Add wildcard expansion of wlist wells in compdat

[larsevj]

No description provided.

[Copilot]

Pull Request Overview

This PR implements wildcard expansion for well names in WLIST when processing COMPDAT keywords. The changes enable WLIST entries containing wildcard patterns (like "*OP") to be expanded to actual well names found in the COMPDAT data at the corresponding date.

• Adds wildcard expansion functionality to WLIST processing by matching patterns against wells defined in COMPDAT
• Refactors WLIST expansion into separate functions for wildcard expansion and action expansion
• Updates function signatures to pass COMPDAT data to WLIST expansion functions

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
src/res2df/compdat.py	Implements wildcard expansion logic and refactors WLIST expansion functions
tests/test_wlist.py	Updates test to use the new function name for WLIST action expansion
tests/test_welopen.py	Adds test case for wildcard expansion in WELOPEN with WLIST

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Pull Request Overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

Comments suppressed due to low confidence (1)

src/res2df/compdat.py:1

• Multiple spelling errors: 'wilcard' should be 'wildcard', 'charachters' should be 'characters' (appears twice).

"""Parser and dataframe generator for the keywords:

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[larsevj]

In test_wlist.py there is the following test case:

        # Wildcard wells should pass through, the dataframe user
        # will have to process it.
        (
            pd.DataFrame(
                [
                    {
                        "NAME": "OP",
                        "ACTION": "NEW",
                        "WELLS": "PROD*",
                        "DATE": datetime.date(1900, 1, 1),
                    },
                ]
            ),
            pd.DataFrame(
                [
                    {
                        "NAME": "OP",
                        "ACTION": "NEW",
                        "WELLS": "PROD*",
                        "DATE": datetime.date(1900, 1, 1),
                    },
                ]
            ),
        ),
    ],

It is unclear to me if this was due to the functionality not being in place at the time, for instance via the get_wells_matching_template function added in #333 or if there are any specific reasons or pitfalls related to supporting this functionality.
Do you remember anything @berland ?

[berland]

89d14d7 indicates that there might be a reason not to process wildcards in WLIST parsing.

[larsevj]

89d14d7 indicates that there might be a reason not to process wildcards in WLIST parsing.

Do you know what the commit message Can be handled in WELOPEN instead., refers to? As I can currently understand, you end up inserting from wlist_df into welopen_df, and afterwards you try and match up the wells in welopen_df and compdat_df, but since they can be wildcarded in wlist_df -> welopen_df you run the risk of them not matching up with compdat_df as seen in https://equinor.slack.com/archives/CMS2PTC3Z/p1759499665007219

But there is an expansion of wildcards in welopen_df, only it happens before wells from wlist_df is inserted into welopen_df. So maybe if you do the wildcards expansion of welopen afterwards it would work, but then I do not see why you would not do it directly in wlist_df. Probably something I am missing here.

[berland]

There is a comment in the commit mentioned above:

        # Wildcard wells are currently not supported
        # (it requires knowledge of all currenly processed wells)

I don't remember details here, but if you are parsing the entire deck when doing this, and thus maintain the state of which wells are currently defined, it could work.