Closed nielsdg closed 8 years ago
As the associations.json file is used in multiple applications (FK-Enrolment, Gandalf) and perhaps also some FK-code, please notify developers before changes are happening.
2015-12-07 23:10 GMT+01:00 Niels De Graef notifications@github.com:
The format for associations seems to be picked at random and is at the brink of being unusable in this state, as it prevents activities from being linked to their respective associations in any way.
Some issues:
- all_activities.json http://student.ugent.be/hydra/api/1.0/all_activities.json uses internal_name, while associations.json http://student.ugent.be/hydra/api/1.0/associations.json uses internalName
- internal names of associations don't match, e.g. FK vs. FKCENTRAAL, or GRAS vs JGroen
- Sometimes full_name is used while displayName equals null, sometimes it's the reverse, and in other cases both are used.
Tagging @mlumingu https://github.com/mlumingu for a follow-up.
— Reply to this email directly or view it on GitHub https://github.com/ZeusWPI/hydra/issues/224.
I think this is also related to our email "Zoneit/Hydra API [GH-0188]" dd wo 1/04/2015 21:37 (michael lumingu, Michiel Van den Berghe; hydra; Tom Naessens), right?
-alex-
On Tue, Dec 8, 2015 at 10:43 AM, Tom Naessens notifications@github.com wrote:
As the associations.json file is used in multiple applications (FK-Enrolment, Gandalf) and perhaps also some FK-code, please notify developers before changes are happening.
2015-12-07 23:10 GMT+01:00 Niels De Graef notifications@github.com:
The format for associations seems to be picked at random and is at the brink of being unusable in this state, as it prevents activities from being linked to their respective associations in any way.
Some issues:
- all_activities.json http://student.ugent.be/hydra/api/1.0/all_activities.json uses internal_name, while associations.json http://student.ugent.be/hydra/api/1.0/associations.json uses internalName
- internal names of associations don't match, e.g. FK vs. FKCENTRAAL, or GRAS vs JGroen
- Sometimes full_name is used while displayName equals null, sometimes it's the reverse, and in other cases both are used.
Tagging @mlumingu https://github.com/mlumingu for a follow-up.
— Reply to this email directly or view it on GitHub https://github.com/ZeusWPI/hydra/issues/224.
— Reply to this email directly or view it on GitHub https://github.com/ZeusWPI/hydra/issues/224#issuecomment-162832195.
I structured this api based on the example json file I was given.
Thanks for the quick reply. Will test as soon as possible!
Can someone confirm what the impact of this API work will be for Zone-it?
@axd1967 The upgrade is backwards compatible, so there shouldn't be any impact for Zone-it.
@Silox
Just to clarify: an upgrade can be backward compatible, but in effect be "dead meat", in other words abandoned; the real "meat" is in the upgrade. Is this the case here?
Have there been evolutions that Zone-it can benefit from? Because if that's the case, we had an agreement to discuss this together.
Hey Alex, I can’t know if there have been any evolutions that Zone-it can benefit from as I don’t know what is already included in Zone-it or how things are handled there. The changes are described in @mlumingu’s post, so you have to check out for yourself if these changes benifit Zone-it.
Lastly, all discussions about the API have been taken in a public place (this topic) where you have replied to, and thus have been included in the conversation (which we will continue to do in the future).
The API is also still publicly available at http://student.ugent.be/hydra/api/ to anyone who wants to use it.
The format for associations seems to be picked at random and is at the brink of being unusable in this state, as it prevents activities from being linked to their respective associations in any way.
Some issues:
internal_name
, while associations.json usesinternalName
FK
vs.FKCENTRAAL
, orGRAS
vsJGroen
full_name
is used whiledisplay_name
equals null, sometimes it's the reverse, and in other cases both are used in associations.json.Tagging @mlumingu for a follow-up.