While `tp_print` was deprecated, Python 3.8+ uses this for
'tp_vectorcall_offset' which wasn't stated in the comment from
efd71aad4f.
Instead of suppressing clang-tidy, use preprocessor a check since
this properly represents the difference between Python versions.
Corrects incorrect usage of contraction for 'it is', when possessive 'its' was required.
Differential Revision: https://developer.blender.org/D9250
Reviewed by Campbell Barton
Define enum `eContextResult` and use its values for returns, instead of
just returning 1, 0, or -1 (and always having some comment that explains
what -1 means).
This also cleans up the mixup between returning `0` and `false`, and `1`
and `true`. An inconsistency was discovered during this cleanup, and
marked with `TODO(sybren)`. It's not fixed here, as it would consititute
a functional change.
The enum isn't used everywhere, as enums in C and C++ can have different
storage sizes. To prevent issues, callback functions are still declared
as returning`int`. To at least make things easier to understand for
humans, I marked those with `int /*eContextResult*/`.
This is a followup of D9090, and is intended to unify how context
callbacks return values. This will make it easier to extend the approach
in D9090 to those functions.
No functional changes.
Differential Revision: https://developer.blender.org/D9095
Printing an evaluated view layer would show:
Evaluated Scene 'Scene'
- Now __repr__ uses the __str__ fallback for evaluated data,
as done in other situations where we can't create a string that
would evaluate to the data.
- __str__ now shows when the data is evaluated.
- __str__ always includes the memory address
(which was previously only shown for structs without a name).
Mpving utils from idcode to idtype proved to be somewhat painful for
some reasons, but now all looks good.
Had to add a fake/empty shell for the special snowflake too,
`ID_LINK_PLACEHOLDER/INDEX_ID_NULL`...
This allows fast access to various arrays in the Python API.
Most notably, `image.pixels` can be accessed much more efficiently now.
**Benchmark**
Below are the results of a benchmark that compares different ways to
set/get all pixel values. I do the tests on 2048x2048 rgba images.
The benchmark tests the following dimensions:
- Byte vs. float per color channel
- Python list vs. numpy array containing floats
- `foreach_set` (new) vs. `image.pixels = ...` (old)
```
Pixel amount: 2048 * 2048 = 4.194.304
Byte buffer size: 16.8 mb
Float buffer size: 67.1 mb
Set pixel colors:
byte - new - list: 271 ms
byte - new - buffer: 29 ms
byte - old - list: 350 ms
byte - old - buffer: 2900 ms
float - new - list: 249 ms
float - new - buffer: 8 ms
float - old - list: 330 ms
float - old - buffer: 2880 ms
Get pixel colors:
byte - list: 128 ms
byte - buffer: 9 ms
float - list: 125 ms
float - buffer: 8 ms
```
**Observations**
The best set and get speed can be achieved with buffers and a float image,
at the cost of higher memory consumption. Furthermore, using buffers when
using `pixels = ...` is incredibly slow, because it is not optimized.
Optimizing this is possible, but might not be trivial (there were multiple
attempts afaik).
Float images are faster due to overhead introduced by the api for byte images.
If I profiled it correctly, a lot of time is spend in the `[0, 1] -> {0, ..., 255}`
conversion. The functions doing that conversion is `unit_float_to_uchar_clamp`.
While I have an idea on how it can be optimized, I do not know if it can be done
without changing its functionality slightly. Performance wise the best solution
would be to not do this conversion at all and accept byte input from the api
user directly, but that seems to be a more involved task as well.
Differential Revision: https://developer.blender.org/D7053
Reviewers: JacquesLucke, mont29
'Private' can be a rather confusing term, especially when considering
its meaning in programming languages.
So now root node trees and master collections are 'embedded' IDs
instead.
Logic for registering and checking properties of registrable classes was
broken, allowing to ignore some errors.
Recent fix rBeb798de101a `broke` the result of the
pyapi_idprop_datablock test, because previously that test would fail
(i.e. suceed, as it is an 'expected to break test') for a reason it was
not designed to check.
This is the problem with that kind of tests - you cannot really check
that they are failing on the expected reason(s)...
This commit solves the bug itself (code was broken when real_id owner of
the private data ID could not be found), and generates a more sensible
representation for all evaluated IDs, makes no sense to display them as
being part of `bpy.data....`!
