From fe41ddbc51b4b7c6ce7d958ea7622983d7805b7e Mon Sep 17 00:00:00 2001
From: Michael Williamson <mike@zwobble.org>
Date: Sat, 17 Feb 2024 13:52:16 +0000
Subject: Support attributes in HTML paths in style mappings

---
 mammoth/styles/parser/html_path_parser.py | 74 ++++++++++++++++++++++++-------
 1 file changed, 57 insertions(+), 17 deletions(-)

(limited to 'mammoth/styles/parser/html_path_parser.py')

diff --git a/mammoth/styles/parser/html_path_parser.py b/mammoth/styles/parser/html_path_parser.py
index 9e155b6..8ed6fed 100644
--- a/mammoth/styles/parser/html_path_parser.py
+++ b/mammoth/styles/parser/html_path_parser.py
@@ -1,6 +1,15 @@
+import cobble
+
 from ... import html_paths
 from .tokeniser import TokenType
-from .token_parser import parse_identifier, parse_string, try_parse_class_name
+from .token_parser import parse_identifier, parse_string
+
+
+@cobble.data
+class _AttributeOrClassName(object):
+    name = cobble.field()
+    value = cobble.field()
+    append = cobble.field()
 
 
 def parse_html_path(tokens):
@@ -12,26 +21,33 @@ def parse_html_path(tokens):
 
 def _parse_html_path_elements(tokens):
     elements = []
-    
+
     if tokens.peek_token_type() == TokenType.IDENTIFIER:
         elements.append(_parse_element(tokens))
-        
+
         while tokens.try_skip_many(((TokenType.WHITESPACE, None), (TokenType.SYMBOL, ">"))):
             tokens.skip(TokenType.WHITESPACE)
             elements.append(_parse_element(tokens))
-        
+
     return elements
 
 
 def _parse_element(tokens):
     tag_names = _parse_tag_names(tokens)
-    class_names = _parse_class_names(tokens)
+    attributes_list = _parse_attribute_or_class_names(tokens)
     is_fresh = _parse_is_fresh(tokens)
     separator = _parse_separator(tokens)
-    
+
+    attributes = {}
+    for attribute in attributes_list:
+        if attribute.append and attributes.get(attribute.name):
+            attributes[attribute.name] += " " + attribute.value
+        else:
+            attributes[attribute.name] = attribute.value
+
     return html_paths.element(
         tag_names,
-        class_names=class_names,
+        attributes=attributes,
         fresh=is_fresh,
         separator=separator,
     )
@@ -39,24 +55,48 @@ def _parse_element(tokens):
 
 def _parse_tag_names(tokens):
     tag_names = [parse_identifier(tokens)]
-    
+
     while tokens.try_skip(TokenType.SYMBOL, "|"):
         tag_names.append(parse_identifier(tokens))
-    
+
     return tag_names
 
 
-def _parse_class_names(tokens):
-    class_names = []
-    
+def _parse_attribute_or_class_names(tokens):
+    attribute_or_class_names = []
+
     while True:
-        class_name = try_parse_class_name(tokens)
-        if class_name is None:
+        attribute_or_class_name = _try_parse_attribute_or_class_name(tokens)
+        if attribute_or_class_name is None:
             break
         else:
-            class_names.append(class_name)
-    
-    return class_names
+            attribute_or_class_names.append(attribute_or_class_name)
+
+    return attribute_or_class_names
+
+
+def _try_parse_attribute_or_class_name(tokens):
+    if tokens.is_next(TokenType.SYMBOL, "["):
+        return _parse_attribute(tokens)
+    if tokens.is_next(TokenType.SYMBOL, "."):
+        return _parse_class_name(tokens)
+    else:
+        return None
+
+
+def _parse_attribute(tokens):
+    tokens.skip(TokenType.SYMBOL, "[")
+    name = parse_identifier(tokens)
+    tokens.skip(TokenType.SYMBOL, "=")
+    value = parse_string(tokens)
+    tokens.skip(TokenType.SYMBOL, "]")
+    return _AttributeOrClassName(name=name, value=value, append=False)
+
+
+def _parse_class_name(tokens):
+    tokens.skip(TokenType.SYMBOL, ".")
+    class_name = parse_identifier(tokens)
+    return _AttributeOrClassName(name="class", value=class_name, append=True)
 
 
 def _parse_is_fresh(tokens):
-- 
cgit v1.2.3