Fix: add tests for categories

Author: lwasser

src/pyosmeta/parse_issues.py

@@ -725,9 +725,14 @@ def get_categories(
             The first comment from the issue split into lines and then the
             lines split as by self.parse_comment()
 
-        fmt : bool
-            Applies some formatting changes to the categories to match what is
-            required for the website.
+        section_str : str
+            The section string to find where the categories live in the review
+            metadata. Example ## Scope contains the package categories such as
+            data viz, etc
+
+        num_vals : int
+            Number of categories expected in the list. for instance
+            3 partner options.
         """
         # Find the starting index of the category section
         # This will be more robust if we use starts_with rather than in i think
@@ -737,8 +742,9 @@ def get_categories(
                 for i, sublist in enumerate(issue_list)
                 if section_str in sublist
             )
-            # Iterate from scope index to first line starting with " - ["
-            # To find list of category check boxes
+            # Iterate starting at the specified section location index
+            # find the first line starting with " - ["
+            # This represents the first category in a list of categories
             for i in range(index + 1, len(issue_list)):
                 if issue_list[i] and issue_list[i][0].startswith("- ["):
                     cat_index = i
@@ -748,10 +754,14 @@ def get_categories(
             return None
 
         # Get checked categories for package
+        # TODO: it would be nice to figure out where the end of the list is
+        # rather than hard coding it
         cat_list = issue_list[cat_index : cat_index + num_vals]
         selected = [
             item[0] for item in cat_list if re.search(r"- \[[xX]\] ", item[0])
         ]
+        # Above returns a list of list elements that are checked
+        # Now, clean the extra markdown and only return the category text
         cleaned = [re.sub(r"- \[[xX]\] ", "", item) for item in selected]
         categories = [
             re.sub(r"(\w+) (\w+)", r"\1-\2", item) for item in cleaned

src/tests/__init__.py

@@ -0,0 +1,67 @@
+import pytest
+
+from pyosmeta.parse_issues import ProcessIssues
+
+checked = [
+    ["Submitting Author", "Nabil Freij (@nabobalis)"],
+    ["- sunpy is a community-developed, free and open-source."],
+    ["## Scope"],
+    ["- Please indicate which category or categories."],
+    ["- [X] Data retrieval"],
+    ["- [ ] Data extraction"],
+    ["- [x] Data Viz"],
+    ["## Domain Specific"],
+    ["something else"],
+]
+
+not_checked = [
+    ["Submitting Author", "Nabil Freij (@nabobalis)"],
+    ["- sunpy is a community-developed, free and open-source."],
+    ["## Scope"],
+    ["- Please indicate which category or categories."],
+    ["- [ ] Data retrieval"],
+    ["- [ ] Data extraction"],
+    ["- [ ] Data Viz"],
+    ["## Domain Specific"],
+    ["something else"],
+]
+
+no_categories = [
+    ["Submitting Author", "Nabil Freij (@nabobalis)"],
+    ["- sunpy is a community-developed, free and open-source."],
+    ["something else"],
+]
+
+
+@pytest.mark.parametrize(
+    "issue_list, expected_categories",
+    [
+        (
+            checked,
+            ["data-retrieval", "data-viz"],
+        ),
+        (
+            not_checked,
+            [],
+        ),
+        (
+            no_categories,
+            None,
+        ),
+    ],
+)
+def test_get_categories(
+    issue_list: list[list[str]], expected_categories: list[str | None]
+):
+    # Initialize your class or use an existing instance
+    issues = ProcessIssues(
+        org="pyopensci",
+        repo_name="software-submission",
+        label_name="presubmission",
+    )
+
+    # Call the get_categories method
+    categories = issues.get_categories(issue_list, "## Scope", 3)
+
+    # Assert the result matches the expected categories
+    assert categories == expected_categories