X-Git-Url: http://demsky.eecs.uci.edu/git/?a=blobdiff_plain;f=folly%2FString.cpp;h=420cf2a19574443e65aa532866f1f055bddcb1fd;hb=7eaace048e3925a712e6a39c2e3d91e32b0cdc9a;hp=bf3fc38cde9fffb2167474f0a84a583d42c0a366;hpb=361946715a2d1a88b98d3f3ab30f7313262df96e;p=folly.git diff --git a/folly/String.cpp b/folly/String.cpp index bf3fc38c..420cf2a1 100644 --- a/folly/String.cpp +++ b/folly/String.cpp @@ -14,80 +14,87 @@ * limitations under the License. */ -#include "folly/String.h" -#include "folly/Format.h" +#include + +#include +#include #include #include #include #include #include +#include #include namespace folly { namespace { -inline void stringPrintfImpl(std::string& output, const char* format, - va_list args) { - // Tru to the space at the end of output for our output buffer. - // Find out write point then inflate its size temporarily to its - // capacity; we will later shrink it to the size needed to represent - // the formatted string. If this buffer isn't large enough, we do a - // resize and try again. - - const auto write_point = output.size(); - auto remaining = output.capacity() - write_point; - output.resize(output.capacity()); - +int stringAppendfImplHelper(char* buf, + size_t bufsize, + const char* format, + va_list args) { va_list args_copy; va_copy(args_copy, args); - int bytes_used = vsnprintf(&output[write_point], remaining, format, - args_copy); + int bytes_used = vsnprintf(buf, bufsize, format, args_copy); va_end(args_copy); + return bytes_used; +} + +void stringAppendfImpl(std::string& output, const char* format, va_list args) { + // Very simple; first, try to avoid an allocation by using an inline + // buffer. If that fails to hold the output string, allocate one on + // the heap, use it instead. + // + // It is hard to guess the proper size of this buffer; some + // heuristics could be based on the number of format characters, or + // static analysis of a codebase. Or, we can just pick a number + // that seems big enough for simple cases (say, one line of text on + // a terminal) without being large enough to be concerning as a + // stack variable. + std::array inline_buffer; + + int bytes_used = stringAppendfImplHelper( + inline_buffer.data(), inline_buffer.size(), format, args); if (bytes_used < 0) { - throw std::runtime_error( - to("Invalid format string; snprintf returned negative " - "with format string: ", format)); - } else if (bytes_used < remaining) { - // There was enough room, just shrink and return. - output.resize(write_point + bytes_used); - } else { - output.resize(write_point + bytes_used + 1); - remaining = bytes_used + 1; - va_list args_copy; - va_copy(args_copy, args); - bytes_used = vsnprintf(&output[write_point], remaining, format, - args_copy); - va_end(args_copy); - if (bytes_used + 1 != remaining) { - throw std::runtime_error( - to("vsnprint retry did not manage to work " - "with format string: ", format)); - } - output.resize(write_point + bytes_used); + throw std::runtime_error(to( + "Invalid format string; snprintf returned negative " + "with format string: ", + format)); + } + + if (static_cast(bytes_used) < inline_buffer.size()) { + output.append(inline_buffer.data(), bytes_used); + return; } + + // Couldn't fit. Heap allocate a buffer, oh well. + std::unique_ptr heap_buffer(new char[bytes_used + 1]); + int final_bytes_used = + stringAppendfImplHelper(heap_buffer.get(), bytes_used + 1, format, args); + // The second call should require the same length, which is 1 less + // than the buffer size (we don't keep the trailing \0 byte in our + // output string). + CHECK(bytes_used == final_bytes_used); + + output.append(heap_buffer.get(), bytes_used); } -} // anon namespace +} // anon namespace std::string stringPrintf(const char* format, ...) { - // snprintf will tell us how large the output buffer should be, but - // we then have to call it a second time, which is costly. By - // guestimating the final size, we avoid the double snprintf in many - // cases, resulting in a performance win. We use this constructor - // of std::string to avoid a double allocation, though it does pad - // the resulting string with nul bytes. Our guestimation is twice - // the format string size, or 32 bytes, whichever is larger. This - // is a hueristic that doesn't affect correctness but attempts to be - // reasonably fast for the most common cases. - std::string ret(std::max(32UL, strlen(format) * 2), '\0'); - ret.resize(0); - va_list ap; va_start(ap, format); - stringPrintfImpl(ret, format, ap); - va_end(ap); + SCOPE_EXIT { + va_end(ap); + }; + return stringVPrintf(format, ap); +} + +std::string stringVPrintf(const char* format, va_list ap) { + std::string ret; + stringAppendfImpl(ret, format, ap); return ret; } @@ -96,17 +103,31 @@ std::string stringPrintf(const char* format, ...) { std::string& stringAppendf(std::string* output, const char* format, ...) { va_list ap; va_start(ap, format); - stringPrintfImpl(*output, format, ap); - va_end(ap); + SCOPE_EXIT { + va_end(ap); + }; + return stringVAppendf(output, format, ap); +} + +std::string& stringVAppendf(std::string* output, + const char* format, + va_list ap) { + stringAppendfImpl(*output, format, ap); return *output; } void stringPrintf(std::string* output, const char* format, ...) { - output->clear(); va_list ap; va_start(ap, format); - stringPrintfImpl(*output, format, ap); - va_end(ap); + SCOPE_EXIT { + va_end(ap); + }; + return stringVPrintf(output, format, ap); +} + +void stringVPrintf(std::string* output, const char* format, va_list ap) { + output->clear(); + stringAppendfImpl(*output, format, ap); }; namespace { @@ -202,7 +223,7 @@ const PrettySuffix kPrettySISuffixes[] = { { "z", 1e-21L }, { "y", 1e-24L }, { " ", 0 }, - { 0, 0} + { 0, 0} }; const PrettySuffix* const kPrettySuffixes[PRETTY_NUM_TYPES] = { @@ -246,7 +267,7 @@ std::string prettyPrint(double val, PrettyType type, bool addSpace) { //TODO: //1) Benchmark & optimize -double prettyToDouble(folly::StringPiece *const prettyString, +double prettyToDouble(folly::StringPiece *const prettyString, const PrettyType type) { double value = folly::to(prettyString); while (prettyString->size() > 0 && std::isspace(prettyString->front())) { @@ -277,13 +298,13 @@ double prettyToDouble(folly::StringPiece *const prettyString, prettyString->toString(), "\"")); } prettyString->advance(longestPrefixLen); - return suffixes[bestPrefixId].val ? value * suffixes[bestPrefixId].val : + return suffixes[bestPrefixId].val ? value * suffixes[bestPrefixId].val : value; } double prettyToDouble(folly::StringPiece prettyString, const PrettyType type){ double result = prettyToDouble(&prettyString, type); - detail::enforceWhitespace(prettyString.data(), + detail::enforceWhitespace(prettyString.data(), prettyString.data() + prettyString.size()); return result; } @@ -307,7 +328,8 @@ fbstring errnoStr(int err) { // https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man3/strerror_r.3.html // http://www.kernel.org/doc/man-pages/online/pages/man3/strerror.3.html -#if defined(__APPLE__) || defined(__FreeBSD__) || \ +#if defined(__APPLE__) || defined(__FreeBSD__) ||\ + defined(__CYGWIN__) || defined(__ANDROID__) ||\ ((_POSIX_C_SOURCE >= 200112L || _XOPEN_SOURCE >= 600) && !_GNU_SOURCE) // Using XSI-compatible strerror_r int r = strerror_r(err, buf, sizeof(buf)); @@ -328,6 +350,166 @@ fbstring errnoStr(int err) { return result; } +StringPiece skipWhitespace(StringPiece sp) { + // Spaces other than ' ' characters are less common but should be + // checked. This configuration where we loop on the ' ' + // separately from oddspaces was empirically fastest. + auto oddspace = [] (char c) { + return c == '\n' || c == '\t' || c == '\r'; + }; + +loop: + for (; !sp.empty() && sp.front() == ' '; sp.pop_front()) { + } + if (!sp.empty() && oddspace(sp.front())) { + sp.pop_front(); + goto loop; + } + + return sp; +} + +namespace { + +void toLowerAscii8(char& c) { + // Branchless tolower, based on the input-rotating trick described + // at http://www.azillionmonkeys.com/qed/asmexample.html + // + // This algorithm depends on an observation: each uppercase + // ASCII character can be converted to its lowercase equivalent + // by adding 0x20. + + // Step 1: Clear the high order bit. We'll deal with it in Step 5. + unsigned char rotated = c & 0x7f; + // Currently, the value of rotated, as a function of the original c is: + // below 'A': 0- 64 + // 'A'-'Z': 65- 90 + // above 'Z': 91-127 + + // Step 2: Add 0x25 (37) + rotated += 0x25; + // Now the value of rotated, as a function of the original c is: + // below 'A': 37-101 + // 'A'-'Z': 102-127 + // above 'Z': 128-164 + + // Step 3: clear the high order bit + rotated &= 0x7f; + // below 'A': 37-101 + // 'A'-'Z': 102-127 + // above 'Z': 0- 36 + + // Step 4: Add 0x1a (26) + rotated += 0x1a; + // below 'A': 63-127 + // 'A'-'Z': 128-153 + // above 'Z': 25- 62 + + // At this point, note that only the uppercase letters have been + // transformed into values with the high order bit set (128 and above). + + // Step 5: Shift the high order bit 2 spaces to the right: the spot + // where the only 1 bit in 0x20 is. But first, how we ignored the + // high order bit of the original c in step 1? If that bit was set, + // we may have just gotten a false match on a value in the range + // 128+'A' to 128+'Z'. To correct this, need to clear the high order + // bit of rotated if the high order bit of c is set. Since we don't + // care about the other bits in rotated, the easiest thing to do + // is invert all the bits in c and bitwise-and them with rotated. + rotated &= ~c; + rotated >>= 2; + + // Step 6: Apply a mask to clear everything except the 0x20 bit + // in rotated. + rotated &= 0x20; + + // At this point, rotated is 0x20 if c is 'A'-'Z' and 0x00 otherwise + + // Step 7: Add rotated to c + c += rotated; +} + +void toLowerAscii32(uint32_t& c) { + // Besides being branchless, the algorithm in toLowerAscii8() has another + // interesting property: None of the addition operations will cause + // an overflow in the 8-bit value. So we can pack four 8-bit values + // into a uint32_t and run each operation on all four values in parallel + // without having to use any CPU-specific SIMD instructions. + uint32_t rotated = c & uint32_t(0x7f7f7f7fL); + rotated += uint32_t(0x25252525L); + rotated &= uint32_t(0x7f7f7f7fL); + rotated += uint32_t(0x1a1a1a1aL); + + // Step 5 involves a shift, so some bits will spill over from each + // 8-bit value into the next. But that's okay, because they're bits + // that will be cleared by the mask in step 6 anyway. + rotated &= ~c; + rotated >>= 2; + rotated &= uint32_t(0x20202020L); + c += rotated; +} + +void toLowerAscii64(uint64_t& c) { + // 64-bit version of toLower32 + uint64_t rotated = c & uint64_t(0x7f7f7f7f7f7f7f7fL); + rotated += uint64_t(0x2525252525252525L); + rotated &= uint64_t(0x7f7f7f7f7f7f7f7fL); + rotated += uint64_t(0x1a1a1a1a1a1a1a1aL); + rotated &= ~c; + rotated >>= 2; + rotated &= uint64_t(0x2020202020202020L); + c += rotated; +} + +} // anon namespace + +void toLowerAscii(char* str, size_t length) { + static const size_t kAlignMask64 = 7; + static const size_t kAlignMask32 = 3; + + // Convert a character at a time until we reach an address that + // is at least 32-bit aligned + size_t n = (size_t)str; + n &= kAlignMask32; + n = std::min(n, length); + size_t offset = 0; + if (n != 0) { + n = std::min(4 - n, length); + do { + toLowerAscii8(str[offset]); + offset++; + } while (offset < n); + } + + n = (size_t)(str + offset); + n &= kAlignMask64; + if ((n != 0) && (offset + 4 <= length)) { + // The next address is 32-bit aligned but not 64-bit aligned. + // Convert the next 4 bytes in order to get to the 64-bit aligned + // part of the input. + toLowerAscii32(*(uint32_t*)(str + offset)); + offset += 4; + } + + // Convert 8 characters at a time + while (offset + 8 <= length) { + toLowerAscii64(*(uint64_t*)(str + offset)); + offset += 8; + } + + // Convert 4 characters at a time + while (offset + 4 <= length) { + toLowerAscii32(*(uint32_t*)(str + offset)); + offset += 4; + } + + // Convert any characters remaining after the last 4-byte aligned group + while (offset < length) { + toLowerAscii8(str[offset]); + offset++; + } +} + namespace detail { size_t hexDumpLine(const void* ptr, size_t offset, size_t size, @@ -384,4 +566,3 @@ size_t hexDumpLine(const void* ptr, size_t offset, size_t size, # undef DMGL_TYPES # undef DMGL_RET_POSTFIX #endif -