Update: Since this post was made, the final version of Update 5 was released and should be used instead.
The ColdFusion 11 Update 5 prerelease build has been refreshed.
For changes please refer this document.
For instructions on how to install this update you can refer the following blog post announcing the Update 5 pre-release: https://coldfusion.adobe.com/coldfusion-11-update-5-prerelease-build-now-available
We continue to look forward to your feedback and suggestions.
NB. Users who had applied ColdFusion 11 Update 5 pre-release build earlier, should uninstall and re-apply this update build. The build number after applying this update should be 11,0,05,293370 (PreRelease).
Newbie
25 posts
@Samson:This issue is fixed and will be available as part of upcoming update of ColdFusion 11. If you need an early access to this hotfix, please reach out to cfsup@adobe.com
This issue only occurs if you have debugging option enabled.
Please let us know if you face this issue even after applying the hotfix.
Still not getting the value from the last output parameter. We have yet to find one CF11 build that’s production ready for us – and now you’re talking about a CF12 pre-release?
I’m on sybase.
I’m being forced to choose between either not upgrading, or turn my output parameters into query results – and I’ve been trying since early December 2014.
I went through the trouble of logging in to add my 2 cents to the bug, and got this error from the bug reporting site;
The web site you are accessing has experienced an unexpected error.
Please contact the website administrator.
The following information is meant for the website developer for debugging purposes.
Error Occurred While Processing Request
You have attempted to dereference a scalar variable of type class java.lang.String as a structure with members.
Resources:
Enable Robust Exception Information to provide greater detail about the source of errors. In the Administrator, click Debugging & Logging > Debug Output Settings, and select the Robust Exception Information option.
Check the ColdFusion documentation to verify that you are using the correct syntax.
Search the Knowledge Base to find a solution to your problem.
Browser Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/44.0.2403.39 Safari/537.36
Remote Address 10.5.14.7
Referrer https://bugbase.adobe.com/index.cfm?event=login&origEvent=bug___id____3971083
Date/Time 17-Jun-15 11:16 PM
Bug ID 3971083 Created and all of this information was included. Please vote / verify BUG so that we can get this addressed. Thank you.
I forgot to change the name of sp in cfm template before posting it.
Yes please log a bug for this issue.
Please share below mentioned details at nimsharm@adobe.com
1. Settings summary
2. Database version
3. OS
@Nimit
(typo in CF code to call SP not sperror but spTest) I was able to recreate error.
Event Name: onRequest
error message: Variable NAME11 is undefined.
error detail:
The contents of the tag stack are:
Template: C:inetpubdev.empowersis.comCF11_test.cfm
Line: 31
ID: ??
Where line 31 is the code #name11#
as you have given it.
Than you for responding to post. Should I create a BUG report for issue?
Hi Matthew,
I have tried re-creating the issue using the following code:
— Table creation
create table Test
( id int,
name varchar2(20),
name1 varchar2(20),
name2 decimal,
name3 varchar2(20),
name4 varchar2(20),
name5 varchar2(20),
name6 varchar2(20),
name7 decimal,
name8 varchar2(20),
name9 varchar2(20),
name10 varchar2(20),
name11 varchar2(20));
— Data insertion
insert into Test values(1, ‘SomeText’, ‘SomeText’, 23432.45, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’, 34.34, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’);
insert into Test values(2, ‘SomeText’, ‘SomeText’, 324.23, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’, 56.5, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’);
insert into Test values(3, ‘SomeText’, ‘SomeText’, 23232.2, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’, 78.6, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’);
insert into Test values(4, ‘SomeText’, ‘SomeText’, 7665.5, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’, 789.5, ‘SomeText’, ‘SomeText’, ‘SomeText’, ‘SomeText’);
–create stored procedure
create or replace procedure spTest (idd in number, name out varchar2, name1 out varchar2, name2 out decimal, name3 out varchar2, name4 out varchar2, name5 out varchar2, name6 out varchar2, name7 out decimal, name8 out varchar2, name9 out varchar2, name10 out varchar2, name11 out varchar2) as
Begin
SELECT name, name1, name2, name3, name4, name5, name6, name7, name8, name9, name10, name11 into name, name1, name2, name3, name4, name5, name6, name7, name8, name9, name10, name11 FROM Test where id = idd;
End;
–Calling Stored ptocedure
#name#
#name1#
#name2#
#name3#
#name4#
#name5#
#name6#
#name7#
#name8#
#name9#
#name10#
#name11#
Please verify if it is working at your end or not.
Share your findings at nimsharm@adobe.com
Matthew,
We have not been able to reproduce the issue so far.
We are still investigating this.
I see update 5 was released and it has caused the same issue for me on our server. It seems that the issue I have explained was not addressed before release. Have any suggestions? Explanations?
Thanks Mathew for sharing the code. We will look into this.
Oh, ok. Yes, to recreate issue. Yes. I am called a Stored Procedure that has more than one OUT parameter. The last OUT parameter variable name is the issue. Sorry, you would need a stored procedure and have more than one output. I can not post code but that explains the situation. EG code looks like such:
return code
Returned variables (from auth_pkg.p_get_user):
vAuthType = #vAuthType#
vAuthId = #vAuthId#
vAuthIdleTimeout = #vAuthIdleTimeout#
vAuthGateway = #vAuthGateway#
vAuthAction = #vAuthAction#
vAuthSuccess = #vAuthSuccess#
vAuthFailure = #vAuthFailure#
vReturnCode = #vReturnCode#
vReturnMsg = #vReturnMsg#
vReturnMsg is the only variable not able to be reference.
Is it happening with any stored procedure with output parameter?
It would be helpful if you can share the sample code using which I can reproduce the issue.
I am sorry I don’t understand “share repro case”?
@Matthew:
Can you please share repro case at nimsharm@adobe.com?
I applied the patch today interested in the CFHTMLtoPDF looping fix. But after applying my site is not working due to returned output calls from . My Prod has one input parameter and two output parameters. where the outpur parameters have variable=”p_erritem” and variable=”p_retval”
My issue is that now the variable is undefined.
” Variable P_RETVAL is undefined.”
on this line of code that immediately follows the call to PROC.
Is the correct place to post such findings? IF not Please advise on where I should post.
@JIm – The update should ship in a few weeks’ time. Please do share additional feedback, if any.
CF Team: Can you give a timeframe as to when the “final” CF 11, Update 5 will be released? Thanks for all your work on this!
@Renee
I don’t like the tone of your comment.
Also, I really wish that you guys would fix the regression introduced with serializeJSON() in U15. https://bugbase.adobe.com/index.cfm?event=bug&id=3941059
We had to switch back to WDDX, to WDDX!!!!
WTH? I just happened to be browsing and saw the comments. what a waste, the majority of it is some guy peeved because of a response with an @ symbol? Your suppose to reply to someone with an @ symbol and I always will. I don’t like # tag because it is used to delimit CFML vars, but I can’t stop the world from doing it…
Hmm. This reminds me of the character Warren in “There’s Something About Mary”, who would be extremely agitated if someone touched his headphones. Would people know that if they approached him? Not likely.
I guess time will tell, Peter, if this sort of concern you raise will become more widely understood and appreciated, and might lead to a shift away from the convention.
Until then, I feel truly disheartened (for the sake of the person and the community) to think that someone would be so offended by what would otherwise to most seem a totally innocent gesture.
charlie wrote:
> Do you perhaps read some other tone in my comment? None was intended.
That’s precisely why I raised it. It’s frustrating every time someone prefixes my name with at symbol, and it increases the likelihood of an otherwise neutral message being taken badly.
> please note that it’s just a convention, used all over the web
> (not just twitter), and intended totally to be a helpful one
Sure, it has spread to the web outside Twitter, but it didn’t happen before – people just used names without the need for a symbol, because everyone who went through pre-school knows what names are.
Using someone’s name can be helpful in any asynchronous discussion medium, (though it can also be merely a turn of phrase, and that’s fine too), but slapping an @ in front is … well, akin to jabbing a finger at someone every time you say their name.
With Twitter it has functionality (though it should only ever have been an input-command; not preserved in the rendered message), and a few other places may copy that, but otherwise it is – at best – unnecessary noise.
> It has never occurred to me (and I’ve never read) that anyone would feel offended by that
I do try not to tell people things they are already aware of… 😉
Now, on to the meat of your Peter’s last comment…
Thanks for clarifying how it was that you were asking (about getting the test cases) from a context of a previous conversation (not here, but in the bugbase). That’s all I was getting at.
Since I (and likely most readers) here wouldn’t have had occasion to have necessarily seen that, it just seemed worth asking, and also to find out whether you meant just this or all their tests.
Like I said, I can see the value in us having the tests (some or all). So again, I was not arguing with or against you. I just asked some simple questions, as always just trying to help carry the conversation forward. Thanks for answering them, ultimately. Hope it helps some readers (and Adobe) to better understand your request.
Peter, what in the world are you carrying on about in your first couple of paragraphs? Are you saying you’re offended because I opened my reply to you with @peter? Is it perhaps because you’re receiving the comment by email and feel you don’t need to be informed who the comment was intended for? Or are you really referring to its use even when viewed here on the web interface?
Either way, please note that it’s just a convention, used all over the web (not just twitter), and intended totally to be a helpful one–not at all insulting! It’s used most typically in blog comments or web forums where it may not be readily obvious to a reader (not just you, but ANYONE) which person in a thread someone may be replying to.
And if you may retort that yours was the first and only comment in the blog post, and so I “didn’t need to bother”, well, fair enough. But sometimes posts can be moderated (for any of several reasons), so it’s possible that a response may not be immediately posted.
So I (and many folks) just regularly preface any reply we make here with the person’s name, and with the @ as a mere convention.
It has never occurred to me (and I’ve never read) that anyone would feel offended by that. Or am I somehow totally misreading what you’ve been offended by? Help a brother out. (And goodness, that’s all I was doing with my comment. Do you perhaps read some other tone in my comment? None was intended.)
Would you have been less offended if I left off the @? Is that really it? If so, ok. To each his own. Please note that I didn’t this time, so as not to offend again. But I may well fall into the habit again.
But more than that, does my reply here change your take on things? Do you see how this has nothing to do with you personally, nor any offense or tone intended at all by that nomenclature?
I’ll reply to the real matter of your comment separately in a moment. (Comments that are too long can be forced into moderation.)
Charlie, this text isn’t invisible to you, and this isn’t Twitter. My name starts with a P, and I’ve had several years of practice in seeing it without needing someone waving a flag saying “hey you, look carefully, there’s a name next to me!”
You’re not (by far) the only one to do this, and I’m sure you aren’t deliberately trying to be insulting, but it *is* annoying to regularly be told I’m as stupid as a dumb piece of software – because that’s the only thing in need of such an indication.
As for the matter of test cases, it’s not a new question – it’s been raised once or twice on this issue, (and I think possibly a few others, but I don’t remember for sure).
For example, see comment on 3910529 …well, the bugbase is too useless to allow linking to comments, but it’s about 60% of the way down; search the page for “Ryan Guill” to get there.
https://bugbase.adobe.com/index.cfm?event=selectBug&CFGRIDKEY=3910529
Given how badly Adobe have handled this specific functionality – implementing, breaking, “fixing”, breaking again, it’s a really pertinent question.
I don’t expect Adobe to ever let us see source code, but test code is not source code, and they are a simple way the community can help to identify flaws (either through spotting missing tests and/or incorrect assumptions), and certainly it’s the only way I’ll have any faith in the functionality having been fixed properly.
It would of course be great to have the full set published, but that would be a bigger piece of work. Just seeing the ones for this functionality seems a reasonable request which shouldn’t take anyone much time.
@Peter, your last question seems to presume we DO get to see them. I never have myself.
Are you asking IF we could (with the understanding that they may not choose to share them)? Or are you referring perhaps to some other conversation with them (that we readers may not know about) where perhaps you were told they could or would be shared?
Don’t get me wrong: my questions are not an attack. I’m just saying this is the first I’ve heard of someone asking to see them, so it doesn’t seem a common expectation. I’m not saying it wouldn’t be interesting. 🙂 But I suspect there may be more to a possible answer (should they release all tests? on a site? only to those who ask? only for those that such folks ask for?)
Just trying to get clarification of this point as a reader here, while you await answers to this and the other question.
So is the null coalescing operator (?:) working yet?
The first version of update 5 pre-release claimed to fix it, yet according to Adam Tuttle it instead broke the partial fixes u4 gave us.
It doesn’t have a ** in the release notes, suggesting it has been left broken, and yet the presence of 3840570 in the release notes suggest you think it’s working.
https://bugbase.adobe.com/index.cfm?event=selectBug&CFGRIDKEY=3840570
https://bugbase.adobe.com/index.cfm?event=selectBug&CFGRIDKEY=3910529
When do we get to see the test cases you’re using to prove the bugs are fixed?
You must be logged in to post a comment.
23 Nov
10AM Pacific
Online
