From 2281823a5515834b4f988891c3ca59ceddb097d1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Niels=20M=C3=B6ller?= Date: Fri, 15 Dec 2017 09:39:16 +0100 Subject: [PATCH] Delete unused code in rtc_base/stringencode.cc Bug: webrtc:6424 Change-Id: I3d72d2a80c502b3afc31af2456d220f48272b348 Reviewed-on: https://webrtc-review.googlesource.com/33006 Commit-Queue: Niels Moller Reviewed-by: Taylor Brandstetter Cr-Commit-Position: refs/heads/master@{#21290} --- rtc_base/stringencode.cc | 307 ------------------------------ rtc_base/stringencode.h | 58 +----- rtc_base/stringencode_unittest.cc | 45 ----- 3 files changed, 2 insertions(+), 408 deletions(-) diff --git a/rtc_base/stringencode.cc b/rtc_base/stringencode.cc index efd9843c46..d77245f5d6 100644 --- a/rtc_base/stringencode.cc +++ b/rtc_base/stringencode.cc @@ -44,127 +44,6 @@ size_t escape(char * buffer, size_t buflen, return bufpos; } -size_t unescape(char * buffer, size_t buflen, - const char * source, size_t srclen, - char escape) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - char ch = source[srcpos++]; - if ((ch == escape) && (srcpos < srclen)) { - ch = source[srcpos++]; - } - buffer[bufpos++] = ch; - } - buffer[bufpos] = '\0'; - return bufpos; -} - -size_t encode(char * buffer, size_t buflen, - const char * source, size_t srclen, - const char * illegal, char escape) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - char ch = source[srcpos++]; - if ((ch != escape) && !::strchr(illegal, ch)) { - buffer[bufpos++] = ch; - } else if (bufpos + 3 >= buflen) { - break; - } else { - buffer[bufpos+0] = escape; - buffer[bufpos+1] = hex_encode((static_cast(ch) >> 4) & 0xF); - buffer[bufpos+2] = hex_encode((static_cast(ch) ) & 0xF); - bufpos += 3; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - -size_t decode(char * buffer, size_t buflen, - const char * source, size_t srclen, - char escape) { - if (buflen <= 0) - return 0; - - unsigned char h1, h2; - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - char ch = source[srcpos++]; - if ((ch == escape) - && (srcpos + 1 < srclen) - && hex_decode(source[srcpos], &h1) - && hex_decode(source[srcpos+1], &h2)) { - buffer[bufpos++] = (h1 << 4) | h2; - srcpos += 2; - } else { - buffer[bufpos++] = ch; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - -const char* unsafe_filename_characters() { - // It might be better to have a single specification which is the union of - // all operating systems, unless one system is overly restrictive. -#if defined(WEBRTC_WIN) - return "\\/:*?\"<>|"; -#else // !WEBRTC_WIN - // TODO(grunell): Should this never be reached? - RTC_NOTREACHED(); - return ""; -#endif // !WEBRTC_WIN -} - -const unsigned char URL_UNSAFE = 0x1; // 0-33 "#$%&+,/:;<=>?@[\]^`{|} 127 -const unsigned char XML_UNSAFE = 0x2; // "&'<> -const unsigned char HTML_UNSAFE = 0x2; // "&'<> - -// ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 6 5 7 8 9 : ; < = > ? -//@ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _ -//` a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ - -const unsigned char ASCII_CLASS[128] = { - 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, - 1,0,3,1,1,1,3,2,0,0,0,1,1,0,0,1,0,0,0,0,0,0,0,0,0,0,1,1,3,1,3,1, - 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0, - 1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,1, -}; - -size_t url_encode(char * buffer, size_t buflen, - const char * source, size_t srclen) { - if (nullptr == buffer) - return srclen * 3 + 1; - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - unsigned char ch = source[srcpos++]; - if ((ch < 128) && (ASCII_CLASS[ch] & URL_UNSAFE)) { - if (bufpos + 3 >= buflen) { - break; - } - buffer[bufpos+0] = '%'; - buffer[bufpos+1] = hex_encode((ch >> 4) & 0xF); - buffer[bufpos+2] = hex_encode((ch ) & 0xF); - bufpos += 3; - } else { - buffer[bufpos++] = ch; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - size_t url_decode(char * buffer, size_t buflen, const char * source, size_t srclen) { if (nullptr == buffer) @@ -254,159 +133,6 @@ size_t utf8_encode(char* buffer, size_t buflen, unsigned long value) { return 0; } -size_t html_encode(char * buffer, size_t buflen, - const char * source, size_t srclen) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - unsigned char ch = source[srcpos]; - if (ch < 128) { - srcpos += 1; - if (ASCII_CLASS[ch] & HTML_UNSAFE) { - const char * escseq = 0; - size_t esclen = 0; - switch (ch) { - case '<': escseq = "<"; esclen = 4; break; - case '>': escseq = ">"; esclen = 4; break; - case '\'': escseq = "'"; esclen = 5; break; - case '\"': escseq = """; esclen = 6; break; - case '&': escseq = "&"; esclen = 5; break; - default: RTC_NOTREACHED(); - } - if (bufpos + esclen >= buflen) { - break; - } - memcpy(buffer + bufpos, escseq, esclen); - bufpos += esclen; - } else { - buffer[bufpos++] = ch; - } - } else { - // Largest value is 0x1FFFFF => � (10 characters) - const size_t kEscseqSize = 11; - char escseq[kEscseqSize]; - unsigned long val; - if (size_t vallen = utf8_decode(&source[srcpos], srclen - srcpos, &val)) { - srcpos += vallen; - } else { - // Not a valid utf8 sequence, just use the raw character. - val = static_cast(source[srcpos++]); - } - size_t esclen = sprintfn(escseq, kEscseqSize, "&#%lu;", val); - if (bufpos + esclen >= buflen) { - break; - } - memcpy(buffer + bufpos, escseq, esclen); - bufpos += esclen; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - -size_t html_decode(char * buffer, size_t buflen, - const char * source, size_t srclen) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - return xml_decode(buffer, buflen, source, srclen); -} - -size_t xml_encode(char * buffer, size_t buflen, - const char * source, size_t srclen) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - unsigned char ch = source[srcpos++]; - if ((ch < 128) && (ASCII_CLASS[ch] & XML_UNSAFE)) { - const char * escseq = 0; - size_t esclen = 0; - switch (ch) { - case '<': escseq = "<"; esclen = 4; break; - case '>': escseq = ">"; esclen = 4; break; - case '\'': escseq = "'"; esclen = 6; break; - case '\"': escseq = """; esclen = 6; break; - case '&': escseq = "&"; esclen = 5; break; - default: RTC_NOTREACHED(); - } - if (bufpos + esclen >= buflen) { - break; - } - memcpy(buffer + bufpos, escseq, esclen); - bufpos += esclen; - } else { - buffer[bufpos++] = ch; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - -size_t xml_decode(char * buffer, size_t buflen, - const char * source, size_t srclen) { - RTC_DCHECK(buffer); // TODO(grunell): estimate output size - if (buflen <= 0) - return 0; - - size_t srcpos = 0, bufpos = 0; - while ((srcpos < srclen) && (bufpos + 1 < buflen)) { - unsigned char ch = source[srcpos++]; - if (ch != '&') { - buffer[bufpos++] = ch; - } else if ((srcpos + 2 < srclen) - && (memcmp(source + srcpos, "lt;", 3) == 0)) { - buffer[bufpos++] = '<'; - srcpos += 3; - } else if ((srcpos + 2 < srclen) - && (memcmp(source + srcpos, "gt;", 3) == 0)) { - buffer[bufpos++] = '>'; - srcpos += 3; - } else if ((srcpos + 4 < srclen) - && (memcmp(source + srcpos, "apos;", 5) == 0)) { - buffer[bufpos++] = '\''; - srcpos += 5; - } else if ((srcpos + 4 < srclen) - && (memcmp(source + srcpos, "quot;", 5) == 0)) { - buffer[bufpos++] = '\"'; - srcpos += 5; - } else if ((srcpos + 3 < srclen) - && (memcmp(source + srcpos, "amp;", 4) == 0)) { - buffer[bufpos++] = '&'; - srcpos += 4; - } else if ((srcpos < srclen) && (source[srcpos] == '#')) { - int int_base = 10; - if ((srcpos + 1 < srclen) && (source[srcpos+1] == 'x')) { - int_base = 16; - srcpos += 1; - } - char * ptr; - // TODO(grunell): Fix hack (ptr may go past end of data) - unsigned long val = strtoul(source + srcpos + 1, &ptr, int_base); - if ((static_cast(ptr - source) < srclen) && (*ptr == ';')) { - srcpos = ptr - source + 1; - } else { - // Not a valid escape sequence. - break; - } - if (size_t esclen = utf8_encode(buffer + bufpos, buflen - bufpos, val)) { - bufpos += esclen; - } else { - // Not enough room to encode the character, or illegal character - break; - } - } else { - // Unrecognized escape sequence. - break; - } - } - buffer[bufpos] = '\0'; - return bufpos; -} - static const char HEX[] = "0123456789abcdef"; char hex_encode(unsigned char val) { @@ -682,37 +408,4 @@ size_t split(const std::string& source, char delimiter, return fields->size(); } -char make_char_safe_for_filename(char c) { - if (c < 32) - return '_'; - - switch (c) { - case '<': - case '>': - case ':': - case '"': - case '/': - case '\\': - case '|': - case '*': - case '?': - return '_'; - - default: - return c; - } -} - -/* -void sprintf(std::string& value, size_t maxlen, const char * format, ...) { - char * buffer = STACK_ARRAY(char, maxlen + 1); - va_list args; - va_start(args, format); - value.assign(buffer, vsprintfn(buffer, maxlen + 1, format, args)); - va_end(args); -} -*/ - -///////////////////////////////////////////////////////////////////////////// - } // namespace rtc diff --git a/rtc_base/stringencode.h b/rtc_base/stringencode.h index d99dd0d4a2..5d675a2598 100644 --- a/rtc_base/stringencode.h +++ b/rtc_base/stringencode.h @@ -23,61 +23,18 @@ namespace rtc { // String Encoding Utilities ////////////////////////////////////////////////////////////////////// -// Convert an unsigned value to it's utf8 representation. Returns the length -// of the encoded string, or 0 if the encoding is longer than buflen - 1. -size_t utf8_encode(char* buffer, size_t buflen, unsigned long value); -// Decode the utf8 encoded value pointed to by source. Returns the number of -// bytes used by the encoding, or 0 if the encoding is invalid. -size_t utf8_decode(const char* source, size_t srclen, unsigned long* value); - +// TODO(nisse): Used only in httpcommon.c. Delete when that file is deleted, or +// possibly if the HttpComposeAttributes funtion can be deleted earlier. // Escaping prefixes illegal characters with the escape character. Compact, but // illegal characters still appear in the string. size_t escape(char * buffer, size_t buflen, const char * source, size_t srclen, const char * illegal, char escape); -// Note: in-place unescaping (buffer == source) is allowed. -size_t unescape(char * buffer, size_t buflen, - const char * source, size_t srclen, - char escape); -// Encoding replaces illegal characters with the escape character and 2 hex -// chars, so it's a little less compact than escape, but completely removes -// illegal characters. note that hex digits should not be used as illegal -// characters. -size_t encode(char * buffer, size_t buflen, - const char * source, size_t srclen, - const char * illegal, char escape); -// Note: in-place decoding (buffer == source) is allowed. -size_t decode(char * buffer, size_t buflen, - const char * source, size_t srclen, - char escape); - -// Returns a list of characters that may be unsafe for use in the name of a -// file, suitable for passing to the 'illegal' member of escape or encode. -const char* unsafe_filename_characters(); - -// url_encode is an encode operation with a predefined set of illegal characters -// and escape character (for use in URLs, obviously). -size_t url_encode(char * buffer, size_t buflen, - const char * source, size_t srclen); // Note: in-place decoding (buffer == source) is allowed. size_t url_decode(char * buffer, size_t buflen, const char * source, size_t srclen); -// html_encode prevents data embedded in html from containing markup. -size_t html_encode(char * buffer, size_t buflen, - const char * source, size_t srclen); -// Note: in-place decoding (buffer == source) is allowed. -size_t html_decode(char * buffer, size_t buflen, - const char * source, size_t srclen); - -// xml_encode makes data suitable for inside xml attributes and values. -size_t xml_encode(char * buffer, size_t buflen, - const char * source, size_t srclen); -// Note: in-place decoding (buffer == source) is allowed. -size_t xml_decode(char * buffer, size_t buflen, - const char * source, size_t srclen); - // Convert an unsigned value from 0 to 15 to the hex character equivalent... char hex_encode(unsigned char val); // ...and vice-versa. @@ -129,9 +86,6 @@ size_t transform(std::string& value, size_t maxlen, const std::string& source, std::string s_transform(const std::string& source, Transform t); // Convenience wrappers. -inline std::string s_url_encode(const std::string& source) { - return s_transform(source, url_encode); -} inline std::string s_url_decode(const std::string& source) { return s_transform(source, url_decode); } @@ -177,10 +131,6 @@ bool tokenize_first(const std::string& source, std::string* token, std::string* rest); -// Safe sprintf to std::string -//void sprintf(std::string& value, size_t maxlen, const char * format, ...) -// PRINTF_FORMAT(3); - // Convert arbitrary values to/from a string. template @@ -217,10 +167,6 @@ static inline T FromString(const T& defaultValue, const std::string& str) { T val(defaultValue); FromString(str, &val); return val; } -// simple function to strip out characters which shouldn't be -// used in filenames -char make_char_safe_for_filename(char c); - ////////////////////////////////////////////////////////////////////// } // namespace rtc diff --git a/rtc_base/stringencode_unittest.cc b/rtc_base/stringencode_unittest.cc index e4b5c960ad..63d8290276 100644 --- a/rtc_base/stringencode_unittest.cc +++ b/rtc_base/stringencode_unittest.cc @@ -15,51 +15,6 @@ namespace rtc { -TEST(Utf8EncodeTest, EncodeDecode) { - const struct Utf8Test { - const char* encoded; - size_t encsize, enclen; - unsigned long decoded; - } kTests[] = { - {"a ", 5, 1, 'a'}, - {"\x7F ", 5, 1, 0x7F}, - {"\xC2\x80 ", 5, 2, 0x80}, - {"\xDF\xBF ", 5, 2, 0x7FF}, - {"\xE0\xA0\x80 ", 5, 3, 0x800}, - {"\xEF\xBF\xBF ", 5, 3, 0xFFFF}, - {"\xF0\x90\x80\x80 ", 5, 4, 0x10000}, - {"\xF0\x90\x80\x80 ", 3, 0, 0x10000}, - {"\xF0\xF0\x80\x80 ", 5, 0, 0}, - {"\xF0\x90\x80 ", 5, 0, 0}, - {"\x90\x80\x80 ", 5, 0, 0}, - {nullptr, 0, 0}, - }; - for (size_t i = 0; kTests[i].encoded; ++i) { - unsigned long val = 0; - ASSERT_EQ(kTests[i].enclen, utf8_decode(kTests[i].encoded, - kTests[i].encsize, - &val)); - unsigned long result = (kTests[i].enclen == 0) ? 0 : kTests[i].decoded; - ASSERT_EQ(result, val); - - if (kTests[i].decoded == 0) { - // Not an interesting encoding test case - continue; - } - - char buffer[5]; - memset(buffer, 0x01, arraysize(buffer)); - ASSERT_EQ(kTests[i].enclen, utf8_encode(buffer, - kTests[i].encsize, - kTests[i].decoded)); - ASSERT_TRUE(memcmp(buffer, kTests[i].encoded, kTests[i].enclen) == 0); - // Make sure remainder of buffer is unchanged - ASSERT_TRUE(memory_check(buffer + kTests[i].enclen, - 0x1, - arraysize(buffer) - kTests[i].enclen)); - } -} - class HexEncodeTest : public testing::Test { public: HexEncodeTest() : enc_res_(0), dec_res_(0) {