All the new fluid simulation parameters need to be added the according structs (i.e. domain settings, flow settings).
Details
Details
- Reviewers
Sergey Sharybin (sergey) Nils Thuerey (n_t) Georg Kohl (GeorgKohl) Bastien Montagne (mont29) - Maniphest Tasks
- T59995: Mantaflow Review
- Commits
- rB9064655a70c6: Mantaflow [Part 10]: Updated fluid modifier settings
rB0be266031393: Mantaflow: Updates to variable names in smoke domain settings
rBe001c3d534c4: Merge branch 'rnaSmoke' into smokeTypes
rBf33519945cdd: Merge branch 'rnaSmoke' into smokeTypes
rB06fd3081030b: Mantaflow: Updated types for outflow
rB85e757257ef8: Mantaflow: Updated fluid modifier settings
Diff Detail
Diff Detail
- Repository
- rB Blender
Event Timeline
Comment Actions
Once again, I do not really see the point of re-using existing smoke data structures here, given the amount of changes, would rather see new data structures, with new modifier type, etc.
Comment Actions
@Bastien Montagne (mont29) This is probably one of the diffs that required only smaller changes:
It's basically everything we had in DNA_smoke_types.h but now with Manta naming, plus a couple of new parameters to account for liquids.
| source/blender/makesdna/DNA_manta_types.h | ||
|---|---|---|
| 127–128 ↗ | (On Diff #19924) | Use anonymous enum. |
| 214 ↗ | (On Diff #19924) | This structure is either split entirely, or to be refactored in a way that it is clear what is a setting (aka, what user actually specifies in the interface and what is saved to disk) and what is a runtime-only fields. This actually applies to other structures in this file. |
| 478 ↗ | (On Diff #19924) | temp is the worst variable name. |
Comment Actions
See a3328a8699ad for changes.
| source/blender/makesdna/DNA_manta_types.h | ||
|---|---|---|
| 214 ↗ | (On Diff #19924) | The order of items in the structs was supposed to give an indication to this. I have added some comments to clarify which fields are being used internally and which are for the user to define. Would this be sufficient? |
| 478 ↗ | (On Diff #19924) | Oh yes .. otherwise big fire hazard! |
| source/blender/makesdna/DNA_manta_types.h | ||
|---|---|---|
| 214 ↗ | (On Diff #19924) | In other areas of DNA the convention is to have explicit field. See Object_Runtime as an example, |
Comment Actions
Discussed here in person.
The runtime thing is already "violated" in SmokeDomainSettings. Bottom line: nice to move those to a dedicated struct, but can happen as an on-going improvement after the merge (since that is not intrusive for users).
There is also discussion here for the DNA/RNA naming, but don't think it helps stopping EVERY patch with something we know we are doing. After that topic is addressed there is nothing i can add here.