Move database backends out of the Django's core sounds great.

Em domingo, 25 de novembro de 2018 15:54:21 UTC-2, Tim Graham escreveu:
> I can't find a past discussion specific to Oracle, but it's not a new 
> proposal. See 
> for "Moving database backends out of the core."
> I think removing Oracle from core would only increase the maintenance 
> burden. Since Oracle has edge cases, it's useful to test those along with 
> new Django features. If the Oracle backend is in a separate repo, then 
> adding new features will often require commits to two repositories and I 
> don't know how we would run the tests with pull request X for Django and 
> pull request Y for the Oracle backend. Then we also have to release the 
> Oracle backend separately.
> isn't reporting any Oracle failures on master. If you've 
> found an issue, please open a ticket with details.
> We don't run the Oracle tests with pull requests because they take about 
> an hour, while other databases take about 10 minutes. It hasn't been 
> difficult to identify which pull requests require running the tests on 
> Oracle and to trigger that build with the trigger phrase.
> On Sunday, November 25, 2018 at 5:21:02 AM UTC-5, Johannes Hoppe wrote:
>> Hi there,
>> I have recently refactored some bits in the database backend and came to 
>> realize that a lot of the complexity in there comes from the poor 
>> implementation of the Oracle backend.
>> Fun fact, did you know that Oracle tests don't run by default and that 
>> the current master, fails on oracle ;)
>> Anyhow, I want to come to a conclusion about the following matter:
>> Should we remove the Oracle database backend from Django core in the 3.0 
>> release?
>> Here are a couple of reasons, why I believe this to be a good idea:
>>    - License
>>    - Oracle is  Proprietary software
>>    - Money
>>       - Oracle is not a sponsor of the Django Foundation, but makes 40bn 
>>       in revenue
>>    - Technical:
>>       - Oracle does not support may features
>>       - due to its lack of features, a lot of edge case handling to the 
>>       base database backend which drives overall complexity
>>    - Development:
>>       - Oracle does not run in the regular CI suite, in fact master is 
>>       broken right now
>>       - entrance barrier for first time contributors is high
>>          - one needs to accept a non open source license
>>          - register with oracle
>>          - go through a very complex setup process
>> Of course there are some users who use Oracle and I don't want to keep 
>> them hanging. I simply believe the database backend should be developed 
>> separately from Django.
>> This could even be helpful for the Oracle community. Since oracle is 
>> enterprise only, they usually looks for longer support cycles than what 
>> Django want's to offer.
>> Ok, I made my case, I am curious, what do you guys think?
>> Best
>> -Joe

You received this message because you are subscribed to the Google Groups 
"Django developers  (Contributions to Django itself)" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
To post to this group, send email to
Visit this group at
To view this discussion on the web visit
For more options, visit

Reply via email to