Hi all,
I have seen in svn the prototype.js was updated to 1.6.
I tried this about a month ago for my site with 4.1.7 and experienced some problems, probably from sript.acoulo.us components having incompatibility problems.
I haven't tried it again, but are there some others experiencing problems after updating to svn 5373?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi Greg,
to show you what happened between script.acoulo.us 1.7.0 (actual use in pgv) and the up-to-date version 1.8.2.
They have seen the need to do some refactoring for prototype 1.6. I will load script.acoulo.us 1.8.2 in my testdrive and check if it works fine.
If yes, it will be usefull to be consistent with library and framework.
Veit
*V1.8.2* (November 18, 2008)
* Update to Prototype 1.6.0.3
* Make sure InPlaceEditor converts HTML entities to text. [Sean Kirby]
* Fix that Builder.node did not return extended elements on IE. Closes #71 and #77.
* Fix a bug in Sortable.destroy to make sure it's called on the referenced Sortable only, which allows for the correct intialization of nested Sortables. Closes Trac #8615. [Leon Chevalier]
* Change Effect.Base#render not to use eval(), so certain JavaScript runtime environments (like Adobe AIR) that do not support eval() work. [King Maxemilian, John-David Dalton]
* Fixed a calculation error in Effect.Transitions.pulse that could lead to flickering, add easing and change it to be a normal 0 to 1 transition that can be used with any effects; Effect.Pulsate now uses its own implementation. [Thomas Fuchs]
* Fixed Effect.ScrollTo. Changeset 8686 had a typo, document.viewport.getScrollOffsets[0] is always undefined. Removed the max check as it is not a cross-browser way to get scroll height and breaks the effect. Depending on scrollTo to do the right thing. Closes #11306. [Nick Stakenburg]
* Update version check so all Prototype versions can be required, not just x.x.x. Closes #10966. [Nick Stakenburg]
* Using $$ in the loader instead of getElementsByTagName to prevent limitations. Closes #9032. [Nick Stakenburg]
* Fix some missing semicolons. [jdalton]
* Fix an issue with Effect.ScrollTo that caused Firefox to scroll to the wrong offset in some situations. Closes #10245. [nik.wakelin]
* Fixes an issue with IE ghosting on non-absolute elements. Closes #10423. [Tanrikut, tdd]
*V1.8.1* (January 3, 2008)
* Fix Element#getStyles in IE. Closes #10563. [Tobie Langel]
* Update to Prototype 1.6.0.1 as of 8551.
* Fix a regression with autocompleters not responding correctly to cursor keys on both Safari and IE. Closes #10107. [thx Zman771]
*V1.8.0* (November 6, 2007)
* Update to Prototype 1.6.0 final
* Ajax.InPlaceEditor now can deal with callbacks that return an object. Closes #10064. [tdd]
* Fix a potential problem with the loader and Firefox 2.0 on the Mac. Closes #9951. [awaters]
* Add duration and distance options to Effect.Shake. Closes #8637. [amiel, rmm5t]
* Update code to use new Hash implemention in Prototype 1.6. Update InPlaceEditor to use new Class.create syntax. [tdd]
*V1.8.0 preview 0* (October 12, 2007)
* Update to new Class.create syntax in Prototype 1.6; update to latest Prototype 1.6 trunk.
* Fix bottom CSS property reassignment and initialization in queues for Effect#SlideUp. Closes #7412, #7761.
* Fix an issue with String#parseStyle that could sometimes cause errors on Safari 2.
* Refactoring to use Prototype 1.6. Remove dependency of Effect.tagifyText from builder. Use Prototype code style rules more consistently.
* Remove the deprecated Effect2 namespace.
* Update to Prototype 1.6.0_rc0.
* Some effect code refactoring to take advantage of new Prototype 1.6 features.
* Effect.ScrollTo is now no longer a core effect (returns a Effect.Tween instance now).
* Add Effect.Tween, a new core effect. This effect tweens between two values and sets a property or calls a method on an object (including DOM elements); or allows for a callback method, which will be automatically bound to the object.
Examples:
var whatever = { blech: 0 };
new Effect.Tween(whatever, 5, 0, 'blech'); // sets property on the object
new Effect.Tween('foo', 10, 20, 'innerHTML'); // sets property on the 'foo' DOM element
new Effect.Tween('foo', 10, 20, 'update'); // method call on 'foo' DOM element
new Effect.Tween('foo', 50, 0, { duration: 2.0 }, function(p){
this.setStyle({border:p.toFixed(3)+'px solid red'});
});
new Effect.Tween(null, 0, 100, function(p){ scrollTo(0,p) }); // scrolls the window
* Revert the workaround for the Firefox issue that makes autocompleter input field values be remembered correctly when the back button is hit to return to a page that contains autocompleted values, as this can cause "Are you sure you want to navigate away?" popups. Closes #8411. [okada]
* Improve the way deactivating the active droppable is handled. Closes #9072. [Karsten Sperling]
* Next version will be 1.8.0. Update to Prototype 1.6.0_pre1.
* Make draggable initialization faster. Closes #8697. [ssinghi]
* Make BDD-style testing compatible with IE6 and IE7. Closes #8972. [steveluscher]
* Add BDD methods to booleans. Closes #9147. [steveluscher]
* Add support for full CSS inheritance in Effect.Morph. Closes #9054. [Tobie]
IMPORTANT: Note that this could potentially change the outcome of classname-based morphs, so be sure to check your morph effects if you use this feature.
For this feature, the new method Element#getStyles is introduced, which returns an objects which properties correspond to the CSS properties found in Element.CSS_PROPERTIES. Note that this method doesn't work seemlessly across browsers for certain non-measurable CSS properties, such as float.
* Add the externalControlOnly option to the in-place editors to allow for external controls to exclusively trigger the in-place editing. Closes #9024. [tdd]
* Complete rewrite of Ajax.InPlaceEditor and Ajax.InPlaceCollectionEditor. [tdd]
THIS UPDATE MIGHT AFFECT BACKWARDS COMPATIBILITY.
Be sure to properly test, especially if you're using multi-line editing and/or monkeypatching.
Read more about all the added goodness in this article: http://mir.aculo.us/2007/7/17/in-place-editing-the-summer-2007-rewrite
* Update to Prototype 1.5.2_rc0 r7076:
- Works around bugs in Safari 1 and Safari 2 Regexp engines that could cause crashes
- Fixes a rendering issue with opacity-based effects and floating elements on Safari
* Mild refactoring of sound.js to take advantage of new Prototype features.
* Add highlight element method as a shortcut to Effect.Highlight.
Example:
$('message').highlight({startcolor:'#ff0000'}).update('KTHXBYE');
* Improve performance of String.prototype.parseStyle by caching the dummy element used.
* Add combined effects as element methods.
Examples:
$('message').appear().update('Oh noes!');
$('login').shake({ duration: 2 });
* Add Effect.Transitions.spring, which simulates a spring which oscillates while coming to rest. This transition is specifically targeted towards use with Effect.Move, for example as a reverteffect when using Draggables, as it will have return values >1 and cause temporary "overshooting" of the effect. Closes #8474. [edg2s]
* Fix effects initialization if no options are given. Fixes #7809. [thx Kroc Camen]
* Fix incorrect placement on drop for absolute ghosting elements. Update to Prototype 1.5.2_rc0 r6955 (required for #8581). Closes #8581. [tdd]
* Make sure the autocompletion happens on the token where the cursor is, not the last token when using tokenized autocompleting text fields. Closes #8588. [tdd]
* Fix that non-Mozilla browsers would display messages with the autocompleter back-button fix introduced in 1.7.1 beta 2. Fixes #7752, #8411. [josh, stetz]
*V1.7.1 beta 3* (May 19, 2007)
* Update to Prototype 1.5.1 final
* New elements and handles options for Sortable.create which take arrays of elements or element ids. These overrule the normal handle and elements finding options on initialization of the sortable, and can yield huge performance increases in situations where these elements or ids are known at call time. [Thomas Fuchs]
* Major performance improvement of Sortable.create when using the handle option. [Tobie Langel]
* Fix id assignment for sounds on Firefox/Windows, fixes #7709 [Robert Wallis]
* Add Sound.enable() and Sound.disable() to globally turn off Sound.play()
* Added Sound.play(url,options) in new sound.js file. scriptaculous.js automatically includes this file.
Based on code by Jules Gravinese, used with permission.
The sound player uses native sound support on IE, and falls back to using <embed> on other browsers,
which means it uses QuickTime for most cases. The recommended format to use is MP3.
Examples:
Sound.play('blah.mp3');
// --> plays sound immediately in 'global' track
Sound.play('blah.mp3',{replace:true});
// --> stop playing all sounds in 'global' track, and play new sound
Sound.play('blah.mp3',{track:'mytrack'});
// --> places the sound in the 'mytrack' track
Sound.play('blah.mp3',{track:'mytrack',replace:true});
// --> stop playing all sounds in 'mytrack' track, and play new sound
The sound effect used in the functional test is "Sword being drawn" by James Greever, released as freeware.
* Various effects engine optimizations [Tobie Langel, Thomas Fuchs]
* Make Ajax.InPlaceEditor more customizable: [thx godlie]
- Add an okLink option to Ajax.InPlaceEditor so you can have a link instead of a button. Uses the 'editor_ok_link' CSS class for styling, fixes #6455
- Add an cancelButton option to Ajax.InPlaceEditor, uses the 'editor_cancel_button' CSS class
- Add textBefore-, textBetween- and textAfterControls (which default to empty strings)
- For consistency, cancel link now also gets 'editor_cancel_link' CSS class
* Add an onDropped callback to Draggables, that fires at the end of a drag when the Draggable was dropped on a Droppable
* Add revert: 'failure' option to Draggables to only revert if not dropped on a valid drop target, fixes #6909 [davidw]
* Update to new Prototype Browser detection
* Fix a memory leak with Effect.Morph in Firefox, fixes #7558 [Malard]
* Add ability to nest single nodes in Builder.node in addition to arrays of nodes.
Examples:
var element = Builder.node('div', Builder.node('span','blah'));
Builder.dump();
var element = DIV(SPAN('blah'));
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It was a little bit tricky. Script.aculo.us has an internal loading mechanism for further js, like effect.js and dragdrop.js. With script.aculo.us 1.8.2 this mechanism only seem to work if prototype.js and scriptacolous.js are placed in the html-header, which isn't done in pgv.
We can workaround this by placing calls to the needed files effects.js and dragdrop.js in the scriptaculous.js.htm.
After doing this all my tests with IE7, IE8, Google Chrome 1.0 and Firefox 3.0.7 went well.
Perhaps someone like to test other browser after checkin to svn.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
"If you want to use both jQuery and Prototype on the same page, you must use jQuery's noConflict mode. If not, you're overwriting Prototype's $ function, and thus breaking about everything in Prototype."
I think this is causing me a greater problem - INDI pages fail to open at first attempt, always requiring a refresh first. This has been a problem for quite sometime, but I've finally given up trying to fix it myself - any suggestions gratefully received. Or, the two issues might be unrelated?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi kiwi,
taking a look on your site I could not reproduce the problem, neither with IE8, nor FF3.
Do you need to be logged in to have this behaviour?
The noconflict mode is taken into account and causes no problems here.
What do you do exactly to produce the error?
Veit
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2009-06-14
Thans Viet, thats interesting. I don't need to be logged in.
Treating the two things as separate issues:
1 - The js error appears on the Welcome page at all times. It disappears if I navigate away to an INDI page. This problem is on FF and IE.
2 - The page oping issue also does not require login. If I click on "Charles Chamar" (Person who lived longest, bottom center iof Welcome page), the page stalls before the tabs are loaded. A page refresh clears the problem and the page opens. This problem is ONLY in FF.
If you don't see the same things, it suggest the issue might be browser related. Perhaps one of the add-ons I have. I'l have to experiment by removing them one at a time.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Kiwi,
I cannot see any errors in FF 3.0.11, but I can see it in IE8. I checked the page code and you are right. The noConflict statement is missing on this page. The 2nd issue I cannot reproduce.
I will check where to include best the statement.
Veit
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Kiwi,
please check if svn 5733 solves your problem.
Veit
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2009-06-14
Thanks Viet, thats done two things:
1 - yes it has fixed the js error (my issue 1. above)
2 - It has confirmed my two issues are not related, as 2. above still remains. This occurs on individual.php only. Looking at the page source I see this near the bottom:
<script type="text/javascript">
//<![CDATA[
var catch_and_ignore; function paste_id(value) {catch_and_ignore = value;}tabswitch(1);if (typeof toggleByClassName == "undefined") {alert("phpgedview.js: A javascript function is missing. Please clear your Web browser cache");}
//]]>
</script>
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2009-06-14
... also, Firebug gives this error when I do a screen refresh to display the page:
doc.location is null
chrome://safe/content/framework.js
Line 1699
So perhaps something I need to change in FF settings?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2009-06-17
Just for the record, I have fixed the problem of pages not opening first time. The conflict in FF3 was with its add-on "Firebug ver 1.3.3". It was permanently enabled in javascript debug mode. No idea why there is a conflict, but having it permanently disabled for all sites has fixed the problem. I can still enable it when I need to use it.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi all,
I have seen in svn the prototype.js was updated to 1.6.
I tried this about a month ago for my site with 4.1.7 and experienced some problems, probably from sript.acoulo.us components having incompatibility problems.
I haven't tried it again, but are there some others experiencing problems after updating to svn 5373?
The update was to fix some reported compatibility issues with Safari.
If the change is causing other problems for anyone, then please say so....
Hi Greg,
to show you what happened between script.acoulo.us 1.7.0 (actual use in pgv) and the up-to-date version 1.8.2.
They have seen the need to do some refactoring for prototype 1.6. I will load script.acoulo.us 1.8.2 in my testdrive and check if it works fine.
If yes, it will be usefull to be consistent with library and framework.
Veit
*V1.8.2* (November 18, 2008)
* Update to Prototype 1.6.0.3
* Make sure InPlaceEditor converts HTML entities to text. [Sean Kirby]
* Fix that Builder.node did not return extended elements on IE. Closes #71 and #77.
* Fix a bug in Sortable.destroy to make sure it's called on the referenced Sortable only, which allows for the correct intialization of nested Sortables. Closes Trac #8615. [Leon Chevalier]
* Change Effect.Base#render not to use eval(), so certain JavaScript runtime environments (like Adobe AIR) that do not support eval() work. [King Maxemilian, John-David Dalton]
* Fixed a calculation error in Effect.Transitions.pulse that could lead to flickering, add easing and change it to be a normal 0 to 1 transition that can be used with any effects; Effect.Pulsate now uses its own implementation. [Thomas Fuchs]
* Fixed Effect.ScrollTo. Changeset 8686 had a typo, document.viewport.getScrollOffsets[0] is always undefined. Removed the max check as it is not a cross-browser way to get scroll height and breaks the effect. Depending on scrollTo to do the right thing. Closes #11306. [Nick Stakenburg]
* Update version check so all Prototype versions can be required, not just x.x.x. Closes #10966. [Nick Stakenburg]
* Using $$ in the loader instead of getElementsByTagName to prevent limitations. Closes #9032. [Nick Stakenburg]
* Fix some missing semicolons. [jdalton]
* Fix an issue with Effect.ScrollTo that caused Firefox to scroll to the wrong offset in some situations. Closes #10245. [nik.wakelin]
* Fixes an issue with IE ghosting on non-absolute elements. Closes #10423. [Tanrikut, tdd]
*V1.8.1* (January 3, 2008)
* Fix Element#getStyles in IE. Closes #10563. [Tobie Langel]
* Update to Prototype 1.6.0.1 as of 8551.
* Fix a regression with autocompleters not responding correctly to cursor keys on both Safari and IE. Closes #10107. [thx Zman771]
*V1.8.0* (November 6, 2007)
* Update to Prototype 1.6.0 final
* Ajax.InPlaceEditor now can deal with callbacks that return an object. Closes #10064. [tdd]
* Fix a potential problem with the loader and Firefox 2.0 on the Mac. Closes #9951. [awaters]
* Add duration and distance options to Effect.Shake. Closes #8637. [amiel, rmm5t]
* Update code to use new Hash implemention in Prototype 1.6. Update InPlaceEditor to use new Class.create syntax. [tdd]
*V1.8.0 preview 0* (October 12, 2007)
* Update to new Class.create syntax in Prototype 1.6; update to latest Prototype 1.6 trunk.
* Fix bottom CSS property reassignment and initialization in queues for Effect#SlideUp. Closes #7412, #7761.
* Fix an issue with String#parseStyle that could sometimes cause errors on Safari 2.
* Refactoring to use Prototype 1.6. Remove dependency of Effect.tagifyText from builder. Use Prototype code style rules more consistently.
* Remove the deprecated Effect2 namespace.
* Update to Prototype 1.6.0_rc0.
* Some effect code refactoring to take advantage of new Prototype 1.6 features.
* Effect.ScrollTo is now no longer a core effect (returns a Effect.Tween instance now).
* Add Effect.Tween, a new core effect. This effect tweens between two values and sets a property or calls a method on an object (including DOM elements); or allows for a callback method, which will be automatically bound to the object.
Examples:
var whatever = { blech: 0 };
new Effect.Tween(whatever, 5, 0, 'blech'); // sets property on the object
new Effect.Tween('foo', 10, 20, 'innerHTML'); // sets property on the 'foo' DOM element
new Effect.Tween('foo', 10, 20, 'update'); // method call on 'foo' DOM element
new Effect.Tween('foo', 50, 0, { duration: 2.0 }, function(p){
this.setStyle({border:p.toFixed(3)+'px solid red'});
});
new Effect.Tween(null, 0, 100, function(p){ scrollTo(0,p) }); // scrolls the window
* Revert the workaround for the Firefox issue that makes autocompleter input field values be remembered correctly when the back button is hit to return to a page that contains autocompleted values, as this can cause "Are you sure you want to navigate away?" popups. Closes #8411. [okada]
* Improve the way deactivating the active droppable is handled. Closes #9072. [Karsten Sperling]
* Next version will be 1.8.0. Update to Prototype 1.6.0_pre1.
* Make draggable initialization faster. Closes #8697. [ssinghi]
* Make BDD-style testing compatible with IE6 and IE7. Closes #8972. [steveluscher]
* Add BDD methods to booleans. Closes #9147. [steveluscher]
* Add support for full CSS inheritance in Effect.Morph. Closes #9054. [Tobie]
IMPORTANT: Note that this could potentially change the outcome of classname-based morphs, so be sure to check your morph effects if you use this feature.
For this feature, the new method Element#getStyles is introduced, which returns an objects which properties correspond to the CSS properties found in Element.CSS_PROPERTIES. Note that this method doesn't work seemlessly across browsers for certain non-measurable CSS properties, such as float.
* Add the externalControlOnly option to the in-place editors to allow for external controls to exclusively trigger the in-place editing. Closes #9024. [tdd]
* Complete rewrite of Ajax.InPlaceEditor and Ajax.InPlaceCollectionEditor. [tdd]
THIS UPDATE MIGHT AFFECT BACKWARDS COMPATIBILITY.
Be sure to properly test, especially if you're using multi-line editing and/or monkeypatching.
Read more about all the added goodness in this article:
http://mir.aculo.us/2007/7/17/in-place-editing-the-summer-2007-rewrite
* Update to Prototype 1.5.2_rc0 r7076:
- Works around bugs in Safari 1 and Safari 2 Regexp engines that could cause crashes
- Fixes a rendering issue with opacity-based effects and floating elements on Safari
* Mild refactoring of sound.js to take advantage of new Prototype features.
* Add highlight element method as a shortcut to Effect.Highlight.
Example:
$('message').highlight({startcolor:'#ff0000'}).update('KTHXBYE');
* Improve performance of String.prototype.parseStyle by caching the dummy element used.
* Add combined effects as element methods.
Examples:
$('message').appear().update('Oh noes!');
$('login').shake({ duration: 2 });
* Add Effect.Transitions.spring, which simulates a spring which oscillates while coming to rest. This transition is specifically targeted towards use with Effect.Move, for example as a reverteffect when using Draggables, as it will have return values >1 and cause temporary "overshooting" of the effect. Closes #8474. [edg2s]
* Fix effects initialization if no options are given. Fixes #7809. [thx Kroc Camen]
* Fix incorrect placement on drop for absolute ghosting elements. Update to Prototype 1.5.2_rc0 r6955 (required for #8581). Closes #8581. [tdd]
* Make sure the autocompletion happens on the token where the cursor is, not the last token when using tokenized autocompleting text fields. Closes #8588. [tdd]
* Fix that non-Mozilla browsers would display messages with the autocompleter back-button fix introduced in 1.7.1 beta 2. Fixes #7752, #8411. [josh, stetz]
*V1.7.1 beta 3* (May 19, 2007)
* Update to Prototype 1.5.1 final
* New elements and handles options for Sortable.create which take arrays of elements or element ids. These overrule the normal handle and elements finding options on initialization of the sortable, and can yield huge performance increases in situations where these elements or ids are known at call time. [Thomas Fuchs]
* Major performance improvement of Sortable.create when using the handle option. [Tobie Langel]
* Fix Builder.node double quote escaping in attributes, fixes #8125. [Aa!]
*V1.7.1 beta 2* (April 28, 2007)
* Update to Prototype 1.5.1_rc4
- Various optimizations and fixes
- http://dev.rubyonrails.org/browser/spinoffs/prototype/trunk/CHANGELOG?rev=6604
* Fix autocompleter indicator not appearing when calling activate() directly. Don't show indicator for local autocompleting. Fixes #7770. [pyro8019]
* Don't clear out autocompleting fields in Mozilla-based browsers when navigating back to a page with the Back button, fixes #7752. [seanc]
*V1.7.1 beta 1* (March 12, 2007)
* Update to Prototype 1.5.1 r6386
- Fixes SlideUp/SlideDown on IE
- Fixes an opacity problem with IE
- Performance optimizations
- See http://dev.rubyonrails.org/browser/spinoffs/prototype/trunk/CHANGELOG?rev=6381 for more information
* Make builder unit tests work on IE
* Fix id assignment for sounds on Firefox/Windows, fixes #7709 [Robert Wallis]
* Add Sound.enable() and Sound.disable() to globally turn off Sound.play()
* Added Sound.play(url,options) in new sound.js file. scriptaculous.js automatically includes this file.
Based on code by Jules Gravinese, used with permission.
The sound player uses native sound support on IE, and falls back to using <embed> on other browsers,
which means it uses QuickTime for most cases. The recommended format to use is MP3.
Examples:
Sound.play('blah.mp3');
// --> plays sound immediately in 'global' track
Sound.play('blah.mp3',{replace:true});
// --> stop playing all sounds in 'global' track, and play new sound
Sound.play('blah.mp3',{track:'mytrack'});
// --> places the sound in the 'mytrack' track
Sound.play('blah.mp3',{track:'mytrack',replace:true});
// --> stop playing all sounds in 'mytrack' track, and play new sound
The sound effect used in the functional test is "Sword being drawn" by James Greever, released as freeware.
* Various effects engine optimizations [Tobie Langel, Thomas Fuchs]
* Make Ajax.InPlaceEditor more customizable: [thx godlie]
- Add an okLink option to Ajax.InPlaceEditor so you can have a link instead of a button. Uses the 'editor_ok_link' CSS class for styling, fixes #6455
- Add an cancelButton option to Ajax.InPlaceEditor, uses the 'editor_cancel_button' CSS class
- Add textBefore-, textBetween- and textAfterControls (which default to empty strings)
- For consistency, cancel link now also gets 'editor_cancel_link' CSS class
* Add an onDropped callback to Draggables, that fires at the end of a drag when the Draggable was dropped on a Droppable
* Add revert: 'failure' option to Draggables to only revert if not dropped on a valid drop target, fixes #6909 [davidw]
* Update to new Prototype Browser detection
* Fix a memory leak with Effect.Morph in Firefox, fixes #7558 [Malard]
* Add ability to nest single nodes in Builder.node in addition to arrays of nodes.
Examples:
var element = Builder.node('div', Builder.node('span','blah'));
Builder.dump();
var element = DIV(SPAN('blah'));
<<I will load script.acoulo.us 1.8.2 in my testdrive>>
Veit - thanks - let us know the outcome, and we'll update SVN if necessary.
Greg
My first findings are that dir_editor.php and the child reordering in edit_interface.php are producing errors without updating script.aculo.us.
Also with script.aculo.us 1.8.2 there seem to be some modifications needed, because of JS errors like "Draggable is not defined".
I will check this and try to deliver a patch.
Veit
It was a little bit tricky. Script.aculo.us has an internal loading mechanism for further js, like effect.js and dragdrop.js. With script.aculo.us 1.8.2 this mechanism only seem to work if prototype.js and scriptacolous.js are placed in the html-header, which isn't done in pgv.
We can workaround this by placing calls to the needed files effects.js and dragdrop.js in the scriptaculous.js.htm.
After doing this all my tests with IE7, IE8, Google Chrome 1.0 and Firefox 3.0.7 went well.
Perhaps someone like to test other browser after checkin to svn.
FYI, I have applied these latest versions to SVN.
Can either of you throw any light on this. In FF3 I get an error on some pages:
element.dispatchEvent is not a function
The warning refers to prototype.js (line 4071)
In IE8 it is similar, but the error reads:
Message: Object doesn't support this property or method
Line: 4073
Char: 9
Code: 0
URI: http://www.our-families.info/js/conio/prototype.js
Googling this issue I found comments like this:
"If you want to use both jQuery and Prototype on the same page, you must use jQuery's noConflict mode. If not, you're overwriting Prototype's $ function, and thus breaking about everything in Prototype."
I think this is causing me a greater problem - INDI pages fail to open at first attempt, always requiring a refresh first. This has been a problem for quite sometime, but I've finally given up trying to fix it myself - any suggestions gratefully received. Or, the two issues might be unrelated?
Hi kiwi,
taking a look on your site I could not reproduce the problem, neither with IE8, nor FF3.
Do you need to be logged in to have this behaviour?
The noconflict mode is taken into account and causes no problems here.
What do you do exactly to produce the error?
Veit
Thans Viet, thats interesting. I don't need to be logged in.
Treating the two things as separate issues:
1 - The js error appears on the Welcome page at all times. It disappears if I navigate away to an INDI page. This problem is on FF and IE.
2 - The page oping issue also does not require login. If I click on "Charles Chamar" (Person who lived longest, bottom center iof Welcome page), the page stalls before the tabs are loaded. A page refresh clears the problem and the page opens. This problem is ONLY in FF.
If you don't see the same things, it suggest the issue might be browser related. Perhaps one of the add-ons I have. I'l have to experiment by removing them one at a time.
Kiwi,
I cannot see any errors in FF 3.0.11, but I can see it in IE8. I checked the page code and you are right. The noConflict statement is missing on this page. The 2nd issue I cannot reproduce.
I will check where to include best the statement.
Veit
And with FF 3.5
Warning: Error in parsing value for 'font-family'. Declaration dropped.
Source File: http://www.myarnolds.com/themes/xenea/style.css
Line: 1002
Warning: Error in parsing value for 'font-family'. Declaration dropped.
Source File: http://www.myarnolds.com/themes/xenea/print.css
Line: 595
Warning: Use of captureEvents() is deprecated. To upgrade your code, use the DOM 2 addEventListener() method. For more help http://developer.mozilla.org/en/docs/DOM:element.addEventListener
Source File: http://www.myarnolds.com/individual.php?pid=I69938&ged=Arnold.ged
Line: 0
None appear to be the Prototype update errors.
Kiwi,
please check if svn 5733 solves your problem.
Veit
Thanks Viet, thats done two things:
1 - yes it has fixed the js error (my issue 1. above)
2 - It has confirmed my two issues are not related, as 2. above still remains. This occurs on individual.php only. Looking at the page source I see this near the bottom:
<script type="text/javascript">
//<![CDATA[
var catch_and_ignore; function paste_id(value) {catch_and_ignore = value;}tabswitch(1);if (typeof toggleByClassName == "undefined") {alert("phpgedview.js: A javascript function is missing. Please clear your Web browser cache");}
//]]>
</script>
... also, Firebug gives this error when I do a screen refresh to display the page:
doc.location is null
chrome://safe/content/framework.js
Line 1699
So perhaps something I need to change in FF settings?
Just for the record, I have fixed the problem of pages not opening first time. The conflict in FF3 was with its add-on "Firebug ver 1.3.3". It was permanently enabled in javascript debug mode. No idea why there is a conflict, but having it permanently disabled for all sites has fixed the problem. I can still enable it when I need to use it.