| Author |
Message
|
| Dharamteja |
 Posted: Wed Jun 27, 2018 12:53 pm Post subject: DFDL Parsing characters with accent marks |
 |
|
Newbie
Joined: 27 Jun 2018
Posts: 2
|
Hi All ,
I'm converting data from pipe delimited to Fixed length but my input data contains some accent mark characters due to that I'm not getting fixed length properly
could you please tell me which Encoding(code page) should I use in DFDL property , I have tried with ISO-8859-1 but no use and I have tried with UTF-8 also but accent characters are changed to another character , my intention is I have to parse the entire data with out any length issue please help me  |
|
| Back to top |
|
 |
| rekarm01 |
| Posted: Wed Jun 27, 2018 5:06 pm Post subject: Re: DFDL Parsing characters with accent marks |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 1415
|
| Dharamteja wrote: |
| I'm converting data from pipe delimited to Fixed length |
Fixed length what? Bits? Bytes? Characters? Something else? "Fixed length" doesn't mean much without also mentioning fixed length units.
| Dharamteja wrote: |
| could you please tell me which Encoding(code page) should I use in DFDL property |
That partly depends on what the receiving application(s) will tolerate. But if the DFDL schema indicates (or implies) that lengthUnits is 'bytes', then it's probably better to stick with a single-byte character set, such as ISO 8859-1.
| Dharamteja wrote: |
| I have tried with ISO-8859-1 but no use |
Tried how? What does "no use" mean? More details would help here.
| Dharamteja wrote: |
| and I have tried with UTF-8 also but accent characters are changed to another character |
Changed how? Are they changed within the message, or is whatever display tool used to view the contents just not displaying them correctly? |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Wed Jun 27, 2018 8:09 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20768
Location: LI,NY
|
Did you try using the InputRoot.Properties.CodedCharSetId for the output?
This way no conversion should be necessary for the characters in question? 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| timber |
| Posted: Thu Jun 28, 2018 2:58 am Post subject: |
|
|
Grand Master
Joined: 25 Aug 2015
Posts: 1292
|
| Quote: |
| I'm converting data from pipe delimited to Fixed length but my input data contains some accent mark characters |
You are not thinking clearly about the problem.
Your message flow needs to map the input data to the output data. Not just the field names but also the *values*. Every character that could possibly occur in the input data needs to be mapped to the output, or else trigger an error.
You have told us almost nothing about your output format. What length units are the 'fixed length' fields using? What character encodings will the receiving application accept? If you send multi-byte characters as part of a fixed-number-of-bytes field, will the receiving application be able to handle them? |
|
| Back to top |
|
|
| Dharamteja |
| Posted: Tue Jul 03, 2018 6:30 am Post subject: |
|
|
Newbie
Joined: 27 Jun 2018
Posts: 2
|
Thanks for your response , My issue got resolved
I have changed UTF-8 encoding in Input dfdl (pipe-delimited) and same was updated in output dfdl(fixed length ) as well and length units I changed from bytes to chars then I got length properly
I have parsed same data (blob) to another flow to insert into AS400 db but I parsed them as blob and converting into chars with ccsid 1208 and I didn't mention encoding then those accent characters also inserting into db as it is (not converted into another char )  |
|
| Back to top |
|
|
|
|
