Parse MODFLOW6 input (partially) #688
Conversation
Introduce StorageBase for Storage packages to avoid repetition
* Use pandas for ~3 times faster reading of list input * Add logic for mimicking parts of Fortran read intrinsic: double formats (to_float), comma and whitespace separation (split_string), weird concatenation operators in text files (read_fortran_deflated_text_array). * Allow parse_option to return more than one option (required for auxiliary). * Add auxiliary and boundnames entries to list input dtype. * Fix delr, delc (both 1D) and top (2D) in read_dis_blockfile. * Intercept errors and attach file name.
* Do not convert to dense (grid) form * Add optional keyword `sparse_to_dense` in list input reading functions * Update test to read Well data
…ring at node 0 for every subset of a grid.
|
In GitLab by @Huite on May 25, 2022, 16:49 requested review from @JoerivanEngelen and @luitjansl |
|
In GitLab by @luitjansl on May 31, 2022, 10:28 Commented on imod/mf6/read_input/common.py line 10 is_empty(line: str) may be a better name. |
|
In GitLab by @luitjansl on May 31, 2022, 10:41 Commented on imod/mf6/out/dis.py line 81 you could define this header size outside this functin, so that the other functions in this source file can use it as well. it comes back a few times, it seems |
|
In GitLab by @luitjansl on May 31, 2022, 11:03 Commented on imod/mf6/out/disu.py line 81 is the hds in the name short for "heads"? if so, it would be clearer to write it out. Also, are these functions going to be re-used for reading concentration? |
|
In GitLab by @luitjansl on May 31, 2022, 11:48 approved this merge request |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:50 Commented on imod/mf6/out/dis.py line 81 I agree, can probably be stored in constants in the module |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:50 Commented on imod/mf6/out/dis.py line 120 See comment https://gitlab.com/deltares/imod/imod-python/-/merge_requests/141#note_965625806 |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:50 Commented on imod/mf6/disu.py line 182 I found this hard to comprehend. |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:50 Commented on imod/mf6/pkgbase.py line 248 What is there still TODO? If nothing, please remove, if so please add. |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:50 Commented on imod/util.py line 1076 Have you added a separate test for this function? I couldn't find it... |
|
In GitLab by @JoerivanEngelen on Jun 7, 2022, 11:52 Looks good, we still have to think of what we want to support with this. For example: What about Groundwater Transport models? And what about Advanced Boundary Condition packages we still have to add? Perhaps raise a specific error for now if somebody tries to load a transport model? |
In GitLab by @Huite on May 25, 2022, 16:49
This is somewhat of a grab-bag of features required for the TKI project.
The primary feature is that it allows reading MODFLOW6 input via
imod.mf6.Modflow6Simulation.open.It also includes a number of methods to convert DIS input to DISU input. The DISU package currently has the form of a
LowLevelUnstructuredDiscretization: low level, since ideally we can store the information in a UGRID 3D kind of class.Finally, it introduces a
StorageBaseclass which stores the common logic for the two classes rather than having it duplicated.Fix https://gitlab.com/deltares/imod/imod-python/-/issues/138