Paging of api-results

Development - Développement
1

I run into some really strange results when I call the API using this (semiofficial?) dhis2 python wrapper

I expected that an unpaged get would give me all the data:
api.get('organisationUnits', params={'paging': False, 'fields': 'id,code,name'})

However (using play.dhis2.org/40.2.2) - the returned JSON string ends like this:

{'code': 'OU_204876', 'name': 'Dia CHP', 'id': 'qHBTf9A89xW'}, {'code': 'OU_222737', 'name': 'Diamei MCHP', 'id':

Note that there is no value for ‘id’ here.

However, if I ask for a large dataset
api.get('organisationUnits', params={'pagesize': 1000, 'fields': 'id,name,code'})

then I get all orgUnites as well as a propper ending of the string:
{'code': 'OU_233393', 'name': 'Baoma MCHP', 'id': 't52CJEyLhch'}]}

Is this on purpose or should it be noted in the docs that you should always page unless your resultset is really small?

2

Hi,

Perhaps this is an issue with the python library, do you see the same issue when you call the API directly? https://dhis2.stoplight.io/docs/dhis2

1 Like
3

What a great resource Stoplight seem to be @Rene: Thanks! I was not aware of it - it seems to be a great supplement (for some a replacement?) for the core docs

I have now tried to call the api from a browser, and indeed there is no truncating of the JSON string there.

So this seem to belong somewhere in python-land …

1 Like
4

Great to hear it’s not the API!

Stoplight is a supplement to the API docs. We do link it from the Developer portal but we could do a better job at sharing the resource!

Perhaps you could submit a PR to the Python library to fix the issue, or raise an issue on the repository page.

A second option would be to use the OpenAPI spec of DHIS2 to generate a python library directly, as the repository does seem to be out of date for a few years

3 Likes