ed/metadesk
1
0
Fork 0
mirror of https://github.com/Ed94/metadesk.git synced 2026-08-07 08:08:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Multi-line string escaping.

Browse Source
This commit is contained in:
Miguel Lechon
2021-03-18 12:28:16 +01:00
parent 96c3c46531
commit f4965c782d
2 changed files with 75 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
source/md_impl.c
+34 -19
View File
@@ -1192,8 +1192,8 @@ MD_Parse_BumpNext(MD_ParseCtx *ctx)
MD_Parse_Bump(ctx, MD_Parse_LexNext(ctx)); MD_Parse_Bump(ctx, MD_Parse_LexNext(ctx));
} }
MD_FUNCTION_IMPL MD_u8 * MD_PRIVATE_FUNCTION_IMPL MD_u8 *
MD_TokenizerScanEscaped(MD_u8 *at, MD_u8 *one_past_last, MD_u8 c) _MD_TokenizerScanEscaped(MD_u8 *at, MD_u8 *one_past_last, MD_u8 c)
{ {
while(at < one_past_last) while(at < one_past_last)
{ {
@@ -1207,13 +1207,35 @@ MD_TokenizerScanEscaped(MD_u8 *at, MD_u8 *one_past_last, MD_u8 c)
return at; return at;
} }
MD_PRIVATE_FUNCTION_IMPL MD_u8 *
_MD_TokenizerScanTripletEscaped(MD_u8 *at, MD_u8 *one_past_last, MD_u8 c)
{
MD_u32 consecutive_c = 0;
while(at < one_past_last && consecutive_c < 3)
{
if(at[0] == c)
{
consecutive_c += 1;
}
else
{
consecutive_c = 0;
if(at[0] == '\\' && at + 1 < one_past_last)
{
if(at[1] == c || at[1] == '\\')
{
at += 1;
}
}
}
at += 1;
}
return at;
}
MD_PRIVATE_FUNCTION_IMPL MD_u32 MD_PRIVATE_FUNCTION_IMPL MD_u32
_MD_ComputeTextLiteralChop(MD_u32 bytes_to_skip, MD_u8 *beg, MD_u8 *end) _MD_ComputeTextLiteralChop(MD_u32 bytes_to_skip, MD_u8 *beg, MD_u8 *end)
{ {
// NOTE(mal): By counting the bytes to chop off text literals and encoding exact token.string and
// token.outer_string we delegate the responsibility of generating errors to the parser
// That way, we can avoid generating errors during peeks and save them for actual token
// consumption. That allows more predictable error ordering.
MD_u32 result = 0; MD_u32 result = 0;
MD_u8 *skip_end = beg + bytes_to_skip; MD_u8 *skip_end = beg + bytes_to_skip;
MD_u8 *chop_beg = end - bytes_to_skip; MD_u8 *chop_beg = end - bytes_to_skip;
@@ -1322,15 +1344,12 @@ MD_Parse_LexNext(MD_ParseCtx *ctx)
if (at + 2 < one_past_last && at[1] == '`' && at[2] == '`') if (at + 2 < one_past_last && at[1] == '`' && at[2] == '`')
{ {
skip_n = 3; skip_n = 3;
at += 3; at = _MD_TokenizerScanTripletEscaped(at+3, one_past_last, '`');
MD_TokenizerScan(!(at + 2 < one_past_last && at[0] == '`' && at[1] == '`' && at[2] == '`'));
at += 3;
} }
else else
{ {
skip_n = 1; skip_n = 1;
at += 1; at = _MD_TokenizerScanEscaped(at+1, one_past_last, '`');
at = MD_TokenizerScanEscaped(at, one_past_last, '`');
} }
chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at); chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at);
}break; }break;
@@ -1342,15 +1361,13 @@ MD_Parse_LexNext(MD_ParseCtx *ctx)
if (at + 2 < one_past_last && at[1] == '"' && at[2] == '"') if (at + 2 < one_past_last && at[1] == '"' && at[2] == '"')
{ {
skip_n = 3; skip_n = 3;
at += 3; at = _MD_TokenizerScanTripletEscaped(at+3, one_past_last, '"');
MD_TokenizerScan(!(at + 2 < one_past_last && at[0] == '"' && at[1] == '"' && at[2] == '"'));
at += 3;
} }
else else
{ {
skip_n = 1; skip_n = 1;
at += 1; at += 1;
at = MD_TokenizerScanEscaped(at, one_past_last, '"'); at = _MD_TokenizerScanEscaped(at, one_past_last, '"');
} }
chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at); chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at);
}break; }break;
@@ -1361,16 +1378,14 @@ MD_Parse_LexNext(MD_ParseCtx *ctx)
{ {
token.kind = MD_TokenKind_StringLiteral; token.kind = MD_TokenKind_StringLiteral;
skip_n = 3; skip_n = 3;
at += 3; at = _MD_TokenizerScanTripletEscaped(at+3, one_past_last, '\'');
MD_TokenizerScan(!(at + 2 < one_past_last && at[0] == '\'' && at[1] == '\'' && at[2] == '\''));
at += 3;
} }
else else
{ {
token.kind = MD_TokenKind_CharLiteral; token.kind = MD_TokenKind_CharLiteral;
skip_n = 1; skip_n = 1;
at += 1; at += 1;
at = MD_TokenizerScanEscaped(at, one_past_last, '\''); at = _MD_TokenizerScanEscaped(at, one_past_last, '\'');
} }
chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at); chop_n = _MD_ComputeTextLiteralChop(skip_n, first, at);
}break; }break;
tests/sanity_tests.c
+40
View File
@@ -539,8 +539,48 @@ int main(void)
slot2 && slot2->value == (void *) 2); slot2 && slot2->value == (void *) 2);
} }
} }
}
Test("String escaping")
{
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("`\\``"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("\\`"), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("``` \\``` ```"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit(" \\``` "), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("`````\\````"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("``\\`"), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("`\\'`"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("\\'"), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("''' \\''' '''"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit(" \\''' "), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("'''''\\''''"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("''\\'"), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("`\\\"`"));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("\\\""), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("\"\"\" \\\"\"\" \"\"\""));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit(" \\\"\"\" "), 0));
}
{
MD_ParseResult parse = MD_ParseOneNode(MD_S8Lit(""), MD_S8Lit("\"\"\"\"\"\\\"\"\"\""));
TestResult(MD_StringMatch(parse.node->string, MD_S8Lit("\"\"\\\""), 0));
}
} }
return 0; return 0;