Merge pull request #201 from pepkit/dev

v0.10.1 release

docs/changelog.md

@@ -2,6 +2,11 @@
 
 This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html) and [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) format.
 
+## [0.10.1] - 2024-08-06
+### Fixed
+- add pipestat summarize and link for pephub backend
+
+
 ## [0.10.0] - 2024-07-18
 ### Fixed
 - allow for bool or boolean in schema [#189](https://github.com/pepkit/pipestat/issues/189)

pipestat/_version.py

@@ -1 +1 @@
-__version__ = "0.10.0"
+__version__ = "0.10.1"

pipestat/backends/pephub_backend/pephubbackend.py

@@ -84,6 +84,15 @@ def check_record_exists(
 
         return bool(query_hit["records"])
 
+    def count_records(self):
+        """
+        Count rows in a selected table
+        :return int: number of records
+        """
+
+        count = self.select_records()["total_size"]
+        return count
+
     def list_results(
         self,
         restrict_to: Optional[List[str]] = None,
@@ -108,7 +117,7 @@ def list_results(
             ]
         )
         try:
-            record = record["records"][0][rid]
+            record = record["records"][0]
         except IndexError:
             return []
 
@@ -339,7 +348,7 @@ def get_status(self, record_identifier: str) -> Optional[str]:
         except RecordNotFoundError:
             return None
         try:
-            status = result["records"][0][record_identifier]["status"]
+            status = result["records"][0]["status"]
         except IndexError or KeyError:
             status = None
 
@@ -433,7 +442,7 @@ def get_operator(op: Literal["eq", "lt", "ge", "gt", "in"]) -> Any:
                 key = filter_condition["key"]
                 value = filter_condition["value"]
 
-                # Create querry for df based on filter conditions
+                # Create query for df based on filter conditions
                 if isinstance(value, list):
                     filter_expression = f"{key} {retrieved_operator} {value}"
                 else:
@@ -451,15 +460,9 @@ def get_operator(op: Literal["eq", "lt", "ge", "gt", "in"]) -> Any:
             else:
                 df = df.query(filter_expression)
 
-            print("done")
-
-        # Once we have the dataframe (filtered or unfiltered), convert to a dict using the sample_name/record_identifier as the primary key
-        df2dict = df.set_index("sample_name").transpose().to_dict(orient="dict")
+        df.rename(columns={"sample_name": "record_identifier"}, inplace=True)
 
-        # Must do this to align output structure with that of db_backend and file_backends
-        records_list = []
-        for key, value in df2dict.items():
-            records_list.append({key: value})
+        records_list = df.to_dict("records")
 
         records_dict = {
             "total_size": total_count,

pipestat/pipestat.py

@@ -908,9 +908,6 @@ def link(self, link_dir) -> Union[str, None]:
         :param str link_dir: path to desired symlink output directory
         :return str | None linked_results_path: path to symlink directory or None
         """
-        if self.cfg["pephub_path"]:
-            _LOGGER.warning(f"Linking results is not supported for PEPHub backend.")
-            return None
 
         self.check_multi_results()
         linked_results_path = self.backend.link(link_dir=link_dir)
@@ -923,20 +920,25 @@ def summarize(
         looper_samples: Optional[list] = None,
         amendment: Optional[str] = None,
         portable: Optional[bool] = False,
+        output_dir: Optional[str] = None,
     ) -> Union[str, None]:
         """
         Builds a browsable html report for reported results.
         :param Iterable[str] looper_samples: list of looper Samples from PEP
         :param Iterable[str] amendment: name indicating amendment to use, optional
         :param bool portable: moves figures and report files to directory for easy sharing
+        :param str output_dir: overrides output_dir set during pipestatManager creation.
         :return str: report_path
 
         """
+
+        if output_dir:
+            self.cfg[OUTPUT_DIR] = output_dir
+
         if self.cfg["pephub_path"]:
-            _LOGGER.warning(
-                f"Summarize not supported for PEPHub backend. Please generate report via PEPHub website."
-            )
-            return None
+            if OUTPUT_DIR not in self.cfg:
+                _LOGGER.warning(f"Output directory is required for pipestat summarize.")
+                return None
 
         self.check_multi_results()
 

pipestat/reports.py

@@ -56,9 +56,14 @@ def __init__(self, prj, portable=False):
 
         results_file_path = getattr(self.prj.backend, "results_file_path", None)
         config_path = self.prj.cfg.get("config_path", None)
-        output_dir = getattr(self.prj.cfg[OUTPUT_DIR], "output_dir", None)
+        output_dir = self.prj.cfg.get(OUTPUT_DIR, None)
         self.output_dir = output_dir or results_file_path or config_path
-        self.output_dir = os.path.dirname(self.output_dir)
+
+        if os.path.isdir(self.output_dir):
+            pass
+        else:
+            self.output_dir = os.path.dirname(self.output_dir)
+
         if not self.portable:
             self.reports_dir = os.path.join(self.output_dir, "reports")
         else:
@@ -1367,7 +1372,7 @@ def get_file_for_table(prj, pipeline_name: str, appendix=None, directory=None) -
     # TODO make determining the output_dir its own small function since we use the same code in HTML report building.
     results_file_path = getattr(prj.backend, "results_file_path", None)
     config_path = prj.cfg.get("config_path", None)
-    output_dir = prj.cfg.get("output_dir", None)
+    output_dir = prj.cfg.get(OUTPUT_DIR, None)
     table_dir = output_dir or results_file_path or config_path
     if not os.path.isdir(table_dir):
         table_dir = os.path.dirname(table_dir)

tests/test_pipestat.py

@@ -2587,7 +2587,7 @@ def test_pephub_backend_retrieve_one(
 
         result = psm.retrieve_one(record_identifier=rec_id)
 
-        assert len(result.keys()) == 1
+        assert len(result.keys()) == 6
 
     def test_pephub_backend_retrieve_many(
         self,
@@ -2687,11 +2687,35 @@ def test_pephub_unsupported_funcs(
 
         assert results is None
 
-        psm.link("somedir")
         psm.list_recent_results()
-        psm.summarize()
 
-    def test_pephub_unsupported_funcs(
+    def test_pephub_backend_summarize(
+        self,
+        config_file_path,
+        schema_file_path,
+    ):
+
+        with TemporaryDirectory() as d:
+            temp_dir = d
+            psm = PipestatManager(pephub_path=PEPHUB_URL, schema_path=schema_file_path)
+            report_path = psm.summarize(output_dir=d)
+
+            assert report_path
+
+    def test_pephub_backend_link(
+        self,
+        config_file_path,
+        schema_file_path,
+    ):
+
+        with TemporaryDirectory() as d:
+            temp_dir = d
+            psm = PipestatManager(pephub_path=PEPHUB_URL, schema_path=schema_file_path)
+            report_path = psm.link(link_dir=d)
+
+            assert report_path
+
+    def test_pephub_bad_path(
         self,
         config_file_path,
         schema_file_path,