Debug vs Standard causing HTTPRequest errors

Hello Flex Community
My worst bug/problem rised up today where everything was running nice and smooth while building my Flex 4 application in Debug mode but when I did a release (as standard .swf) the application failed constantly straight after any httpRequest would fire.
After searching the net for a bit I found this:
http://stackoverflow.com/questions/1137223/flex-app-not-working-without-debug-version-of-f lash-player
This looked like the problem as I had just switched from using xmlDecode functions to serialize the data to the serialization filer and had some getStackTrace functions that would potentially fail if it were not in Debug mode.
My question is can there be any other debug only methods causing this? or possibly something to do with the newer serializationFilter?
It would be worth noting these are https requests
Any feedback would be appreciated
Thanks
Tyrone

getStackTrace
is not a problem really, problem is that it returns null in regular FP and if you do some noughty operations on it you get NPE,
but since that will only happen in non debug version of FP you get no clue, and hence app haltes,
so rule of thumb is to protect possible NPE generating code, for example
getStackTrace().substring... something will definately halt your system in non debug version of FP player since you can not do substring on null
easies way is to put it inside try {} catch .

Similar Messages

Form debugging - Which line is causing the error?

I have a form that I'm working on, and when I run the form like normal, just as the end user would, I keep getting an error, "FRM-41039: Invalid Alert ID 0." I can't figure out where this error is coming from in my code. It seems that it must have something to do with timing, because in trying to track down the error, when I run the form in debug mode, obviously the timing is slower as I go line-by-line in the code to see which line may be causing the error, but when I do that, I don't get the same error. I'm not close to being an expert when it comes to design-time debugging in forms, and I was wondering if someone with more experience could point me in the right direction. Is there any way that I can somehow determine which line of code is causing this error to raise? If I could identify the line of code, I could maybe find out which alert it's looking for, right? Then I could maybe narrow down the problem a bit more. Anyone have any ideas? Thanks in advance.
YEX
<)))><

Like it says :"FRM-41039: Invalid Alert ID 0." Alert does not exist. So try to find line in your code which contain something like:
Show_Alert("<name of alert>")
Or Alert with this name dows not exist or you are misspelling the name of existing Alert.

DESADV01 E1EDP07 segment per line item causes EDIFACT error

When a Sales Order has the PO Number + Date populated, the subsequent outbound Delivery IDoc (DESADV01) contains a parent E1EDP07 segment for every child line item segment (E1EDP09). For a SO without PO Date, the IDoc is created with a single E1EDP07 with multiple child E1EDP09 segments. This is standard SAP - do you need to code around this?
The first scenario causes an error in EDIFACT mapping, as a limit of 9 RFF segments is allowed (E1EDP07 is a level 1 segment). Partner systems reject this.

Hi,
Segment field should be optional, so check the posting keys in OB41, check the GL account which hit this posting and check the FIeld status in OBC4 and make the segment field as optional.
while posting the transactions you can check in more then Profit center and segment details will be updated in the postings.
try out, this might help. check the new ledger document types also.
Regards,
Padma

Acrobat standard 9.0 error while enabling commets option using Java script object

I'm using the below java script code in my VB 6.0 application for enabling commets options in the PDF file
Dim oJSO As Object
Set oJSO = mPDDoc.GetJSObject
oJSO.Collab.showAnnotToolsWhenNoCollab = True
It is working fine till Acrobat 8.0. Now I'm trying with Acrobat 9.0 standard edition, it is throwing scripting error 'An unhandled Win32 exception occured in Acrobat.exe[308].
Can anyone suggest the way to enable 'comment' options in Acrobat Standard 9.0 using Java Script Object?
Thanks

I get much the same problem with VB6, Acrobat 9 Pro and WinXP SP3. When I single step the line:
oJSO.Collab.showAnnotToolsWhenNoCollab = True
in debug I go to the error handler for my Sub. Inspecting the VB Err object has Number = -2147417851 and Description = "Automation error
The server threw an exception." I get a Microsoft crash notice about Acrobat a short while later.
I've experimented with the JavaScript debugger in Acrobat and you can execute Collab.showAnnotToolsWhenNoCollab = True and reference the value of Collab.showAnnotToolsWhenNoCollab in the console. The problem appears only when accessing Acrobat Javascript through the Interapplication Communication API.
If anyone can help it would be appreciated. I tried Adobe's Acrobat support and they were no help at all.

What could cause "Internal error: java.lang.NullPointerException: can't not add null to SWF frame" when using compc?

using flex sdk 4.6, and air sdk 15.0.0.356
we get the following error when using compc in our ant build:
[java] Internal error: java.lang.NullPointerException: can't not add null to SWF frame
[java] at com.adobe.flash.swf.SWFFrame.addTag(SWFFrame.java:77)
[java] at com.adobe.flash.compiler.internal.units.requests.SWFTagsRequestResult.addToFrame(SWFTagsR equestResult.java:114)
[java] at com.adobe.flash.compiler.internal.targets.SWFTarget.addCompilationUnitsAndDependenciesToF rame(SWFTarget.java:398)
[java] at com.adobe.flash.compiler.internal.targets.SWFTarget$FramesInformation.createFrame(SWFTarg et.java:838)
[java] at com.adobe.flash.compiler.internal.targets.SWFTarget$FramesInformation.createFrames(SWFTar get.java:875)
[java] at com.adobe.flash.compiler.internal.targets.SWFTarget.build(SWFTarget.java:243)
[java] at com.adobe.flash.compiler.internal.targets.SWCTarget.buildLibrarySWF(SWCTarget.java:348)
[java] at com.adobe.flash.compiler.internal.targets.SWCTarget.build(SWCTarget.java:128)
[java] at com.adobe.flash.compiler.clients.COMPC.buildArtifact(COMPC.java:144)
[java] at com.adobe.flash.compiler.clients.MXMLC.compile(MXMLC.java:541)
[java] at com.adobe.flash.compiler.clients.MXMLC.mainNoExit(MXMLC.java:230)
[java] at com.adobe.flash.compiler.clients.MXMLC.mainNoExit(MXMLC.java:184)
[java] at com.adobe.flash.compiler.clients.COMPC.staticMainNoExit(COMPC.java:75)
[java] at com.adobe.flash.compiler.clients.COMPC.main(COMPC.java:62)
which seems to be rather vague, and offers no clue as to what caused the problem. so, my question is, what could cause this error?
these are the compiler options we use in the build file:
<java jar="${flex.compc.jar}" dir="." fork="true" failonerror="true">
      <jvmarg value="-Xmx512m" />
      <arg line="@{options} ${_compc.libpath} ${_compc.externspath} ${_compc.includespath}" />
      <arg value="+flexlib=${flex.frameworks.dir}" />
      <arg value="-load-config=@{config}" />
      <arg value="-source-path=@{src.dir}" />
      <arg value="-include-sources=@{src.dir}" />
      <arg value="-output=@{dest}" />
      <arg value="-swf-version=@{swf.version}"/>
      <arg value="-debug=${flash.debug}"/>
      <arg value="-advanced-telemetry"/>
</java>
assume that the properties resolve to correct values, since all projects (except one) call this task and compile successfully. the error must come from this one project that fails, but we don't know why.
any ideas?

You could be running out of memory, try increasing the -Xmx512m value to -Xmx1024m or another value that you prefer

Debug a standard prg

Hi Friends,
the funcational consultant doing a work with t.code kks2.
there is one errror message is coming.
error message : Item 00001 : Cost element 1276101000 does not exist in CO area US01 for 11/05/2007 .
so how to find out this error in program debugging...
where i can catch this error in prg...
pls. explain how to debug the standard prg. and tcode.
regards,

Best way to debug is :
Take the message number and message id .
now goto Program ( When you go transaction -> system->status -> here you get program name).
click on find -> enter message number ( Select main program ) -> here you get message number ,now keep the break point ).
some times they call messages as like dynamically.
goto transaction -> enter all the data -> /h in command bar -> now create watch point -> use varaible as message id and value as message number ) ,operator is =
now it stops exact area
Thanks
Seshu

Comiling the standard form gives errors

Hi,
When I open the standard form APXPMTCH.fmb in forms builder and compile, I receive the below error.
"identifier AP_AMT_MATCH_PKH.AP_AMT_MATCH must be declared"
when I looked for this procedure it exists in the instance. I am not sure what is causing this error? Are the number of parameters or parameters types changes if we open the form in form builders in this procedure call?
Thanks,
HC

Please post the details of the application release, database version, OS, forms version.
Have you completed the steps in this thread? -- URGENT: Forms 6i error
Can you reproduce the issue when you open other form (for example, TEMPLATE.fmb)?
Thanks,
Hussein

Standard I/O error window popup

hello :
I am facing with a desperate trouble. I have completed the program code and ".uir" design, and when I Compile and execute them, there is no error appear, however when bringing datas and processing them use it, it would popup a standard I/O error window . what's wrong with them?
hopping for your answer.

CVI can use a console (sort of command prompt window) to output messages from specific functions. This console is called Standard I/O window and is exactly the one you are seeing running your app. Now, printf () is the most probable cause of this window to appear, but there are several other instructions that can cause this to happen: this old knowledgebase article lists some of them. You may check if your app is using any of them.
Standard I/O behaviour can be modified: see this document for reference; look also SetStdIOWindowOptions () command. Consider also this forum thread, which can be useful if calls to those functions is included in some external library you are using that you cannot modify.
Proud to use LW/CVI from 3.1 on.
My contributions to the Developer Zone Community
If I have helped you, why not giving me a kudos?

HT4076 What causes this error?

I get this error every couple of weeks. This completely defeats the purpose of running Time Machine. Anybody know what might cause this error?
My setup is a Mac 21.5" running OSX 10.7.3 connected via WiFi to an Airport Extreme (firmware v7.6.1). The Airport Extreme is connected via Ethernet-cable to a Synology DS411j (DSM 4.0) on which the Time Machine sparsebundle resides. Another Mac also uses the SAN for Time Machine backups and hasn't had any troubles sofar.

Download and use 5.6 utility for Lion.. v6 in Lion can run alongside 5.6 to provide pretty network pictures.
http://support.apple.com/kb/DL1482
You will want to download your local version in Dutch??
So the cabled computer has no problems.. definitely a wireless issue then.
Could a '.' in the name be a problem?
Yes, if you are going to fix names fix them properly. No spaces no non-alphanumeric.. it is not that some special characters cannot be used.. it is to reduce the chance of problems. Wireless names (SSID) does not contribute to security at all.. except that it is good to use a name different to standard. ie if all brandx modems and routers use wireless name brandx it allows people to precompile lists of brute force attack. If the name is different they need to create it on the spot.. but in neither case is it a big issue.
I name the TC.. TC3 for example.. call the 2.4ghz wireless TC24ghz and the 5ghz TC5ghz. Just use something that makes good sense to you so you can identify it as your signal not your neighbours.
BTW, does that mean I'll have two networks?
No both 2.4ghz and 5ghz will share the same ip range and can talk to each other.
It is just like having two different ethernet cables. The problem with wireless, is it is single channel half-duplex.. and like the old thin ethernet, everybody shares the same band.. and divides the bandwith. Using two different bands allows two computers to talk at the same time.

I am trying to install current Firefox 3.6, as I am currently running an older Firefox on my Windows ME, but I keep getting this error message: "Setup has caused an error in SETUP.EXE Setup will now close."--Please help?

I have downloaded all IP changer Privacy software I have purchased and am trying to install it, and although it is made to work with WindowsME that I am currently using on my computer, it now says it requires Firefox 3 or above to be able to install the software. Therefore, I am trying to update my Firefox to current 3.6 but once Firefox 3.6 downloads it doesn't install. Instead I just keep getting the same error message: Setup has caused an error in SETUP.EXE. Set up will now close. Several times, I have also tried restarting the computer as the error message also recommended But I just still keep getting the same error message when I try to update/install Firefox 3.5. Please help/advise on how I can fix this problem. Many thanks in advance!

Firefox 2.0.0.20 is the last ever version of Firefox for Win98 / SE / ME.
You can get it from here, but you are better off using K-Meleon (derived from Firefox) or Opera instead an old version of Firefox:
[ftp://ftp.mozilla.org/pub/firefox/releases/2.0.0.20/win32/en-US/]
You can try KernelEx to get Firefox 3+ to work, but it would be very slow with an old PC of less than 750Mhz. 3.5 and later supposedly doesn't work.
http://sourceforge.net/projects/kernelex/

A view, function and TO_DATE causing an error.

I have the following statement which calls a view, VW_DIST_RPT_WORK_LIST which in turn calls a function which returns either 'Null' or a date string e.g. '07 Oct 2003' as a VARCHAR2 (alias PROJECTED_DELIVERY_DATE).
Statement:
SELECT CUSTOMER_NAME, PROTOCOL_REFERENCE, SHIPPING_REFERENCE, CUSTOMER_REFERENCE, COUNTRY, PROJECTED_DELIVERY_DATE, STATUS, NOTES,
TO_DATE(PROJECTED_DELIVERY_DATE)
FROM VW_DIST_RPT_WORK_LIST
WHERE EXPECTED_DESP_DT IS NOT NULL
AND UPPER(PROJECTED_DELIVERY_DATE) NOT LIKE('NULL%')
AND EXPECTED_DESP_DT <= TO_DATE('07/10/2003', 'DD/MM/YYYY')
AND TO_DATE(PROJECTED_DELIVERY_DATE) <= TO_DATE('31/12/2003', 'DD/MM/YYYY') --< Problem here
I need to be able to specify a date filter on the PROJECTED_DELIVERY_DATE field and hence used the TO_DATE(PROJECTED_DELIVERY_DATE) <= TO_DATE('31/12/2003', 'DD/MM/YYYY') but this is generating an ORA-01858: a non-numeric character was found where a numeric character was expected.
I think the problem lies with the fact that this field can contain 'Null' which cannot be converted to a date using TO_DATE. I've tried adding a NOT LIKE ('NULL%') statement to catch any nulls which may be creeping in bu this doesn't solve the problem.
I've added TO_DATE(PROJECTED_DELIVERY_DATE) to the select above to determine if the nulls are being caught and if the TO_DATE in performing the conversion correctly which it is on both counts.
Any ideas anyone ?

The answer provided above by Monika will work for this situation. However, you should seriously think whether you should be using a string for date datatype. Ideally, you should rewrite the function that returns PROJECTED_DELIVERY_DATE and change the return type to DATE. The least you should do is to return NULL (instead of the string 'NULL') from the function. Oracle handles nulls perfectly, there is no reason you should write code to handle nulls;
One more thing. Looking at the type of error you are receiving, it seems that you are using rule based optimizer. Why do I think so? Because, in rule based optimizer, the conditions are evaluated in a specific order (viz, bottoms-up for AND clauses). To show this, look at the following simple demonstration. I did this in Oracle 8.1.6 (also in 9.2.0.4.0 on Windows).
-- Check the database version
select * from v$version;
BANNER
Oracle8i Enterprise Edition Release 8.1.6.1.0 - Production
PL/SQL Release 8.1.6.1.0 - Production
CORE 8.1.6.0.0 Production
TNS for Solaris: Version 8.1.6.0.0 - Production
NLSRTL Version 3.4.0.0.0 - Production
-- Create the test table
create table test (a number(2));
insert into test(a) values (0);
insert into test(a) values (1);
insert into test(a) values (2);
insert into test(a) values (3);
insert into test(a) values (4);
insert into test(a) values (5);
insert into test(a) values (6);
insert into test(a) values (7);
commit;
-- See that I have not analyzed the table. This will make use of RULE based optimizer
select * from test
where a > 0
and 1/a < .25;
and 1/a < .25
ERROR at line 3:
ORA-01476: divisor is equal to zero
-- Look at the query clause. Even though I specifically asked for records where a is positive
-- the evaluation path of rule based optimizer started at the bottom and as it evaluated the
-- first row with a=0, and caused an error.
-- Now look at the query below. I just re-arranged the conditions so that a > 0 is evaluated
-- first. As a result, the row with a=0 is ignored and the query executes without any problem.
select * from test
where 1/a < .25
and a > 0;
A
5
6
7
-- Now I analyze the table to create statistics. This will make the query use the
-- cost based optimizer (since optimizer goal is set to CHOOSE)
analyze table test compute statistics;
Table analyzed.
-- Now I issue the erring query. See it executes without any problem. This indicates that
-- the cost based optimizer was intelligent enough to evaluate the proper path instead of
-- looking only at the syntax.
select * from test
where a > 0
and 1/a < .25;
A
5
6
7
Does the above example seem familiar to your case? Even though you had the AND UPPER(PROJECTED_DELIVERY_DATE) NOT LIKE('NULL%') in your query, a record with PROJECTED_DELIVERY_DATE = 'NULL' was evaluated first and that caused the error.
Summary
1. Use dates for dates and strings for strings
2. Use cost based optimizer
Thanks
Suman

Photoshop CS4 for Mac What can cause the error: Plug-in failed to load - NO VERSION

With Photoshop CS4 for Mac I install the Kodak Plug-in DIGITAL GEM Airbrush Professional v2.1.0. When Photoshop starts I get an error:
Plug-ins that failed to load:
   DIGITAL GEM Airbrush Professional v2.1.0 NO VERSION - - from the file “DigitalAIRProv2.plugin”Flash:
   Connections
   Kuler
This plug-in works fine on my Mac Pro with Photoshop CS4 and OS 10.6 but fails on a new Macbook Air.
Reinstalling the plug-in did not make any difference. What can cause this error?

Thanks Chris, That was my first thought too but this same plug-in works on another installation of CS4. This is a Mac installation and the plug-in is Universal so it should work. Is there any other possible cause for this error? Would Photoshop refuse to load a plug-in in a low memory situation?
Also, The Flash extensions:Connections and Kuler are working OK, I just copied a little more out of the System Info report than I needed to.

Does anyone know what is causing this error please?

I recently ran sfc /scannow. It completed without giving me any notifications other than a number of:
"Files that are required for Windows to run properly must be copied to the DLL cache."
I was prompted to insert my XP SP3 installation CD and click on 'retry'.
The scan lasted for around 40 minutes then closed.......at no stage was I advised that it had been unsuccessful.
When I restarted my computer directly afterwards, half way through booting a 'please wait' appeared, as if something was taking place.
I thought all had gone well until I looked in Event Viewer later. There are approximately 90 Windows File Protection entries there mostly with the following information:
The system file ........ could not be copied into the DLL cache. The specific error code is 0x800b0100 [No signature was present in the subject]. This file is necessary to maintain system stability.
Does anyone know what causes that error and what I need to do to resolve it please?

Thank you very much for helping.
I have already tried to run the Microsoft Fix it from there for XP SP3, the os I have>>>>>>>"This troubleshooter does not apply to your system. We're sorry, but this troubleshooter is not compatible with your current system's configuration."
That was the only Microsoft information I could find on the specific error code I received: 0x800b0100......although the article applies to Microsoft Updates, I thought it might cover this too.
I have now run SFC three times, but errors still showed in Event Viewer.
I regularly run Chkdsk, last few times no errors.
I am actually having my computer checked for malware right now by specialist malware removers......all the malware scans I regularly run have given me the all clear for some time (McAfee Stinger, Hitman Pro, Malwarebytes - MBAM, TDSSKiller) but there has
been malware on my computer in the past, which was removed (not serious I was told).
In two of the scans the techs asked me to carry out there are
hundreds of dll cache entries under
Files/Folders - Created Within 30 Days on the date I run SFC.
I am waiting for this to be confirmed, but it looks to me as if SFC was actually successful to a large extent and the entries in Event Viewer are some failures (not all as I previously thought)?
I also have a serious problem in relation to Microsoft Updates (please see my other question).

[svn:fx-trunk] 11999: Fixed: ASC-3889 - Using setting in a for loop causes Verify error

Revision: 11999
Revision: 11999
Author:   [email protected]
Date:     2009-11-19 11:37:09 -0800 (Thu, 19 Nov 2009)
Log Message:
Fixed: ASC-3889 - Using setting in a for loop causes Verify error
Notes: emit pop after callstatic to a void function to balance the stack.
Reviewer: jodyer+
Testing: asc,tamarin,flex checkin tests
Ticket Links:
    http://bugs.adobe.com/jira/browse/ASC-3889
Modified Paths:
    flex/sdk/trunk/modules/asc/src/java/macromedia/asc/semantics/CodeGenerator.java

Blacklisting the ahci module does indeed get rid of the error. Would be nice to figure out why it was conflicting with the ahci module though.
I have not yet tried the 173xx drivers, but the latest drivers for the Quadro FX 580 are the 256.53 drivers according to nvidia.
Posted at the nV forums (http://www.nvnews.net/vbulletin/showthread.php?t=155282), so we'll see what they suggest.

[svn:fx-trunk] 11530: Fix ASC-3790 ( conditional expression in for loop causes verifier error) r=jodyer

Revision: 11530
Author:   [email protected]
Date:     2009-11-06 13:23:05 -0800 (Fri, 06 Nov 2009)
Log Message:
Fix ASC-3790 (conditional expression in for loop causes verifier error) r=jodyer
Ticket Links:
    http://bugs.adobe.com/jira/browse/ASC-3790
Modified Paths:
    flex/sdk/trunk/modules/asc/src/java/macromedia/asc/parser/ConditionalExpressionNode.java

Debug vs Standard causing HTTPRequest errors

Similar Messages

Maybe you are looking for