Skip to content

⚡ Optimize EquationParser by pre-compiling RegExps #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
smjxpro wants to merge 1 commit into
base: main
Choose a base branch
from
Open

⚡ Optimize EquationParser by pre-compiling RegExps #2

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
113 changes: 48 additions & 65 deletions lib/src/equation_parser.dart
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -306,6 +306,37 @@ class EquationParser {
),
};

// ═════════════════════════════════════════════════════════════════════════
// PRE-COMPILED REGULAR EXPRESSIONS (Optimization)
// ═════════════════════════════════════════════════════════════════════════

static final _textRegExp = RegExp(r'\\text\{([^}]*)\}');
static final _styleRegExp = RegExp(
r'\\(mathrm|mathbf|mathit|mathsf|mathtt|bold|invisible)\{([^}]*)\}',
);
static final _bracketRegExp = RegExp(r'\\(left|right|big|Big|bigg|Bigg)');
static final _decorationRegExp = RegExp(
r'\\(vec|hat|bar|dot|ddot|tilde)\{([^}]*)\}',
);
static final _decorationSpacedRegExp = RegExp(
r'\\(vec|hat|bar|dot|ddot|tilde)\s+([a-zA-Z])',
);
static final _limRegExp = RegExp(r'\\lim_\{([^}]*)\}');
static final _exponentRegExp = RegExp(r'\^{([^}]*)}');
static final _subscriptRegExp = RegExp(r'_\{([^}]*)\}');
static final _fracRegExp = RegExp(r'\\frac\{([^}]*)\}\{([^}]*)\}');
static final _sqrtNRegExp = RegExp(r'\\sqrt\[([^\]]*)\]\{([^}]*)\}');
static final _sqrtRegExp = RegExp(r'\\sqrt\{([^}]*)\}');
static final _deltaRegExp = RegExp(r'(\b[Dd]elta)\s+([a-zA-Z])');
static final _iRegExp = RegExp(r'(?<![a-zA-Z])i(?![a-zA-Z])');
static final _implicitParenRegExp = RegExp(r'\)([a-zA-Z0-9(])');
static final _implicitDigitParenRegExp = RegExp(r'(\d)\(');

/// Combined regex for common LaTeX math functions.
static final _functionRegExp = RegExp(
r'\\(asin|acos|atan|sinh|cosh|tanh|sin|cos|tan|sqrt|exp|log|abs|sec|csc|cot|lim|pow)(?![a-zA-Z])',
);

/// Returns the constant definition for a given name.
static NaturalConstant? getConstant(String name) {
// Ultra-Strict lookup: ONLY look for keys in naturalConstants.
Expand Down Expand Up @@ -469,11 +500,11 @@ class EquationParser {
// would be expected but isn't supported by the parser.
// Cases: )( , )letter, )digit, digit(, letter(
substitutedExpr = substitutedExpr.replaceAllMapped(
RegExp(r'\)([a-zA-Z0-9(])'),
_implicitParenRegExp,
(m) => ')*${m.group(1)}',
);
substitutedExpr = substitutedExpr.replaceAllMapped(
RegExp(r'(\d)\('),
_implicitDigitParenRegExp,
(m) => '${m.group(1)}*(',
);

Expand Down Expand Up @@ -630,29 +661,18 @@ class EquationParser {

// 1. Remove common LaTeX style commands
// \text{...} -> remove entirely (content often units/labels)
expr = expr.replaceAllMapped(RegExp(r'\\text\{([^}]*)\}'), (m) => '');
expr = expr.replaceAllMapped(_textRegExp, (m) => '');
// Others -> keep content
expr = expr.replaceAllMapped(
RegExp(
r'\\(mathrm|mathbf|mathit|mathsf|mathtt|bold|invisible)\{([^}]*)\}',
),
(m) => m[2]!,
);
expr = expr.replaceAll(RegExp(r'\\(left|right|big|Big|bigg|Bigg)'), '');
expr = expr.replaceAllMapped(_styleRegExp, (m) => m[2]!);
expr = expr.replaceAll(_bracketRegExp, '');
expr = expr.replaceAll(r'\,', ' ');
expr = expr.replaceAll(r'\:', ' ');
expr = expr.replaceAll(r'\;', ' ');
expr = expr.replaceAll(r'\!', '');

// 1.5 Remove decorations: \vec{x}, \hat{x}, \bar{x}, \dot{x}, \ddot{x}
expr = expr.replaceAllMapped(
RegExp(r'\\(vec|hat|bar|dot|ddot|tilde)\{([^}]*)\}'),
(m) => '${m[2]}',
);
expr = expr.replaceAllMapped(
RegExp(r'\\(vec|hat|bar|dot|ddot|tilde)\s+([a-zA-Z])'),
(m) => '${m[2]}',
);
expr = expr.replaceAllMapped(_decorationRegExp, (m) => '${m[2]}');
expr = expr.replaceAllMapped(_decorationSpacedRegExp, (m) => '${m[2]}');

// 2. Constants/Operators
expr = expr.replaceAll(r'\cdot', '*');
Expand All @@ -665,64 +685,33 @@ class EquationParser {
expr = expr.replaceAll(r'\rightarrow', ' ');

// 3. Limits
expr = expr.replaceAllMapped(RegExp(r'\\lim_\{([^}]*)\}'), (m) => 'lim ');
expr = expr.replaceAllMapped(_limRegExp, (m) => 'lim ');
expr = expr.replaceAll(r' \lim ', ' lim ');
expr = expr.replaceAll(r'\lim ', 'lim ');

// 4. Exponents and Subscripts (removes {} - do this BEFORE fractions to handle nested braces in exp)
expr = expr.replaceAllMapped(RegExp(r'\^{([^}]*)}'), (m) => '^(${m[1]})');
expr = expr.replaceAllMapped(RegExp(r'_\{([^}]*)\}'), (m) => '_${m[1]}');
expr = expr.replaceAllMapped(_exponentRegExp, (m) => '^(${m[1]})');
expr = expr.replaceAllMapped(_subscriptRegExp, (m) => '_${m[1]}');

// 5. Fractions (Recursive, removes {})
final fracPattern = RegExp(r'\\frac\{([^}]*)\}\{([^}]*)\}');
while (expr.contains(r'\frac{')) {
final oldExpr = expr;
expr = expr.replaceAllMapped(fracPattern, (m) => '(${m[1]})/(${m[2]})');
expr = expr.replaceAllMapped(_fracRegExp, (m) => '(${m[1]})/(${m[2]})');
if (oldExpr == expr) break;
}

// 6. Roots (removes {})
// \sqrt[n]{x}
expr = expr.replaceAllMapped(
RegExp(r'\\sqrt\[([^\]]*)\]\{([^}]*)\}'),
_sqrtNRegExp,
(m) => '(${m[2]})^(1/${m[1]})',
);
// \sqrt{x}
expr = expr.replaceAllMapped(
RegExp(r'\\sqrt\{([^}]*)\}'),
(m) => 'sqrt(${m[1]})',
);
expr = expr.replaceAllMapped(_sqrtRegExp, (m) => 'sqrt(${m[1]})');

// 7. Replace literal functions (e.g. \sin -> sin)
// Note: \sin is often space separated, but might be attached if braces were removed?
// Usually \sin x or \sin(x).
for (final f in [
'asin',
'acos',
'atan',
'sinh',
'cosh',
'tanh',
'sin',
'cos',
'tan',
'sqrt',
'exp',
'log',
'abs',
'sec',
'csc',
'cot',
'lim',
'pow',
]) {
expr = expr.replaceAll('\\$f ', '$f ');
expr = expr.replaceAll('\\$f(', '$f(');
// Also catch cases where space might have been removed or not present
// e.g. \sinx -> sinx (then split).
// Use regex to be safe
expr = expr.replaceAllMapped(RegExp('\\\\$f(?![a-zA-Z])'), (m) => f);
}
// Use optimized combined regex for all common math functions.
expr = expr.replaceAllMapped(_functionRegExp, (m) => m.group(1)!);

// Greek letters
final greekMap = {
Expand Down Expand Up @@ -764,10 +753,7 @@ class EquationParser {
greekMap.forEach((k, v) => expr = expr.replaceAll(k, v));

// Special case for Delta/delta as prefix for change (Delta t)
expr = expr.replaceAllMapped(
RegExp(r'(\b[Dd]elta)\s+([a-zA-Z])'),
(m) => '${m[1]}${m[2]}',
);
expr = expr.replaceAllMapped(_deltaRegExp, (m) => '${m[1]}${m[2]}');

// Implicit Multiplication

Expand All @@ -777,10 +763,7 @@ class EquationParser {

// Normalize standalone 'i' to 'IN' (imaginary unit internal token)
// This ensures consistency between extractVariables and substitution
expr = expr.replaceAllMapped(
RegExp(r'(?<![a-zA-Z])i(?![a-zA-Z])'),
(m) => 'IN',
);
expr = expr.replaceAllMapped(_iRegExp, (m) => 'IN');

return expr;
}
Expand Down