securityos/node_modules/@jridgewell/trace-mapping/dist/trace-mapping.umd.js

601 lines
25 KiB
JavaScript
Raw Permalink Normal View History

2024-09-06 15:32:35 +00:00
(function (global, factory) {
typeof exports === 'object' && typeof module !== 'undefined' ? factory(exports, require('@jridgewell/sourcemap-codec'), require('@jridgewell/resolve-uri')) :
typeof define === 'function' && define.amd ? define(['exports', '@jridgewell/sourcemap-codec', '@jridgewell/resolve-uri'], factory) :
(global = typeof globalThis !== 'undefined' ? globalThis : global || self, factory(global.traceMapping = {}, global.sourcemapCodec, global.resolveURI));
})(this, (function (exports, sourcemapCodec, resolveUri) { 'use strict';
function resolve(input, base) {
// The base is always treated as a directory, if it's not empty.
// https://github.com/mozilla/source-map/blob/8cb3ee57/lib/util.js#L327
// https://github.com/chromium/chromium/blob/da4adbb3/third_party/blink/renderer/devtools/front_end/sdk/SourceMap.js#L400-L401
if (base && !base.endsWith('/'))
base += '/';
return resolveUri(input, base);
}
/**
* Removes everything after the last "/", but leaves the slash.
*/
function stripFilename(path) {
if (!path)
return '';
const index = path.lastIndexOf('/');
return path.slice(0, index + 1);
}
const COLUMN = 0;
const SOURCES_INDEX = 1;
const SOURCE_LINE = 2;
const SOURCE_COLUMN = 3;
const NAMES_INDEX = 4;
const REV_GENERATED_LINE = 1;
const REV_GENERATED_COLUMN = 2;
function maybeSort(mappings, owned) {
const unsortedIndex = nextUnsortedSegmentLine(mappings, 0);
if (unsortedIndex === mappings.length)
return mappings;
// If we own the array (meaning we parsed it from JSON), then we're free to directly mutate it. If
// not, we do not want to modify the consumer's input array.
if (!owned)
mappings = mappings.slice();
for (let i = unsortedIndex; i < mappings.length; i = nextUnsortedSegmentLine(mappings, i + 1)) {
mappings[i] = sortSegments(mappings[i], owned);
}
return mappings;
}
function nextUnsortedSegmentLine(mappings, start) {
for (let i = start; i < mappings.length; i++) {
if (!isSorted(mappings[i]))
return i;
}
return mappings.length;
}
function isSorted(line) {
for (let j = 1; j < line.length; j++) {
if (line[j][COLUMN] < line[j - 1][COLUMN]) {
return false;
}
}
return true;
}
function sortSegments(line, owned) {
if (!owned)
line = line.slice();
return line.sort(sortComparator);
}
function sortComparator(a, b) {
return a[COLUMN] - b[COLUMN];
}
let found = false;
/**
* A binary search implementation that returns the index if a match is found.
* If no match is found, then the left-index (the index associated with the item that comes just
* before the desired index) is returned. To maintain proper sort order, a splice would happen at
* the next index:
*
* ```js
* const array = [1, 3];
* const needle = 2;
* const index = binarySearch(array, needle, (item, needle) => item - needle);
*
* assert.equal(index, 0);
* array.splice(index + 1, 0, needle);
* assert.deepEqual(array, [1, 2, 3]);
* ```
*/
function binarySearch(haystack, needle, low, high) {
while (low <= high) {
const mid = low + ((high - low) >> 1);
const cmp = haystack[mid][COLUMN] - needle;
if (cmp === 0) {
found = true;
return mid;
}
if (cmp < 0) {
low = mid + 1;
}
else {
high = mid - 1;
}
}
found = false;
return low - 1;
}
function upperBound(haystack, needle, index) {
for (let i = index + 1; i < haystack.length; index = i++) {
if (haystack[i][COLUMN] !== needle)
break;
}
return index;
}
function lowerBound(haystack, needle, index) {
for (let i = index - 1; i >= 0; index = i--) {
if (haystack[i][COLUMN] !== needle)
break;
}
return index;
}
function memoizedState() {
return {
lastKey: -1,
lastNeedle: -1,
lastIndex: -1,
};
}
/**
* This overly complicated beast is just to record the last tested line/column and the resulting
* index, allowing us to skip a few tests if mappings are monotonically increasing.
*/
function memoizedBinarySearch(haystack, needle, state, key) {
const { lastKey, lastNeedle, lastIndex } = state;
let low = 0;
let high = haystack.length - 1;
if (key === lastKey) {
if (needle === lastNeedle) {
found = lastIndex !== -1 && haystack[lastIndex][COLUMN] === needle;
return lastIndex;
}
if (needle >= lastNeedle) {
// lastIndex may be -1 if the previous needle was not found.
low = lastIndex === -1 ? 0 : lastIndex;
}
else {
high = lastIndex;
}
}
state.lastKey = key;
state.lastNeedle = needle;
return (state.lastIndex = binarySearch(haystack, needle, low, high));
}
// Rebuilds the original source files, with mappings that are ordered by source line/column instead
// of generated line/column.
function buildBySources(decoded, memos) {
const sources = memos.map(buildNullArray);
for (let i = 0; i < decoded.length; i++) {
const line = decoded[i];
for (let j = 0; j < line.length; j++) {
const seg = line[j];
if (seg.length === 1)
continue;
const sourceIndex = seg[SOURCES_INDEX];
const sourceLine = seg[SOURCE_LINE];
const sourceColumn = seg[SOURCE_COLUMN];
const originalSource = sources[sourceIndex];
const originalLine = (originalSource[sourceLine] || (originalSource[sourceLine] = []));
const memo = memos[sourceIndex];
// The binary search either found a match, or it found the left-index just before where the
// segment should go. Either way, we want to insert after that. And there may be multiple
// generated segments associated with an original location, so there may need to move several
// indexes before we find where we need to insert.
let index = upperBound(originalLine, sourceColumn, memoizedBinarySearch(originalLine, sourceColumn, memo, sourceLine));
memo.lastIndex = ++index;
insert(originalLine, index, [sourceColumn, i, seg[COLUMN]]);
}
}
return sources;
}
function insert(array, index, value) {
for (let i = array.length; i > index; i--) {
array[i] = array[i - 1];
}
array[index] = value;
}
// Null arrays allow us to use ordered index keys without actually allocating contiguous memory like
// a real array. We use a null-prototype object to avoid prototype pollution and deoptimizations.
// Numeric properties on objects are magically sorted in ascending order by the engine regardless of
// the insertion order. So, by setting any numeric keys, even out of order, we'll get ascending
// order when iterating with for-in.
function buildNullArray() {
return { __proto__: null };
}
const AnyMap = function (map, mapUrl) {
const parsed = parse(map);
if (!('sections' in parsed)) {
return new TraceMap(parsed, mapUrl);
}
const mappings = [];
const sources = [];
const sourcesContent = [];
const names = [];
const ignoreList = [];
recurse(parsed, mapUrl, mappings, sources, sourcesContent, names, ignoreList, 0, 0, Infinity, Infinity);
const joined = {
version: 3,
file: parsed.file,
names,
sources,
sourcesContent,
mappings,
ignoreList,
};
return presortedDecodedMap(joined);
};
function parse(map) {
return typeof map === 'string' ? JSON.parse(map) : map;
}
function recurse(input, mapUrl, mappings, sources, sourcesContent, names, ignoreList, lineOffset, columnOffset, stopLine, stopColumn) {
const { sections } = input;
for (let i = 0; i < sections.length; i++) {
const { map, offset } = sections[i];
let sl = stopLine;
let sc = stopColumn;
if (i + 1 < sections.length) {
const nextOffset = sections[i + 1].offset;
sl = Math.min(stopLine, lineOffset + nextOffset.line);
if (sl === stopLine) {
sc = Math.min(stopColumn, columnOffset + nextOffset.column);
}
else if (sl < stopLine) {
sc = columnOffset + nextOffset.column;
}
}
addSection(map, mapUrl, mappings, sources, sourcesContent, names, ignoreList, lineOffset + offset.line, columnOffset + offset.column, sl, sc);
}
}
function addSection(input, mapUrl, mappings, sources, sourcesContent, names, ignoreList, lineOffset, columnOffset, stopLine, stopColumn) {
const parsed = parse(input);
if ('sections' in parsed)
return recurse(...arguments);
const map = new TraceMap(parsed, mapUrl);
const sourcesOffset = sources.length;
const namesOffset = names.length;
const decoded = decodedMappings(map);
const { resolvedSources, sourcesContent: contents, ignoreList: ignores } = map;
append(sources, resolvedSources);
append(names, map.names);
if (contents)
append(sourcesContent, contents);
else
for (let i = 0; i < resolvedSources.length; i++)
sourcesContent.push(null);
if (ignores)
for (let i = 0; i < ignores.length; i++)
ignoreList.push(ignores[i] + sourcesOffset);
for (let i = 0; i < decoded.length; i++) {
const lineI = lineOffset + i;
// We can only add so many lines before we step into the range that the next section's map
// controls. When we get to the last line, then we'll start checking the segments to see if
// they've crossed into the column range. But it may not have any columns that overstep, so we
// still need to check that we don't overstep lines, too.
if (lineI > stopLine)
return;
// The out line may already exist in mappings (if we're continuing the line started by a
// previous section). Or, we may have jumped ahead several lines to start this section.
const out = getLine(mappings, lineI);
// On the 0th loop, the section's column offset shifts us forward. On all other lines (since the
// map can be multiple lines), it doesn't.
const cOffset = i === 0 ? columnOffset : 0;
const line = decoded[i];
for (let j = 0; j < line.length; j++) {
const seg = line[j];
const column = cOffset + seg[COLUMN];
// If this segment steps into the column range that the next section's map controls, we need
// to stop early.
if (lineI === stopLine && column >= stopColumn)
return;
if (seg.length === 1) {
out.push([column]);
continue;
}
const sourcesIndex = sourcesOffset + seg[SOURCES_INDEX];
const sourceLine = seg[SOURCE_LINE];
const sourceColumn = seg[SOURCE_COLUMN];
out.push(seg.length === 4
? [column, sourcesIndex, sourceLine, sourceColumn]
: [column, sourcesIndex, sourceLine, sourceColumn, namesOffset + seg[NAMES_INDEX]]);
}
}
}
function append(arr, other) {
for (let i = 0; i < other.length; i++)
arr.push(other[i]);
}
function getLine(arr, index) {
for (let i = arr.length; i <= index; i++)
arr[i] = [];
return arr[index];
}
const LINE_GTR_ZERO = '`line` must be greater than 0 (lines start at line 1)';
const COL_GTR_EQ_ZERO = '`column` must be greater than or equal to 0 (columns start at column 0)';
const LEAST_UPPER_BOUND = -1;
const GREATEST_LOWER_BOUND = 1;
class TraceMap {
constructor(map, mapUrl) {
const isString = typeof map === 'string';
if (!isString && map._decodedMemo)
return map;
const parsed = (isString ? JSON.parse(map) : map);
const { version, file, names, sourceRoot, sources, sourcesContent } = parsed;
this.version = version;
this.file = file;
this.names = names || [];
this.sourceRoot = sourceRoot;
this.sources = sources;
this.sourcesContent = sourcesContent;
this.ignoreList = parsed.ignoreList || parsed.x_google_ignoreList || undefined;
const from = resolve(sourceRoot || '', stripFilename(mapUrl));
this.resolvedSources = sources.map((s) => resolve(s || '', from));
const { mappings } = parsed;
if (typeof mappings === 'string') {
this._encoded = mappings;
this._decoded = undefined;
}
else {
this._encoded = undefined;
this._decoded = maybeSort(mappings, isString);
}
this._decodedMemo = memoizedState();
this._bySources = undefined;
this._bySourceMemos = undefined;
}
}
/**
* Typescript doesn't allow friend access to private fields, so this just casts the map into a type
* with public access modifiers.
*/
function cast(map) {
return map;
}
/**
* Returns the encoded (VLQ string) form of the SourceMap's mappings field.
*/
function encodedMappings(map) {
var _a;
var _b;
return ((_a = (_b = cast(map))._encoded) !== null && _a !== void 0 ? _a : (_b._encoded = sourcemapCodec.encode(cast(map)._decoded)));
}
/**
* Returns the decoded (array of lines of segments) form of the SourceMap's mappings field.
*/
function decodedMappings(map) {
var _a;
return ((_a = cast(map))._decoded || (_a._decoded = sourcemapCodec.decode(cast(map)._encoded)));
}
/**
* A low-level API to find the segment associated with a generated line/column (think, from a
* stack trace). Line and column here are 0-based, unlike `originalPositionFor`.
*/
function traceSegment(map, line, column) {
const decoded = decodedMappings(map);
// It's common for parent source maps to have pointers to lines that have no
// mapping (like a "//# sourceMappingURL=") at the end of the child file.
if (line >= decoded.length)
return null;
const segments = decoded[line];
const index = traceSegmentInternal(segments, cast(map)._decodedMemo, line, column, GREATEST_LOWER_BOUND);
return index === -1 ? null : segments[index];
}
/**
* A higher-level API to find the source/line/column associated with a generated line/column
* (think, from a stack trace). Line is 1-based, but column is 0-based, due to legacy behavior in
* `source-map` library.
*/
function originalPositionFor(map, needle) {
let { line, column, bias } = needle;
line--;
if (line < 0)
throw new Error(LINE_GTR_ZERO);
if (column < 0)
throw new Error(COL_GTR_EQ_ZERO);
const decoded = decodedMappings(map);
// It's common for parent source maps to have pointers to lines that have no
// mapping (like a "//# sourceMappingURL=") at the end of the child file.
if (line >= decoded.length)
return OMapping(null, null, null, null);
const segments = decoded[line];
const index = traceSegmentInternal(segments, cast(map)._decodedMemo, line, column, bias || GREATEST_LOWER_BOUND);
if (index === -1)
return OMapping(null, null, null, null);
const segment = segments[index];
if (segment.length === 1)
return OMapping(null, null, null, null);
const { names, resolvedSources } = map;
return OMapping(resolvedSources[segment[SOURCES_INDEX]], segment[SOURCE_LINE] + 1, segment[SOURCE_COLUMN], segment.length === 5 ? names[segment[NAMES_INDEX]] : null);
}
/**
* Finds the generated line/column position of the provided source/line/column source position.
*/
function generatedPositionFor(map, needle) {
const { source, line, column, bias } = needle;
return generatedPosition(map, source, line, column, bias || GREATEST_LOWER_BOUND, false);
}
/**
* Finds all generated line/column positions of the provided source/line/column source position.
*/
function allGeneratedPositionsFor(map, needle) {
const { source, line, column, bias } = needle;
// SourceMapConsumer uses LEAST_UPPER_BOUND for some reason, so we follow suit.
return generatedPosition(map, source, line, column, bias || LEAST_UPPER_BOUND, true);
}
/**
* Iterates each mapping in generated position order.
*/
function eachMapping(map, cb) {
const decoded = decodedMappings(map);
const { names, resolvedSources } = map;
for (let i = 0; i < decoded.length; i++) {
const line = decoded[i];
for (let j = 0; j < line.length; j++) {
const seg = line[j];
const generatedLine = i + 1;
const generatedColumn = seg[0];
let source = null;
let originalLine = null;
let originalColumn = null;
let name = null;
if (seg.length !== 1) {
source = resolvedSources[seg[1]];
originalLine = seg[2] + 1;
originalColumn = seg[3];
}
if (seg.length === 5)
name = names[seg[4]];
cb({
generatedLine,
generatedColumn,
source,
originalLine,
originalColumn,
name,
});
}
}
}
function sourceIndex(map, source) {
const { sources, resolvedSources } = map;
let index = sources.indexOf(source);
if (index === -1)
index = resolvedSources.indexOf(source);
return index;
}
/**
* Retrieves the source content for a particular source, if its found. Returns null if not.
*/
function sourceContentFor(map, source) {
const { sourcesContent } = map;
if (sourcesContent == null)
return null;
const index = sourceIndex(map, source);
return index === -1 ? null : sourcesContent[index];
}
/**
* Determines if the source is marked to ignore by the source map.
*/
function isIgnored(map, source) {
const { ignoreList } = map;
if (ignoreList == null)
return false;
const index = sourceIndex(map, source);
return index === -1 ? false : ignoreList.includes(index);
}
/**
* A helper that skips sorting of the input map's mappings array, which can be expensive for larger
* maps.
*/
function presortedDecodedMap(map, mapUrl) {
const tracer = new TraceMap(clone(map, []), mapUrl);
cast(tracer)._decoded = map.mappings;
return tracer;
}
/**
* Returns a sourcemap object (with decoded mappings) suitable for passing to a library that expects
* a sourcemap, or to JSON.stringify.
*/
function decodedMap(map) {
return clone(map, decodedMappings(map));
}
/**
* Returns a sourcemap object (with encoded mappings) suitable for passing to a library that expects
* a sourcemap, or to JSON.stringify.
*/
function encodedMap(map) {
return clone(map, encodedMappings(map));
}
function clone(map, mappings) {
return {
version: map.version,
file: map.file,
names: map.names,
sourceRoot: map.sourceRoot,
sources: map.sources,
sourcesContent: map.sourcesContent,
mappings,
ignoreList: map.ignoreList || map.x_google_ignoreList,
};
}
function OMapping(source, line, column, name) {
return { source, line, column, name };
}
function GMapping(line, column) {
return { line, column };
}
function traceSegmentInternal(segments, memo, line, column, bias) {
let index = memoizedBinarySearch(segments, column, memo, line);
if (found) {
index = (bias === LEAST_UPPER_BOUND ? upperBound : lowerBound)(segments, column, index);
}
else if (bias === LEAST_UPPER_BOUND)
index++;
if (index === -1 || index === segments.length)
return -1;
return index;
}
function sliceGeneratedPositions(segments, memo, line, column, bias) {
let min = traceSegmentInternal(segments, memo, line, column, GREATEST_LOWER_BOUND);
// We ignored the bias when tracing the segment so that we're guarnateed to find the first (in
// insertion order) segment that matched. Even if we did respect the bias when tracing, we would
// still need to call `lowerBound()` to find the first segment, which is slower than just looking
// for the GREATEST_LOWER_BOUND to begin with. The only difference that matters for us is when the
// binary search didn't match, in which case GREATEST_LOWER_BOUND just needs to increment to
// match LEAST_UPPER_BOUND.
if (!found && bias === LEAST_UPPER_BOUND)
min++;
if (min === -1 || min === segments.length)
return [];
// We may have found the segment that started at an earlier column. If this is the case, then we
// need to slice all generated segments that match _that_ column, because all such segments span
// to our desired column.
const matchedColumn = found ? column : segments[min][COLUMN];
// The binary search is not guaranteed to find the lower bound when a match wasn't found.
if (!found)
min = lowerBound(segments, matchedColumn, min);
const max = upperBound(segments, matchedColumn, min);
const result = [];
for (; min <= max; min++) {
const segment = segments[min];
result.push(GMapping(segment[REV_GENERATED_LINE] + 1, segment[REV_GENERATED_COLUMN]));
}
return result;
}
function generatedPosition(map, source, line, column, bias, all) {
var _a;
line--;
if (line < 0)
throw new Error(LINE_GTR_ZERO);
if (column < 0)
throw new Error(COL_GTR_EQ_ZERO);
const { sources, resolvedSources } = map;
let sourceIndex = sources.indexOf(source);
if (sourceIndex === -1)
sourceIndex = resolvedSources.indexOf(source);
if (sourceIndex === -1)
return all ? [] : GMapping(null, null);
const generated = ((_a = cast(map))._bySources || (_a._bySources = buildBySources(decodedMappings(map), (cast(map)._bySourceMemos = sources.map(memoizedState)))));
const segments = generated[sourceIndex][line];
if (segments == null)
return all ? [] : GMapping(null, null);
const memo = cast(map)._bySourceMemos[sourceIndex];
if (all)
return sliceGeneratedPositions(segments, memo, line, column, bias);
const index = traceSegmentInternal(segments, memo, line, column, bias);
if (index === -1)
return GMapping(null, null);
const segment = segments[index];
return GMapping(segment[REV_GENERATED_LINE] + 1, segment[REV_GENERATED_COLUMN]);
}
exports.AnyMap = AnyMap;
exports.GREATEST_LOWER_BOUND = GREATEST_LOWER_BOUND;
exports.LEAST_UPPER_BOUND = LEAST_UPPER_BOUND;
exports.TraceMap = TraceMap;
exports.allGeneratedPositionsFor = allGeneratedPositionsFor;
exports.decodedMap = decodedMap;
exports.decodedMappings = decodedMappings;
exports.eachMapping = eachMapping;
exports.encodedMap = encodedMap;
exports.encodedMappings = encodedMappings;
exports.generatedPositionFor = generatedPositionFor;
exports.isIgnored = isIgnored;
exports.originalPositionFor = originalPositionFor;
exports.presortedDecodedMap = presortedDecodedMap;
exports.sourceContentFor = sourceContentFor;
exports.traceSegment = traceSegment;
}));
//# sourceMappingURL=trace-mapping.umd.js.map