feat(mdx-storage): Phase2 Task2.2 Figure → ac:image 변환 구현

confluence-mdx/bin/mdx_to_storage/emitter.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import os
 import re
 from typing import Optional
 
@@ -69,6 +70,9 @@ def emit_block(block: Block, context: Optional[dict] = None) -> str:
     if block.type == "callout":
         return _emit_callout(block, context)
 
+    if block.type == "figure":
+        return _emit_figure(block)
+
     return ""
 
 
@@ -168,3 +172,24 @@ def _parse_callout_children_from_content(content: str) -> list[Block]:
         return []
     from .parser import parse_mdx
     return parse_mdx(inner_text)
+
+
+def _emit_figure(block: Block) -> str:
+    src = block.attrs.get("src", "").strip()
+    if not src:
+        return ""
+
+    filename = os.path.basename(src)
+    width = block.attrs.get("width", "").strip()
+    caption = block.attrs.get("caption", "").strip()
+
+    attrs = ['ac:align="center"']
+    if width:
+        attrs.append(f'ac:width="{width}"')
+
+    parts = [f"<ac:image {' '.join(attrs)}>"]
+    parts.append(f'<ri:attachment ri:filename="{filename}"></ri:attachment>')
+    if caption:
+        parts.append(f"<ac:caption><p>{convert_inline(caption)}</p></ac:caption>")
+    parts.append("</ac:image>")
+    return "".join(parts)

confluence-mdx/bin/mdx_to_storage/parser.py

@@ -232,9 +232,20 @@ def _parse_figure_attrs(content: str) -> dict:
     for key, v1, v2 in _FIGURE_IMG_ATTR_PATTERN.findall(img_match.group(1)):
         value = v1 or v2
         attrs[key] = value
+    caption = _extract_figure_caption(content)
+    if caption:
+        attrs["caption"] = caption
     return attrs
 
 
+def _extract_figure_caption(content: str) -> str:
+    match = re.search(r"<figcaption[^>]*>(.*?)</figcaption>", content, flags=re.DOTALL)
+    if not match:
+        return ""
+    inner = re.sub(r"<[^>]+>", "", match.group(1))
+    return inner.strip()
+
+
 def _parse_html_block(lines: list[str], start: int) -> tuple[Block, int]:
     i = start + 1
     while i < len(lines):

confluence-mdx/tests/test_mdx_to_storage/test_emitter.py

@@ -221,3 +221,76 @@ def test_emit_callout_body_multiple_paragraphs():
     assert "<p>First paragraph.</p>" in xhtml
     assert "<p>Second paragraph.</p>" in xhtml
     assert '<ac:structured-macro ac:name="info">' in xhtml
+
+
+def test_emit_figure_to_ac_image_with_width():
+    mdx = """<figure>
+  <img src="/images/path/sample.png" alt="Sample" width="700" data-layout="center">
+</figure>
+"""
+    xhtml = emit_document(parse_mdx(mdx))
+    assert (
+        xhtml
+        == '<ac:image ac:align="center" ac:width="700"><ri:attachment ri:filename="sample.png"></ri:attachment></ac:image>'
+    )
+
+
+def test_emit_figure_with_caption():
+    mdx = """<figure>
+  <img src="/images/path/sample.png" alt="Sample">
+  <figcaption>This is **caption**</figcaption>
+</figure>
+"""
+    xhtml = emit_document(parse_mdx(mdx))
+    assert '<ac:image ac:align="center">' in xhtml
+    assert '<ri:attachment ri:filename="sample.png"></ri:attachment>' in xhtml
+    assert "<ac:caption><p>This is <strong>caption</strong></p></ac:caption>" in xhtml
+
+
+def test_emit_figure_without_src_is_skipped():
+    block = Block(type="figure", content="<figure></figure>", attrs={})
+    assert emit_block(block) == ""
+
+
+def test_emit_figure_without_width_minimal():
+    """Figure with src only (no width, no caption) → ac:image without ac:width."""
+    mdx = """<figure>
+  <img src="/images/sample.png" alt="Alt">
+</figure>
+"""
+    xhtml = emit_document(parse_mdx(mdx))
+    assert (
+        xhtml
+        == '<ac:image ac:align="center"><ri:attachment ri:filename="sample.png"></ri:attachment></ac:image>'
+    )
+    assert "ac:width" not in xhtml
+
+
+def test_emit_figure_in_mixed_document():
+    """Figure block integrated with other blocks in a full document."""
+    mdx = """---
+title: "Page"
+---
+
+# Page
+
+## Overview
+
+Some text.
+
+<figure>
+  <img src="/images/path/diagram.png" alt="Diagram" width="600">
+  <figcaption>Architecture diagram</figcaption>
+</figure>
+
+More text after image.
+"""
+    xhtml = emit_document(parse_mdx(mdx))
+    assert "<h1>Overview</h1>" in xhtml
+    assert "<p>Some text.</p>" in xhtml
+    assert '<ac:image ac:align="center" ac:width="600">' in xhtml
+    assert '<ri:attachment ri:filename="diagram.png">' in xhtml
+    assert "<ac:caption><p>Architecture diagram</p></ac:caption>" in xhtml
+    assert "<p>More text after image.</p>" in xhtml
+    # frontmatter and page title should be excluded
+    assert "Page" not in xhtml

confluence-mdx/tests/test_mdx_to_storage/test_parser.py

@@ -63,6 +63,29 @@ def test_parse_figure_block_and_img_attrs():
     }
 
 
+def test_parse_figure_with_caption():
+    text = """<figure>
+  <img src="/images/sample.png" alt="Sample" width="700" data-layout="center">
+  <figcaption>Caption <strong>text</strong></figcaption>
+</figure>
+"""
+    blocks = parse_mdx(text)
+    assert blocks[0].type == "figure"
+    assert blocks[0].attrs["caption"] == "Caption text"
+
+
+def test_parse_figure_self_closing_img():
+    """Self-closing <img ... /> tag should parse identically."""
+    text = """<figure>
+  <img src="/images/sample.png" alt="Sample" width="500" />
+</figure>
+"""
+    blocks = parse_mdx(text)
+    assert blocks[0].type == "figure"
+    assert blocks[0].attrs["src"] == "/images/sample.png"
+    assert blocks[0].attrs["width"] == "500"
+
+
 def test_parse_paragraph_fallback():
     blocks = parse_mdx("one\ntwo\n\n")
     assert blocks[0].type == "paragraph"