@gmod/trix 0.2.0 → 2.0.0

package/CHANGELOG.md

@@ -0,0 +1,19 @@
+- Fix issue with infinite loop
+- Add abortsignal support
+- Only query first word when string with multiple words is entered
+
+# v1.0.0
+
+- Change result format from just the "result" string returned to be "term,result"
+
+# v0.2.1
+
+- Fix error when identifiers contain commas
+
+# v0.2.0
+
+- Improve performance of fetches with sequential chunk parsing
+
+# v0.1.1
+
+- Initial release

package/README.md

@@ -1,4 +1,7 @@
+[![Build Status](https://img.shields.io/github/workflow/status/GMOD/trix-js/Push/main?logo=github&style=for-the-badge)](https://github.com/GMOD/trix-js/actions?query=branch%3Amain+workflow%3APush+)
+
 # trix-js
+
 Read UCSC Trix indexes in pure JavaScript
 
 ## Usage
@@ -9,37 +12,43 @@ import { RemoteFile } from 'generic-filehandle'
 
 // any filehandle object that supports the Nodejs FileHandle API will work.
 // We use generic-filehandle here to demonstrate searching files on remote servers.
-const ixxFile = new RemoteFile('https://hgdownload.soe.ucsc.edu/gbdb/hg38/knownGene.ixx');
-const ixFile = new RemoteFile('https://hgdownload.soe.ucsc.edu/gbdb/hg38/knownGene.ix');
+const ixxFile = new RemoteFile(
+  'https://hgdownload.soe.ucsc.edu/gbdb/hg38/knownGene.ixx',
+)
+const ixFile = new RemoteFile(
+  'https://hgdownload.soe.ucsc.edu/gbdb/hg38/knownGene.ix',
+)
 
-const trix = new Trix(ixxFile, ixFile);
+const trix = new Trix(ixxFile, ixFile)
 
 async function doStuff() {
-  const results = await trix.search('oca');
-  console.log(results);
+  const results = await trix.search('oca')
+  console.log(results)
 }
-doStuff();
-
+doStuff()
 ```
 
 ## Documentation
+
 ### Trix constructor
+
 The Trix class constructor accepts arguments:
+
 - `ixxFile` - a filehandle object for the trix .ixx file
 - `ixFile` - a filehandle object for the trix .ix file
 - `maxResults = 20` - an optional number specifying the maximum number of results to return on `trix.search()`
 
-
 ### Trix search
+
 **Search the index files for a term and find its keys.**<br>
 **In the case of searching with multiple words, `trix.search()` finds the intersection of the result sets.**<br>
 The Trix search function accepts argument:
+
 - `searchString` - a string of space-separated words for what to search the index file and find keys for<br>
-  
+
 The Trix search function returns: <br>
-- `Promise<string[]>` - a promised array of strings where each string is an itemId result
-  
 
+- `Promise<[term,result][] as [string,string][]>` - an array of [term, result] pairs where each term is the left column in the trix and the right column is the trix match
 
 ## Examples
 
@@ -47,49 +56,37 @@ The Trix search function returns: <br>
 import { LocalFile } from 'generic-filehandle'
 import Trix from '@gmod/trix'
 
-const ixxFile = new LocalFile('out.ixx');
-const ixFile = new LocalFile('out.ix');
+const ixxFile = new LocalFile('out.ixx')
+const ixFile = new LocalFile('out.ix')
 
 // limit maxResults to 5
-const trix = new Trix(ixxFile, ixFile, 5);
+const trix = new Trix(ixxFile, ixFile, 5)
 
 async function doStuff() {
-const results1 = await trix.search('herc');
-  console.log(results1);
+  const results1 = await trix.search('herc')
+  console.log(results1)
 
   // increase maxResults to 30
-  trix.maxResults = 30;
+  trix.maxResults = 30
 
-  const results2 = await trix.search('linc');
-  console.log(results2);
+  const results2 = await trix.search('linc')
+  console.log(results2)
 }
 
-doStuff();
+doStuff()
 ```
-<br><br>
-
-
-
-
-
-
-
-
-
-
-
-
-
 
+<br><br>
 
 ## Development
 
-
 ### Test trix-js
+
 First, clone this repo and install npm packages. <br>
 Then, run `npm test`. <br>
 
 ### Test the USCS TrixSearch - Requires Linux
+
 First, clone this repo.
 To run test searches on a track hub using the USCS `TrixSearch`, navigate to `tests/testdata/test#` and run `bash test#script.sh` where # is the test number.
 To change search terms, edit `searchterms.txt`.
@@ -97,7 +94,9 @@ To change search terms, edit `searchterms.txt`.
 **Wondering what to search for?**<br>
 Open up `tests/testdata/test#/input.txt`.
 
-
 **How to test my own .gff.gz data?**<br>
 Navigate to `/test/rawGenomes` and create a directory with your .gff.gz file in it. From within that directory, run `bash ../../programs/gff3ToInput.sh <.gff3.gz FILE> <OUTPUT NAME>`.
 
+## Reference
+
+See https://genome.ucsc.edu/goldenPath/help/trix.html for basic concepts of trix and https://github.com/GMOD/ixixx-js for a javascript implementation of the ixIxx command

package/dist/index.d.ts

@@ -1,63 +1,13 @@
-/// <reference types="node" />
-import type { FileHandle } from 'fs/promises';
-import type { LocalFile, RemoteFile, BlobFile } from 'generic-filehandle';
-declare type AnyFile = LocalFile | RemoteFile | BlobFile | FileHandle;
-export default class Trix {
-    private index;
-    private ixFile;
-    maxResults: number;
-    /**
-     * @param ixxFile [anyFile] the second-level trix index file produced by ixIxx.
-     * @param ixFile [anyFile] the first-level trix index file produced by ixIxx.
-     * @param maxResults [number] the maximum number of results to return. Default is set to 20.
-     */
-    constructor(ixxFile: AnyFile, ixFile: AnyFile, maxResults?: number);
-    /**
-     * Search trix for the given searchWord(s). Return up to {this.maxResults} results.
-     * This method matches each index prefix against each searchWord. It does not do fuzzy matching.
-     *
-     * @param searchString [string] term(s) separated by spaces to search for id(s).
-     * @returns results [Array<string>] where each string is a corresponding itemId.
-     */
-    search(searchString: string): Promise<string[]>;
-    /**
-     * Seek ahead to the correct position in the .ix file,
-     * then load that chunk of .ix into a buffer.
-     *
-     * @param searchWord [string]
-     * @returns a Buffer holding the sections we want to search.
-     */
-    private _getBuffer;
-    /**
-     * Given the end position of the last buffer,
-     * load the next chunk  of .ix data into a buffer and return it.
-     *
-     * @param seekPosStart [number] where to start loading data into the new buffer.
-     * @returns a Buffer holding the chunk we want to search.
-     */
-    private _getNextChunk;
-    /**
-   * Create and return a buffer given the start and end position
-   * of what to load from the .ix file.
-   *
-   * @param seekPosStart [number] byte the buffer should start reading from file.
-   * @param seekPosEnd [number] byte the buffer should stop reading from file.
-   * @returns a Buffer holding the chunk of data.
-   */
-    private _createBuffer;
-    /**
-     * Takes in a hit string and returns an array of result terms.
-     *
-     * @param line [string] The line of .ix that is a hit.
-     * @returns results [Array<hit>]. Each hit contains the itemId [string], and wordPos [number].
-     */
-    private _parseHitString;
-    /**
-     * Parses ixx file and constructs a map of {word: ixFileLocation}
-     *
-     * @param ixxFile [anyFile] second level index that is produced by ixIxx.
-     * @returns a ParsedIxx map.
-     */
-    private _parseIxx;
-}
-export {};
+import type { GenericFilehandle } from 'generic-filehandle';
+export default class Trix {
+    private ixFile;
+    private ixxFile;
+    maxResults: number;
+    constructor(ixxFile: GenericFilehandle, ixFile: GenericFilehandle, maxResults?: number);
+    search(searchString: string, opts?: {
+        signal?: AbortSignal;
+    }): Promise<string[][]>;
+    private getIndex;
+    private _getBuffer;
+    private _parseIxx;
+}

package/dist/index.js

@@ -1,8 +1,280 @@
-
-'use strict'
-
-if (process.env.NODE_ENV === 'production') {
-  module.exports = require('./trix.cjs.production.min.js')
-} else {
-  module.exports = require('./trix.cjs.development.js')
-}
+"use strict";
+var __assign = (this && this.__assign) || function () {
+    __assign = Object.assign || function(t) {
+        for (var s, i = 1, n = arguments.length; i < n; i++) {
+            s = arguments[i];
+            for (var p in s) if (Object.prototype.hasOwnProperty.call(s, p))
+                t[p] = s[p];
+        }
+        return t;
+    };
+    return __assign.apply(this, arguments);
+};
+var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
+    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
+    return new (P || (P = Promise))(function (resolve, reject) {
+        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
+        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
+        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
+        step((generator = generator.apply(thisArg, _arguments || [])).next());
+    });
+};
+var __generator = (this && this.__generator) || function (thisArg, body) {
+    var _ = { label: 0, sent: function() { if (t[0] & 1) throw t[1]; return t[1]; }, trys: [], ops: [] }, f, y, t, g;
+    return g = { next: verb(0), "throw": verb(1), "return": verb(2) }, typeof Symbol === "function" && (g[Symbol.iterator] = function() { return this; }), g;
+    function verb(n) { return function (v) { return step([n, v]); }; }
+    function step(op) {
+        if (f) throw new TypeError("Generator is already executing.");
+        while (_) try {
+            if (f = 1, y && (t = op[0] & 2 ? y["return"] : op[0] ? y["throw"] || ((t = y["return"]) && t.call(y), 0) : y.next) && !(t = t.call(y, op[1])).done) return t;
+            if (y = 0, t) op = [op[0] & 2, t.value];
+            switch (op[0]) {
+                case 0: case 1: t = op; break;
+                case 4: _.label++; return { value: op[1], done: false };
+                case 5: _.label++; y = op[1]; op = [0]; continue;
+                case 7: op = _.ops.pop(); _.trys.pop(); continue;
+                default:
+                    if (!(t = _.trys, t = t.length > 0 && t[t.length - 1]) && (op[0] === 6 || op[0] === 2)) { _ = 0; continue; }
+                    if (op[0] === 3 && (!t || (op[1] > t[0] && op[1] < t[3]))) { _.label = op[1]; break; }
+                    if (op[0] === 6 && _.label < t[1]) { _.label = t[1]; t = op; break; }
+                    if (t && _.label < t[2]) { _.label = t[2]; _.ops.push(op); break; }
+                    if (t[2]) _.ops.pop();
+                    _.trys.pop(); continue;
+            }
+            op = body.call(thisArg, _);
+        } catch (e) { op = [6, e]; y = 0; } finally { f = t = 0; }
+        if (op[0] & 5) throw op[1]; return { value: op[0] ? op[1] : void 0, done: true };
+    }
+};
+var __read = (this && this.__read) || function (o, n) {
+    var m = typeof Symbol === "function" && o[Symbol.iterator];
+    if (!m) return o;
+    var i = m.call(o), r, ar = [], e;
+    try {
+        while ((n === void 0 || n-- > 0) && !(r = i.next()).done) ar.push(r.value);
+    }
+    catch (error) { e = { error: error }; }
+    finally {
+        try {
+            if (r && !r.done && (m = i["return"])) m.call(i);
+        }
+        finally { if (e) throw e.error; }
+    }
+    return ar;
+};
+var __spreadArray = (this && this.__spreadArray) || function (to, from, pack) {
+    if (pack || arguments.length === 2) for (var i = 0, l = from.length, ar; i < l; i++) {
+        if (ar || !(i in from)) {
+            if (!ar) ar = Array.prototype.slice.call(from, 0, i);
+            ar[i] = from[i];
+        }
+    }
+    return to.concat(ar || Array.prototype.slice.call(from));
+};
+var __values = (this && this.__values) || function(o) {
+    var s = typeof Symbol === "function" && Symbol.iterator, m = s && o[s], i = 0;
+    if (m) return m.call(o);
+    if (o && typeof o.length === "number") return {
+        next: function () {
+            if (o && i >= o.length) o = void 0;
+            return { value: o && o[i++], done: !o };
+        }
+    };
+    throw new TypeError(s ? "Object is not iterable." : "Symbol.iterator is not defined.");
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+var trixPrefixSize = 5;
+var CHUNKSIZE = 65536;
+// Define this object with .ixx and .ix files.
+// Then use the trixSearch() function to search for a word.
+var Trix = /** @class */ (function () {
+    function Trix(ixxFile, ixFile, maxResults) {
+        if (maxResults === void 0) { maxResults = 20; }
+        this.ixFile = ixFile;
+        this.ixxFile = ixxFile;
+        this.maxResults = maxResults;
+    }
+    Trix.prototype.search = function (searchString, opts) {
+        return __awaiter(this, void 0, void 0, function () {
+            var resultArr, searchWords, _loop_1, this_1, w;
+            return __generator(this, function (_a) {
+                switch (_a.label) {
+                    case 0:
+                        resultArr = [];
+                        searchWords = searchString.split(' ');
+                        _loop_1 = function (w) {
+                            var searchWord, done, res, prevLen, _loop_2, state_1;
+                            return __generator(this, function (_b) {
+                                switch (_b.label) {
+                                    case 0:
+                                        searchWord = searchWords[w].toLowerCase();
+                                        done = false;
+                                        return [4 /*yield*/, this_1._getBuffer(searchWord, opts)];
+                                    case 1:
+                                        res = _b.sent();
+                                        prevLen = void 0;
+                                        _loop_2 = function () {
+                                            var seekPosEnd, buffer, foundSomething, str, lines, hits, res_1;
+                                            return __generator(this, function (_c) {
+                                                switch (_c.label) {
+                                                    case 0:
+                                                        seekPosEnd = res.seekPosEnd, buffer = res.buffer;
+                                                        foundSomething = false;
+                                                        str = buffer.toString();
+                                                        lines = str
+                                                            .slice(0, str.lastIndexOf('\n'))
+                                                            .split('\n')
+                                                            .filter(function (f) { return !!f; });
+                                                        hits = lines
+                                                            .filter(function (line) {
+                                                            var word = line.split(' ')[0];
+                                                            var match = word.startsWith(searchString);
+                                                            if (!foundSomething && match) {
+                                                                foundSomething = true;
+                                                            }
+                                                            else if (foundSomething && !match) {
+                                                                done = true;
+                                                            }
+                                                            else if (word > searchString) {
+                                                                done = true;
+                                                            }
+                                                            return match;
+                                                        })
+                                                            .map(function (line) {
+                                                            var _a = __read(line.split(' ')), term = _a[0], parts = _a.slice(1);
+                                                            return parts.map(function (elt) { return [term, elt.split(',')[0]]; });
+                                                        })
+                                                            .flat();
+                                                        if (!hits.length) {
+                                                            done = true;
+                                                        }
+                                                        if (prevLen === hits.length) {
+                                                            done = true;
+                                                        }
+                                                        if (!(resultArr.length + hits.length < this_1.maxResults && !done)) return [3 /*break*/, 2];
+                                                        return [4 /*yield*/, this_1.ixFile.read(Buffer.alloc(CHUNKSIZE), 0, CHUNKSIZE, seekPosEnd, opts)
+                                                            //early break if empty response
+                                                        ];
+                                                    case 1:
+                                                        res_1 = _c.sent();
+                                                        //early break if empty response
+                                                        if (!res_1.bytesRead) {
+                                                            resultArr = resultArr.concat(hits);
+                                                            return [2 /*return*/, "break"];
+                                                        }
+                                                        buffer = Buffer.concat([buffer, res_1.buffer]);
+                                                        seekPosEnd += CHUNKSIZE;
+                                                        prevLen = hits.length;
+                                                        return [3 /*break*/, 3];
+                                                    case 2:
+                                                        if (resultArr.length + hits.length >= this_1.maxResults || done) {
+                                                            resultArr = resultArr.concat(hits);
+                                                            return [2 /*return*/, "break"];
+                                                        }
+                                                        _c.label = 3;
+                                                    case 3: return [2 /*return*/];
+                                                }
+                                            });
+                                        };
+                                        _b.label = 2;
+                                    case 2:
+                                        if (!(res && !done)) return [3 /*break*/, 4];
+                                        return [5 /*yield**/, _loop_2()];
+                                    case 3:
+                                        state_1 = _b.sent();
+                                        if (state_1 === "break")
+                                            return [3 /*break*/, 4];
+                                        return [3 /*break*/, 2];
+                                    case 4: return [2 /*return*/];
+                                }
+                            });
+                        };
+                        this_1 = this;
+                        w = 0;
+                        _a.label = 1;
+                    case 1:
+                        if (!(w < searchWords.length)) return [3 /*break*/, 4];
+                        return [5 /*yield**/, _loop_1(w)];
+                    case 2:
+                        _a.sent();
+                        _a.label = 3;
+                    case 3:
+                        w++;
+                        return [3 /*break*/, 1];
+                    case 4: return [2 /*return*/, __spreadArray([], __read(resultArr), false).slice(0, this.maxResults)];
+                }
+            });
+        });
+    };
+    Trix.prototype.getIndex = function (opts) {
+        return this._parseIxx(this.ixxFile, opts);
+    };
+    Trix.prototype._getBuffer = function (searchWord, opts) {
+        return __awaiter(this, void 0, void 0, function () {
+            var seekPosStart, seekPosEnd, indexes, indexes_1, indexes_1_1, _a, key, value, trimmedKey, len, res;
+            var e_1, _b;
+            return __generator(this, function (_c) {
+                switch (_c.label) {
+                    case 0:
+                        seekPosStart = 0;
+                        seekPosEnd = -1;
+                        return [4 /*yield*/, this.getIndex(opts)];
+                    case 1:
+                        indexes = _c.sent();
+                        try {
+                            for (indexes_1 = __values(indexes), indexes_1_1 = indexes_1.next(); !indexes_1_1.done; indexes_1_1 = indexes_1.next()) {
+                                _a = __read(indexes_1_1.value, 2), key = _a[0], value = _a[1];
+                                trimmedKey = key.slice(0, searchWord.length);
+                                if (trimmedKey >= searchWord) {
+                                    break;
+                                }
+                                else {
+                                    seekPosStart = value;
+                                    seekPosEnd = value + 65536;
+                                }
+                            }
+                        }
+                        catch (e_1_1) { e_1 = { error: e_1_1 }; }
+                        finally {
+                            try {
+                                if (indexes_1_1 && !indexes_1_1.done && (_b = indexes_1.return)) _b.call(indexes_1);
+                            }
+                            finally { if (e_1) throw e_1.error; }
+                        }
+                        len = seekPosEnd - seekPosStart;
+                        if (len < 0) {
+                            return [2 /*return*/, undefined];
+                        }
+                        return [4 /*yield*/, this.ixFile.read(Buffer.alloc(len), 0, len, seekPosStart, opts)];
+                    case 2:
+                        res = _c.sent();
+                        return [2 /*return*/, __assign(__assign({}, res), { seekPosEnd: seekPosEnd })];
+                }
+            });
+        });
+    };
+    Trix.prototype._parseIxx = function (ixxFile, opts) {
+        return __awaiter(this, void 0, void 0, function () {
+            var file;
+            return __generator(this, function (_a) {
+                switch (_a.label) {
+                    case 0: return [4 /*yield*/, ixxFile.readFile(__assign({ encoding: 'utf8' }, opts))];
+                    case 1:
+                        file = (_a.sent());
+                        return [2 /*return*/, new Map(file
+                                .split('\n')
+                                .filter(function (f) { return !!f; })
+                                .map(function (line) {
+                                var prefix = line.slice(0, trixPrefixSize);
+                                var posStr = line.slice(trixPrefixSize);
+                                var pos = Number.parseInt(posStr, 16);
+                                return [prefix, pos];
+                            }))];
+                }
+            });
+        });
+    };
+    return Trix;
+}());
+exports.default = Trix;
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAEA,IAAM,cAAc,GAAG,CAAC,CAAA;AAExB,IAAM,SAAS,GAAG,KAAK,CAAA;AAEvB,8CAA8C;AAC9C,2DAA2D;AAC3D;IAKE,cACE,OAA0B,EAC1B,MAAyB,EACzB,UAAe;QAAf,2BAAA,EAAA,eAAe;QAEf,IAAI,CAAC,MAAM,GAAG,MAAM,CAAA;QACpB,IAAI,CAAC,OAAO,GAAG,OAAO,CAAA;QACtB,IAAI,CAAC,UAAU,GAAG,UAAU,CAAA;IAC9B,CAAC;IAEK,qBAAM,GAAZ,UAAa,YAAoB,EAAE,IAA+B;;;;;;wBAC5D,SAAS,GAAG,EAAgB,CAAA;wBAC1B,WAAW,GAAG,YAAY,CAAC,KAAK,CAAC,GAAG,CAAC,CAAA;4CAClC,CAAC;;;;;wCACF,UAAU,GAAG,WAAW,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,CAAA;wCAC3C,IAAI,GAAG,KAAK,CAAA;wCACJ,qBAAM,OAAK,UAAU,CAAC,UAAU,EAAE,IAAI,CAAC,EAAA;;wCAA7C,GAAG,GAAG,SAAuC;wCAC/C,OAAO,SAAA,CAAA;;;;;;wDAGH,UAAU,GAAa,GAAG,WAAhB,EAAE,MAAM,GAAK,GAAG,OAAR,CAAQ;wDAC5B,cAAc,GAAG,KAAK,CAAA;wDACpB,GAAG,GAAG,MAAM,CAAC,QAAQ,EAAE,CAAA;wDAIvB,KAAK,GAAG,GAAG;6DACd,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,WAAW,CAAC,IAAI,CAAC,CAAC;6DAC/B,KAAK,CAAC,IAAI,CAAC;6DACX,MAAM,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,CAAC,EAAH,CAAG,CAAC,CAAA;wDAEb,IAAI,GAAG,KAAK;6DACf,MAAM,CAAC,UAAA,IAAI;4DACV,IAAM,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAA;4DAC/B,IAAM,KAAK,GAAG,IAAI,CAAC,UAAU,CAAC,YAAY,CAAC,CAAA;4DAC3C,IAAI,CAAC,cAAc,IAAI,KAAK,EAAE;gEAC5B,cAAc,GAAG,IAAI,CAAA;6DACtB;iEAAM,IAAI,cAAc,IAAI,CAAC,KAAK,EAAE;gEACnC,IAAI,GAAG,IAAI,CAAA;6DACZ;iEAAM,IAAI,IAAI,GAAG,YAAY,EAAE;gEAC9B,IAAI,GAAG,IAAI,CAAA;6DACZ;4DACD,OAAO,KAAK,CAAA;wDACd,CAAC,CAAC;6DACD,GAAG,CAAC,UAAA,IAAI;4DACD,IAAA,KAAA,OAAmB,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAA,EAAjC,IAAI,QAAA,EAAK,KAAK,cAAmB,CAAA;4DACxC,OAAO,KAAK,CAAC,GAAG,CAAC,UAAA,GAAG,IAAI,OAAA,CAAC,IAAI,EAAE,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,EAAzB,CAAyB,CAAC,CAAA;wDACpD,CAAC,CAAC;6DACD,IAAI,EAAwB,CAAA;wDAE/B,IAAI,CAAC,IAAI,CAAC,MAAM,EAAE;4DAChB,IAAI,GAAG,IAAI,CAAA;yDACZ;wDACD,IAAI,OAAO,KAAK,IAAI,CAAC,MAAM,EAAE;4DAC3B,IAAI,GAAG,IAAI,CAAA;yDACZ;6DAEG,CAAA,SAAS,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,GAAG,OAAK,UAAU,IAAI,CAAC,IAAI,CAAA,EAAzD,wBAAyD;wDAC/C,qBAAM,OAAK,MAAM,CAAC,IAAI,CAChC,MAAM,CAAC,KAAK,CAAC,SAAS,CAAC,EACvB,CAAC,EACD,SAAS,EACT,UAAU,EACV,IAAI,CACL;4DAED,+BAA+B;0DAF9B;;wDANK,QAAM,SAMX;wDAED,+BAA+B;wDAC/B,IAAI,CAAC,KAAG,CAAC,SAAS,EAAE;4DAClB,SAAS,GAAG,SAAS,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;;yDAEnC;wDACD,MAAM,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,KAAG,CAAC,MAAM,CAAC,CAAC,CAAA;wDAC5C,UAAU,IAAI,SAAS,CAAA;wDACvB,OAAO,GAAG,IAAI,CAAC,MAAM,CAAA;;;wDAChB,IAAI,SAAS,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,IAAI,OAAK,UAAU,IAAI,IAAI,EAAE;4DACpE,SAAS,GAAG,SAAS,CAAC,MAAM,CAAC,IAAI,CAAC,CAAA;;yDAEnC;;;;;;;;6CA1DI,CAAA,GAAG,IAAI,CAAC,IAAI,CAAA;;;;;;;;;;;;wBANZ,CAAC,GAAG,CAAC;;;6BAAE,CAAA,CAAC,GAAG,WAAW,CAAC,MAAM,CAAA;sDAA7B,CAAC;;;;;wBAA8B,CAAC,EAAE,CAAA;;4BAoE3C,sBAAO,yBAAI,SAAS,UAAE,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,UAAU,CAAC,EAAA;;;;KAChD;IAEO,uBAAQ,GAAhB,UAAiB,IAA+B;QAC9C,OAAO,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,OAAO,EAAE,IAAI,CAAC,CAAA;IAC3C,CAAC;IAEa,yBAAU,GAAxB,UACE,UAAkB,EAClB,IAA+B;;;;;;;wBAE3B,YAAY,GAAG,CAAC,CAAA;wBAChB,UAAU,GAAG,CAAC,CAAC,CAAA;wBACH,qBAAM,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,EAAA;;wBAAnC,OAAO,GAAG,SAAyB;;4BACzC,KAA2B,YAAA,SAAA,OAAO,CAAA,qFAAE;gCAAzB,KAAA,4BAAY,EAAX,GAAG,QAAA,EAAE,KAAK,QAAA;gCACd,UAAU,GAAG,GAAG,CAAC,KAAK,CAAC,CAAC,EAAE,UAAU,CAAC,MAAM,CAAC,CAAA;gCAClD,IAAI,UAAU,IAAI,UAAU,EAAE;oCAC5B,MAAK;iCACN;qCAAM;oCACL,YAAY,GAAG,KAAK,CAAA;oCACpB,UAAU,GAAG,KAAK,GAAG,KAAK,CAAA;iCAC3B;6BACF;;;;;;;;;wBAGK,GAAG,GAAG,UAAU,GAAG,YAAY,CAAA;wBACrC,IAAI,GAAG,GAAG,CAAC,EAAE;4BACX,sBAAO,SAAS,EAAA;yBACjB;wBACW,qBAAM,IAAI,CAAC,MAAM,CAAC,IAAI,CAChC,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,EACjB,CAAC,EACD,GAAG,EACH,YAAY,EACZ,IAAI,CACL,EAAA;;wBANK,GAAG,GAAG,SAMX;wBACD,4CACK,GAAG,KACN,UAAU,YAAA,KACX;;;;KACF;IAEa,wBAAS,GAAvB,UACE,OAA0B,EAC1B,IAA+B;;;;;4BAEjB,qBAAM,OAAO,CAAC,QAAQ,YAClC,QAAQ,EAAE,MAAM,IACb,IAAI,EACP,EAAA;;wBAHI,IAAI,GAAG,CAAC,SAGZ,CAAW;wBACb,sBAAO,IAAI,GAAG,CACZ,IAAI;iCACD,KAAK,CAAC,IAAI,CAAC;iCACX,MAAM,CAAC,UAAA,CAAC,IAAI,OAAA,CAAC,CAAC,CAAC,EAAH,CAAG,CAAC;iCAChB,GAAG,CAAC,UAAA,IAAI;gCACP,IAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,cAAc,CAAC,CAAA;gCAC5C,IAAM,MAAM,GAAG,IAAI,CAAC,KAAK,CAAC,cAAc,CAAC,CAAA;gCACzC,IAAM,GAAG,GAAG,MAAM,CAAC,QAAQ,CAAC,MAAM,EAAE,EAAE,CAAC,CAAA;gCACvC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAA;4BACtB,CAAC,CAAC,CACL,EAAA;;;;KACF;IACH,WAAC;AAAD,CAAC,AApJD,IAoJC"}

package/esm/index.d.ts

@@ -0,0 +1,12 @@
+import type { GenericFilehandle } from 'generic-filehandle';
+export default class Trix {
+    private ixFile;
+    private ixxFile;
+    maxResults: number;
+    constructor(ixxFile: GenericFilehandle, ixFile: GenericFilehandle, maxResults?: number);
+    search(searchString: string, opts?: {
+        signal?: AbortSignal;
+    }): Promise<[string, string][]>;
+    private getIndex;
+    private _getBuffer;
+}

package/esm/index.js

@@ -0,0 +1,120 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+const TRIX_PREFIX_SIZE = 5;
+const CHUNK_SIZE = 65536;
+// https://stackoverflow.com/a/9229821/2129219
+function uniqBy(a, key) {
+    let seen = new Set();
+    return a.filter(item => {
+        let k = key(item);
+        return seen.has(k) ? false : seen.add(k);
+    });
+}
+class Trix {
+    constructor(ixxFile, ixFile, maxResults = 20) {
+        this.ixFile = ixFile;
+        this.ixxFile = ixxFile;
+        this.maxResults = maxResults;
+    }
+    async search(searchString, opts) {
+        let resultArr = [];
+        const searchWords = searchString.split(' ');
+        // we only search one word at a time
+        const searchWord = searchWords[0].toLowerCase();
+        const res = await this._getBuffer(searchWord, opts);
+        if (!res) {
+            return [];
+        }
+        let { seekPosEnd, buffer } = res;
+        let done = false;
+        while (!done) {
+            let foundSomething = false;
+            const str = buffer.toString();
+            // slice to lastIndexOf('\n') to make sure we get complete records
+            // since the buffer fetch could get halfway into a record
+            const lines = str
+                .slice(0, str.lastIndexOf('\n'))
+                .split('\n')
+                .filter(f => !!f);
+            const hits = lines
+                // eslint-disable-next-line @typescript-eslint/no-loop-func
+                .filter(line => {
+                const word = line.split(' ')[0];
+                const match = word.startsWith(searchWord);
+                if (!foundSomething && match) {
+                    foundSomething = true;
+                }
+                // we are done scanning if we are lexicographically greater than the
+                // search string
+                if (word > searchWord) {
+                    done = true;
+                }
+                return match;
+            })
+                .map(line => {
+                const [term, ...parts] = line.split(' ');
+                return parts.map(elt => [term, elt.split(',')[0]]);
+            })
+                .flat();
+            // if we are not done, and we haven't filled up maxResults with hits yet,
+            // then refetch
+            if (resultArr.length + hits.length < this.maxResults && !done) {
+                // eslint-disable-next-line no-await-in-loop
+                const res2 = await this.ixFile.read(Buffer.alloc(CHUNK_SIZE), 0, CHUNK_SIZE, seekPosEnd, opts);
+                // early break if empty response
+                if (!res2.bytesRead) {
+                    resultArr = resultArr.concat(hits);
+                    break;
+                }
+                buffer = Buffer.concat([buffer, res2.buffer]);
+                seekPosEnd += CHUNK_SIZE;
+            }
+            // if we have filled up the hits, or we are detected to be done via the
+            // filtering, then return
+            else if (resultArr.length + hits.length >= this.maxResults || done) {
+                resultArr = resultArr.concat(hits);
+                break;
+            }
+        }
+        // deduplicate results based on the detail column (resultArr[1])
+        return uniqBy(resultArr, elt => elt[1]).slice(0, this.maxResults);
+    }
+    async getIndex(opts) {
+        const file = await this.ixxFile.readFile({
+            encoding: 'utf8',
+            ...opts,
+        });
+        return file
+            .split('\n')
+            .filter(f => !!f)
+            .map(line => {
+            const prefix = line.slice(0, TRIX_PREFIX_SIZE);
+            const posStr = line.slice(TRIX_PREFIX_SIZE);
+            const pos = Number.parseInt(posStr, 16);
+            return [prefix, pos];
+        });
+    }
+    async _getBuffer(searchWord, opts) {
+        let seekPosStart = 0;
+        let seekPosEnd = -1;
+        const indexes = await this.getIndex(opts);
+        indexes.forEach(([key, value]) => {
+            const trimmedKey = key.slice(0, searchWord.length);
+            if (trimmedKey < searchWord) {
+                seekPosStart = value;
+                seekPosEnd = value + 65536;
+            }
+        });
+        // Return the buffer and its end position in the file.
+        const len = seekPosEnd - seekPosStart;
+        if (len < 0) {
+            return undefined;
+        }
+        const res = await this.ixFile.read(Buffer.alloc(len), 0, len, seekPosStart, opts);
+        return {
+            ...res,
+            seekPosEnd,
+        };
+    }
+}
+exports.default = Trix;