Closed henryk closed 1 year ago
It should be enough to drop the XSD file, as long as the new version does not add any required fields or removes any fields. In this case, we'd need code changes to adjust the content. However, a simple test case should make that clear.
comdirect needs pain.001.003.03
, and yes, I fixed it by adding https://github.com/hbci4j/hbci4java/blob/master/src/main/resources/pain.001.003.03.xsd in https://github.com/raphaelm/python-sepaxml/tree/master/sepaxml/schemas to make SEPA transfer work.
Can you just add it to the package or are there some legal issues?
Thank you!
I think it's fine, feel free to create a PR
Ok, I checked that a bit more in detail, the only difference between pain.001.001.03.xsd and pain.001.003.03.xsd is the naming of complexType ServiceLevel (and the occurences). I the old version pain.001.003.03 it is called ServiceLevelSEPA.
As I cannot find it in https://www.iso20022.org/catalogue-messages/iso-20022-messages-archive?search=pain.001.003.03 , your pain.001.001.03.xsd is in big parts different to pain.001.001.03.xsd from hbci4java and pain.001.003.03 seems to be just an old standard - I just copy your pain.001.001.03.xsd to pain.001.003.03.xsd und set the namespace. For comdirect that works, too.
@raphaelm If I understand it correctly the library is missing the latest pain
that can be downloaded at https://www.iso20022.org/iso-20022-message-definitions?search=pain
because BIC is no longer required since 2016.
ref: https://en.wikipedia.org/wiki/ISO_9362
I've added a pain.001.003.03 I found online to the repo now, but I continue to be confused since that version does not exist at the link supplied by @1oglop1
I continue to be confused since that version does not exist at the link supplied by @1oglop1
As written above - I think it is deprecated (and just still used by some banks)... Maybe that's why they don't list it anymore. But still confusing, that 001.001.03 is the current version and 001.003.03 is the old version.
@raphaelm There is a possibility that I was referencing pain.001.001.XX
- it took me a couple of days to understand the versioning scheme. Of the whole message set and realize that each message is versioned separately from the message set.
I wish there was a simple description of their versioning scheme. Here is my understanding pain.AAA.BBB.CC
. A
- message type, B
message sub-type, C
- version. But I could be worng.
This is all very confusing. I've cleaned up versions recently and now added a PR to add the newest ones #59
What is the primary obstacle to increasing the set of supported SEPA PAIN versions? For example https://github.com/raphaelm/python-fints/blob/18e361ab2c5beeddf126a601bfcc7753cf84efec/fints/client.py#L732 may choose
pain.001.003.03
which I guess worked before schema validation was introduced, but fails now, because that schema is not inschemas/
. Is it enough to drop the XSD file there, or are there more required steps?