airGR issueshttps://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues2021-08-04T14:02:11+02:00https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/62Embed Oudin's Fortran code2021-08-04T14:02:11+02:00Delaigue OlivierEmbed Oudin's Fortran codeThe [PREMHYCE](https://gitlab.irstea.fr/HYCAR-Hydro/PREMHYCE) project need a faster version of the `PE_Oudin()` function.
The vectorization of the code is not satisfactory, it would be preferable to integrate the Fortran version of the ...The [PREMHYCE](https://gitlab.irstea.fr/HYCAR-Hydro/PREMHYCE) project need a faster version of the `PE_Oudin()` function.
The vectorization of the code is not satisfactory, it would be preferable to integrate the Fortran version of the code. I think that is a good idea to keep the R version of the code to facilitate the comprehension by the users. In this case we can add an argument to switch from the R code to the Fortran code.v1.6.10https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/86Gitlab-CI: correctly handle WARNING and NOTE in checks2021-06-22T08:53:02+02:00Dorchies DavidGitlab-CI: correctly handle WARNING and NOTE in checksCurrently, check fails only if in an ERROR occurs especially because some warnings or notes occur because of reasons outside the scope of the package itself (dependency to 'qpdf' or curl issues like in https://gitlab.irstea.fr/HYCAR-Hydr...Currently, check fails only if in an ERROR occurs especially because some warnings or notes occur because of reasons outside the scope of the package itself (dependency to 'qpdf' or curl issues like in https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/jobs/142218).
airGRiwrm now implements a new CI procedure that correcly handle WARNING in check by putting the job and the pipeline in failure state (See: https://gitlab.irstea.fr/in-wop/airGRiwrm/-/commits/master/.gitlab-ci.yml). This way of doing seems reasonable because of the CRAN's requirements which doesn't allow any WARNING in the check of a package submission.
This procedure can be implemented in airGR, especially the part using R tidyverse docker image hosted on the server hosted at INRAE Lyon.
A new concept "don't stop on failure" exists in Gitlab-CI (See: https://docs.gitlab.com/ee/ci/yaml/#allow_failure) that allows to continue a pipeline on a job failure and to display corresponding job and pipeline in an "orange warning state". That could be useful if we want to be aware of NOTEs in checks without setting all the check process in a red failure state.v1.6.11Dorchies DavidDorchies Davidhttps://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/80Use user-defined macros in the manual2021-01-12T14:54:58+01:00Delaigue OlivierUse user-defined macros in the manualIt is possible to use user-defines macros in order not to write a new time the package title, the package description, etc. For references, it is also possible to replace the `href` mark by the `doi` mark.
* `\packageTitle`
* `\packageD...It is possible to use user-defines macros in order not to write a new time the package title, the package description, etc. For references, it is also possible to replace the `href` mark by the `doi` mark.
* `\packageTitle`
* `\packageDescription`
* `\packageAuthor`
* `\packageMaintainer`
* `\packageDESCRIPTION`
* `\packageIndices`
* `\doi`
See *[Writing R Extensions](https://cloud.r-project.org/doc/manuals/R-exts.html#User_002ddefined-macros)*[ (User-defined macros)](https://cloud.r-project.org/doc/manuals/R-exts.html#User_002ddefined-macros)
This avoids, for example, possible differences between the title of the description file and the title of the package's help page.v1.6.10https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/70Improve explanations of GR5 outputs values2021-01-11T11:54:38+01:00Delaigue OlivierImprove explanations of GR5 outputs valuesthe explanations of the outputs of the GR5* models must be improved, because they refer to the use of 2 unit hydrographs whereas there is only one in GR5.
This concerns:
- RunModel_CemaNeigeGR5H
- RunModel_CemaNeigeGR5J
- RunModel_GR5H...the explanations of the outputs of the GR5* models must be improved, because they refer to the use of 2 unit hydrographs whereas there is only one in GR5.
This concerns:
- RunModel_CemaNeigeGR5H
- RunModel_CemaNeigeGR5J
- RunModel_GR5H
- RunModel_GR5J
|value| explanation |
|-----|---------------------------------------------|
| $Q9 | [numeric] series of UH1 outflow (Q9) [mm/d] |
| $Q1 | [numeric] series of UH2 outflow (Q1) [mm/d] |v1.6.10https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/76Group release history of non submited versions2021-01-08T04:57:39+01:00Delaigue OlivierGroup release history of non submited versionsExcept for the very first versions of the packages. Only the package versions submitted to CRAN is subject to a history. Therefore, in the NEWS file, the information for the version 1.6.3.73 and the latest version should be grouped toget...Except for the very first versions of the packages. Only the package versions submitted to CRAN is subject to a history. Therefore, in the NEWS file, the information for the version 1.6.3.73 and the latest version should be grouped together (nota: the latest version on the CRAN is the 1.4.3.65.)v1.6.10https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/25Add a regime calculation function2021-01-06T11:43:24+01:00Thirel GuillaumeAdd a regime calculation function(transfer from ticket https://forge.irstea.fr/issues/5445)
It could be nice to propose a function to calculate regime (for P, T, PE, Q, SnowPack, SCA) from an `OutputsModel `object and Qobs (or SWE obs or SCA obs) and also to plot them....(transfer from ticket https://forge.irstea.fr/issues/5445)
It could be nice to propose a function to calculate regime (for P, T, PE, Q, SnowPack, SCA) from an `OutputsModel `object and Qobs (or SWE obs or SCA obs) and also to plot them.
This function could have as arguments the aggregation time step and a sliding window length.v1.6.10https://gitlab.irstea.fr/HYCAR-Hydro/airgr/-/issues/41Use of SeriesAggreg when TimeLag >= 36002021-01-06T10:06:11+01:00Delaigue OlivierUse of SeriesAggreg when TimeLag >= 3600The `SeriesAggreg()` function does not work when `TimeLag >= 3600`.
```
Error in `*tmp*`[1:(length(Ind2) - 1)] :
only 0's may be mixed with negative subscripts
```The `SeriesAggreg()` function does not work when `TimeLag >= 3600`.
```
Error in `*tmp*`[1:(length(Ind2) - 1)] :
only 0's may be mixed with negative subscripts
```v1.6.10