| Author |
Message
|
| Vitor |
 Posted: Fri May 18, 2012 6:36 am Post subject: |
 |
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| mqjeff wrote: |
| kimbert wrote: |
| I have a theory... |
It could be bunnies? |
Once More, With Feeling. 
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
 |
| kimbert |
| Posted: Tue May 22, 2012 12:27 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
@msynder: Any progress on this since last week? My current assumptions are:
- the nil handling problem was down to the 'xsd' prefix being incorrectly applied to the 'nillable' attribute
- the parsing of element2 as 'elem2' instead of 'elem2 ' is probably down to the trimming options on the element. I'm fairly sure that it is not treating the space character as a delimiter.
- I noticed that element3 was correctly parsed as a nil value because it was empty. I wonder whether the incorrect parsing of element4 was caused by the separatorPolicy='suppressed' setting. Can you confirm that element4 is still parsed as a nil value when separatorPolicy is changed to 'suppressedAtEnd'? |
|
| Back to top |
|
|
| msnyder |
| Posted: Tue May 22, 2012 11:10 am Post subject: |
|
|
Apprentice
Joined: 24 May 2002
Posts: 32
Location: Jacksonville, FL
|
I opened a PMR (03659 442 000) last week that was forwarded to L3 and haven't had anyone contact me yet on it. The only way I've been able to get this to parse correctly is to change the spaces to characters. The suggestion to change the global suppressPolicy didn't have any effect.
I've also run into another parsing issue where the 2nd element of an array is being identified as empty by the parser even though it contains a value. I don't have an example I can share on this forum, but steps to recreate as well as associated files are in the PMR.
Any idea when the 1st fix pack will be available? |
|
| Back to top |
|
|
| smdavies99 |
| Posted: Tue May 22, 2012 11:28 am Post subject: |
|
|
Jedi Council
Joined: 10 Feb 2003
Posts: 6076
Location: Somewhere over the Rainbow this side of Never-never land.
|
| msnyder wrote: |
Any idea when the 1st fix pack will be available? |
If what you have found is a new defect then there is (AFAIK) no chance that it will be in the problems addressed by 8.0.0.1 unless FP1 is delayed by several months.
IF you are given a fix then make sure that you get versions of the fix FOR ALL FP levels you are using.
_________________
WMQ User since 1999
MQSI/WBI/WMB/'Thingy' User since 2002
Linux user since 1995
Every time you reinvent the wheel the more square it gets (anon). If in doubt think and investigate before you ask silly questions. |
|
| Back to top |
|
|
| kimbert |
| Posted: Tue May 22, 2012 1:35 pm Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
OK - I reproduced this. There is a defect in the parser, so you should expect an update to your PMR soon.
Meanwhile, I have a workaround for you. I know this works, because I have just tested it out.
- Go to the default format for your xsd ( same place as for separatorPolicy).
- Find the property 'Trim Kind' under the section 'Text Content'
- Change its value from 'padChar' to 'none'.
- Save your xsd and re-test. |
|
| Back to top |
|
|
| msnyder |
| Posted: Thu May 24, 2012 7:33 am Post subject: |
|
|
Apprentice
Joined: 24 May 2002
Posts: 32
Location: Jacksonville, FL
|
For the benefit of others . . .
Although this did help to properly parse the example I provided for this forum, it does not work for the real message model and associated data. The element padded with spaces does get correctly parsed. However, the parser identifies a subsequent element as empty when it contains a value.
This is similar to another issue being encountered with elements in an array being identified as empty when they contain values. |
|
| Back to top |
|
|
| gene |
| Posted: Fri Sep 28, 2012 5:20 am Post subject: |
|
|
Newbie
Joined: 21 Aug 2012
Posts: 6
|
I got a little lost in this thread since it seems like it might have slightly changed topics halfway through but I have a question similar to the first one here.
| Quote: |
This is a known defect in the version of IBM DFDL that ships with WMB v8.0. If you open a PMR then you will get the fix as soon as it is available.
In the mean time, fjb_saper's suggestion is probably your best option. You will need:
- xs:nillable="true"
- dfdl:nilKind="nilLiteralValue"
- dfdl:nilValue="%ES;"
which will cause any zero-length element to be put into the message tree with a value of NULL. |
If I have the same situation that is described and I am currently on FP1 of V8 is this still the correct implementation? I have a set amount of pipe delimited fields per record that I am reading in and want || to be recognized as a null field. Similary I have a situation where I want the serialized output to place || in a file for a field that is null. Thanks for the help! |
|
| Back to top |
|
|
| kimbert |
| Posted: Fri Sep 28, 2012 6:27 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
| As far as I'm aware, all of the problems mentioned in this thread ( including the last one ) were eventually dealt with via PMRs. If you open a PMR and quote this thread as background information then you should get the fixes that you require. |
|
| Back to top |
|
|
|
|
