Improve xpath patterns for spaces after coma in xpath

emmetify/__init__.py

@@ -10,21 +10,21 @@ def emmetify_html(content, format="html", **options):
     return emmetifier.emmetify(content)
 
 
-def emmetify_compact_html(content, **options):
+def emmetify_compact_html(content):
     """Convenience function for quick HTML conversion with simplified tags and attributes"""
-    emmetifier = Emmetifier.create(
+    emmetifier = Emmetifier(
         format="html",
         config={
             "html": {
                 "skip_tags": True,
                 "prioritize_attributes": True,
                 "simplify_classes": True,
                 "simplify_images": True,
+                # LLM agents works better when they know the relative links, otherwise they will start looping on redirects
                 "simplify_relative_links": False,
                 "simplify_absolute_links": True,
             }
         },
-        **options,
     )
     return emmetifier.emmetify(content)
 

emmetify/utils/xpath.py

@@ -10,18 +10,18 @@ def restore_attribute_in_xpath(xpath: str, tag: str, attr: str, replace_map: dic
         rf"@{attr}=(?P<quote>['\"])(?P<value>.*?)(?P=quote)",
         rf"@{attr.lower()}=(?P<quote>['\"])(?P<value>.*?)(?P=quote)",
         # Contains function
-        rf"contains\(@{attr},(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"contains\(@{attr}\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
         # Normalize-space function
         rf"normalize-space\(@{attr}\)=(?P<quote>['\"])(?P<value>.*?)(?P=quote)",
         # Functions with normalize-space
-        rf"contains\(normalize-space\(@{attr}\),(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"contains\(normalize-space\(@{attr}\)\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
         # Nested functions with normalize-space
-        rf"ends-with\(normalize-space\(@{attr}\),(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
-        rf"starts-with\(normalize-space\(@{attr}\),(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"ends-with\(normalize-space\(@{attr}\)\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"starts-with\(normalize-space\(@{attr}\)\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
         # Other functions
-        rf"starts-with\(@{attr},(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
-        rf"ends-with\(@{attr},(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
-        rf"matches\(@{attr},(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"starts-with\(@{attr}\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"ends-with\(@{attr}\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
+        rf"matches\(@{attr}\s*,\s*(?P<quote>['\"])(?P<value>.*?)(?P=quote)\)",
     ]
 
     # Compile patterns with case-insensitive matching

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "emmetify"
-version = "0.0.1-rc.4"
+version = "0.0.1-rc.5"
 description = "Cut LLM costs and boost processing speed by transforming verbose HTML into efficient Emmet notation"
 authors = ["Mariusz Korzekwa <[email protected]>"]
 repository = "https://github.com/emmetify/emmetify-py"

tests/utils/xpath_restore_classes_test.py

@@ -95,6 +95,13 @@ def test_class_with_multiple_predicates_and_functions(self):
         result = restore_classes_in_xpath(xpath, replace_map)
         self.assertEqual(result, expected)
 
+    def test_class_with_multiple_predicates_and_functions_with_space(self):
+        xpath = "//*[@id='test' and contains(@class, 'john')]"
+        replace_map = {"john": "example-class"}
+        expected = "//*[@id='test' and contains(@class, 'example-class')]"
+        result = restore_classes_in_xpath(xpath, replace_map)
+        self.assertEqual(result, expected)
+
     def test_complex_xpath_expression(self):
         xpath = "//div//*[@class='john' and @data-id='123']//span[text()='Click here']"
         replace_map = {"john": "example-class"}
@@ -158,6 +165,13 @@ def test_class_with_function_inside_predicate(self):
         result = restore_classes_in_xpath(xpath, replace_map)
         self.assertEqual(result, expected)
 
+    def test_class_with_function_inside_predicate_with_space(self):
+        xpath = "//*[@class=concat('john', 'doe')]"
+        replace_map = {"johndoe": "example-class"}
+        expected = "//*[@class=concat('john', 'doe')]"
+        result = restore_classes_in_xpath(xpath, replace_map)
+        self.assertEqual(result, expected)
+
     def test_class_with_variable_in_predicate(self):
         xpath = "//*[@class=$john]"
         replace_map = {"$john": "example-class"}
@@ -179,6 +193,13 @@ def test_class_with_comment_in_predicate(self):
         result = restore_classes_in_xpath(xpath, replace_map)
         self.assertEqual(result, expected)
 
+    def test_class_with_double_slash_and_index(self):
+        xpath = "//footer//div[contains(@class, 'omi')]/text()[2]"
+        replace_map = {"omi": "example-class"}
+        expected = "//footer//div[contains(@class, 'example-class')]/text()[2]"
+        result = restore_classes_in_xpath(xpath, replace_map)
+        self.assertEqual(result, expected)
+
     def test_class_with_case_sensitive_tag_name(self):
         xpath = "//DIV[@class='john']"
         replace_map = {"john": "example-class"}