Merge pull request #41 from ewdlop/ewdlop-patch-33-Create-merge_html_elements.py

Create merge_html_elements.py

Author: ewdlop

Python/merge_html_elements.py

@@ -0,0 +1,62 @@
+from lxml import html, etree
+from itertools import zip_longest
+
+def merge_elements(a: etree._Element, b: etree._Element):
+    """
+    Lazy-merge 兩個 <Element>，相同標籤才往下比對。
+    每次合併完成後立即 yield，讓呼叫端自行處理。
+    """
+    if a is None and b is None:
+        return                     # 兩邊都沒東西
+    if a is None:
+        yield b                    # 只有右邊
+        return
+    if b is None:
+        yield a                    # 只有左邊
+        return
+
+    # 標籤一致才真正合併；否則視為差異，各自輸出
+    if a.tag != b.tag:
+        yield a
+        yield b
+        return
+
+    # ---------- 1) 建立新的節點 ----------
+    merged = etree.Element(a.tag)
+
+    # ---------- 2) 合併屬性 ----------
+    merged.attrib.update(a.attrib)         # 先左
+    merged.attrib.update(b.attrib)         # 再右（右邊覆蓋）
+
+    # ---------- 3) 合併文字 ----------
+    merged.text = (a.text or '') + (b.text or '')
+
+    # ---------- 4) 逐子節點遞迴 ----------
+    for left_child, right_child in zip_longest(list(a), list(b)):
+        for sub in merge_elements(left_child, right_child):
+            merged.append(sub)
+
+    # ---------- 5) 把結果交出去 ----------
+    yield merged
+
+
+# --------- 使用範例 ---------
+html1 = """<ul>
+  <li id="a">apple</li>
+  <li id="b">banana</li>
+</ul>"""
+
+html2 = """<ul>
+  <li id="b">BANANA</li>
+  <li id="c">cherry</li>
+</ul>"""
+
+dom1 = html.fromstring(html1)
+dom2 = html.fromstring(html2)
+
+# 只取第一層 <ul>
+merged_ul, = merge_elements(dom1, dom2)    # 注意逗號 (unpack generator)
+
+print(html.tostring(merged_ul,
+                    encoding="unicode",
+                    pretty_print=True))