refactor: make less messy

src/markdown/inlines.py

@@ -0,0 +1,92 @@
+from typing import Callable
+
+from src.markdown.extract import extract_markdown_images, extract_markdown_links
+from src.nodes.text import TextNode, TextType
+
+
+def split_nodes_delimiter(
+    old_nodes: list[TextNode], delimiter: str, text_type: TextType
+) -> list[TextNode]:
+    result = []
+
+    for node in old_nodes:
+        if node.text_type != TextType.PLAIN:
+            result.append(node)
+            continue
+
+        parts = node.text.split(delimiter)
+
+        if len(parts) % 2 == 0:
+            raise RuntimeError("No matching delimiter found")
+
+        for i in range(len(parts)):
+            current_text = parts[i]
+
+            if i % 2 == 0:
+                result.append(TextNode(current_text, TextType.PLAIN))
+            else:
+                result.append(TextNode(current_text, text_type))
+
+    return result
+
+
+def split_nodes_image(old_nodes: list[TextNode]) -> list[TextNode]:
+    return split_nodes(
+        old_nodes,
+        extracter=extract_markdown_images,
+        pattern_builder=lambda t: f"![{t[0]}]({t[1]})",
+        node_creator=lambda t: TextNode(t[0], TextType.IMAGE, t[1]),
+    )
+
+
+def split_nodes_link(old_nodes: list[TextNode]) -> list[TextNode]:
+    return split_nodes(
+        old_nodes,
+        extracter=extract_markdown_links,
+        pattern_builder=lambda t: f"[{t[0]}]({t[1]})",
+        node_creator=lambda t: TextNode(t[0], TextType.LINK, t[1]),
+    )
+
+
+def split_nodes(
+    old_nodes: list[TextNode],
+    extracter: Callable[[str], list[tuple[str, str]]],
+    pattern_builder: Callable[[tuple[str, str]], str],
+    node_creator: Callable[[tuple[str, str]], TextNode],
+) -> list[TextNode]:
+    result = []
+
+    for node in old_nodes:
+        if node.text_type != TextType.PLAIN:
+            result.append(node)
+            continue
+
+        extracted = extracter(node.text)
+
+        node_text = node.text
+
+        for match in extracted:
+            pattern = pattern_builder(match)
+            parts = node_text.split(pattern, maxsplit=1)
+
+            result.append(TextNode(parts[0], TextType.PLAIN))
+
+            matched_node = node_creator(match)
+            result.append(matched_node)
+
+            node_text = parts[1]
+
+        if len(node_text) > 0:
+            result.append(TextNode(node_text, TextType.PLAIN))
+
+    return result
+
+
+def text_to_text_node(text: str) -> list[TextNode]:
+    old_nodes = [TextNode(text, TextType.PLAIN)]
+    old_nodes = split_nodes_delimiter(old_nodes, "**", TextType.BOLD)
+    old_nodes = split_nodes_delimiter(old_nodes, "_", TextType.ITALIC)
+    old_nodes = split_nodes_delimiter(old_nodes, "`", TextType.CODE)
+    old_nodes = split_nodes_image(old_nodes)
+    old_nodes = split_nodes_link(old_nodes)
+    return old_nodes