v0.5.2

Что то сделал

word2md/omml.py

@@ -0,0 +1,149 @@
+"""Best-effort OMML → readable LaTeX-like string for $$ / $ fences."""
+
+from __future__ import annotations
+
+_OMML_NS = "http://schemas.openxmlformats.org/officeDocument/2006/math"
+
+
+def _local(tag: str) -> str:
+    if "}" in tag:
+        return tag.rsplit("}", 1)[-1]
+    return tag
+
+
+def omml_to_latex(element) -> str:
+    """Flatten OMML tree to a readable formula string (not perfect LaTeX)."""
+    if element is None:
+        return ""
+    parts: list[str] = []
+    _walk(element, parts)
+    text = "".join(parts)
+    text = " ".join(text.split())
+    return text.strip()
+
+
+def _walk(node, parts: list[str]) -> None:
+    tag = _local(node.tag)
+
+    if tag == "t":
+        parts.append(node.text or "")
+        return
+
+    if tag == "f":  # fraction
+        num = _child(node, "num")
+        den = _child(node, "den")
+        parts.append(r"\frac{")
+        if num is not None:
+            _walk(num, parts)
+        parts.append("}{")
+        if den is not None:
+            _walk(den, parts)
+        parts.append("}")
+        return
+
+    if tag == "sSup":
+        base = _child(node, "e")
+        sup = _child(node, "sup")
+        if base is not None:
+            _walk(base, parts)
+        parts.append("^{")
+        if sup is not None:
+            _walk(sup, parts)
+        parts.append("}")
+        return
+
+    if tag == "sSub":
+        base = _child(node, "e")
+        sub = _child(node, "sub")
+        if base is not None:
+            _walk(base, parts)
+        parts.append("_{")
+        if sub is not None:
+            _walk(sub, parts)
+        parts.append("}")
+        return
+
+    if tag == "sSubSup":
+        base = _child(node, "e")
+        sub = _child(node, "sub")
+        sup = _child(node, "sup")
+        if base is not None:
+            _walk(base, parts)
+        parts.append("_{")
+        if sub is not None:
+            _walk(sub, parts)
+        parts.append("}^{")
+        if sup is not None:
+            _walk(sup, parts)
+        parts.append("}")
+        return
+
+    if tag == "rad":
+        deg = _child(node, "deg")
+        e = _child(node, "e")
+        if deg is not None and "".join(deg.itertext()).strip():
+            parts.append(r"\sqrt[")
+            _walk(deg, parts)
+            parts.append("]{")
+        else:
+            parts.append(r"\sqrt{")
+        if e is not None:
+            _walk(e, parts)
+        parts.append("}")
+        return
+
+    if tag == "nary":
+        for child in node:
+            _walk(child, parts)
+        return
+
+    if tag in ("oMath", "oMathPara", "e", "num", "den", "sup", "sub", "deg",
+               "fPr", "ctrlPr", "rPr", "sSupPr", "sSubPr", "sSubSupPr", "radPr",
+               "naryPr", "dPr", "boxPr", "argSz"):
+        for child in node:
+            _walk(child, parts)
+        return
+
+    if tag == "r":
+        for child in node:
+            _walk(child, parts)
+        return
+
+    if tag == "d":  # delimiter
+        parts.append("(")
+        for child in node:
+            if _local(child.tag) != "dPr":
+                _walk(child, parts)
+        parts.append(")")
+        return
+
+    if tag in ("pPr", "rFonts", "sz", "szCs", "color", "jc"):
+        return
+
+    for child in node:
+        _walk(child, parts)
+    if node.tail:
+        parts.append(node.tail)
+
+
+def _child(node, local_name: str):
+    for child in node:
+        if _local(child.tag) == local_name:
+            return child
+    return None
+
+
+def find_omml(paragraph_element) -> list:
+    """Return all m:oMath / m:oMathPara elements under a paragraph."""
+    if paragraph_element is None:
+        return []
+    found = []
+    for el in paragraph_element.iter():
+        tag = _local(el.tag)
+        if tag in ("oMath", "oMathPara"):
+            found.append(el)
+    return found
+
+
+def paragraph_has_omml(paragraph_element) -> bool:
+    return bool(find_omml(paragraph_element))