carbon-language/carbon-lang
1
0
Fork 0
mirror of https://github.com/carbon-language/carbon-lang.git synced 2026-10-02 22:02:51 +01:00
Code Issues Packages Projects Releases 10 Wiki Activity

Reject invalid string literal whitespace on unescape (#793)

Browse Source 
This is based on discussion on #732: that we should probably parse the invalid whitespace, then reject it as part of string validation, rather than having different parses. I worry the question of "how is this parsed" may lead to subtly unexpected results if we aren't consistent, so I'm switching the logic from the lexer to the unescape library (and also adjusting the list of rejected whitespace).
This commit is contained in:
Jon Meow
2021-08-30 15:22:03 -07:00
committed by GitHub
parent 32f5845e7b
commit 49013ae1cc
7 changed files with 83 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/string_helpers.cpp
+61 -51
View File
@@ -27,58 +27,68 @@ auto UnescapeStringLiteral(llvm::StringRef source)
size_t i = 0;
while (i < source.size()) {
char c = source[i];
if (c == '\\') {
++i;
if (i == source.size()) {
return std::nullopt;
}
switch (source[i]) {
case 'n':
ret.push_back('\n');
break;
case 'r':
ret.push_back('\r');
break;
case 't':
ret.push_back('\t');
break;
case '0':
if (i + 1 < source.size() && llvm::isDigit(source[i + 1])) {
// \0[0-9] is reserved.
return std::nullopt;
}
ret.push_back('\0');
break;
case '"':
ret.push_back('"');
break;
case '\'':
ret.push_back('\'');
break;
case '\\':
ret.push_back('\\');
break;
case 'x': {
i += 2;
if (i >= source.size()) {
return std::nullopt;
}
std::optional<char> c1 = FromHex(source[i - 1]);
std::optional<char> c2 = FromHex(source[i]);
if (c1 == std::nullopt || c2 == std::nullopt) {
return std::nullopt;
}
ret.push_back(16 * *c1 + *c2);
break;
}
case 'u':
FATAL() << "\\u is not yet supported in string literals";
default:
// Unsupported.
switch (c) {
case '\\':
++i;
if (i == source.size()) {
return std::nullopt;
}
} else {
ret.push_back(c);
}
switch (source[i]) {
case 'n':
ret.push_back('\n');
break;
case 'r':
ret.push_back('\r');
break;
case 't':
ret.push_back('\t');
break;
case '0':
if (i + 1 < source.size() && llvm::isDigit(source[i + 1])) {
// \0[0-9] is reserved.
return std::nullopt;
}
ret.push_back('\0');
break;
case '"':
ret.push_back('"');
break;
case '\'':
ret.push_back('\'');
break;
case '\\':
ret.push_back('\\');
break;
case 'x': {
i += 2;
if (i >= source.size()) {
return std::nullopt;
}
std::optional<char> c1 = FromHex(source[i - 1]);
std::optional<char> c2 = FromHex(source[i]);
if (c1 == std::nullopt || c2 == std::nullopt) {
return std::nullopt;
}
ret.push_back(16 * *c1 + *c2);
break;
}
case 'u':
FATAL() << "\\u is not yet supported in string literals";
default:
// Unsupported.
return std::nullopt;
}
break;
case '\t':
// Disallow non-` ` horizontal whitespace:
// https://github.com/carbon-language/carbon-lang/blob/trunk/docs/design/lexical_conventions/whitespace.md
// TODO: This doesn't handle unicode whitespace.
return std::nullopt;
default:
ret.push_back(c);
break;
}
++i;
}