**Please try this version** Report problems here or on the issues
list.
Our plan is to make this the *last beta* test. b3 will be available
briefly for sanity checks and then become 1.1.0. (Unless its horrible,
then forget you read this).
To ensure that we move toward stability on 1.1 we have decided to drop
some features that could stretch out the 1.1 timeline. As
compensation we will start 1.2 directly and these features will
available in 1.2a.
Features backed out:
---------------------
BreakOnTopLevel -- much better implementation for 1.2 planned.
showEvalSources (will be true always in 1.1) -- part of 1.2
performance enhancement.
Bug Score Card:
-----------------
Fix 283 JS warning flood
Fix 281 error preview in statusbar
Fix 244, 240, 48 self closing tags.
Fix 290 Open Editor local files
Fix 143 console.log formatted flag ** Affects API for logging **
Fix 297 firebug-service dump OS independent
Fix 269 UTF-16 contrib by pointervoid
fix stepping in calling frames
workaround for jsdIScript.functionSource crash on OSX when /packer/ JS
compressor is used
Reported 291 to Mozilla, fix is in 1.8.
Diagnosed, not fixed 68, 123, 125, 157, 213, 289 CSS positioning
Diagnosed, not fixed 186
Diagnosed, not fixed 246
No action (Does not happen for me) 170, 184, 260 Kidnapped Keys
No Action Regexp 192, 242
Small features:
--------------
add "Open With Editor" to browser content area context menu
Aptana Debugger integration targets to build.xml and svn:external link
to its sources
Alphabetize FBTrace Options
Support for FF3a8 forced security on update.
First of all, thank you for the continuing efforts in improving this
great extension.
Secondly, I'm having a REAL issue with the console.info.
It seems to NOT deal with anything other than strings/integers
properly and only deals with the first parameter. This will completely
mess up my debugging/logging.
I have the following template for all my methods/functions ...
function x() {
if (b_Debug) { console.group('x(', arguments, ')'); }
...
if (b_Debug) { console.groupEnd(); }
}
This is great as it will show all the arguments as an object for
clicking on and it all used to work.
With this beta, I only get the 'x(' part.
Even worse, if I use console.info(1,2,3), I only get 1.
This is a significant change.
I'm going back to the official release as for me this is a MAJOR
feature when I'm writing/testing code. This technique allows me to
instantly see the call trace and what calls what and with what
parameters.
Regards,
Richard Quadling.
--
-----
Richard Quadling
Zend Certified Engineer : http://zend.com/zce.php?c=ZEND002498&r=213474731
"Standing on the shoulders of some very clever giants!"
For the code ...
<html>
<head>
<title>Testing arguments</title>
</head>
<body>
Requires FireBug or FireBugLite.
<script>
function fn_None() { console.info('None', arguments); }
function fn_Args(a,b,c,d) { console.info('Args', arguments, 'Args'); }
function fn_Call() { console.group('Call', arguments);
fn_None(arguments); fn_Args(arguments); console.groupEnd();}
fn_None();
fn_Args();
fn_Args(1,2,3);
fn_Args(document);
fn_Call();
fn_Call(1,2,3);
fn_Call(document);
</script>
</body>
</html>
I've attached 3 images. The first is the console from 1.0.5. Then the
console from 1.1b2 and then the DOM page if you select an object in
the console for 1.1b2.
If you select the object (document) from the console in 1.0.5, you get
all the details as expected.
On Sep 27, 8:05 am, "Richard Quadling" <rquadl...@googlemail.com>
wrote:
> On 27/09/2007, Richard Quadling <rquadl...@googlemail.com> wrote:
>
>
>
> > On 27/09/2007, John J Barton <johnjbar...@johnjbarton.com> wrote:
>
> > > Firebug 1.1.0b2 is available onhttp://fireclipse.xucia.com. (It will
> FB1.05.png
> 14KViewDownload
>
> FB1.1b2.gif
> 24KViewDownload
>
> FB1.1b2 DOM.gif
> 19KViewDownload
Back on Firebug 1.0, any object with '.length' was treated as an Array
object. The funky object 'arguments' has '.length' but is not an
Array. So in Firebug 1.0 your code would print:
None [ ]
You can get the Firebug 1.0 behavior by passing an Array:
function copyArgs(argumentsObject){
var args = [];
for (var i = 0; i < argumentsObject.length; i++)
args[i] = argumentsObject[i];
return args;
}
function fn_None() {
console.formatted = true; // emulate old API
console.info('None', copyArgs(arguments));
}
Now the output will be
None []
On Sep 27, 9:26 am, John J Barton <johnjbar...@johnjbarton.com> wrote:
> Thanks. It would be helpful if you can put your test case and
> observations in a bug report on the issues listhttp://code.google.com/p/fbug/issues/list
> > I don't understand your example. 19KViewDownload
Thanks for that. I've added a new method to my library. As it is
dependent upon prototype I'm using ...
args: function(o_Arguments) {
var a_Return = [];
$A(o_Arguments).each(function(o_Arg){
a_Return.push(o_Arg);
});
return a_Return;
},
which is working a treat with ...
if (b_Debug) { console.group('RAQLib.require(', RAQLib.args(arguments), ')'); }
Not a significant change for me, but I think arguments SHOULD be a special case.
An object with a length property suggests that it should be enumerable
and therefore should be able to be examined like an array. But if the
object is a shape, "length" means something different.Hmmm.
Anyway, thanks for the fix. I'll continue using the beta now with this
patch. If you amend the beta to allow for arguments, then I can retest
this without any great difficulty.
Thanks,
I tried reinstall extension but it is still same.
DiGi
On Oct 2, 5:07 pm, "Richard Quadling" <rquadl...@googlemail.com>
wrote:
There was two options... First was "Enable Firebug" and secondary
"Enable Firebug for this site". Now I have only "Enable Firebug".
See original shoots (without first option) http://www.getfirebug.com/using.html
DiGi