feat!: Support semicolon-separated lists in Performer fields (#141)

altamisa/isatab/models.py

@@ -460,7 +460,7 @@ class Process:
     #: Process date
     date: Optional[Union[datetime.date, Literal[""]]]
     #: Performer of process
-    performer: Optional[str]
+    performer: Optional[Tuple[str, ...]]
     #: Tuple of parameters values
     parameter_values: Tuple[ParameterValue, ...]
     #: Tuple of process comments

altamisa/isatab/parse_assay_study.py

@@ -442,7 +442,7 @@ def build(self, line: List[str]) -> models.Process:
         else:
             date = None
         if self.performer_header:
-            performer = line[self.performer_header.col_no]
+            performer = tuple(self._token_with_escape(line[self.performer_header.col_no]))
         else:
             performer = None
         comments = tuple(

altamisa/isatab/write_assay_study.py

@@ -411,7 +411,9 @@ def _extract_process(self, node: Process) -> dict:
         if node.protocol_ref != TOKEN_UNKNOWN:
             attributes[table_headers.PROTOCOL_REF] = node.protocol_ref
         if node.performer is not None:
-            attributes[table_headers.PERFORMER] = node.performer
+            attributes[table_headers.PERFORMER] = ";".join(
+                [p.replace(";", "\\;") if p else "" for p in node.performer]
+            )
         if node.date is not None:
             attributes[table_headers.DATE] = node.date
         for parameter in node.parameter_values:

docs/index.rst

@@ -53,10 +53,10 @@ Special Extensions
 In addition to the original ISA-Tab format specifications, AltamISA supports
  the following special modifications to improve specific use cases:
 
-- **List of values** in ``Characterics``, ``Parameter Value``, or ``Factor Value`` fields by using
-  semicolon-separators (";"). Note, for ontology terms the same number of
-  splits is expected in the associated field ``Term Source REF`` and
-  ``Term Accession Number``.
+- **List of values** in ``Characterics``, ``Parameter Value``, ``Factor Value``, 
+  and ``Performer`` fields by using semicolon-separators (";"). Note: For 
+  ontology terms the same number of splits is expected in the associated field 
+  ``Term Source REF`` and ``Term Accession Number``.
 - **Material name** ``Library Name`` for improved library
   annotation in nucleotide sequencing assays.
 

tests/test_parse_study.py

@@ -257,7 +257,7 @@ def test_study_row_reader_small_study(small_investigation_file, small_study_file
         None,
         None,
         date(2018, 2, 2),
-        "John Doe",
+        ("John Doe",),
         (models.ParameterValue("instrument", ["scalpel"], None),),
         (),
         None,
@@ -298,7 +298,7 @@ def test_study_row_reader_small_study(small_investigation_file, small_study_file
         None,
         None,
         date(2018, 2, 2),
-        "John Doe",
+        ("John Doe",),
         (models.ParameterValue("instrument", ["scalpel type A", "scalpel type B"], None),),
         (),
         None,
@@ -507,7 +507,7 @@ def test_study_reader_small_study(
         None,
         None,
         date(2018, 2, 2),
-        "John Doe",
+        ("John Doe",),
         (models.ParameterValue("instrument", ["scalpel"], None),),
         (),
         None,
@@ -522,7 +522,7 @@ def test_study_reader_small_study(
         None,
         None,
         date(2018, 2, 2),
-        "John Doe",
+        ("John Doe",),
         (models.ParameterValue("instrument", ["scalpel type A", "scalpel type B"], None),),
         (),
         None,
@@ -537,7 +537,7 @@ def test_study_reader_small_study(
         None,
         None,
         date(2018, 2, 2),
-        "John Doe",
+        ("John Doe",),
         (models.ParameterValue("instrument", ["scalpel"], None),),
         (),
         None,