Release 0.15.1

pypiper/ngstk.py

@@ -559,22 +559,22 @@ def temp_func(input_files=input_files, output_files=output_files, paired_end=pai
                 [int(self.count_reads(input_file, paired_end)) for input_file in input_files]
             )
             raw_reads = int(total_reads / n_input_files)
-            self.pm.pipestat.report(values={"Raw_reads": str(raw_reads)})
+            self.pm.report_result("Raw_reads", str(raw_reads))
 
             total_fastq_reads = sum(
                 [int(self.count_reads(output_file, paired_end)) for output_file in output_files]
             )
             fastq_reads = int(total_fastq_reads / n_output_files)
 
-            self.pm.pipestat.report(values={"Fastq_reads": fastq_reads})
+            self.pm.report_result("Fastq_reads", fastq_reads)
             input_ext = self.get_input_ext(input_files[0])
             # We can only assess pass filter reads in bam files with flags.
             if input_ext == ".bam":
                 num_failed_filter = sum(
                     [int(self.count_fail_reads(f, paired_end)) for f in input_files]
                 )
                 pf_reads = int(raw_reads) - num_failed_filter
-                self.pm.pipestat.report(values={"PF_reads": str(pf_reads)})
+                self.pm.report_result("PF_reads", str(pf_reads))
             if fastq_reads != int(raw_reads):
                 raise Exception(
                     "Fastq conversion error? Number of input reads doesn't number of output reads."
@@ -610,9 +610,9 @@ def temp_func():
                 print("WARNING: specified paired-end but no R2 file")
 
             n_trim = float(self.count_reads(trimmed_fastq, paired_end))
-            self.pm.pipestat.report(values={"Trimmed_reads": int(n_trim)})
+            self.pm.report_result("Trimmed_reads", int(n_trim))
             try:
-                rr = float(self.pm.pipestat.retrieve("Raw_reads"))
+                rr = float(self.pm.get_stat("Raw_reads"))
             except Exception:
                 print("Can't calculate trim loss rate without raw read result.")
             else:
@@ -626,27 +626,17 @@ def temp_func():
                 self.pm.run(cmd, lock_name="trimmed_fastqc", nofail=True)
                 fname, ext = os.path.splitext(os.path.basename(trimmed_fastq))
                 fastqc_html = os.path.join(fastqc_folder, fname + "_fastqc.html")
-                self.pm.pipestat.report(
-                    values={
-                        "FastQC_report_R1": {
-                            "path": fastqc_html,
-                            "title": "FastQC report R1",
-                        }
-                    }
+                self.pm.report_result(
+                    "FastQC_report_R1", {"path": fastqc_html, "title": "FastQC report R1"}
                 )
 
                 if paired_end and trimmed_fastq_R2:
                     cmd = self.fastqc(trimmed_fastq_R2, fastqc_folder)
                     self.pm.run(cmd, lock_name="trimmed_fastqc_R2", nofail=True)
                     fname, ext = os.path.splitext(os.path.basename(trimmed_fastq_R2))
                     fastqc_html = os.path.join(fastqc_folder, fname + "_fastqc.html")
-                    self.pm.pipestat.report(
-                        values={
-                            "FastQC_report_R2": {
-                                "path": fastqc_html,
-                                "title": "FastQC report R2",
-                            }
-                        }
+                    self.pm.report_result(
+                        "FastQC_report_R2", {"path": fastqc_html, "title": "FastQC report R2"}
                     )
 
         return temp_func
@@ -1886,25 +1876,21 @@ def get_read_type(self, bam_file: str, n: int = 10) -> tuple[str, int]:
         else:
             return "SE", read_length
 
-    def parse_bowtie_stats(self, stats_file: str) -> Any:
+    def parse_bowtie_stats(self, stats_file: str) -> dict:
         """
-        Parses Bowtie2 stats file, returns series with values.
+        Parses Bowtie2 stats file, returns dict with values.
 
         Args:
             stats_file (str): Bowtie2 output file with alignment statistics.
         """
-        import pandas as pd
-
-        stats = pd.Series(
-            index=[
-                "readCount",
-                "unpaired",
-                "unaligned",
-                "unique",
-                "multiple",
-                "alignmentRate",
-            ]
-        )
+        stats = {
+            "readCount": None,
+            "unpaired": None,
+            "unaligned": None,
+            "unique": None,
+            "multiple": None,
+            "alignmentRate": None,
+        }
         try:
             with open(stats_file) as handle:
                 content = handle.readlines()  # list of strings per line
@@ -1938,16 +1924,14 @@ def parse_bowtie_stats(self, stats_file: str) -> Any:
             pass
         return stats
 
-    def parse_duplicate_stats(self, stats_file: str) -> Any:
+    def parse_duplicate_stats(self, stats_file: str) -> dict:
         """
-        Parses sambamba markdup output, returns series with values.
+        Parses sambamba markdup output, returns dict with values.
 
         Args:
             stats_file (str): sambamba output file with duplicate statistics.
         """
-        import pandas as pd
-
-        series = pd.Series()
+        series = {}
         try:
             with open(stats_file) as handle:
                 content = handle.readlines()  # list of strings per line
@@ -1968,17 +1952,15 @@ def parse_duplicate_stats(self, stats_file: str) -> Any:
             pass
         return series
 
-    def parse_qc(self, qc_file: str) -> Any:
+    def parse_qc(self, qc_file: str) -> dict:
         """
         Parse phantompeakqualtools (spp) QC table and return quality metrics.
 
         Args:
             qc_file (str): Path to phantompeakqualtools output file, which
                 contains sample quality measurements.
         """
-        import pandas as pd
-
-        series = pd.Series()
+        series = {}
         try:
             with open(qc_file) as handle:
                 line = handle.readlines()[0].strip().split("\t")  # list of strings per line
@@ -2001,17 +1983,15 @@ def get_peak_number(self, sample: Any) -> Any:
         sample["peakNumber"] = re.sub(r"\D.*", "", out)
         return sample
 
-    def get_frip(self, sample: Any) -> Any:
+    def get_frip(self, sample: Any) -> dict:
         """
         Calculates the fraction of reads in peaks for a given sample.
 
         Args:
             sample (pipelines.Sample): Sample object with "peaks" attribute.
         """
-        import pandas as pd
-
         with open(sample.frip, "r") as handle:
             content = handle.readlines()
         reads_in_peaks = int(re.sub(r"\D", "", content[0]))
         mapped_reads = sample["readCount"] - sample["unaligned"]
-        return pd.Series(reads_in_peaks / mapped_reads, index="FRiP")
+        return {"FRiP": reads_in_peaks / mapped_reads}

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "piper"
-version = "0.15.0"
+version = "0.15.1"
 description = "A lightweight python toolkit for gluing together restartable, robust command line pipelines"
 readme = "README.md"
 license = "BSD-2-Clause"
@@ -24,12 +24,11 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Bio-Informatics",
 ]
 dependencies = [
-    "logmuse>=0.2.4",
+    "logmuse>=0.3.0",
     "psutil",
-    "pandas",
-    "ubiquerg>=0.9.0",
-    "yacman>=0.9.5",
-    "pipestat>=0.13.0",
+    "ubiquerg>=0.9.1",
+    "yacman>=1.0.0",
+    "pipestat>=0.13.1",
 ]
 
 [project.urls]

tests/conftest.py

@@ -65,15 +65,15 @@ def cleanup_legacy_output_folders(request):
 
 @pytest.fixture
 def get_pipe_manager(tmpdir):
-    """Provide safe creation of pipeline manager, with multi=True."""
+    """Provide safe creation of pipeline manager."""
 
     def get_mgr(**kwargs):
         if "outfolder" in kwargs:
             kwd_args = kwargs
         else:
             kwd_args = copy.deepcopy(kwargs)
             kwd_args["outfolder"] = tmpdir.strpath
-        return PipelineManager(multi=True, **kwd_args)
+        return PipelineManager(**kwd_args)
 
     return get_mgr
 

tests/helpers.py

@@ -64,7 +64,7 @@ class SafeTestPipeline(Pipeline):
     """Pipeline for tests that protects against bad file descriptor."""
 
     def __init__(self, *args, **kwargs):
-        kwd_args = {"multi": True}  # Like interactive mode.
+        kwd_args = {"multi": True}  # Allow shared results file in tests.
         kwd_args.update(kwargs)
         super(SafeTestPipeline, self).__init__(*args, **kwd_args)
 

tests/pipeline_manager/test_pipeline_manager.py

@@ -30,7 +30,7 @@ def single_pipeline_manager(tmpdir, request):
     """
     outfolder = str(tmpdir.mkdir("pipeline_output"))
     pp = pypiper.PipelineManager(
-        "sample_pipeline", outfolder=outfolder, multi=True, pipestat_schema=TEST_SCHEMA_PATH
+        "sample_pipeline", outfolder=outfolder, pipestat_schema=TEST_SCHEMA_PATH
     )
 
     def cleanup():
@@ -509,7 +509,6 @@ def test_pipestat_validate_results_default_with_schema(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
         )
         assert pp.pipestat.cfg["validate_results"] is True
@@ -521,7 +520,6 @@ def test_pipestat_validate_results_default_without_schema(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
         )
         assert pp.pipestat.cfg["validate_results"] is False
         pp.stop_pipeline()
@@ -532,7 +530,6 @@ def test_pipestat_validate_results_explicit_false(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
             pipestat_validate_results=False,
         )
@@ -545,7 +542,6 @@ def test_pipestat_validate_results_explicit_true(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
             pipestat_validate_results=True,
         )
@@ -558,7 +554,6 @@ def test_pipestat_additional_properties_default(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
         )
         # Default per JSON Schema spec should be True
@@ -571,7 +566,6 @@ def test_pipestat_additional_properties_explicit_false(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
             pipestat_additional_properties=False,
         )
@@ -584,7 +578,6 @@ def test_pipestat_additional_properties_explicit_true(self, tmpdir):
         pp = pypiper.PipelineManager(
             "test_pipeline",
             outfolder=outfolder,
-            multi=True,
             pipestat_schema=TEST_SCHEMA_PATH,
             pipestat_additional_properties=True,
         )

tests/test_context_manager.py

@@ -12,34 +12,34 @@ class TestContextManager:
 
     def test_context_manager_clean_exit(self, tmpdir):
         """On clean exit, stop_pipeline() is called and status is completed."""
-        with PipelineManager("test_cm", tmpdir.strpath, multi=True) as pm:
+        with PipelineManager("test_cm", tmpdir.strpath) as pm:
             pass
         assert pm._has_exit_status
 
     def test_context_manager_exception_sets_failed(self, tmpdir):
         """Exception inside with block calls fail_pipeline() and propagates."""
         with pytest.raises(ValueError, match="test error"):
-            with PipelineManager("test_cm_fail", tmpdir.strpath, multi=True) as pm:
+            with PipelineManager("test_cm_fail", tmpdir.strpath) as pm:
                 raise ValueError("test error")
         assert pm._failed
 
     def test_context_manager_yields_self(self, tmpdir):
         """The yielded object is the PipelineManager instance."""
-        with PipelineManager("test_cm_self", tmpdir.strpath, multi=True) as pm:
+        with PipelineManager("test_cm_self", tmpdir.strpath) as pm:
             assert isinstance(pm, PipelineManager)
             pm_ref = pm
         assert pm_ref is pm
 
     def test_context_manager_run_command(self, tmpdir):
         """End-to-end: run a command with target inside with, verify target created."""
         target = os.path.join(tmpdir.strpath, "output.txt")
-        with PipelineManager("test_cm_run", tmpdir.strpath, multi=True) as pm:
+        with PipelineManager("test_cm_run", tmpdir.strpath) as pm:
             pm.run(f"echo 'hello' > {target}", target=target)
         assert os.path.isfile(target)
 
     def test_context_manager_keyboard_interrupt(self, tmpdir):
         """KeyboardInterrupt triggers fail_pipeline() and propagates."""
         with pytest.raises(KeyboardInterrupt):
-            with PipelineManager("test_cm_kb", tmpdir.strpath, multi=True) as pm:
+            with PipelineManager("test_cm_kb", tmpdir.strpath) as pm:
                 raise KeyboardInterrupt()
         assert pm._failed