v2.0 docs update

.gitignore

@@ -6,6 +6,7 @@ data_export/
 .DS_Store
 cumulus_library_columns.json
 output.sql
+*generated.md
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

cumulus_library/cli.py

@@ -226,6 +226,22 @@ def generate_study_sql(
             parser=self.db.parser(),
         )
 
+    def generate_study_markdown(
+        self,
+        target: pathlib.Path,
+    ) -> None:
+        """Materializes study sql from templates
+
+        :param target: A path to the study directory
+        """
+        studyparser = study_parser.StudyManifestParser(target)
+        studyparser.run_generate_markdown(
+            self.cursor,
+            self.schema_name,
+            verbose=self.verbose,
+            parser=self.db.parser(),
+        )
+
 
 def get_abs_posix_path(path: str) -> pathlib.Path:
     """Convenience method for handling abs vs rel paths"""
@@ -373,6 +389,10 @@ def run_cli(args: dict):
             elif args["action"] == "generate-sql":
                 for target in args["target"]:
                     runner.generate_study_sql(study_dict[target])
+
+            elif args["action"] == "generate-md":
+                for target in args["target"]:
+                    runner.generate_study_markdown(study_dict[target])
         finally:
             db_backend.close()
 

cumulus_library/cli_parser.py

@@ -229,4 +229,15 @@ def get_parser() -> argparse.ArgumentParser:
     add_target_argument(generate)
     add_study_dir_argument(generate)
     add_db_config(generate)
+
+    # Generate markdown tables for documentation
+    markdown = actions.add_parser(
+        "generate-md", help="Generates markdown tables for study documentation"
+    )
+    add_target_argument(markdown)
+    add_study_dir_argument(markdown)
+    add_data_path_argument(markdown)
+    add_db_config(markdown)
+    add_verbose_argument(markdown)
+
     return parser

cumulus_library/studies/core/core_templates/observation.sql.jinja

@@ -36,7 +36,6 @@ WITH temp_observation AS (
         odvcc.code AS valuecodeableconcept_code,
         odvcc.code_system AS valuecodeableconcept_code_system,
         odvcc.display AS valuecodeableconcept_display,
-        o.component,
         o.referencerange,
         o.valuequantity,
         o.subject.reference AS subject_ref,
@@ -52,15 +51,12 @@ SELECT
     id,
     category_code,
     category_code_system,
-    component,
     status,
     observation_code,
     observation_code_system,
     interpretation_code,
     interpretation_code_system,
     interpretation_display,
-    referencerange,
-    valuequantity,
     valuecodeableconcept_code,
     valuecodeableconcept_code_system,
     valuecodeableconcept_display,

cumulus_library/studies/core/observation_type.sql

@@ -29,13 +29,10 @@ SELECT
     co.valuecodeableconcept_code,
     co.valuecodeableconcept_code_system,
     co.valuecodeableconcept_display,
-    co.component,
     co.status,
     co.interpretation_code,
     co.interpretation_code_system,
     co.interpretation_display,
-    co.referencerange,
-    co.valuequantity,
     co.obs_date,
     co.obs_week,
     co.obs_month,

cumulus_library/study_parser.py

@@ -5,7 +5,10 @@
 import inspect
 import pathlib
 import sys
+import typing
 
+import pandas
+import pytablewriter
 import toml
 from rich.progress import Progress, TaskID, track
 
@@ -552,6 +555,56 @@ def run_generate_sql(
                 verbose=verbose,
             )
 
+    def run_generate_markdown(
+        self,
+        cursor: databases.DatabaseCursor,
+        schema: str,
+        parser: databases.DatabaseParser = None,
+        verbose: bool = False,
+    ) -> None:
+        """Generates reference SQL from builders listed in the manifest
+
+        :param cursor: A DatabaseCursor object
+        :param schema: The name of the schema to write tables to
+        :param verbose: toggle from progress bar to query output
+        """
+
+        query = base_templates.get_show_tables(
+            schema_name=schema, prefix=f"{self.get_study_prefix()}__"
+        )
+
+        tables = [x[0] for x in cursor.execute(query).fetchall()]
+        query = base_templates.get_column_datatype_query(
+            schema_name=schema, table_name=tables, include_table_names=True
+        )
+        study_df = pandas.DataFrame(
+            cursor.execute(query).fetchall(), columns=["Table", "Column", "Type"]
+        )
+        with open(
+            self._study_path / f"{self.get_study_prefix()}_generated.md", "w"
+        ) as f:
+            table_list = sorted(study_df["Table"].unique())
+            count_tables = [t for t in table_list if "__count_" in t]
+            base_tables = [t for t in table_list if "__count_" not in t]
+            if len(count_tables) > 0:
+                f.write(f"## {self.get_study_prefix()} count tables\n\n")
+                for table in count_tables:
+                    self._write_md_table(table, study_df, f)
+            if len(base_tables) > 0:
+                f.write(f"## {self.get_study_prefix()} base tables\n\n")
+                for table in base_tables:
+                    self._write_md_table(table, study_df, f)
+
+    def _write_md_table(self, name: str, df: pandas.DataFrame, file: typing.IO):
+        table_df = df[df["Table"] == name].drop("Table", axis=1)
+        table_df = table_df.assign(Description="")
+        writer = pytablewriter.MarkdownTableWriter(dataframe=table_df)
+        writer.table_name = name
+        writer.set_indent_level(2)
+        writer.stream = file
+        writer.write_table()
+        file.write("\n")
+
     def build_study(
         self,
         cursor: databases.DatabaseCursor,

cumulus_library/template_sql/base_templates.py

@@ -90,13 +90,22 @@ def get_codeable_concept_denormalize_query(
     )
 
 
-def get_column_datatype_query(schema_name: str, table_name: str, column_names: list):
+# TODO: change test of table_name to plural
+def get_column_datatype_query(
+    schema_name: str,
+    table_name: str | list,
+    column_names: list | None = None,
+    include_table_names: bool | None = False,
+):
     """Gets the in-database data representation of a given column"""
+    if isinstance(table_name, str):
+        table_name = [table_name]
     return get_base_template(
         "column_datatype",
         schema_name=schema_name,
         table_name=table_name,
         column_names=column_names,
+        include_table_names=include_table_names,
     )
 
 

cumulus_library/template_sql/column_datatype.sql.jinja

@@ -1,8 +1,13 @@
 SELECT
+    {%- if include_table_names %}
+    table_name,
+    {% endif %}
     column_name,
     data_type
 FROM information_schema.columns
 WHERE
     table_schema = '{{ schema_name }}'
-    AND table_name = '{{ table_name|lower }}'
+    AND table_name IN ('{{ table_name|join("', '")|lower }}')
+    {%- if column_names %}
     AND LOWER(column_name) IN ('{{ column_names|join("', '")|lower }}') --noqa: LT05
+    {% endif %}