diff options
Diffstat (limited to 'source/compiler-core/slang-json-source-map-util.cpp')
| -rw-r--r-- | source/compiler-core/slang-json-source-map-util.cpp | 464 |
1 files changed, 464 insertions, 0 deletions
diff --git a/source/compiler-core/slang-json-source-map-util.cpp b/source/compiler-core/slang-json-source-map-util.cpp new file mode 100644 index 000000000..f7c1711f3 --- /dev/null +++ b/source/compiler-core/slang-json-source-map-util.cpp @@ -0,0 +1,464 @@ +#include "slang-json-source-map-util.h" + +#include "../../slang-com-helper.h" + +#include "../core/slang-string-util.h" + +#include "slang-json-native.h" + +namespace Slang { + +/* +Support for source maps. Source maps provide a standardized mechanism to associate a location in one output file +with another. + +* [Source Map Proposal](https://docs.google.com/document/d/1U1RGAehQwRypUTovF1KRlpiOFze0b-_2gc6fAH0KY0k/edit?hl=en_US&pli=1&pli=1) +* [Chrome Source Map post](https://developer.chrome.com/blog/sourcemaps/) +* [Base64 VLQs in Source Maps](https://www.lucidchart.com/techblog/2019/08/22/decode-encoding-base64-vlqs-source-maps/) + +Example... + +{ +"version" : 3, +"file": "out.js", +"sourceRoot": "", +"sources": ["foo.js", "bar.js"], +"sourcesContent": [null, null], +"names": ["src", "maps", "are", "fun"], +"mappings": "A,AAAB;;ABCDE;" +} +*/ + +namespace { // anonymous + +struct JSONSourceMap +{ + /// File version (always the first entry in the object) and must be a positive integer. + int32_t version = 3; + /// An optional name of the generated code that this source map is associated with. + String file; + /// An optional source root, useful for relocating source files on a server or removing repeated values in + /// the “sources” entry. This value is prepended to the individual entries in the “source” field. + String sourceRoot; + /// A list of original sources used by the “mappings” entry. + List<UnownedStringSlice> sources; + /// An optional list of source content, useful when the “source” can’t be hosted. The contents are listed in the same order as the sources in line 5. + /// “null” may be used if some original sources should be retrieved by name. + /// Because could be a string or nullptr, we use JSONValue to hold value. + List<JSONValue> sourcesContent; + /// A list of symbol names used by the “mappings” entry. + List<UnownedStringSlice> names; + /// A string with the encoded mapping data. + UnownedStringSlice mappings; + + static const StructRttiInfo g_rttiInfo; +}; + +} // anonymous + +static const StructRttiInfo _makeJSONSourceMap_Rtti() +{ + JSONSourceMap obj; + + StructRttiBuilder builder(&obj, "SourceMap", nullptr); + + builder.addField("version", &obj.version); + builder.addField("file", &obj.file); + builder.addField("sourceRoot", &obj.sourceRoot, StructRttiInfo::Flag::Optional); + builder.addField("sources", &obj.sources); + builder.addField("sourcesContent", &obj.sourcesContent, StructRttiInfo::Flag::Optional); + builder.addField("names", &obj.names, StructRttiInfo::Flag::Optional); + builder.addField("mappings", &obj.mappings); + + return builder.make(); +} +/* static */const StructRttiInfo JSONSourceMap::g_rttiInfo = _makeJSONSourceMap_Rtti(); + +// Encode a 6 bit value to VLQ encoding +static const char g_vlqEncodeTable[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; + +struct VlqDecodeTable +{ + VlqDecodeTable() + { + ::memset(map, -1, sizeof(map)); + for (Index i = 0; i < SLANG_COUNT_OF(g_vlqEncodeTable); ++i) + { + map[g_vlqEncodeTable[i]] = int8_t(i); + } + } + /// Returns a *negative* value if invalid + SLANG_FORCE_INLINE int8_t operator[](char c) const { return (c & ~char(0x7f)) ? -1 : map[c]; } + + int8_t map[128]; +}; + +static const VlqDecodeTable g_vlqDecodeTable; + +/* +https://docs.google.com/document/d/1U1RGAehQwRypUTovF1KRlpiOFze0b-_2gc6fAH0KY0k/edit?hl=en_US&pli=1&pli=1# +The VLQ is a Base64 value, where the most significant bit (the 6th bit) is used as the continuation +bit, and the “digits” are encoded into the string least significant first, and where the least significant +bit of the first digit is used as the sign bit. */ + +static SlangResult _decode(UnownedStringSlice& ioEncoded, Index& out) +{ + Index v = 0; + + Index shift = 0; + const char* cur = ioEncoded.begin(); + const char* end = ioEncoded.end(); + + // Must have some chars + if (cur >= end) + { + return SLANG_FAIL; + } + + for (; cur < end; ++cur) + { + const Index value = g_vlqDecodeTable[*cur]; + if (value < 0) + { + return SLANG_FAIL; + } + + v += (value & 0x1f) << shift; + + // If the continuation bit is not set we are done + if (( value & 0x20) == 0) + { + ++cur; + break; + } + + shift += 5; + } + + // Save out the remaining part + ioEncoded = UnownedStringSlice(cur, end); + + // Handle negating + out = (v & 1) ? -(v >> 1) : (v >> 1); + return SLANG_OK; +} + +void _encode(Index v, StringBuilder& out) +{ + // Double to free up low bit to hold the sign + v += v; + + // We want to make v always positive to encode + // we use the last bit to indicate negativity + v = (v < 0) ? (1 - v) : v; + + // We'll use a simple buffer, so as to not have to constantly update he StringBuffer + char dst[8]; + char* cur = dst; + + do + { + // Encode it + const auto nextV = v >> 5; + + // Encode 5 bits + char c = g_vlqEncodeTable[(v & 0x1f)]; + + // See what bits are remaining + v = (v >> 5); + + // Set the continuation bit's if there is more to encode + c |= v ? 0x20 : 0; + + // Save the char + *cur++ = c; + } + while (v); + + out.append(dst, cur); +} + +/* static */SlangResult JSONSourceMapUtil::decode(JSONContainer* container, JSONValue root, DiagnosticSink* sink, RefPtr<SourceMap>& out) +{ + RefPtr<SourceMap> sourceMap; + + // Let's try and decode the JSON into native types to make this easier... + RttiTypeFuncsMap typeMap = JSONNativeUtil::getTypeFuncsMap(); + + // Convert to native + JSONSourceMap native; + { + JSONToNativeConverter converter(container, &typeMap, sink); + + // Convert to the native type + SLANG_RETURN_ON_FAIL(converter.convert(root, GetRttiInfo<JSONSourceMap>::get(), &native)); + } + + sourceMap->m_file = native.file; + sourceMap->m_sourceRoot = native.sourceRoot; + + const Count sourcesCount = native.sources.getCount(); + + // These should all be unique, but for simplicity, we build a table + sourceMap->m_sources.setCount(sourcesCount); + for (Index i = 0; i < sourcesCount; ++i) + { + sourceMap->m_sources[i] = sourceMap->m_slicePool.add(native.sources[i]); + } + + Count sourcesContentCount = native.sourcesContent.getCount(); + sourcesContentCount = std::min(sourcesContentCount, sourcesCount); + + sourceMap->m_sourcesContent.setCount(sourcesContentCount); + for (auto& cur : sourceMap->m_sourcesContent) + { + cur = StringSlicePool::kNullHandle; + } + + for (Index i = 0; i < sourcesContentCount; ++i) + { + auto value = native.sourcesContent[i]; + + if (value.type != JSONValue::Type::Null) + { + if (value.getKind() == JSONValue::Kind::String) + { + auto stringValue = container->getString(value); + sourceMap->m_sourcesContent[i] = sourceMap->m_slicePool.add(stringValue); + } + } + } + + List<UnownedStringSlice> lines; + StringUtil::split(native.mappings, ';', lines); + + List<UnownedStringSlice> segments; + + // Index into sources + Index sourceFileIndex = 0; + + Index sourceLine = 0; + Index sourceColumn = 0; + Index nameIndex = 0; + + const Count linesCount = lines.getCount(); + + sourceMap->m_lineStarts.setCount(linesCount + 1); + + for (Index generatedLine = 0; generatedLine < linesCount; ++generatedLine) + { + const auto line = lines[generatedLine]; + + sourceMap->m_lineStarts[generatedLine] = sourceMap->m_lineEntries.getCount(); + + // If it's empty move to next line + if (line.getLength() == 0) + { + continue; + } + + // Split the line into segments + segments.clear(); + StringUtil::split(line, ',', segments); + + Index generatedColumn = 0; + + for (auto segment : segments) + { + Index colDelta; + SLANG_RETURN_ON_FAIL(_decode(segment, colDelta)); + + generatedColumn += colDelta; + SLANG_ASSERT(generatedColumn >= 0); + + // It can be 4 or 5 parts + if (segment.getLength()) + { + /* If present, an zero-based index into the "sources" list. This field is a base 64 VLQ relative to the previous occurrence of this field, unless this is the first occurrence of this field, in which case the whole value is represented. + If present, the zero-based starting line in the original source represented. This field is a base 64 VLQ relative to the previous occurrence of this field, unless this is the first occurrence of this field, in which case the whole value is represented. Always present if there is a source field. + If present, the zero-based starting column of the line in the source represented. This field is a base 64 VLQ relative to the previous occurrence of this field, unless this is the first occurrence of this field, in which case the whole value is represented. Always present if there is a source field. + */ + + Index sourceFileDelta; + Index sourceLineDelta; + Index sourceColumnDelta; + + SLANG_RETURN_ON_FAIL(_decode(segment, sourceFileDelta)); + SLANG_RETURN_ON_FAIL(_decode(segment, sourceLineDelta)); + SLANG_RETURN_ON_FAIL(_decode(segment, sourceColumnDelta)); + + sourceFileIndex += sourceFileDelta; + sourceLine += sourceLineDelta; + sourceColumn += sourceColumnDelta; + + SLANG_ASSERT(sourceFileIndex >= 0); + SLANG_ASSERT(sourceLine >= 0); + SLANG_ASSERT(sourceColumn >= 0); + + // 5 parts + if (segment.getLength() > 0) + { + /* If present, the zero - based index into the "names" list associated with this segment. + This field is a base 64 VLQ relative to the previous occurrence of this field, unless this is the first occurrence + of this field, in which case the whole value is represented. + */ + + Index nameDelta; + SLANG_RETURN_ON_FAIL(_decode(segment, nameDelta)); + + nameIndex += nameDelta; + SLANG_ASSERT(nameIndex >= 0); + } + } + + SourceMap::Entry entry; + entry.generatedColumn = generatedColumn; + entry.sourceColumn = sourceColumn; + entry.sourceLine = sourceLine; + entry.sourceFileIndex = sourceFileIndex; + entry.nameIndex = nameIndex; + + sourceMap->m_lineEntries.add(entry); + } + } + + // Mark the end + sourceMap->m_lineStarts[linesCount] = sourceMap->m_lineEntries.getCount(); + + out.swapWith(sourceMap); + + return SLANG_OK; +} + +SlangResult JSONSourceMapUtil::encode(SourceMap* sourceMap, JSONContainer* container, DiagnosticSink* sink, JSONValue& outValue) +{ + // Convert to native + JSONSourceMap native; + + native.file = sourceMap->m_file; + native.sourceRoot = sourceMap->m_sourceRoot; + + // Copy over the sources + { + const auto count = sourceMap->m_sources.getCount(); + native.sources.setCount(count); + for (Index i = 0; i < count; ++i) + { + native.sources[i] = sourceMap->m_slicePool.getSlice(sourceMap->m_sources[i]); + } + } + + // Copy out the sourcesContent, care is needed around handling null + { + const auto count = sourceMap->m_sourcesContent.getCount(); + native.sourcesContent.setCount(count); + for (Index i = 0; i < count; ++i) + { + const auto srcValue = sourceMap->m_sourcesContent[i]; + + const JSONValue dstValue = (srcValue == StringSlicePool::kNullHandle) ? + native.sourcesContent[i] = JSONValue::makeNull() : + container->createString(sourceMap->m_slicePool.getSlice(srcValue)); + + native.sourcesContent[i] = dstValue; + } + } + + // Copy out the names + { + const auto count = sourceMap->m_names.getCount(); + native.names.setCount(count); + for (Index i = 0; i < count; ++i) + { + native.names[i] = sourceMap->m_slicePool.getSlice(sourceMap->m_names[i]); + } + } + + StringBuilder mappings; + + // Do the encoding! + { + const Count linesCount = sourceMap->getGeneratedLineCount(); + + Index sourceFileIndex = 0; + + Index sourceLine = 0; + Index sourceColumn = 0; + Index nameIndex = 0; + + for (Index i = 0; i < linesCount; ++i) + { + // Add the semicolon to start the line + if (i > 0) + { + mappings.appendChar(';'); + } + + const auto entries = sourceMap->getEntriesForLine(i); + const auto entriesCount = entries.getCount(); + + if (entriesCount == 0) + { + continue; + } + + // We reset the generated column index at the start of each new generated line + Index generatedColumn = 0; + + for (Index j = 0; j < entriesCount; ++j) + { + auto entry = entries[j]; + + if (j > 0) + { + mappings.appendChar(','); + } + + Index generatedDelta = entry.generatedColumn - generatedColumn; + generatedColumn = entry.generatedColumn; + + _encode(generatedDelta, mappings); + + // See if there any other deltas we need to handle + const Index sourceFileDelta = entry.sourceFileIndex - sourceFileIndex; + const Index sourceLineDelta = entry.sourceLine - sourceLine; + const Index sourceColumnDelta = entry.sourceColumn - sourceColumn; + const Index nameIndexDelta = entry.nameIndex - nameIndex; + + if (sourceFileDelta || sourceLineDelta || sourceColumnDelta || nameIndex) + { + // Okay we have to encode all these deltae + _encode(sourceFileDelta, mappings); + _encode(sourceLineDelta, mappings); + _encode(sourceColumnDelta, mappings); + + // Update these values + sourceFileIndex = entry.sourceFileIndex; + sourceLine = entry.sourceLine; + sourceColumn = entry.sourceColumn; + + if (nameIndexDelta) + { + _encode(nameIndexDelta, mappings); + nameIndex = entry.nameIndex; + } + } + } + } + } + + // Set the mappings + native.mappings = mappings.getUnownedSlice(); + + // Write it out + { + RttiTypeFuncsMap typeMap = JSONNativeUtil::getTypeFuncsMap(); + + NativeToJSONConverter converter(container, &typeMap, sink); + SLANG_RETURN_ON_FAIL(converter.convert(GetRttiInfo<JSONSourceMap>::get(), &native, outValue)); + } + + return SLANG_OK; +} + +} // namespace Slang |