xhrGet({
>> onError: function(errorData, request){
>> if(request.xhr.status == 403){
>> alert("Your session expired!");
>> }
>> }});
>>
>> But i have different dojo xhr requests (get, post, put, delete) throughout
>> the page, also data stores uses ajax to fetch data from server. So i need to
>> do it at one place. What is the best/right way to implement it? Seems like
>> dojo pipelines is the way to go, but i don't know how to use it, could you
>> please provide an example code.
>> Thanks in advance.
>> _______________________________________________
>> FAQ: http://dojotoolkit.org/support/faq
>> Book: http://docs.dojocampus.org
>> Dojo-interest@...
>> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>>
>>
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest
>

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Re: Custom build performance

2009-08-27T10:48:43Z

you can include your dojo.js and customer layers as script tags right
before </body> and still use parseOnLoad: true as far as I know. The
dojo.require()s aren't necessary if you have everything in your custom
layer (but if you forgot something you'll end up with an error of
course).

Dustin

On Aug 27, 2009, at 1:44 PM, Jose Noheda wrote:

> Hi,
>
> I was wondering that if I include my custom build js file at the end
> of the body tag and add a dojo.addOnLoad (in HEAD) to parse things
> later then could I remove all the dojo.require calls entirely? I'm
> assuming that the build includes all the needed dependencies here.
> My problem is that I have dozens of repeated dojo.requires (each
> inside an inline script tag) due to the nature of the JSP tag files
> and I'm trying to reduce the perceived speed as much as I can
> (currently not that stellar, you know?)
>
> Regards
> _______________________________________________
> FAQ: http://dojotoolkit.org/support/faq
> Book: http://docs.dojocampus.org
> Dojo-interest@...
> http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Custom build performance

2009-08-27T10:44:43Z

Hi,

I was wondering that if I include my custom build js file at the end of the body tag and add a dojo.addOnLoad (in HEAD) to parse things later then could I remove all the dojo.require calls entirely? I'm assuming that the build includes all the needed dependencies here. My problem is that I have dozens of repeated dojo.requires (each inside an inline script tag) due to the nature of the JSP tag files and I'm trying to reduce the perceived speed as much as I can (currently not that stellar, you know?)

Regards

_______________________________________________
FAQ: http://dojotoolkit.org/support/faq
Book: http://docs.dojocampus.org
Dojo-interest@...
http://mail.dojotoolkit.org/mailman/listinfo/dojo-interest

Handling dojoX cometd unsuccessful subscription

2009-08-27T09:49:55Z

Hi,

I had already posted this question in Stackoverflow, but since I had no response I'm re-posting it here.

I'm using a Bayeux server implementation for Netty and the latest version of dojoX cometd. When I refuse a subscription a /meta/subscription message is sent with successful set to false. Response:

[{"channel":"/meta/subscribe","clientId":"abf5fb2408bf4f8e","successful":false,"subscription":"/badChannel","id":"3","timestamp":"2009-08-26T12:08:40"}]

In response to this, dojoX library sends a disconnect request followed by connect request. Should it this be the behaviour? I believe that it would make sense to use the already existent callback (http://api.dojotoolkit.org/jsdoc/1.3/dojox.cometd.Connection.subscribe, example 4) to notify the clients of the subscription failure.

Is the observed behaviour the correct one? How should I handle the unsuccessful subscription? Does my proposal makes sense?

Thanks,
Luís M. Costa