diff options
| -rw-r--r-- | libmisc/fmt.c | 30 | ||||
| -rw-r--r-- | libmisc/tests/test_fmt.c | 64 | ||||
| -rw-r--r-- | libmisc/utf8.c | 18 |
3 files changed, 102 insertions, 10 deletions
diff --git a/libmisc/fmt.c b/libmisc/fmt.c index a8baa84..7c18ef5 100644 --- a/libmisc/fmt.c +++ b/libmisc/fmt.c @@ -71,9 +71,33 @@ void fmt_print_ptr(lo_interface fmt_dest w, void *ptr) { */ void fmt_print_qbyte(lo_interface fmt_dest w, uint8_t b) { fmt_print_byte(w, '\''); - if (' ' <= b && b <= '~') { - if (b == '\'' || b == '\\') - fmt_print_byte(w, '\\'); + if (b == '\0' || + b == '\b' || + b == '\f' || + b == '\n' || + b == '\r' || + b == '\t' || + b == '\v' || + b == '\\' || + b == '\'' || + b == '"' || + b == '?') { + fmt_print_byte(w, '\\'); + switch (b) { + case '\0': fmt_print_byte(w, '0'); break; + case '\a': fmt_print_byte(w, 'a'); break; + case '\b': fmt_print_byte(w, 'b'); break; + case '\f': fmt_print_byte(w, 'f'); break; + case '\n': fmt_print_byte(w, 'n'); break; + case '\r': fmt_print_byte(w, 'r'); break; + case '\t': fmt_print_byte(w, 't'); break; + case '\v': fmt_print_byte(w, 'v'); break; + case '\\': fmt_print_byte(w, '\\'); break; + case '\'': fmt_print_byte(w, '\''); break; + case '"': fmt_print_byte(w, '"'); break; + case '?': fmt_print_byte(w, '?'); break; + } + } else if (' ' <= b && b <= '~') { fmt_print_byte(w, b); } else { fmt_print_byte(w, '\\'); diff --git a/libmisc/tests/test_fmt.c b/libmisc/tests/test_fmt.c index 6a6eb7c..a9157d6 100644 --- a/libmisc/tests/test_fmt.c +++ b/libmisc/tests/test_fmt.c @@ -62,6 +62,26 @@ int main() { test_assert(strcmp(str, "\"hell\"") == 0); memset(str, 0, sizeof(str)); + do_print((qstr, "hello\xFFworld🚧")); + test_assert(strcmp(str, "\"hello\\xFFworld\\U0001F6A7\"") == 0); + memset(str, 0, sizeof(str)); + + do_print((qstr, "¡hello world!")); + test_assert(strcmp(str, "\"\\u00A1hello world!\"") == 0); + memset(str, 0, sizeof(str)); + + do_print((qmem, "🚧", 3)); /* truncated UTF-8 */ + test_assert(strcmp(str, "\"\\xF0\\x9F\\x9A\"") == 0); + memset(str, 0, sizeof(str)); + + do_print((qmem, "\xF7\xBF\xBF\xBF", 4)); /* over unicode_max */ + test_assert(strcmp(str, "\"\\xF7\\xBF\\xBF\\xBF\"") == 0); + memset(str, 0, sizeof(str)); + + do_print((qmem, "\xE0\xA0", 2)); /* non-optimal encoding (of ' ') */ + test_assert(strcmp(str, "\"\\xE0\\xA0\"") == 0); + memset(str, 0, sizeof(str)); + do_print((byte, 'h'), (byte, 'w')); test_assert(strcmp(str, "hw") == 0); memset(str, 0, sizeof(str)); @@ -70,6 +90,26 @@ int main() { test_assert(strcmp(str, "'h''w'") == 0); memset(str, 0, sizeof(str)); + do_print((qbyte, 0)); + test_assert(strcmp(str, "'\\0'") == 0); + memset(str, 0, sizeof(str)); + + do_print((qbyte, '\\')); + test_assert(strcmp(str, "'\\\\'") == 0); + memset(str, 0, sizeof(str)); + + do_print((qbyte, '\'')); + test_assert(strcmp(str, "'\\''") == 0); + memset(str, 0, sizeof(str)); + + do_print((qbyte, '\n')); + test_assert(strcmp(str, "'\\n'") == 0); + memset(str, 0, sizeof(str)); + + do_print((qbyte, 0xff)); + test_assert(strcmp(str, "'\\xFF'") == 0); + memset(str, 0, sizeof(str)); + do_print("zero ", 0); test_assert(strcmp(str, "zero 0") == 0); memset(str, 0, sizeof(str)); @@ -166,5 +206,29 @@ int main() { test_assert(strcmp(str, " 1x") == 0); memset(str, 0, sizeof(str)); + do_print((base16_u8_, 1)); + test_assert(strcmp(str, "0x01") == 0); + memset(str, 0, sizeof(str)); + + do_print((base16_u16_, 1)); + test_assert(strcmp(str, "0x0001") == 0); + memset(str, 0, sizeof(str)); + + do_print((base16_u32_, 1)); + test_assert(strcmp(str, "0x00000001") == 0); + memset(str, 0, sizeof(str)); + + do_print((base16_u64_, 1)); + test_assert(strcmp(str, "0x0000000000000001") == 0); + memset(str, 0, sizeof(str)); + + do_print((hbyte, 1)); + test_assert(strcmp(str, "0x01") == 0); + memset(str, 0, sizeof(str)); + + do_print((hmem, "hello", 6)); + test_assert(strcmp(str, "{0x68,0x65,0x6C,0x6C,0x6F,0x00}") == 0); + memset(str, 0, sizeof(str)); + return 0; } diff --git a/libmisc/utf8.c b/libmisc/utf8.c index 5f91021..28357f0 100644 --- a/libmisc/utf8.c +++ b/libmisc/utf8.c @@ -9,17 +9,21 @@ void utf8_decode_codepoint(const uint8_t *str, size_t len, uint32_t *ret_ch, uint8_t *ret_chlen) { uint32_t ch; uint8_t chlen; - if ((str[0] & 0b10000000) == 0b00000000) { ch = str[0] & 0b01111111; chlen = 1; } - else if ((str[0] & 0b11100000) == 0b11000000) { ch = str[0] & 0b00011111; chlen = 2; } - else if ((str[0] & 0b11110000) == 0b11100000) { ch = str[0] & 0b00001111; chlen = 3; } - else if ((str[0] & 0b11111000) == 0b11110000) { ch = str[0] & 0b00000111; chlen = 4; } + uint32_t chmin; + if ((str[0] & 0b10000000) == 0b00000000) { ch = str[0] & 0b01111111; chlen = 1; chmin = 0; } /* bits=7+(0*6)= 7 */ + else if ((str[0] & 0b11100000) == 0b11000000) { ch = str[0] & 0b00011111; chlen = 2; chmin = UINT32_C(1)<< 7; } /* bits=5+(1*6)=11 */ + else if ((str[0] & 0b11110000) == 0b11100000) { ch = str[0] & 0b00001111; chlen = 3; chmin = UINT32_C(1)<<11; } /* bits=4+(2*6)=16 */ + else if ((str[0] & 0b11111000) == 0b11110000) { ch = str[0] & 0b00000111; chlen = 4; chmin = UINT32_C(1)<<16; } /* bits=3+(3*6)=21 */ else goto invalid; - if ((ch == 0 && chlen != 1) || chlen > len) goto invalid; + if (chlen > len) + goto invalid; for (uint8_t i = 1; i < chlen; i++) { - if ((str[i] & 0b11000000) != 0b10000000) goto invalid; + if ((str[i] & 0b11000000) != 0b10000000) + goto invalid; ch = (ch << 6) | (str[i] & 0b00111111); } - if (ch > 0x10FFFF) goto invalid; + if (ch > 0x10FFFF || ch < chmin) + goto invalid; *ret_ch = ch; *ret_chlen = chlen; return; |