add copy to move_and_delete #233
Conversation
| @@ -5,6 +5,21 @@ | |||
| # It looks if the result of a task exists in the project catalog before executing the task. | |||
| # The workflow does NOT automatically remove intermediate files. You might run out of space. | |||
|
|
|||
|
|
|||
There was a problem hiding this comment.
@vindelico Does this text answer the questions you had about the config ?
Is there more that should be included ?
As I said in #232, adding every possible arguments to functions in sections that are automatically passed seems overkill to me. You don't normally pass argument with just their defaults...
| @@ -193,7 +200,7 @@ def _infer_freq_from_meta(ind): | |||
| if (out[c].attrs != ds[c].attrs) and (out[c].sizes == ds[c].sizes): | |||
| out[c].attrs = ds[c].attrs | |||
|
|
|||
| if "time" in out.dims: | |||
| if restrict_years and "time" in out.dims: | |||
There was a problem hiding this comment.
Could you add a test? What happens with something like AS-JUL ?
There was a problem hiding this comment.
I added a test with as-jul to show.
Basically, with xclim, you have a nan in the first and the last position.
The first time is the year before the start of the input.
The last time is the same year as the end of the input.
With restrict_years= False, we keep the time steps for the 2 nans.
With restrict_years= True, we cut the first nan.
Co-authored-by: RondeauG <[email protected]>
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
| @@ -1,14 +1,43 @@ | |||
| # example of a config for the workflow | |||
| # Example of a config file for the workflow | |||
There was a problem hiding this comment.
@juliettelavoie
I added quite some detail here.
My suggestion would be to comment the first task similarly detailed. After that, subsequent tasks all follow a similar scheme of iteration loops, anyway.
|
For comments on template docs, see PR#235. |
Fine with me! |
Pull Request Checklist:
number) and pull request (:pull:number) has been added.What kind of change does this PR introduce?
*Many small change to help my workflow work better.
copytomove_and_deleteis useful for taking everything in a directory and adding files inside an already existing structure with other files.var_as_strtoget_cat_attrsis useful to automatically write paths without parenthesis.** Before:
('tasmin',)_CanESM5.zarr**After: tasmin_CanESM5.zarr`
restrict_yearstocompute_indicatorsis useful to reproduce exactly xclim's output. I want to redo exactly somehting that was done pre-xscen. The default is still to cut the nans, but this makes it possible to keep them.Does this PR introduce a breaking change?
no ( if you had a workflow writting a path with the tuple, it will be broken but I doubt anyone did that).
Other information: