sofar: Upcoming changes in SOFA standard and conventions
评估
这个 Issue 还没有评估数据。
描述
The standard convention will get an update soon, which mainly enforces a stricter handling of variable shapes. Before the update a variable could have shapes of different dimenstions, e.g., IC or ECM (I: single dimension, C: 3 coordinate points, E: number of emitters, M: number of measurements). After the update this is not allowed anymore and shapes would have to be ICI or ECM for the example above.
This leads to many cases where variables can have leading and trailing single dimensions, e.g., ICI or even II, which might be a little annoying. Currently trailing single dimensions are squeezed away as described in the notes of sofa_read. This can change as it is, but it shoule be tested if
- multidimensional arrays containing only single dimensions, e.g., an array of
shape=(1,1)are squeezed as is currently already done for arrays of shape(1, ) - String Variables of new shapes (see below) are correctly handled.
- Make sure that all possible shapes that a sofa variable can have are of the same dimension (verification for convention)
- Revisit checking the shapes (at the moment trailing single dimensions are added until the dimension equals 4). Is this still fine.?- [ ] Requires https://github.com/pyfar/sofa_conventions/tree/SOFAtoolbox_develop_conventions
- Cherry pick changes in rules/upgrade.json
- Discard changes in conventions folder (chaged due to manual copy of new convensions from SOFAtoolbox devlop branch)
- Properly update conventions from sofaconventions.org/conventions using
sofar.update_conventions - run tests again
- update submodule in sofar
Questions:
- SOFAtoolbox and sofaconventions.org/conventions list GeneralTF_1.0 and GeneralTF_2.0 on the develop branch. Shouldn't the older one be deprecated?
- Should free field directivity 1.1 be increased to 1.2 because the name 'Description' changed to 'Descriptions'?
Changes in shapes (dimensions) of updated conventions:
- EmitterDescriptions (string) to shapes (MSI, ESI, MES)
- from (MS, ES, MES)
- FreeFieldDirectivityTF_1.1
- from (ES, ESM) - wild
- SingleRoomMIMOSRIR_1.0
- from (MS, ES, MES)
- EmitterPosition (double) to shapes (eCI, ECM)
- from (eC, eCM)
- AnnotatedEmitterAudio_0.2
- AnnotatedReceiverAudio_0.2
- FreeFieldDirectivityTF_1.1
- GeneralTF_2.0
- from (eC, eCM)
- EmitterPosition (double) to shapes (ICI, ECI, ECM)
- from (IC, ECI, ECM)
- FreeFieldHRIR_1.0
- FreeFieldHRTF_1.0
- SingleRoomMIMOSRIR_1.0
- from (IC, EC, ECM)
- GeneralFIR-E_2.0
- GeneralTF-E_1.0
- from (IC, ECI, ECM)
- ReceiverDescriptions (string) to shapes (MSI, RSI, MRS)
- from shapes (RS, RSM) - wild
- SingleRoomMIMOSRIR_1.0
- from shapes (RS, RSM) - wild
- Receiver position (double) to shapes (RCI, RCM)
- from (RC, RCM)
- AnnotatedReceiverAudio_0.2
- AnnotatedEmitterAudio_0.2
- from (RC, RCM)
- Receiver position (double) to shapes (ICI, RCI, RCM)
- from (IC, RC, RCM)
- FreeFieldDirectivityTF_1.1
- GeneralFIR-E_2.0
- GeneralFIR_1.0
- GeneralSOS_1.0
- GeneralTF-E_1.0
- GeneralTF_2.0
- from (IC, RCI, RCM)
- SingleRoomMIMOSRIR_1.0
- from (IC, RC, RCM)
- Room Volume (double) to shapes (II, MI)
- SingleRoomSRIR_1.0
- SingleRoomSRIR_1.1
- SingleRoomMIMOSRIR_1.0
Other changes:
- FreeFieldDirectivityTF_1.1
- Renamed Description to Descriptions
- SimpleHeadphoneIR_1.1 added 's' to end of string variables:
- SourceManufacturer renamed to SourceManufacturers
- SourceModel renamed to SourceModels
- ReceiverDescription renamed to ReceiverDescriptions
- EmitterDescription renamed to EmitterDescriptions
- SingleRoomSRIR 1.1 added variables
- ReceiverDescriptions (string) of shape MSI, RSI, MRS
- EmitterDescriptions (string) of shape MSI, ESI, MES
- MeasurementDate (double) of shape M (???)
- Changes of the default coordinate system (not logged here)
- Updated comments (not logged here)
- Removed mandatory flag for some fields (not logged here)
- Changed default values of some coordinates (not logged here)
- 主要语言
- Python
- 星标
- 37
- 派生
- 4
- PR 合并指标
- 30 天内没有已合并 PR
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
pyfar/sofar 的其他 Issue
-
bug
难度 2/5 1-3 小时 新手友好度 68/100
-
难度 4/5 3-5 天 新手友好度 45/100
-
invalid SOFA files for UTF-8 strings可能已有人在做 @f-brinkmann 于 28 天前认领。 未关闭bug
难度 3/5 1-2 天 新手友好度 55/100
-
Move sofar-related pyfar functionality to sofar可能重新可做 @f-brinkmann 于 92 天前认领,目前没有进行中的 PR。 未关闭feature
-
[MAINT] Remove restrcition for Numpy 2.5可能已有人在做 @f-brinkmann 于 28 天前认领。 未关闭
难度 3/5 1-2 天 新手友好度 56/100
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 85/100
Kludex/python-multipart#375 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 65/100
维护者通常 3 天内回复
-
enhancement
难度 2/5 1-3 小时 新手友好度 74/100
维护者通常 11 天内回复
-
难度 2/5 1-3 小时 新手友好度 80/100
FuRongJun-1999/dsh-memory#1000 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 68/100
gcivil-nyu-org/team5-mon-fall26#154 ·
维护者通常 1 天内回复