Since I updated to the latest version of Safari (3.1) this week I'm
getting RJS Erros when trying to update a record. This does NOT
happen in Firefox 2.0.0.12 (Latest version thats not beta 3). I'm
linking screenshots of the two parts of the error messages:
PS. Pretty sure it due to the latest update in Safari, because
another project already in production using ActiveScaffold which I
havent touched in a while is now showing those errors too.
This is also happening when creating new records. The RJS error
message only pops up when running the application in localhost.
Otherwise in production it only sits there doing nothing after you
click update/create
On Mar 20, 8:39 pm, Jose Fernandez <fernandez.joser...@gmail.com>
wrote:
> Since I updated to the latest version of Safari (3.1) this week I'm
> getting RJS Erros when trying to update a record. This does NOT
> happen in Firefox 2.0.0.12 (Latest version thats not beta 3). I'm
> linking screenshots of the two parts of the error messages:
> PS. Pretty sure it due to the latest update in Safari, because
> another project already in production using ActiveScaffold which I
> havent touched in a while is now showing those errors too.
> This is also happening when creating new records. The RJS error
> message only pops up when running the application in localhost.
> Otherwise in production it only sits there doing nothing after you
> click update/create
> On Mar 20, 8:39 pm, Jose Fernandez <fernandez.joser...@gmail.com>
> wrote:
> > Since I updated to the latest version of Safari (3.1) this week I'm
> > getting RJS Erros when trying to update a record. This does NOT
> > happen in Firefox 2.0.0.12 (Latest version thats not beta 3). I'm
> > linking screenshots of the two parts of the error messages:
> > PS. Pretty sure it due to the latest update in Safari, because
> > another project already in production using ActiveScaffold which I
> > havent touched in a while is now showing those errors too.
> On Mar 21, 9:43 am, Jose Fernandez <fernandez.joser...@gmail.com>
> wrote:
> > This is also happening when creating new records. The RJS error
> > message only pops up when running the application in localhost.
> > Otherwise in production it only sits there doing nothing after you
> > click update/create
> > On Mar 20, 8:39 pm, Jose Fernandez <fernandez.joser...@gmail.com>
> > wrote:
> > > Since I updated to the latest version of Safari (3.1) this week I'm
> > > getting RJS Erros when trying to update a record. This does NOT
> > > happen in Firefox 2.0.0.12 (Latest version thats not beta 3). I'm
> > > linking screenshots of the two parts of the error messages:
> > > PS. Pretty sure it due to the latest update in Safari, because
> > > another project already in production using ActiveScaffold which I
> > > havent touched in a while is now showing those errors too.
> I can confirm this behavior on my Safari 3.1 as well. It's pretty sure
> there is something new in 3.1 that accounts for that, but I am not
> sure yet.
> On Mar 22, 11:43 am, ash <ashc...@gmail.com> wrote:
> > Does happen in Firefox 3 Beta 4.
> > On Mar 21, 9:43 am, Jose Fernandez <fernandez.joser...@gmail.com>
> > wrote:
> > > This is also happening when creating new records. The RJS error
> > > message only pops up when running the application in localhost.
> > > Otherwise in production it only sits there doing nothing after you
> > > click update/create
> > > On Mar 20, 8:39 pm, Jose Fernandez <fernandez.joser...@gmail.com>
> > > wrote:
> > > > Since I updated to the latest version of Safari (3.1) this week I'm
> > > > getting RJS Erros when trying to update a record. This does NOT
> > > > happen in Firefox 2.0.0.12 (Latest version thats not beta 3). I'm
> > > > linking screenshots of the two parts of the error messages:
> > > > PS. Pretty sure it due to the latest update in Safari, because
> > > > another project already in production using ActiveScaffold which I
> > > > havent touched in a while is now showing those errors too.
The guys at caboose (http://blog.caboo.se/articles/2008/3/21/ safari-3-1-breaks-your-app) figured out that both Safari and Firefox
implements a new getElementsByClassName() function that breaks
compatibility. The solution (there are variants) is this:
> I'm having a tough time seeing whats going on with '.link'. Anyone
> else?
> It's a problem in update.rjs too. Are 'cancel' links used anywhere
> else?
> This is somewhat important. Safari 3.1 is out and Firefox 3 is weeks
> away.
Seems not to fix the issue to me... I can't understand why, but
changing the js file at line 266 in public/javascript , made no
difference at all... hmmm! Caching problem, I'll look at it!
> Thanks for this fix. I would note that anybody who cuts and pastes
> this should add a right parenthical after (elem} and before the right
> bracket.
> On Mar 25, 8:53 am, AkitaOnRails <fabioak...@gmail.com> wrote:
> In this case I am replacing with Prototype's select() mechanism. This
> worked for me onSafari3.1.
> On Mar 24, 9:13 pm, David Richardson <david.richard...@channel200.net>
> wrote:
> > I'm having a tough time seeing whats going on with '.link'. Anyone
> > else?
> > It's a problem in update.rjstoo. Are 'cancel' links used anywhere
> > else?
> > This is somewhat important.Safari3.1 is out and Firefox 3 is weeks
> > away.
Why hasn't the trunk been updated yet. It's been a month and I'm
getting irritated fixing this problem for each app. Do we need to
submit a bug report to the project leaders?
On Apr 8, 5:47 pm, broox <derek.br...@gmail.com> wrote:
> > In this case I am replacing with Prototype's select() mechanism. This
> > worked for me onSafari3.1.
> > On Mar 24, 9:13 pm, David Richardson <david.richard...@channel200.net>
> > wrote:
> > > I'm having a tough time seeing whats going on with '.link'. Anyone
> > > else?
> > > It's a problem in update.rjstoo. Are 'cancel' links used anywhere
> > > else?
> > > This is somewhat important.Safari3.1 is out and Firefox 3 is weeks
> > > away.
> Why hasn't the trunk been updated yet. It's been a month and I'm > getting irritated fixing this problem for each app. Do we need to > submit a bug report to the project leaders?
> On Apr 8, 5:47 pm, broox <derek.br...@gmail.com> wrote: >> You rule. Good find.
>> On Mar 25, 8:53 am, AkitaOnRails <fabioak...@gmail.com> wrote:
>>> Ok, I figured out the problem.
>>> On active_scaffold.js (make sure you're looking at the copy on your >>> public/javascript folder), around line #266 there is this code:
>>> The guys at caboose (http://blog.caboo.se/articles/2008/3/21/safari-3-1-breaks-your-app >>> ) figured out that bothSafariand Firefox >>> implements a new getElementsByClassName() function that breaks >>> compatibility. The solution (there are variants) is this:
>>> In this case I am replacing with Prototype's select() mechanism. >>> This >>> worked for me onSafari3.1.
>>> On Mar 24, 9:13 pm, David Richardson >>> <david.richard...@channel200.net> >>> wrote:
>>>> I'm having a tough time seeing whats going on with '.link'. Anyone >>>> else? >>>> It's a problem in update.rjstoo. Are 'cancel' links used anywhere >>>> else?
>>>> This is somewhat important.Safari3.1 is out and Firefox 3 is weeks >>>> away.
> Should be fixed here (i think). Otherwise you can fork the rep and
> make your own fixed copy, thats what im about to do now.
> Jose
> On Apr 18, 2008, at 3:26 PM, MZBPhotoartl...@gmail.com wrote:
> > Why hasn't the trunk been updated yet. It's been a month and I'm
> > getting irritated fixing this problem for each app. Do we need to
> > submit a bug report to the project leaders?
> > On Apr 8, 5:47 pm, broox <derek.br...@gmail.com> wrote:
> >> You rule. Good find.
> >> On Mar 25, 8:53 am, AkitaOnRails <fabioak...@gmail.com> wrote:
> >>> Ok, I figured out the problem.
> >>> On active_scaffold.js (make sure you're looking at the copy on your
> >>> public/javascript folder), around line #266 there is this code:
> >>> The guys at caboose (http://blog.caboo.se/articles/2008/3/21/safari-3-1-breaks-your-app > >>> ) figured out that bothSafariand Firefox
> >>> implements a new getElementsByClassName() function that breaks
> >>> compatibility. The solution (there are variants) is this:
> >>> In this case I am replacing with Prototype's select() mechanism.
> >>> This
> >>> worked for me onSafari3.1.
> >>> On Mar 24, 9:13 pm, David Richardson
> >>> <david.richard...@channel200.net>
> >>> wrote:
> >>>> I'm having a tough time seeing whats going on with '.link'. Anyone
> >>>> else?
> >>>> It's a problem in update.rjstoo. Are 'cancel' links used anywhere
> >>>> else?
> >>>> This is somewhat important.Safari3.1 is out and Firefox 3 is weeks
> >>>> away.
Second to this. It doesnt work on Safari 3.1, even changing row 266 to
select doesnt work. It fixed the problem on failed create (with
validation failure) but fails again on a successful create! Help.....
On Apr 21, 7:23 pm, chetanm <den1...@gmail.com> wrote:
> Second to this. It doesnt work on Safari 3.1, even changing row 266 to
> select doesnt work. It fixed the problem on failed create (with
> validation failure) but fails again on a successful create! Help.....
> On Apr 21, 7:23 pm, chetanm <den1...@gmail.com> wrote:
> > yes !! even I get RJS errors in FF3b5 and Safari3.1
> > On Apr 21, 9:47 am, David Herman <dst...@purpleasphalt.com> wrote:
> I want to know if there is anyone have clues on how to fix this? > Otherwise I am just not comfortable to use active_scaffold for now...
> On May 9, 11:45 pm, goodwill <william.yeung...@gmail.com> wrote: >> Second to this. It doesnt work on Safari 3.1, even changing row 266 >> to >> select doesnt work. It fixed the problem on failed create (with >> validation failure) but fails again on a successful create! Help.....
>> On Apr 21, 7:23 pm, chetanm <den1...@gmail.com> wrote:
>>> yes !! even I get RJS errors in FF3b5 and Safari3.1
>>> On Apr 21, 9:47 am, David Herman <dst...@purpleasphalt.com> wrote:
>>>> I'm also getting this in FF 3b5
>>>> On Apr 20, 1:49 pm, Sonar <serp...@gmail.com> wrote:
>>>>> RJSerror(the same as described above or a new one, reappeared >>>>> inSafari3.1.1)
> The latest version from github fixes this error.
> On May 10, 2008, at 9:40 AM, goodwill <william.yeung...@gmail.com>
> wrote:
> > I want to know if there is anyone have clues on how to fix this?
> > Otherwise I am just not comfortable to use active_scaffold for now...
> > On May 9, 11:45 pm, goodwill <william.yeung...@gmail.com> wrote:
> >> Second to this. It doesnt work on Safari 3.1, even changing row 266
> >> to
> >> select doesnt work. It fixed the problem on failed create (with
> >> validation failure) but fails again on a successful create! Help.....
> >> On Apr 21, 7:23 pm, chetanm <den1...@gmail.com> wrote:
> >>> yes !! even I get RJS errors in FF3b5 and Safari3.1
> >>> On Apr 21, 9:47 am, David Herman <dst...@purpleasphalt.com> wrote:
> >>>> I'm also getting this in FF 3b5
> >>>> On Apr 20, 1:49 pm, Sonar <serp...@gmail.com> wrote:
> >>>>> RJSerror(the same as described above or a new one, reappeared
> >>>>> inSafari3.1.1)
I should say I am completely new on git- how could I install the
plugin from there? if I just download the gz file, is there any method
to install that directly via the download without creating my own svn
store?
On May 11, 11:29 am, goodwill <william.yeung...@gmail.com> wrote:
> On May 11, 3:05 am, S <webi...@gmail.com> wrote:
> > The latest version from github fixes this error.
> > On May 10, 2008, at 9:40 AM, goodwill <william.yeung...@gmail.com>
> > wrote:
> > > I want to know if there is anyone have clues on how to fix this?
> > > Otherwise I am just not comfortable to use active_scaffold for now...
> > > On May 9, 11:45 pm, goodwill <william.yeung...@gmail.com> wrote:
> > >> Second to this. It doesnt work on Safari 3.1, even changing row 266
> > >> to
> > >> select doesnt work. It fixed the problem on failed create (with
> > >> validation failure) but fails again on a successful create! Help.....
> You just have to change a line of javascript. You can compare the last > two changes to see the difference. This might just be the line 266 > change. It fixed the rjs error in ie7. I have not verified it in > safari.
> On May 10, 2008, at 11:36 PM, goodwill <william.yeung...@gmail.com> > wrote:
>> I should say I am completely new on git- how could I install the >> plugin from there? if I just download the gz file, is there any >> method >> to install that directly via the download without creating my own svn >> store?
>> On May 11, 11:29 am, goodwill <william.yeung...@gmail.com> wrote: >>> oh really, can be installed via script/plugin?
>>> On May 11, 3:05 am, S <webi...@gmail.com> wrote:
>>>> The latest version from github fixes this error.
>>>> On May 10, 2008, at 9:40 AM, goodwill <william.yeung...@gmail.com> >>>> wrote:
>>>>> I want to know if there is anyone have clues on how to fix this? >>>>> Otherwise I am just not comfortable to use active_scaffold for >>>>> now...
>>>>> On May 9, 11:45 pm, goodwill <william.yeung...@gmail.com> wrote: >>>>>> Second to this. It doesnt work on Safari 3.1, even changing row >>>>>> 266 >>>>>> to >>>>>> select doesnt work. It fixed the problem on failed create (with >>>>>> validation failure) but fails again on a successful create! >>>>>> Help.....
>>>>>> On Apr 21, 7:23 pm, chetanm <den1...@gmail.com> wrote:
>>>>>>> yes !! even I get RJS errors in FF3b5 and Safari3.1
>>>>>>> On Apr 21, 9:47 am, David Herman <dst...@purpleasphalt.com> >>>>>>> wrote:
>>>>>>>> I'm also getting this in FF 3b5
>>>>>>>> On Apr 20, 1:49 pm, Sonar <serp...@gmail.com> wrote:
>>>>>>>>> RJSerror(the same as described above or a new one, reappeared >>>>>>>>> inSafari3.1.1)
I can confirm this issue. I've tried two earlier mentioned AS versions from github and also change manually javascript files but no success. $$("#admin__admins-update-2-form a.cancel").first().link return undefined value. I wonder if .link is attribute introduced by "pure" javascript, prototype or just active scaffold js? I found one place where value is set to .link attribute. It's in function register_cancel_hooks:
register_cancel_hooks: function() { // anything in the insert with a class of cancel gets the closer method, and a reference to this object for good measure var self = this; this.adapter.select('.cancel').each(function(elem) { elem.observe('click', this.close_handler.bind(this)); elem.link = self; }.bind(this)) },
It looks like .link is some helper attribute used only by AS but isn't properly set. Can anyone fix it please?
|
