res2df 1.3.11__tar.gz → 1.3.12__tar.gz

{res2df-1.3.11 → res2df-1.3.12}/.github/workflows/res2df.yml

@@ -10,7 +10,7 @@ on:
 
   schedule:
     # Run CI every night and check that tests are working with latest dependencies
-    - cron: '0 0 * * *'
+    - cron: "0 0 * * *"
 
 env:
   ERT_SHOW_BACKTRACE: 1
@@ -21,16 +21,23 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ['3.11', '3.12', '3.13', '3.14']
+        python-version: ["3.11", "3.12", "3.13", "3.14"]
+        pandas-version: [""]
         include:
           # For one of the Python versions we
           # install the extra dependency ert
           # (in order to not double the job matrix)
-          - python-version: '3.11'
+          - python-version: "3.11"
             install-ert: true
+          - python-version: "3.13"
+            pandas-version: ">=2,<3"
+          - python-version: "3.13"
+            pandas-version: ">=3.0.0"
+        exclude:
+          - python-version: "3.13"
+            pandas-version: ""
 
     steps:
-
       - name: Install Ubuntu dependencies
         run: |
           sudo apt-get update
@@ -59,11 +66,14 @@ jobs:
         run: |
           uv pip install ".[tests, docs]"
 
-
       - name: Install ert
         if: matrix.install-ert
         run: uv pip install ".[ert]"
 
+      - name: Pin pandas version
+        if: matrix.pandas-version
+        run: uv pip install "pandas${{ matrix.pandas-version }}"
+
       - name: List all installed packages
         run: uv pip freeze
 
@@ -84,22 +94,22 @@ jobs:
       - name: Update GitHub pages
         if: github.repository_owner == 'equinor' && github.ref == 'refs/heads/master' && matrix.python-version == '3.11'
         run: |
-            cp -R ./build/sphinx/html ../html
+          cp -R ./build/sphinx/html ../html
 
-            git config --local user.email "res2df-github-action"
-            git config --local user.name "res2df-github-action"
-            git fetch origin gh-pages
-            git checkout --track origin/gh-pages
-            git clean -f -f -d -x
-            git rm -r *
+          git config --local user.email "res2df-github-action"
+          git config --local user.name "res2df-github-action"
+          git fetch origin gh-pages
+          git checkout --track origin/gh-pages
+          git clean -f -f -d -x
+          git rm -r *
 
-            cp -R ../html/* .
+          cp -R ../html/* .
 
-            git add .
+          git add .
 
-            if git diff-index --quiet HEAD; then
-              echo "No changes in documentation. Skip documentation deploy."
-            else
-              git commit -m "Update Github Pages"
-              git push "https://${{ github.actor }}:${{ secrets.GITHUB_TOKEN }}@github.com/${{ github.repository }}.git" gh-pages
-            fi
+          if git diff-index --quiet HEAD; then
+            echo "No changes in documentation. Skip documentation deploy."
+          else
+            git commit -m "Update Github Pages"
+            git push "https://${{ github.actor }}:${{ secrets.GITHUB_TOKEN }}@github.com/${{ github.repository }}.git" gh-pages
+          fi

{res2df-1.3.11 → res2df-1.3.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: res2df
-Version: 1.3.11
+Version: 1.3.12
 Summary: Convert reservoir simulator input and output to DataFrames
 Author-email: Håvard Berland <havb@equinor.com>
 License-Expression: GPL-3.0-only
@@ -26,7 +26,7 @@ Requires-Dist: resfo
 Requires-Dist: networkx
 Requires-Dist: numpy
 Requires-Dist: opm>=2020.10.2
-Requires-Dist: pandas
+Requires-Dist: pandas>=2
 Requires-Dist: pyarrow
 Requires-Dist: pyyaml>=5.1
 Requires-Dist: treelib

{res2df-1.3.11 → res2df-1.3.12}/pyproject.toml

@@ -40,7 +40,7 @@ dependencies= [
 "networkx",
 "numpy",
 "opm>=2020.10.2",
-"pandas",
+"pandas >= 2",
 "pyarrow",
 "pyyaml>=5.1",
 "treelib",
@@ -128,24 +128,44 @@ line-length = 88
 
 [tool.ruff.lint]
 select = [
-  "E",  # pycodestyle
-  "W",  # pycodestyle
-  "I",  # isort
+  "AIR",  # airflow
+  "ANN",  # flake8-annotations
+  "ASYNC",  # flake8-async
   "B",  # flake-8-bugbear
-  "SIM",  # flake-8-simplify
+  "C4",  # flake8-comprehensions
+  "DJ",  # django
+  "E",  # pycodestyle
   "F",  # pyflakes
-  "PL",  # pylint
+  "FA",  # future-annotations
+  "FAST",  # fastapi
+  "FIX", # fixme
+  "FLY", # flynt
+  "FURB",  # refurb
+  "I",  # isort
+  "ICN",  # flake8-import-conventions
+  "INT", # gettext
+  "ISC",  # flake8-implicit-str-concat
   "NPY",  # numpy specific rules
-  "C4",  # flake8-comprehensions
   "PD",  # pandas-vet
-  "RUF",  # ruff specific rules
-  "UP",  # pyupgrade
-  "ANN",  # flake8-annotations
-  "FURB",  # refurb
   "PERF",  # perflint
+  "PGH",  # pygrep-hooks
+  "PIE",  # flake8-pie
+  "PL",  # pylint
+  "PLC",  # pylint convention
+  "PLE",  # pylint error
+  "PLR",  # pylint refactor
+  "PLW",  # pylint warning
   "PTH",  # flake8-use-pathlib
-  "ICN",  # flake8-import-conventions
-  "ISC",  # flake8-implicit-str-concat
+  "PYI",  # flake8-pyi
+  "Q",  # quotes
+  "RSE",  # raise
+  "RUF",  # ruff specific rules
+  "SIM",  # flake-8-simplify
+  "SLOT", # flake8-slots
+  "TD",  # todos
+  "UP",  # pyupgrade
+  "W",  # pycodestyle
+  "YTT",  # year 2020
 ]
 preview = true
 ignore = [

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/common.py

@@ -653,7 +653,7 @@ def generic_deck_table(
 
     # Start building the string we are to return:
     string = keyword + "\n"
-    if comment is not None and comment:
+    if comment:
         string += "\n".join(["-- " + line for line in comment.splitlines()]) + "\n"
 
     # Empty tables are ok with Eclipse (at least sometimes)
@@ -664,7 +664,7 @@ def generic_deck_table(
     # Pandas make a pretty txt table:
     dframe = dframe.copy()
 
-    # Column names are pr. ec2ldf standard, redo to opm.common in order to use
+    # Column names are pr. res2df standard, redo to opm.common in order to use
     # sorting from that:
     if renamer is not None:
         inv_renamer = {value: key for key, value in renamer.items()}
@@ -696,11 +696,14 @@ def generic_deck_table(
     dframe = dframe.fillna(value="1*")
 
     if drop_trailing_columns:
+        columns_to_drop = []
         for col_name in reversed(relevant_columns):
-            if set(dframe[col_name].to_numpy()) == {"1*"}:
-                del dframe[col_name]
+            if (dframe[col_name] == "1*").all():
+                columns_to_drop.append(col_name)
             else:
                 break
+        if columns_to_drop:
+            dframe = dframe.drop(columns=columns_to_drop)
 
     # It is critical for opm.common, maybe also E100 to have integers printed
     # as integers, for correct parsing. Ensure these are integer where the json
@@ -708,13 +711,15 @@ def generic_deck_table(
     integer_cols = {
         item["name"]
         for item in OPMKEYWORDS[keyword]["items"]
-        if item["value_type"] == "INT"  # and item["name"] in col_headers
+        if item["value_type"] == "INT"
     }
     for int_col in integer_cols.intersection(dframe.columns):
-        defaulted_rows = dframe[int_col] == "1*"
-        integer_values = dframe.loc[~defaulted_rows, int_col].astype(int)
+        mask = dframe[int_col] != "1*"
         dframe[int_col] = dframe[int_col].astype(str)
-        dframe.loc[~defaulted_rows, int_col] = integer_values
+        if mask.any():
+            dframe.loc[mask, int_col] = (
+                dframe.loc[mask, int_col].astype(float).astype(int).astype(str)
+            )
 
     # Quote all string data. This is not always needed, but needed
     # for some colums, for example well-names containing a slash.
@@ -726,9 +731,8 @@ def generic_deck_table(
     for str_col in string_cols.intersection(dframe.columns):
         # Ensure 1* is not quoted.
         non_defaulted_rows = dframe[str_col] != "1*"
-        dframe.loc[non_defaulted_rows, str_col].str.replace("'", "")
         dframe.loc[non_defaulted_rows, str_col] = (
-            "'" + dframe.loc[non_defaulted_rows, str_col] + "'"
+            "'" + dframe.loc[non_defaulted_rows, str_col].str.replace("'", "") + "'"
         )
 
     # Now rename again to have prettier column names:
@@ -739,8 +743,7 @@ def generic_deck_table(
     tablestring = dframe.to_string(header=True, index=False)
     # Indent all lines with two spaces:
     tablestring = "\n".join(
-        ["  " + line.strip().replace("  /", " /") for line in tablestring.splitlines()]
-        # The replace() in there is needed for py36/pandas==1.1.5 only.
+        "  " + line.strip().replace("  /", " /") for line in tablestring.splitlines()
     )
     # Eclipse comment for the header line:
     tablestring = "--" + tablestring[1:]
@@ -836,7 +839,7 @@ def stack_on_colnames(
     # Drop rows stemming from the NaNs in the second tuple-element for
     # static columns:
     dframe = dframe.dropna(axis="index", subset=["DATE"])
-    del dframe["level_0"]
+    dframe = dframe.drop(columns="level_0")
     dframe.index.name = ""
     return dframe
 
@@ -912,11 +915,11 @@ def parse_lyrfile(filename: str | Path) -> list[dict[str, Any]] | None:
                     zonedict["to_layer"] = to_layer
                 else:
                     logger.error("From_layer higher than to_layer")
-                    raise ValueError()
+                    raise ValueError
             elif len(numbers) == 1:
                 zonedict["span"] = int(numbers[0])
             else:
-                raise ValueError()
+                raise ValueError
             lyrlist.append(zonedict)
         except ValueError:
             logger.error("Could not parse lyr file %s", filename)
@@ -967,7 +970,7 @@ def get_wells_matching_template(template: str, wells: list[str]) -> list[str]:
     Returns:
         List of matched wells
     """
-    if template.startswith("*") or template.startswith("?"):
+    if template.startswith(("*", "?")):
         raise ValueError(
             "Well template not allowed to start with a wildcard character: "
             f"Must be preceded with a \\: {template}"

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/compdat.py

@@ -751,7 +751,7 @@ def expand_complump_in_welopen_df(
                 exp_welopens.append(cell_row)
 
     dframe = pd.DataFrame(exp_welopens)
-    return dframe.astype(object).where(pd.notna(dframe), None)  # type: ignore[call-overload]
+    return dframe.astype(object).replace({np.nan: None})
 
 
 def expand_wlist_in_welopen_df(
@@ -785,7 +785,7 @@ def expand_wlist_in_welopen_df(
             # Explicit wellname was used, no expansion to happen:
             exp_welopens.append(row)
     dframe = pd.DataFrame(exp_welopens)
-    return dframe.astype(object).where(pd.notna(dframe), None)  # type: ignore[call-overload]
+    return dframe.astype(object).replace({np.nan: None})
 
 
 def applywelopen(
@@ -844,7 +844,7 @@ def applywelopen(
                 "The WLIST dataframe must be expanded through expand_wlist()"
             )
 
-    welopen_df = welopen_df.astype(object).where(pd.notna(welopen_df), None)  # type: ignore[call-overload]
+    welopen_df = welopen_df.astype(object).replace({np.nan: None})
     if wlist_df is not None:
         welopen_df = expand_wlist_in_welopen_df(welopen_df, wlist_df)
     if complump_df is not None:

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/grid.py

@@ -47,7 +47,7 @@ def get_available_rst_dates(resdatafiles: ResdataFiles) -> list[datetime.date]:
     )
     return [
         resdatafiles.get_rstfile().iget_restart_sim_time(index).date()
-        for index in range(0, len(report_indices))
+        for index in range(len(report_indices))
     ]
 
 

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/gruptree.py

@@ -9,9 +9,7 @@ import warnings
 from typing import Any
 
 import numpy as np
-
-# Needed for mypy
-import opm.io
+import opm
 import pandas as pd
 import treelib
 
@@ -63,6 +61,9 @@ def df(
     date: datetime.date | None
     date = startdate if startdate is not None else None
 
+    if not isinstance(deck, (ResdataFiles, opm.opmcommon_python.Deck)):
+        raise TypeError("Input deck must be either ResdataFiles or an opm Deck.")
+
     if isinstance(deck, ResdataFiles):
         deck = deck.get_deck()
 
@@ -178,7 +179,6 @@ def df(
     # This happens with WELSPECS if both GRUPTREE and BRANPROP is defined
     # at the same timestep. And when a node is redirected to a new parent node
     dframe = dframe.drop_duplicates(subset=["DATE", "CHILD", "KEYWORD"], keep="last")
-    print(dframe)
     return dframe
 
 

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/parameters.py

@@ -5,7 +5,7 @@ import json
 import logging
 import warnings
 from pathlib import Path
-from typing import Any
+from typing import Any, cast
 
 import pandas as pd
 import yaml
@@ -79,11 +79,12 @@ def load_parameterstxt(filename: str | Path) -> dict[str, Any]:
                 engine="python",
                 names=["KEY", "VALUE"],
                 index_col=False,
+                dtype={"KEY": str},
             )
         except pd.errors.ParserWarning as txt_exc:
             raise pd.errors.ParserError(txt_exc) from txt_exc
 
-    return dframe.set_index("KEY")["VALUE"].to_dict()
+    return cast(dict[str, Any], dframe.set_index("KEY")["VALUE"].to_dict())
 
 
 def load_all(

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/summary.py

@@ -39,6 +39,7 @@ PD_FREQ_MNEMONICS: dict[str, str] = {
 See
 https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#dateoffset-objects
 """
+PANDAS_MAJOR_VERSION = int(pd.__version__.split(".")[0])
 
 
 def date_range(
@@ -617,7 +618,9 @@ def _fix_dframe_for_resdata(dframe: pd.DataFrame) -> pd.DataFrame:
     dframe = dframe.copy()
     if "DATE" in dframe.columns:
         # Infer datatype (Pandas cannot answer it) based on the first element:
-        if isinstance(dframe["DATE"].to_numpy()[0], str):
+        if PANDAS_MAJOR_VERSION >= 3:
+            dframe["DATE"] = pd.to_datetime(dframe["DATE"])
+        elif isinstance(dframe["DATE"].to_numpy()[0], str):
             # Do not use pd.Series.apply() here, Pandas would try to convert it to
             # datetime64[ns] which is limited at year 2262.
             dframe["DATE"] = pd.Series(
@@ -625,7 +628,7 @@ def _fix_dframe_for_resdata(dframe: pd.DataFrame) -> pd.DataFrame:
                 dtype="object",
                 index=dframe.index,
             )
-        if isinstance(dframe["DATE"].to_numpy()[0], dt.date):
+        elif isinstance(dframe["DATE"].to_numpy()[0], dt.date):
             dframe["DATE"] = pd.Series(
                 [
                     dt.datetime.combine(dateobj, dt.datetime.min.time())
@@ -650,9 +653,7 @@ def _fix_dframe_for_resdata(dframe: pd.DataFrame) -> pd.DataFrame:
     if "Unnamed: 0" in dframe:
         dframe = dframe.drop("Unnamed: 0", axis="columns")
 
-    block_columns = [
-        col for col in dframe.columns if (col.startswith("B") or col.startswith("LB"))
-    ]
+    block_columns = [col for col in dframe.columns if (col.startswith(("B", "LB")))]
     if block_columns:
         dframe = dframe.drop(columns=block_columns)
         logger.warning(

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '1.3.11'
-__version_tuple__ = version_tuple = (1, 3, 11)
+__version__ = version = '1.3.12'
+__version_tuple__ = version_tuple = (1, 3, 12)
 
-__commit_id__ = commit_id = 'g6a680e0c8'
+__commit_id__ = commit_id = 'gef2e91cb5'

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/vfp/_vfpcommon.py

@@ -123,7 +123,7 @@ def _stack_vfptable2df(
     no_indices = len(index_names_list)
 
     # insert index values as first columns in dataframe
-    for i in range(0, no_indices):
+    for i in range(no_indices):
         df_vfptable.insert(i, index_names_list[i], index_values_list[i])
 
     #  create multi-index for columns

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/vfp/_vfpinj.py

@@ -14,19 +14,15 @@ import numbers
 from typing import Any
 
 import numpy as np
+
+# Needed for mypy
+import opm.io
 import pandas as pd
 import pyarrow as pa
 
-try:
-    # Needed for mypy
-
-    import opm.io
-
-    # This import is seemingly not used, but necessary for some attributes
-    # to be included in DeckItem objects.
-    from opm.io.deck import DeckKeyword  # noqa: F401
-except ImportError:
-    pass
+# This import is seemingly not used, but necessary for some attributes
+# to be included in DeckItem objects.
+from opm.io.deck import DeckKeyword  # noqa: F401
 
 from ..common import comment_formatter, parse_opmio_deckrecord
 from ._vfpcommon import (
@@ -276,7 +272,7 @@ def basic_data2pyarrow(
 
     # Column metadata is index in THP array
     col_metadata_list = []
-    for i in range(0, len(thp_values)):
+    for i in range(len(thp_values)):
         col_name = str(i)
         col_dtype = pa.float64()
         col_metadata = {
@@ -397,7 +393,7 @@ def pyarrow2basic_data(pa_table: pa.Table) -> dict[str, Any]:
     # Extract index data from colum metadata
     thp_indices = [
         int(pa_table.schema.field(i).metadata[b"thp_idx"])
-        for i in range(0, pa_table.num_columns)
+        for i in range(pa_table.num_columns)
     ]
 
     # Extract table data as numpy.array
@@ -602,7 +598,7 @@ def _write_table_records(
     else:
         table = table.reshape(no_records, no_flow_values)
 
-    for row in range(0, no_records):
+    for row in range(no_records):
         thp = thp_indices[row]
         deck_str += f"{thp:2d}"
         for n, value in enumerate(table[row, :]):

{res2df-1.3.11 → res2df-1.3.12}/src/res2df/vfp/_vfpprod.py

@@ -14,18 +14,15 @@ import numbers
 from typing import Any
 
 import numpy as np
+
+# Needed for mypy
+import opm.io
 import pandas as pd
 import pyarrow as pa
 
-try:
-    # Needed for mypy
-    import opm.io
-
-    # This import is seemingly not used, but necessary for some attributes
-    # to be included in DeckItem objects.
-    from opm.io.deck import DeckKeyword  # noqa: F401
-except ImportError:
-    pass
+# This import is seemingly not used, but necessary for some attributes
+# to be included in DeckItem objects.
+from opm.io.deck import DeckKeyword  # noqa: F401
 
 from ..common import comment_formatter, parse_opmio_deckrecord
 from ._vfpcommon import (
@@ -382,7 +379,7 @@ def basic_data2pyarrow(
     # Column metadata is list of indices (THP,WFR,GFR,ALQ)
     col_metadata_list = []
     num_records = len(thp_values) * len(wfr_values) * len(gfr_values) * len(alq_values)
-    for i in range(0, num_records):
+    for i in range(num_records):
         thp_idx = thp_indices[i]
         wfr_idx = wfr_indices[i]
         gfr_idx = gfr_indices[i]
@@ -597,7 +594,7 @@ def pyarrow2basic_data(pa_table: pa.Table) -> dict[str, Any]:
     wfr_indices = []
     gfr_indices = []
     alq_indices = []
-    for i in range(0, pa_table.num_columns):
+    for i in range(pa_table.num_columns):
         thp_indices.append(int(pa_table.schema.field(i).metadata[b"thp_idx"]))
         wfr_indices.append(int(pa_table.schema.field(i).metadata[b"wfr_idx"]))
         gfr_indices.append(int(pa_table.schema.field(i).metadata[b"gfr_idx"]))
@@ -880,7 +877,7 @@ def _write_table_records(
     else:
         table = table.reshape(no_records, no_flow_values)
 
-    for row in range(0, no_records):
+    for row in range(no_records):
         thp = thp_indices[row]
         wfr = wfr_indices[row]
         gfr = gfr_indices[row]

{res2df-1.3.11 → res2df-1.3.12}/src/res2df.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: res2df
-Version: 1.3.11
+Version: 1.3.12
 Summary: Convert reservoir simulator input and output to DataFrames
 Author-email: Håvard Berland <havb@equinor.com>
 License-Expression: GPL-3.0-only
@@ -26,7 +26,7 @@ Requires-Dist: resfo
 Requires-Dist: networkx
 Requires-Dist: numpy
 Requires-Dist: opm>=2020.10.2
-Requires-Dist: pandas
+Requires-Dist: pandas>=2
 Requires-Dist: pyarrow
 Requires-Dist: pyyaml>=5.1
 Requires-Dist: treelib

{res2df-1.3.11 → res2df-1.3.12}/src/res2df.egg-info/requires.txt

@@ -3,7 +3,7 @@ resfo
 networkx
 numpy
 opm>=2020.10.2
-pandas
+pandas>=2
 pyarrow
 pyyaml>=5.1
 treelib

{res2df-1.3.11 → res2df-1.3.12}/tests/test_common.py

@@ -5,7 +5,6 @@ import os
 from pathlib import Path
 
 import numpy as np
-import packaging.version
 import pandas as pd
 import pytest
 
@@ -447,10 +446,4 @@ def test_generic_deck_table(
         renamer=renamer,
         drop_trailing_columns=drop_trailing_columns,
     )
-    # Pandas 1.1.5 gives a different amount of whitespace than what
-    # these tests are written for. If so, be more slack about whitespace.
-    if packaging.version.parse(pd.__version__) < packaging.version.parse("1.2.0"):
-        stringtable = " ".join(stringtable.split())
-        assert stringtable == " ".join(expected.split())
-    else:
-        assert stringtable == expected
+    assert stringtable == expected

{res2df-1.3.11 → res2df-1.3.12}/tests/test_gruptree.py

@@ -417,6 +417,15 @@ def test_emptytree_strdeck():
     assert not treelibtree
 
 
+def test_non_deck_input_raises():
+    """Test giving in non valid input to df gives
+    reasonable error message"""
+    with pytest.raises(
+        TypeError, match="Input deck must be either ResdataFiles or an opm Deck"
+    ):
+        _ = gruptree.df("this should not be valid")
+
+
 def test_emptytree_commandlinetool(tmp_path, mocker, caplog):
     """Test the command line tool on a .DATA file which is empty"""
     os.chdir(tmp_path)