Move html path reading into separate module

1 files changed, 79 insertions, 0 deletions

diff --git a/mammoth/style_reader/html_path_reader.py b/mammoth/style_reader/html_path_reader.py
new file mode 100644
index 0000000..0de6e8c
--- /dev/null
+++ b/mammoth/style_reader/html_path_reader.py
@@ -0,0 +1,79 @@
+from parsimonious.grammar import Grammar
+
+from .. import html_paths
+
+
+def read_html_path(string):
+    path_node = _grammar.parse(string)
+    return _read_path_node(path_node)
+    
+
+
+def _read_path_node(path_node):
+    elements = [
+        _read_element_node(child)
+        for child in _repeated_children_with_separator(path_node, has_whitespace=True)
+    ]
+    return html_paths.path(elements)
+
+
+def _read_element_node(node):
+    tag_names = _read_tag_names_node(node.children[0])
+    class_names = _read_class_names_node(node.children[1])
+    fresh = _read_fresh_node(node.children[2])
+    return html_paths.element(tag_names, class_names=class_names, fresh=fresh)
+
+
+def _read_tag_names_node(node):
+    return [
+        child.text
+        for child in _repeated_children_with_separator(node, has_whitespace=False)
+    ]
+
+
+def _read_class_names_node(class_names_node):
+    return [
+        _read_class_name_node(node)
+        for node in class_names_node.children
+    ]
+
+
+def _read_class_name_node(node):
+    return node.children[1].text
+
+
+def _read_fresh_node(node):
+    return len(node.children) > 0
+
+
+def _repeated_children_with_separator(node, has_whitespace):
+    yield node.children[0]
+    
+    if has_whitespace:
+        sequence_node_index = 2
+    else:
+        sequence_node_index = 1
+    
+    sequence_node = node.children[sequence_node_index]
+    for child in sequence_node.children:
+        yield child.children[sequence_node_index]
+
+
+_grammar = Grammar(r"""
+
+path = element whitespace* (">" whitespace* element)*
+
+element = tag_names class_name* fresh?
+
+tag_names = identifier ("|" identifier)*
+
+class_name = "." identifier
+
+fresh = ":fresh"
+
+identifier = ~"[A-Z0-9]*"i
+
+whitespace = ~"\s"*
+
+""")
+