Add Unicode math formatter

2026-09-11 13:55:54 +02:00
parent 5e938fe456
commit 6cb34ead26
1 changed files with 134 additions and 0 deletions
@@ -0,0 +1,134 @@
+from __future__ import annotations
+
+import re
+
+GREEK_AND_SYMBOLS = {
+    r"\alpha": "α",
+    r"\beta": "β",
+    r"\gamma": "γ",
+    r"\delta": "δ",
+    r"\epsilon": "ε",
+    r"\varepsilon": "ε",
+    r"\zeta": "ζ",
+    r"\eta": "η",
+    r"\theta": "θ",
+    r"\vartheta": "ϑ",
+    r"\iota": "ι",
+    r"\kappa": "κ",
+    r"\lambda": "λ",
+    r"\mu": "μ",
+    r"\nu": "ν",
+    r"\xi": "ξ",
+    r"\pi": "π",
+    r"\rho": "ρ",
+    r"\sigma": "σ",
+    r"\tau": "τ",
+    r"\upsilon": "υ",
+    r"\phi": "φ",
+    r"\varphi": "φ",
+    r"\chi": "χ",
+    r"\psi": "ψ",
+    r"\omega": "ω",
+    r"\Gamma": "Γ",
+    r"\Delta": "Δ",
+    r"\Theta": "Θ",
+    r"\Lambda": "Λ",
+    r"\Xi": "Ξ",
+    r"\Pi": "Π",
+    r"\Sigma": "Σ",
+    r"\Phi": "Φ",
+    r"\Psi": "Ψ",
+    r"\Omega": "Ω",
+    r"\infty": "∞",
+    r"\leq": "≤",
+    r"\le": "≤",
+    r"\geq": "≥",
+    r"\ge": "≥",
+    r"\neq": "≠",
+    r"\ne": "≠",
+    r"\approx": "≈",
+    r"\times": "×",
+    r"\cdot": "·",
+    r"\pm": "±",
+    r"\mp": "∓",
+    r"\to": "→",
+    r"\rightarrow": "→",
+    r"\leftarrow": "←",
+    r"\Rightarrow": "⇒",
+    r"\Leftrightarrow": "⇔",
+    r"\forall": "∀",
+    r"\exists": "∃",
+    r"\in": "∈",
+    r"\notin": "∉",
+    r"\subset": "⊂",
+    r"\subseteq": "⊆",
+    r"\cup": "∪",
+    r"\cap": "∩",
+    r"\emptyset": "∅",
+    r"\mathbb{R}": "ℝ",
+    r"\mathbb{C}": "ℂ",
+    r"\mathbb{N}": "ℕ",
+    r"\mathbb{Z}": "ℤ",
+    r"\mathbb{Q}": "ℚ",
+    r"\int": "∫",
+    r"\sum": "∑",
+    r"\prod": "∏",
+    r"\partial": "∂",
+    r"\nabla": "∇",
+}
+
+SUPERSCRIPT = str.maketrans({
+    "0": "⁰", "1": "¹", "2": "²", "3": "³", "4": "⁴",
+    "5": "⁵", "6": "⁶", "7": "⁷", "8": "⁸", "9": "⁹",
+    "+": "⁺", "-": "⁻", "=": "⁼", "(": "⁽", ")": "⁾",
+    "n": "ⁿ", "i": "ⁱ",
+})
+
+SUBSCRIPT = str.maketrans({
+    "0": "₀", "1": "₁", "2": "₂", "3": "₃", "4": "₄",
+    "5": "₅", "6": "₆", "7": "₇", "8": "₈", "9": "₉",
+    "+": "₊", "-": "₋", "=": "₌", "(": "₍", ")": "₎",
+    "a": "ₐ", "e": "ₑ", "h": "ₕ", "i": "ᵢ", "j": "ⱼ", "k": "ₖ",
+    "l": "ₗ", "m": "ₘ", "n": "ₙ", "o": "ₒ", "p": "ₚ", "r": "ᵣ",
+    "s": "ₛ", "t": "ₜ", "u": "ᵤ", "v": "ᵥ", "x": "ₓ",
+})
+
+_GROUP_PATTERN = re.compile(r"([_^])\(([^()]+)\)|([_^])\{([^{}]+)\}|([_^])([A-Za-z0-9+\-=])")
+
+
+def _translate_script(value: str, marker: str) -> str:
+    table = SUPERSCRIPT if marker == "^" else SUBSCRIPT
+    converted = value.translate(table)
+    return converted if converted != value else marker + value
+
+
+def _replace_script(match: re.Match[str]) -> str:
+    marker = match.group(1) or match.group(3) or match.group(5)
+    value = match.group(2) or match.group(4) or match.group(6)
+    return _translate_script(value, marker)
+
+
+def format_math_text(text: str, mode: str = "plain") -> str:
+    r"""Convert common UnicodeMath/LaTeX-like linear math to Unicode plain text.
+
+    This is meant for Word normal text insertion. It makes expressions such as
+    z_1, x^2, \alpha and \infty render as z₁, x², α and ∞ without requiring the
+    user to manually open Word's equation editor for every expression.
+    """
+    if mode == "plain":
+        return text
+    if mode not in {"unicode", "unicode_math"}:
+        raise ValueError("math_text_format doit être 'plain' ou 'unicode'")
+
+    formatted = text
+    # Replace longer commands first so \subseteq wins before \subset.
+    for command, replacement in sorted(GREEK_AND_SYMBOLS.items(), key=lambda item: len(item[0]), reverse=True):
+        formatted = formatted.replace(command, replacement)
+
+    formatted = re.sub(r"\\vec\(([^()]+)\)", lambda match: match.group(1) + "⃗", formatted)
+    formatted = formatted.replace(r"\sqrt", "√")
+    previous = None
+    while previous != formatted:
+        previous = formatted
+        formatted = _GROUP_PATTERN.sub(_replace_script, formatted)
+    return formatted