Merge branch 'main' into fix/charge_molecules_cli_formatting

docs/installation.rst

@@ -378,18 +378,25 @@ Now the CLI tool should work as well ::
      --log PATH  logging configuration file
      -h, --help  Show this message and exit.
 
-   Setup Commands:
-     atommapping        Check the atom mapping of a given pair of ligands
-     plan-rhfe-network  Plan a relative hydration free energy network, saved in a
-                        dir with multiple JSON files
-     plan-rbfe-network  Plan a relative binding free energy network, saved in a
-                        dir with multiple JSON files.
+   Network Planning Commands:
+     plan-rhfe-network    Plan a relative hydration free energy network, saved as
+                          JSON files for the quickrun command.
+     plan-rbfe-network    Plan a relative binding free energy network, saved as
+                          JSON files for the quickrun command.
+     view-ligand-network  Visualize a ligand network
 
-   Simulation Commands:
-     gather    Gather DAG result jsons for network of RFE results into single TSV
-               file
+   Quickrun Executor Commands:
+     gather    Gather result jsons for network of RFE results into a TSV file
      quickrun  Run a given transformation, saved as a JSON file
 
+   Miscellaneous Commands:
+     fetch             Fetch tutorial or other resource.
+     charge-molecules  Generate partial charges for a set of molecules.
+     test              Run the OpenFE test suite
+
+
+
+
 To make sure everything is working, run the tests ::
 
   $ pytest --pyargs openfe openfecli

news/add_filepaths_support.rst

@@ -0,0 +1,23 @@
+**Added:**
+
+* ``openfe gather`` now accepts any number of filepaths and/or directories containing results JSON files, instead of only accepting one results directory.
+
+**Changed:**
+
+* <news item>
+
+**Deprecated:**
+
+* <news item>
+
+**Removed:**
+
+* <news item>
+
+**Fixed:**
+
+* <news item>
+
+**Security:**
+
+* <news item>

openfecli/commands/gather.py

@@ -4,7 +4,7 @@
 import click
 import os
 import pathlib
-from typing import Callable, Literal
+from typing import Callable, Literal, List
 import warnings
 
 from openfecli import OFECommandPlugin
@@ -370,8 +370,9 @@ def _write_dg_mle(legs:dict, writer:Callable, allow_partial:bool):
     'gather',
     short_help="Gather result jsons for network of RFE results into a TSV file"
 )
-@click.argument('rootdir',
-                type=click.Path(dir_okay=True, file_okay=False,
+@click.argument('results',
+                nargs=-1,  # accept any number of results
+                type=click.Path(dir_okay=True, file_okay=True,
                                 path_type=pathlib.Path),
                 required=True)
 @click.option(
@@ -395,14 +396,15 @@ def _write_dg_mle(legs:dict, writer:Callable, allow_partial:bool):
         "(Skip those edges and issue warning instead.)"
     )
 )
-def gather(rootdir:os.PathLike|str,
+def gather(results:List[os.PathLike|str],
            output:os.PathLike|str,
            report:Literal['dg','ddg','raw'],
            allow_partial:bool
            ):
-    """Gather simulation result jsons of relative calculations to a tsv file
+    """Gather simulation result JSON files of relative calculations to a tsv file.
 
-    This walks ROOTDIR recursively and finds all result JSON files from the
+
+    This walks RESULTS recursively and finds all result JSON files from the
     quickrun command (these files must end in .json). Each of these contains
     the results of a separate leg from a relative free energy thermodynamic
     cycle.
@@ -427,8 +429,19 @@ def gather(rootdir:os.PathLike|str,
     import glob
     import csv
 
+    results = sorted(results)  # not necessary, but ensures reproducibility
+    def collect_jsons(results:List[os.PathLike]):
+        all_jsons = []
+        for p in results:
+            if str(p).endswith('json'):
+                all_jsons.append(p)
+            elif p.is_dir():
+                all_jsons.extend(glob.glob(f"{p}/**/*json", recursive=True))
+        
+        return all_jsons
+
     # 1) find all possible jsons
-    json_fns = glob.glob(str(rootdir) + '/**/*json', recursive=True)
+    json_fns = collect_jsons(results)
 
     # 2) filter only result jsons
     result_fns = filter(is_results_json, json_fns)

openfecli/tests/commands/test_gather.py

@@ -1,3 +1,4 @@
+import glob
 from click.testing import CliRunner
 import os
 import pathlib
@@ -230,7 +231,8 @@ def rbfe_result_dir()->pathlib.Path:
 @pytest.mark.skipif(not os.path.exists(POOCH_CACHE) and not HAS_INTERNET,reason="Internet seems to be unavailable and test data is not cached locally.")
 @pytest.mark.parametrize('dataset', ['rbfe_results_serial_repeats', 'rbfe_results_parallel_repeats'])
 @pytest.mark.parametrize('report', ["", "dg", "ddg", "raw"])
-def test_gather(rbfe_result_dir, dataset, report):
+@pytest.mark.parametrize('input_mode', ['directory','filepaths'])
+def test_gather(rbfe_result_dir, dataset, report, input_mode):
 
     expected = {
         "": _EXPECTED_DG,
@@ -245,12 +247,18 @@ def test_gather(rbfe_result_dir, dataset, report):
     else:
         args = []
 
-    results_dir = rbfe_result_dir(dataset)
-    result = runner.invoke(gather, [str(results_dir)] + args + ['-o', '-'])
+    results = rbfe_result_dir(dataset)
+    if input_mode == 'directory':
+        results = [str(results)]
+    elif input_mode == 'filepaths':
+        results = glob.glob(f"{results}/*", recursive=True)
+        assert len(results) > 1  # sanity check to make sure we're passing in multiple paths
 
-    assert_click_success(result)
+    cli_result = runner.invoke(gather, results + args + ['-o', '-'])
 
-    actual_lines = set(result.stdout_bytes.split(b'\n'))
+    assert_click_success(cli_result)
+
+    actual_lines = set(cli_result.stdout_bytes.split(b'\n'))
     assert set(expected.split(b'\n')) == actual_lines
 
 @pytest.mark.skipif(not os.path.exists(POOCH_CACHE) and not HAS_INTERNET,reason="Internet seems to be unavailable and test data is not cached locally.")