Refactor of Transformation Definitions in constants, add reverse transformations #133
Conversation
…ing of variables in other modules and tests Signed-off-by: SGV-Geodesy <56060078+SGV-Geodesy@users.noreply.github.com>
geodepy/tests/test_transform.py
Outdated
| mga94_to_mga2020, | ||
| mga2020_to_mga94) | ||
| from geodepy.constants import itrf14togda20, gda94_to_gda2020 | ||
| from geodepy.constants import itrf2014_to_gda2020, gda1994_to_gda2020 |
There was a problem hiding this comment.
I think it best use stick with common usage. So, GDA94 but GDA2020. Similarly, the ITRF versions pre-2000 have only two digit years (ITRF92), while those from 2000 on have four digits (ITRF2014).
geodepy/transform.py
Outdated
|
|
||
|
|
||
| def atrf2014_to_gda2020(x, y, z, epoch_from, vcv=None): | ||
| def conform_atrf2014_to_gda2020(x, y, z, epoch_from, vcv=None): |
There was a problem hiding this comment.
transform_atrf2014_to_gda2020 may be a clearer function name
There was a problem hiding this comment.
@harry093 good idea, I've also extended this to add transform_ prefix to all datum-to-datum functions in the transform module for clarity
harry093
left a comment
harry093
left a comment
There was a problem hiding this comment.
@BatchelorJ thanks for this. I am open to being convinced otherwise, but I think we should follow the standard names - as frustrating as they are. See my comment in test_transform.py
Thanks @harry093, I thought I'd throw this code up to see what we thought about it so thanks for the feedback. I'm happy to stick to the rule of years <2000 using 2-digit suffixes while >=2000 using 4-digit suffixes as long as it's consistently applied. Most in the geodetic and survey community understand this convention so we should be able to get away with it. |
…datum-specific transformation functions with transform_ prefix Signed-off-by: SGV-Geodesy <56060078+SGV-Geodesy@users.noreply.github.com>
geodepy/constants.py
Outdated
| sc=-1.854, | ||
| rx=0.008, ry=-0.557, rz=-0.178) | ||
|
|
||
| gda19_to_itrf2008 = -itrf2008_to_gda94 |
There was a problem hiding this comment.
Should be gda94_to_itrf2008? Same typo made below as well
There was a problem hiding this comment.
@harry093 well spotted! All fixed up in latest commit
harry093
left a comment
harry093
left a comment
There was a problem hiding this comment.
@BatchelorJ looks great mate. I think there's a small typo in the constants module and once that's fixed we are good to merge
Signed-off-by: SGV-Geodesy <56060078+SGV-Geodesy@users.noreply.github.com>
harry093
left a comment
harry093
left a comment
There was a problem hiding this comment.
@BatchelorJ all good! Thanks
This PR standardizes the naming of transformation definitions in the constants module (resolves #129) and adds reverse transformations utilizing the same naming convention (resolves #127). Several other modules and tests that use some of these transformation objects have been refactored, and the names of functions that convert GDA2020 <-> ATRF2014 have been changed so that they have different names to the transformation objects themselves.