diff options
author | fanquake <fanquake@gmail.com> | 2022-06-11 14:29:09 +0100 |
---|---|---|
committer | fanquake <fanquake@gmail.com> | 2022-06-15 12:56:44 +0100 |
commit | d873ff96e51a3e7f2fdc3fdd1baee2bbe7583e06 (patch) | |
tree | d75c5d34033c6c709985a4697053a31def926533 /src/univalue/include | |
parent | e2aa7047f9b94232ff4958381321e6867d24b379 (diff) | |
download | bitcoin-d873ff96e51a3e7f2fdc3fdd1baee2bbe7583e06.tar.xz |
refactor: cleanups post unsubtree'ing univalue
Mostly changes to remove src/univalue exceptions from the various linters,
and the required code changes to make them happy. As well as minor doc
changes.
Diffstat (limited to 'src/univalue/include')
-rw-r--r-- | src/univalue/include/univalue.h | 6 | ||||
-rw-r--r-- | src/univalue/include/univalue_escapes.h | 261 | ||||
-rw-r--r-- | src/univalue/include/univalue_utffilter.h | 119 |
3 files changed, 383 insertions, 3 deletions
diff --git a/src/univalue/include/univalue.h b/src/univalue/include/univalue.h index f0d4de2035..22be0311e8 100644 --- a/src/univalue/include/univalue.h +++ b/src/univalue/include/univalue.h @@ -3,8 +3,8 @@ // Distributed under the MIT software license, see the accompanying // file COPYING or https://opensource.org/licenses/mit-license.php. -#ifndef __UNIVALUE_H__ -#define __UNIVALUE_H__ +#ifndef BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_H +#define BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_H #include <charconv> #include <cstdint> @@ -194,4 +194,4 @@ extern const UniValue NullUniValue; const UniValue& find_value( const UniValue& obj, const std::string& name); -#endif // __UNIVALUE_H__ +#endif // BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_H diff --git a/src/univalue/include/univalue_escapes.h b/src/univalue/include/univalue_escapes.h new file mode 100644 index 0000000000..83767e8ac5 --- /dev/null +++ b/src/univalue/include/univalue_escapes.h @@ -0,0 +1,261 @@ +#ifndef BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_ESCAPES_H +#define BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_ESCAPES_H +static const char *escapes[256] = { + "\\u0000", + "\\u0001", + "\\u0002", + "\\u0003", + "\\u0004", + "\\u0005", + "\\u0006", + "\\u0007", + "\\b", + "\\t", + "\\n", + "\\u000b", + "\\f", + "\\r", + "\\u000e", + "\\u000f", + "\\u0010", + "\\u0011", + "\\u0012", + "\\u0013", + "\\u0014", + "\\u0015", + "\\u0016", + "\\u0017", + "\\u0018", + "\\u0019", + "\\u001a", + "\\u001b", + "\\u001c", + "\\u001d", + "\\u001e", + "\\u001f", + nullptr, + nullptr, + "\\\"", + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + "\\\\", + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + "\\u007f", + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, + nullptr, +}; +#endif // BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_ESCAPES_H diff --git a/src/univalue/include/univalue_utffilter.h b/src/univalue/include/univalue_utffilter.h new file mode 100644 index 0000000000..f688eaaa30 --- /dev/null +++ b/src/univalue/include/univalue_utffilter.h @@ -0,0 +1,119 @@ +// Copyright 2016 Wladimir J. van der Laan +// Distributed under the MIT software license, see the accompanying +// file COPYING or https://opensource.org/licenses/mit-license.php. +#ifndef BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_UTFFILTER_H +#define BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_UTFFILTER_H + +#include <string> + +/** + * Filter that generates and validates UTF-8, as well as collates UTF-16 + * surrogate pairs as specified in RFC4627. + */ +class JSONUTF8StringFilter +{ +public: + explicit JSONUTF8StringFilter(std::string &s): + str(s), is_valid(true), codepoint(0), state(0), surpair(0) + { + } + // Write single 8-bit char (may be part of UTF-8 sequence) + void push_back(unsigned char ch) + { + if (state == 0) { + if (ch < 0x80) // 7-bit ASCII, fast direct pass-through + str.push_back(ch); + else if (ch < 0xc0) // Mid-sequence character, invalid in this state + is_valid = false; + else if (ch < 0xe0) { // Start of 2-byte sequence + codepoint = (ch & 0x1f) << 6; + state = 6; + } else if (ch < 0xf0) { // Start of 3-byte sequence + codepoint = (ch & 0x0f) << 12; + state = 12; + } else if (ch < 0xf8) { // Start of 4-byte sequence + codepoint = (ch & 0x07) << 18; + state = 18; + } else // Reserved, invalid + is_valid = false; + } else { + if ((ch & 0xc0) != 0x80) // Not a continuation, invalid + is_valid = false; + state -= 6; + codepoint |= (ch & 0x3f) << state; + if (state == 0) + push_back_u(codepoint); + } + } + // Write codepoint directly, possibly collating surrogate pairs + void push_back_u(unsigned int codepoint_) + { + if (state) // Only accept full codepoints in open state + is_valid = false; + if (codepoint_ >= 0xD800 && codepoint_ < 0xDC00) { // First half of surrogate pair + if (surpair) // Two subsequent surrogate pair openers - fail + is_valid = false; + else + surpair = codepoint_; + } else if (codepoint_ >= 0xDC00 && codepoint_ < 0xE000) { // Second half of surrogate pair + if (surpair) { // Open surrogate pair, expect second half + // Compute code point from UTF-16 surrogate pair + append_codepoint(0x10000 | ((surpair - 0xD800)<<10) | (codepoint_ - 0xDC00)); + surpair = 0; + } else // Second half doesn't follow a first half - fail + is_valid = false; + } else { + if (surpair) // First half of surrogate pair not followed by second - fail + is_valid = false; + else + append_codepoint(codepoint_); + } + } + // Check that we're in a state where the string can be ended + // No open sequences, no open surrogate pairs, etc + bool finalize() + { + if (state || surpair) + is_valid = false; + return is_valid; + } +private: + std::string &str; + bool is_valid; + // Current UTF-8 decoding state + unsigned int codepoint; + int state; // Top bit to be filled in for next UTF-8 byte, or 0 + + // Keep track of the following state to handle the following section of + // RFC4627: + // + // To escape an extended character that is not in the Basic Multilingual + // Plane, the character is represented as a twelve-character sequence, + // encoding the UTF-16 surrogate pair. So, for example, a string + // containing only the G clef character (U+1D11E) may be represented as + // "\uD834\uDD1E". + // + // Two subsequent \u.... may have to be replaced with one actual codepoint. + unsigned int surpair; // First half of open UTF-16 surrogate pair, or 0 + + void append_codepoint(unsigned int codepoint_) + { + if (codepoint_ <= 0x7f) + str.push_back((char)codepoint_); + else if (codepoint_ <= 0x7FF) { + str.push_back((char)(0xC0 | (codepoint_ >> 6))); + str.push_back((char)(0x80 | (codepoint_ & 0x3F))); + } else if (codepoint_ <= 0xFFFF) { + str.push_back((char)(0xE0 | (codepoint_ >> 12))); + str.push_back((char)(0x80 | ((codepoint_ >> 6) & 0x3F))); + str.push_back((char)(0x80 | (codepoint_ & 0x3F))); + } else if (codepoint_ <= 0x1FFFFF) { + str.push_back((char)(0xF0 | (codepoint_ >> 18))); + str.push_back((char)(0x80 | ((codepoint_ >> 12) & 0x3F))); + str.push_back((char)(0x80 | ((codepoint_ >> 6) & 0x3F))); + str.push_back((char)(0x80 | (codepoint_ & 0x3F))); + } + } +}; + +#endif // BITCOIN_UNIVALUE_INCLUDE_UNIVALUE_UTFFILTER_H |