parser: fix comment parsing

priv-kweihmann · priv-kweihmann · commit bf3a9764fb67 · 2025-09-02T10:14:42.000+02:00
for comments that will transition right into a different kind of block, like functions or python blocks. This will also merge consecutive comments into single comment blocks Closes #317 Signed-off-by: Konrad Weihmann <kweihmann@outlook.com>
diff --git a/oelint_parser/parser.py b/oelint_parser/parser.py
@@ -36,6 +36,7 @@
 __regex_export_wval = regex.compile(r"^\s*?export(\s+|\t+)(?P<name>.+)\s*=\s*\"(?P<value>.*)\"")
 __regex_export_woval = regex.compile(r"^\s*?export(\s+|\t+)(?P<name>.+)\s*$")
 __regex_comments = regex.compile(r"^(\s|\t)*#+\s*(?P<body>.*)")
+__regex_comments_pure = regex.compile(r"^#+\s*(?P<body>.*)")
 __regex_python = regex.compile(r"^(\s*|\t*)def(\s+|\t+)(?P<funcname>[a-z0-9_\-]+)(\s*|\t*)\(.*\)\:")
 __regex_include = regex.compile(r"^(\s*|\t*)(?P<statement>include|require)(\s+|\t+)(?P<incname>[A-za-z0-9\-\./\$\{\}]+)")
 __regex_addtask = regex.compile(
@@ -120,13 +121,28 @@ def iterate(_iter: Iterable, buffer: str) -> tuple[str, str]:
                     _, line = _iter.__next__()
                 except StopIteration:
                     stopiter = True
+        elif RegexRpl.search(__regex_comments_pure, res):
+            stopiter = False
+            while not stopiter:
+                try:
+                    _, line = _iter.__next__()
+                    if not RegexRpl.search(__regex_comments_pure, line):
+                        stopiter = True
+                        next_ = line.strip()
+                    else:
+                        res += line
+                        next_ = ''
+                except StopIteration:
+                    stopiter = True
+                    next_ = ''
         elif res.strip().startswith("def "):
             stopiter = False
             while not stopiter:
                 try:
                     _, line = _iter.__next__()
                 except StopIteration:
                     stopiter = True
+                    next_ = ''
                 if stopiter:
                     break
                 elif RegexRpl.match(__valid_func_name_regex__, line):
diff --git a/tests/pattern/pattern_comment_py.bb b/tests/pattern/pattern_comment_py.bb
@@ -0,0 +1,12 @@
+# Some comment
+# with more than a single line
+def some_python_code():
+    result = ''
+    return result
+
+# Just a comment
+# across multiple lines
+
+do_example() {
+    bbwarn "This is an example warning"
+}
diff --git a/tests/pattern/pattern_comment_py.bb.expect b/tests/pattern/pattern_comment_py.bb.expect
@@ -0,0 +1,52 @@
+[
+    {
+        "Line": 1,
+        "Raw": "# Some comment\n# with more than a single line\n",
+        "InFileLine": 1,
+        "RealRaw": "# Some comment\n# with more than a single line\n",
+        "IsFromClass": false,
+        "OverrideDelimiter": "_",
+        "IsNewStyleOverrideSyntax": false,
+        "InlineBlocks": []
+    },
+    {
+        "FuncName": "some_python_code",
+        "Line": 4,
+        "Raw": "def some_python_code():    result = ''\n    return result\n\n",
+        "InFileLine": 4,
+        "RealRaw": "def some_python_code():    result = ''\n    return result\n\n",
+        "IsFromClass": false,
+        "OverrideDelimiter": "_",
+        "IsNewStyleOverrideSyntax": false,
+        "InlineBlocks": []
+    },
+    {
+        "Line": 8,
+        "Raw": "# Just a comment\n# across multiple lines\n",
+        "InFileLine": 8,
+        "RealRaw": "# Just a comment\n# across multiple lines\n",
+        "IsFromClass": false,
+        "OverrideDelimiter": "_",
+        "IsNewStyleOverrideSyntax": false,
+        "InlineBlocks": []
+    },
+    {
+        "IsPython": false,
+        "IsFakeroot": false,
+        "FuncName": "do_example",
+        "FuncNameComplete": "do_example",
+        "SubItem": "",
+        "SubItems": [],
+        "FuncBody": "    bbwarn \"This is an example warning\"",
+        "FuncBodyStripped": "bbwarn \"This is an example warning\"",
+        "FuncBodyRaw": "\nbbwarn \"This is an example warning\"\n",
+        "Line": 10,
+        "Raw": "do_example() {\n    bbwarn \"This is an example warning\"\n}\n",
+        "InFileLine": 10,
+        "RealRaw": "do_example() {\n    bbwarn \"This is an example warning\"\n}\n",
+        "IsFromClass": false,
+        "OverrideDelimiter": "_",
+        "IsNewStyleOverrideSyntax": false,
+        "InlineBlocks": []
+    }
+]
diff --git a/tests/test_parser.py b/tests/test_parser.py
@@ -491,14 +491,14 @@ def test_append_to_stash(self):
 
         _stash = self.__stash.GetItemsFor(classifier=Comment.CLASSIFIER)
 
-        assert len(_stash) == 2
+        assert len(_stash) == 1
 
         new_comment = Comment('foo', 1, 1, '# abc', '# def', [])
         self.__stash.Append(new_comment)
 
         _stash = self.__stash.GetItemsFor(classifier=Comment.CLASSIFIER)
 
-        assert len(_stash) == 3
+        assert len(_stash) == 2
 
     def test_inherit(self):
         from oelint_parser.cls_item import Inherit
diff --git a/tests/test_pattern.py b/tests/test_pattern.py
@@ -9,6 +9,7 @@
                              'pattern_var.bb',
                              'pattern_python.bb',
                              'pattern_function.bb',
+                             'pattern_comment_py.bb',
                          ],
                          )
 def test_pattern(file):

Original file line number	Diff line number	Diff line change
`@@ -9,6 +9,7 @@`
`9`	`9`	`'pattern_var.bb',`
`10`	`10`	`'pattern_python.bb',`
`11`	`11`	`'pattern_function.bb',`
	`12`	`+ 'pattern_comment_py.bb',`
`12`	`13`	`],`
`13`	`14`	`)`
`14`	`15`	`def test_pattern(file):`