I was thinking "column" fields would make sense but it clashes with the 
internal concept of columns (Col types), and is only name-appropriate for a 
database backend. I'm not sure these are actually problems worth 
considering though.

On Friday, 1 August 2014 01:07:21 UTC+10, Collin Anderson wrote:
>
> This is awesome. I've always wished that django.contrib.admin used only 
> documented django APIs. It's unfair to 3rd party apps. :) I'm also thankful 
> I won't need to import bitflags just to use this API.
>  
>
>> The Options API is at the core of Django, it enables introspection of 
>> Django Models with the rest of the system. This enables lookups, queries, 
>> forms, admin to understand the capabilities of every model. The Options API 
>> is hidden under the _meta attribute of each model class. Options has always 
>> been a private API, but Django developers have always been using it in 
>> their own projects, in a non-official way.  As part of my SoC project, I am 
>> exposing the new API for public use. As we are also formalizing a new 
>> API that has always existed, It is important to revisit the terminology and 
>> document it.
>
>  
> The word "Options" doesn't make much sense to me. I realize that both 
> "meta options" and the fields are stored on the same object. I think "Model 
> Meta API" or "Model Schema API" or "Model Fields API" or "Model Class API" 
> or "Model Definition API" or "Model Structure API" would make more sense.
>
> In the current docs, we're saying that meta options are "anything that's 
> not a field", and it may make sense to keep "options" to referring to that.
> https://docs.djangoproject.com/en/dev/topics/db/models/#meta-options
>
> I prefer "concrete" to "data" as it makes the parallel with "virtual" more 
>> obvious.
>
>
> Does this include m2m? We could call 'em "column" fields, as in fields 
> that actually map to a column on the database table.
>
>>
>>    - "related_objects" > "reverse_rel"
>>    - "related_m2m" > "reverse_m2m"
>>    - What about reverse o2o, do they currently fall under 
>>    "related_objects"?
>>    - The main issue I have with "related" is that it doesn't provide a 
>>    sense of direction, after all both sides of a FK have related objects 
>>    waiting on the other end. I also like the symmetry between "m2m" and 
>>    "reverse_m2m".
>>    - Django isn't always consistent (and sometime actually wrong) in 
>>    naming relations, especially indjango/db/models/fields/related.py. 
>>    Now that we start documenting all those things, I see value in getting it 
>>    right.
>>
>> Maybe also "hidden" -> "reverse_hidden" to be consistent?
>
> Maybe export_map -> return_names. or return_names_map.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/django-developers.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/django-developers/f2e17f7b-8901-46fc-8e78-d560f4e4e40b%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to