Improve affine.rst: document unit behavior, limitations, and examples (#3902) #4630
Conversation
rouault
left a comment
rouault
left a comment
There was a problem hiding this comment.
I strongly suspect you have used some LLM/AI tool to generate that. There's a lot of uselessly verbose and inaccurate statements.
| When :code:`+proj=affine` is used on angular coordinates (longitude/latitude): | ||
|
|
||
| * **Input angular values are interpreted in radians** | ||
| * **Output angular values are expressed in degrees** | ||
|
|
There was a problem hiding this comment.
This is not true. affine is not unit aware at all.
The unit conversion is done here by cs2cs and we could argue this is a bug of cs2cs to accept +proj=affine. Normal manual usage of +proj=affine should be done with the cct utility
| Example of valid usage inside a pipeline: | ||
|
|
||
| proj pipeline +step +proj=affine +xoff=1 | ||
|
|
There was a problem hiding this comment.
the proj utility isn't appropriate to execute pipelines. That should be done with cct
| **3. Recommended usage** | ||
|
|
||
| The affine transformation is generally intended for: | ||
|
|
||
| * Modifying projected coordinates | ||
| * Applying small adjustments inside pipelines | ||
| * Implementing EPSG parametric and similarity transforms | ||
|
|
||
| When applied to geographic coordinates, users must explicitly handle/unit-convert | ||
| angles as needed. | ||
|
|
|
Thank you for the clarification, and apologies for the inaccuracies. |
2 participants
This PR addresses issue #3902 by expanding the documentation for +proj=affine.
Summary of changes
proj +proj=affinecannot be used directly and will raise an error.Checklist