Skip to content
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

Correct Regular Expressions Behavior Related to Annex B #58320

Merged
merged 12 commits into from
May 29, 2024
Merged

Correct Regular Expressions Behavior Related to Annex B #58320

Show file tree
Hide file tree
Changes from 7 commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
e049438
Correct RegExp Behavior Related to Annex B
graphemecluster Apr 25, 2024
358eb30
Fix Comment
graphemecluster Apr 25, 2024
8facb0a
Correct Quantifiability of `{1` in Annex B
graphemecluster Apr 27, 2024
f5c0b60
`\c` is an Annex B thing
graphemecluster Apr 27, 2024
cff993f
Add Tests
graphemecluster Apr 27, 2024
603c3cf
Merge from main
rbuckton May 16, 2024
2e62d25
Apply Suggested Changes
graphemecluster May 21, 2024
ed08ef7
Apply Suggested Changes
graphemecluster May 24, 2024
8b67d77
Merge branch 'main' into regex-syntax-check-annexB
graphemecluster May 24, 2024
b48f0d0
Provide User-Friendly Errors for Decimal Escapes Inside Character Cla…
graphemecluster May 24, 2024
c72f92f
Fix Behavior Related to `namedCaptureGroups`
graphemecluster May 28, 2024
70a3214
Clarify why nested character classes should not be considered in the …
graphemecluster May 28, 2024
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
72 changes: 46 additions & 26 deletions src/compiler/scanner.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1556,9 +1556,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
tokenFlags |= TokenFlags.ContainsInvalidEscape;
if (isRegularExpression || shouldEmitInvalidEscapeError) {
const code = parseInt(text.substring(start + 1, pos), 8);
if (isRegularExpression !== "annex-b") {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Since this was the only place we check for "annex-b", we can make isRegularExpression a boolean again.

Copy link
Contributor Author

@graphemecluster graphemecluster May 23, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I thought the same too, but there is a isRegularExpression === true at the bottom

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This probably is worth being refactored, since shouldEmitInvalidEscapeError does not explain itself but is taken over to mean anyUnicodeMode for regular expressions. But this will be yet another PR I guess.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you use an enum for such a case?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, I found that I once created a enum named ClassSetExpressionType.

error(Diagnostics.Octal_escape_sequences_are_not_allowed_Use_the_syntax_0, start, pos - start, "\\x" + code.toString(16).padStart(2, "0"));
}
error(Diagnostics.Octal_escape_sequences_are_not_allowed_Use_the_syntax_0, start, pos - start, "\\x" + code.toString(16).padStart(2, "0"));
return String.fromCharCode(code);
}
return text.substring(start, pos);
Expand Down Expand Up @@ -2426,6 +2424,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
// Quickly get to the end of regex such that we know the flags
let p = tokenStart + 1;
let inEscape = false;
let namedCaptureGroups = false;
// Although nested character classes are allowed in Unicode Sets mode,
// an unescaped slash is nevertheless invalid even in a character class in Unicode mode.
// Additionally, parsing nested character classes will misinterpret regexes like `/[[]/`
Expand Down Expand Up @@ -2469,6 +2468,15 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
else if (ch === CharacterCodes.closeBracket) {
inCharacterClass = false;
}
else if (
ch === CharacterCodes.openParen
&& charCodeUnchecked(p + 1) === CharacterCodes.question
&& charCodeUnchecked(p + 2) === CharacterCodes.lessThan
&& charCodeUnchecked(p + 3) !== CharacterCodes.equals
&& charCodeUnchecked(p + 3) !== CharacterCodes.exclamation
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I missed this in my last review. These need to be range checked:

Suggested change
&& charCodeUnchecked(p + 1) === CharacterCodes.question
&& charCodeUnchecked(p + 2) === CharacterCodes.lessThan
&& charCodeUnchecked(p + 3) !== CharacterCodes.equals
&& charCodeUnchecked(p + 3) !== CharacterCodes.exclamation
&& charCodeChecked(p + 1) === CharacterCodes.question
&& charCodeChecked(p + 2) === CharacterCodes.lessThan
&& charCodeChecked(p + 3) !== CharacterCodes.equals
&& charCodeChecked(p + 3) !== CharacterCodes.exclamation

) {
namedCaptureGroups = true;
}
p++;
}
const isUnterminated = !!(tokenFlags & TokenFlags.Unterminated);
Expand Down Expand Up @@ -2505,7 +2513,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
const saveEnd = end;
pos = tokenStart + 1;
end = endOfBody;
scanRegularExpressionWorker(regExpFlags, isUnterminated, /*annexB*/ true);
scanRegularExpressionWorker(regExpFlags, isUnterminated, /*annexB*/ true, namedCaptureGroups);
tokenStart = saveTokenStart;
tokenFlags = saveTokenFlags;
pos = savePos;
Expand All @@ -2517,7 +2525,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
return token;
}

function scanRegularExpressionWorker(regExpFlags: RegularExpressionFlags, isUnterminated: boolean, annexB: boolean) {
function scanRegularExpressionWorker(regExpFlags: RegularExpressionFlags, isUnterminated: boolean, annexB: boolean, namedCaptureGroups: boolean) {
// Why var? It avoids TDZ checks in the runtime which can be costly.
// See: https://github.com/microsoft/TypeScript/issues/52924
/* eslint-disable no-var */
Expand All @@ -2527,10 +2535,8 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
/** Grammar parameter */
var unicodeMode = !!(regExpFlags & RegularExpressionFlags.UnicodeMode);

if (unicodeMode) {
// Annex B treats any unicode mode as the strict syntax.
annexB = false;
}
// Annex B treats any unicode mode as the strict syntax.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The comment explained why we set annexB to false, but doesn't apply to this whole declaration. I would change it to be more explanatory, i.e.:

Suggested change
// Annex B treats any unicode mode as the strict syntax.
// Regular expressions are checked more strictly when either in 'u' or 'v' mode, or
// when not using the looser interpretation of the syntax from ECMA-262 Annex B.

var anyUnicodeModeOrNonAnnexB = unicodeMode || !annexB;

/** @see {scanClassSetExpression} */
var mayContainStrings = false;
Expand Down Expand Up @@ -2626,7 +2632,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
case CharacterCodes.exclamation:
pos++;
// In Annex B, `(?=Disjunction)` and `(?!Disjunction)` are quantifiable
isPreviousTermQuantifiable = annexB;
isPreviousTermQuantifiable = !anyUnicodeModeOrNonAnnexB;
break;
case CharacterCodes.lessThan:
const groupNameStart = pos;
Expand Down Expand Up @@ -2675,6 +2681,10 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
const digitsStart = pos;
scanDigits();
const min = tokenValue;
if (!anyUnicodeModeOrNonAnnexB && !min) {
isPreviousTermQuantifiable = true;
break;
}
if (charCodeChecked(pos) === CharacterCodes.comma) {
pos++;
scanDigits();
Expand All @@ -2684,26 +2694,32 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
error(Diagnostics.Incomplete_quantifier_Digit_expected, digitsStart, 0);
}
else {
if (unicodeMode) {
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, start, 1, String.fromCharCode(ch));
}
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, start, 1, String.fromCharCode(ch));
isPreviousTermQuantifiable = true;
break;
}
}
if (max && Number.parseInt(min) > Number.parseInt(max)) {
else if (max && Number.parseInt(min) > Number.parseInt(max) && (anyUnicodeModeOrNonAnnexB || text.charCodeAt(pos) === CharacterCodes.closeBrace)) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
else if (max && Number.parseInt(min) > Number.parseInt(max) && (anyUnicodeModeOrNonAnnexB || text.charCodeAt(pos) === CharacterCodes.closeBrace)) {
else if (max && Number.parseInt(min) > Number.parseInt(max) && (anyUnicodeModeOrNonAnnexB || charCodeChecked(pos) === CharacterCodes.closeBrace)) {

error(Diagnostics.Numbers_out_of_order_in_quantifier, digitsStart, pos - digitsStart);
}
}
else if (!min) {
if (unicodeMode) {
if (anyUnicodeModeOrNonAnnexB) {
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, start, 1, String.fromCharCode(ch));
}
isPreviousTermQuantifiable = true;
break;
}
scanExpectedChar(CharacterCodes.closeBrace);
pos--;
if (charCodeChecked(pos) !== CharacterCodes.closeBrace) {
if (anyUnicodeModeOrNonAnnexB) {
error(Diagnostics._0_expected, pos, 0, String.fromCharCode(CharacterCodes.closeBrace));
pos--;
}
else {
isPreviousTermQuantifiable = true;
break;
}
}
// falls through
case CharacterCodes.asterisk:
case CharacterCodes.plus:
Expand Down Expand Up @@ -2744,7 +2760,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
// Assume what starting from the character to be outside of the regex
return;
}
if (unicodeMode || ch === CharacterCodes.closeParen) {
if (anyUnicodeModeOrNonAnnexB || ch === CharacterCodes.closeParen) {
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, pos, 1, String.fromCharCode(ch));
}
pos++;
Expand Down Expand Up @@ -2801,7 +2817,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
scanGroupName(/*isReference*/ true);
scanExpectedChar(CharacterCodes.greaterThan);
}
else if (unicodeMode) {
else if (namedCaptureGroups) {
error(Diagnostics.k_must_be_followed_by_a_capturing_group_name_enclosed_in_angle_brackets, pos - 2, 2);
}
break;
Expand Down Expand Up @@ -2851,10 +2867,10 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
pos++;
return String.fromCharCode(ch & 0x1f);
}
if (unicodeMode) {
if (anyUnicodeModeOrNonAnnexB) {
error(Diagnostics.c_must_be_followed_by_an_ASCII_letter, pos - 2, 2);
}
else if (atomEscape && annexB) {
else if (atomEscape) {
// Annex B treats
//
// ExtendedAtom : `\` [lookahead = `c`]
Expand Down Expand Up @@ -2887,7 +2903,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
return "\\";
}
pos--;
return scanEscapeSequence(/*shouldEmitInvalidEscapeError*/ unicodeMode, /*isRegularExpression*/ annexB ? "annex-b" : true);
return scanEscapeSequence(/*shouldEmitInvalidEscapeError*/ unicodeMode, /*isRegularExpression*/ anyUnicodeModeOrNonAnnexB || "annex-b");
rbuckton marked this conversation as resolved.
Show resolved Hide resolved
}
}

Expand Down Expand Up @@ -2936,12 +2952,12 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
if (isClassContentExit(ch)) {
return;
}
if (!minCharacter && !annexB) {
if (!minCharacter && anyUnicodeModeOrNonAnnexB) {
error(Diagnostics.A_character_class_range_must_not_be_bounded_by_another_character_class, minStart, pos - 1 - minStart);
}
const maxStart = pos;
const maxCharacter = scanClassAtom();
if (!maxCharacter && !annexB) {
if (!maxCharacter && anyUnicodeModeOrNonAnnexB) {
error(Diagnostics.A_character_class_range_must_not_be_bounded_by_another_character_class, maxStart, pos - maxStart);
continue;
}
Expand Down Expand Up @@ -3437,9 +3453,13 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
error(Diagnostics.Unicode_property_value_expressions_are_only_available_when_the_Unicode_u_flag_or_the_Unicode_Sets_v_flag_is_set, start, pos - start);
}
}
else if (unicodeMode) {
else if (anyUnicodeModeOrNonAnnexB) {
error(Diagnostics._0_must_be_followed_by_a_Unicode_property_value_expression_enclosed_in_braces, pos - 2, 2, String.fromCharCode(ch));
}
else {
pos--;
return false;
}
return true;
}
return false;
Expand Down Expand Up @@ -3483,7 +3503,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
forEach(decimalEscapes, escape => {
// in AnnexB, if a DecimalEscape is greater than the number of capturing groups then it is treated as
// either a LegacyOctalEscapeSequence or IdentityEscape
if (!annexB && escape.value > numberOfCapturingGroups) {
if (anyUnicodeModeOrNonAnnexB && escape.value > numberOfCapturingGroups) {
if (numberOfCapturingGroups) {
error(Diagnostics.This_backreference_refers_to_a_group_that_does_not_exist_There_are_only_0_capturing_groups_in_this_regular_expression, escape.pos, escape.end - escape.pos, numberOfCapturingGroups);
}
Expand Down
Loading