trackparse 0.1.0 → 0.2.0

package/dist/index.cjs

@@ -1,7 +1,7 @@
 'use strict';
 
 // src/data.generated.ts
-var SPEC_VERSION = "0.1.0";
+var SPEC_VERSION = "0.2.0";
 var descriptorsData = {
   "descriptors": [
     "radio",
@@ -469,11 +469,11 @@ function isAllDigits(s) {
   for (let i = 0; i < s.length; i++) if (!isDigit(s[i])) return false;
   return true;
 }
-var ALPHABETIC = /^\p{Alphabetic}$/u;
+var LETTER = /^\p{L}$/u;
 function isLetterAt(s, index) {
   const cp = s.codePointAt(index);
   if (cp === void 0) return false;
-  return ALPHABETIC.test(String.fromCodePoint(cp));
+  return LETTER.test(String.fromCodePoint(cp));
 }
 function asciiLower(s) {
   let out = "";
@@ -705,8 +705,15 @@ function byPos(items) {
 }
 
 // src/tables.ts
+function stringList(value) {
+  return Array.isArray(value) ? value.filter((s) => typeof s === "string") : [];
+}
+function stringEntries(value) {
+  if (typeof value !== "object" || value === null || Array.isArray(value)) return [];
+  return Object.entries(value).filter((e) => typeof e[1] === "string");
+}
 function lowerKeys(list) {
-  return (list ?? []).map((s) => asciiLower(s).trim()).filter((s) => s.length > 0);
+  return stringList(list).map((s) => asciiLower(s).trim()).filter((s) => s.length > 0);
 }
 function buildTables(keywords) {
   const featMarkers = new PhraseTable(
@@ -722,7 +729,7 @@ function buildTables(keywords) {
   );
   const junkEntries = [
     ...Object.entries(junkData.phrases),
-    ...Object.entries(keywords?.junk ?? {})
+    ...stringEntries(keywords?.junk)
   ];
   const genres = [...genresData.genres, ...lowerKeys(keywords?.genres)];
   const junkPhrases = new PhraseTable(junkEntries);
@@ -732,7 +739,7 @@ function buildTables(keywords) {
   ]);
   const headEntries = [
     ...Object.entries(versionKeywordsData.heads),
-    ...Object.entries(keywords?.versionHeads ?? {})
+    ...stringEntries(keywords?.versionHeads)
   ];
   const heads = new PhraseTable(headEntries);
   const runInfo = /* @__PURE__ */ new Map();
@@ -966,7 +973,7 @@ function normalize(input) {
 
 // src/context.ts
 function makeCtx(t, options) {
-  const known = (options?.knownArtists ?? []).filter((k) => typeof k === "string");
+  const known = stringList(options?.knownArtists);
   const knownKeys = new Set(known.map(dedupKey).filter((k) => k.length > 0));
   const knownMaxLen = known.reduce((m, k) => Math.max(m, k.length), 0);
   const splitAnd = options?.splitAnd ?? "auto";
@@ -2088,21 +2095,28 @@ function extractUnbracketedCredits(input, out, ctx) {
   return skel;
 }
 function stripTrailingJunk(input, out, ctx) {
-  let skel = trimSkel(input);
+  const skel = trimSkel(input);
+  const spans = wordSpans(skel.text);
+  const words = spans.map((s) => s.text);
+  let end = spans.length;
   for (; ; ) {
-    const hit = trailingJunkPhrase(skel.text, ctx);
-    if (!hit) return skel;
-    let before = trimSkel(sliceSkel(skel, 0, hit.start));
-    const lastChar = before.text[before.text.length - 1];
-    const lone = before.text.length === 1 || before.text[before.text.length - 2] === " ";
-    if ((lastChar === "-" || lastChar === "|") && lone) {
-      before = trimSkel(sliceSkel(before, 0, before.text.length - 1));
-    }
-    if (before.text.length === 0) return skel;
-    const raw = collapseSpaces(skel.text.slice(hit.start));
-    out.junk.push({ raw, kind: hit.kind, pos: skel.pos[hit.start] ?? 0 });
-    skel = before;
+    const m = ctx.t.junkPhrases.matchEnding(words, end);
+    if (!m) break;
+    const first = end - m.length;
+    let keep = first;
+    if (keep > 0 && isLoneEdge(words[keep - 1])) keep--;
+    if (keep === 0) break;
+    const from = spans[first]?.start ?? 0;
+    const to = spans[end - 1]?.end ?? skel.text.length;
+    out.junk.push({
+      raw: collapseSpaces(skel.text.slice(from, to)),
+      kind: m.entry.value,
+      pos: skel.pos[from] ?? 0
+    });
+    end = keep;
   }
+  if (end === spans.length) return skel;
+  return trimSkel(sliceSkel(skel, 0, spans[end - 1]?.end ?? 0));
 }
 function isLoneEdge(word) {
   return word === "-" || word === "|";
@@ -2242,16 +2256,21 @@ function createParser(base = {}) {
   };
 }
 function mergeKeywords(a, b) {
-  const x = a.keywords ?? {};
-  const y = b.keywords ?? {};
+  const x = isRecord(a.keywords) ? a.keywords : {};
+  const y = isRecord(b.keywords) ? b.keywords : {};
+  const map = (k) => Object.fromEntries([...stringEntries(x[k]), ...stringEntries(y[k])]);
+  const list = (k) => [...stringList(x[k]), ...stringList(y[k])];
   return {
-    versionHeads: { ...x.versionHeads, ...y.versionHeads },
-    descriptors: [...x.descriptors ?? [], ...y.descriptors ?? []],
-    genres: [...x.genres ?? [], ...y.genres ?? []],
-    junk: { ...x.junk, ...y.junk },
-    featMarkers: [...x.featMarkers ?? [], ...y.featMarkers ?? []]
+    versionHeads: map("versionHeads"),
+    descriptors: list("descriptors"),
+    genres: list("genres"),
+    junk: map("junk"),
+    featMarkers: list("featMarkers")
   };
 }
+function isRecord(v) {
+  return typeof v === "object" && v !== null && !Array.isArray(v);
+}
 var defaultParser = createParser();
 function parse(input, options) {
   return defaultParser.parse(input, options);

package/dist/index.d.cts

@@ -1,4 +1,4 @@
-declare const SPEC_VERSION = "0.1.0";
+declare const SPEC_VERSION = "0.2.0";
 
 /** Public types. They mirror spec/schema/parsed-track.schema.json and options.schema.json. */
 type Mode = "clean" | "youtube" | "filename";

package/dist/index.d.ts

@@ -1,4 +1,4 @@
-declare const SPEC_VERSION = "0.1.0";
+declare const SPEC_VERSION = "0.2.0";
 
 /** Public types. They mirror spec/schema/parsed-track.schema.json and options.schema.json. */
 type Mode = "clean" | "youtube" | "filename";

package/dist/index.js

@@ -1,5 +1,5 @@
 // src/data.generated.ts
-var SPEC_VERSION = "0.1.0";
+var SPEC_VERSION = "0.2.0";
 var descriptorsData = {
   "descriptors": [
     "radio",
@@ -467,11 +467,11 @@ function isAllDigits(s) {
   for (let i = 0; i < s.length; i++) if (!isDigit(s[i])) return false;
   return true;
 }
-var ALPHABETIC = /^\p{Alphabetic}$/u;
+var LETTER = /^\p{L}$/u;
 function isLetterAt(s, index) {
   const cp = s.codePointAt(index);
   if (cp === void 0) return false;
-  return ALPHABETIC.test(String.fromCodePoint(cp));
+  return LETTER.test(String.fromCodePoint(cp));
 }
 function asciiLower(s) {
   let out = "";
@@ -703,8 +703,15 @@ function byPos(items) {
 }
 
 // src/tables.ts
+function stringList(value) {
+  return Array.isArray(value) ? value.filter((s) => typeof s === "string") : [];
+}
+function stringEntries(value) {
+  if (typeof value !== "object" || value === null || Array.isArray(value)) return [];
+  return Object.entries(value).filter((e) => typeof e[1] === "string");
+}
 function lowerKeys(list) {
-  return (list ?? []).map((s) => asciiLower(s).trim()).filter((s) => s.length > 0);
+  return stringList(list).map((s) => asciiLower(s).trim()).filter((s) => s.length > 0);
 }
 function buildTables(keywords) {
   const featMarkers = new PhraseTable(
@@ -720,7 +727,7 @@ function buildTables(keywords) {
   );
   const junkEntries = [
     ...Object.entries(junkData.phrases),
-    ...Object.entries(keywords?.junk ?? {})
+    ...stringEntries(keywords?.junk)
   ];
   const genres = [...genresData.genres, ...lowerKeys(keywords?.genres)];
   const junkPhrases = new PhraseTable(junkEntries);
@@ -730,7 +737,7 @@ function buildTables(keywords) {
   ]);
   const headEntries = [
     ...Object.entries(versionKeywordsData.heads),
-    ...Object.entries(keywords?.versionHeads ?? {})
+    ...stringEntries(keywords?.versionHeads)
   ];
   const heads = new PhraseTable(headEntries);
   const runInfo = /* @__PURE__ */ new Map();
@@ -964,7 +971,7 @@ function normalize(input) {
 
 // src/context.ts
 function makeCtx(t, options) {
-  const known = (options?.knownArtists ?? []).filter((k) => typeof k === "string");
+  const known = stringList(options?.knownArtists);
   const knownKeys = new Set(known.map(dedupKey).filter((k) => k.length > 0));
   const knownMaxLen = known.reduce((m, k) => Math.max(m, k.length), 0);
   const splitAnd = options?.splitAnd ?? "auto";
@@ -2086,21 +2093,28 @@ function extractUnbracketedCredits(input, out, ctx) {
   return skel;
 }
 function stripTrailingJunk(input, out, ctx) {
-  let skel = trimSkel(input);
+  const skel = trimSkel(input);
+  const spans = wordSpans(skel.text);
+  const words = spans.map((s) => s.text);
+  let end = spans.length;
   for (; ; ) {
-    const hit = trailingJunkPhrase(skel.text, ctx);
-    if (!hit) return skel;
-    let before = trimSkel(sliceSkel(skel, 0, hit.start));
-    const lastChar = before.text[before.text.length - 1];
-    const lone = before.text.length === 1 || before.text[before.text.length - 2] === " ";
-    if ((lastChar === "-" || lastChar === "|") && lone) {
-      before = trimSkel(sliceSkel(before, 0, before.text.length - 1));
-    }
-    if (before.text.length === 0) return skel;
-    const raw = collapseSpaces(skel.text.slice(hit.start));
-    out.junk.push({ raw, kind: hit.kind, pos: skel.pos[hit.start] ?? 0 });
-    skel = before;
+    const m = ctx.t.junkPhrases.matchEnding(words, end);
+    if (!m) break;
+    const first = end - m.length;
+    let keep = first;
+    if (keep > 0 && isLoneEdge(words[keep - 1])) keep--;
+    if (keep === 0) break;
+    const from = spans[first]?.start ?? 0;
+    const to = spans[end - 1]?.end ?? skel.text.length;
+    out.junk.push({
+      raw: collapseSpaces(skel.text.slice(from, to)),
+      kind: m.entry.value,
+      pos: skel.pos[from] ?? 0
+    });
+    end = keep;
   }
+  if (end === spans.length) return skel;
+  return trimSkel(sliceSkel(skel, 0, spans[end - 1]?.end ?? 0));
 }
 function isLoneEdge(word) {
   return word === "-" || word === "|";
@@ -2240,16 +2254,21 @@ function createParser(base = {}) {
   };
 }
 function mergeKeywords(a, b) {
-  const x = a.keywords ?? {};
-  const y = b.keywords ?? {};
+  const x = isRecord(a.keywords) ? a.keywords : {};
+  const y = isRecord(b.keywords) ? b.keywords : {};
+  const map = (k) => Object.fromEntries([...stringEntries(x[k]), ...stringEntries(y[k])]);
+  const list = (k) => [...stringList(x[k]), ...stringList(y[k])];
   return {
-    versionHeads: { ...x.versionHeads, ...y.versionHeads },
-    descriptors: [...x.descriptors ?? [], ...y.descriptors ?? []],
-    genres: [...x.genres ?? [], ...y.genres ?? []],
-    junk: { ...x.junk, ...y.junk },
-    featMarkers: [...x.featMarkers ?? [], ...y.featMarkers ?? []]
+    versionHeads: map("versionHeads"),
+    descriptors: list("descriptors"),
+    genres: list("genres"),
+    junk: map("junk"),
+    featMarkers: list("featMarkers")
   };
 }
+function isRecord(v) {
+  return typeof v === "object" && v !== null && !Array.isArray(v);
+}
 var defaultParser = createParser();
 function parse(input, options) {
   return defaultParser.parse(input, options);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "trackparse",
-  "version": "0.1.0",
+  "version": "0.2.0",
   "description": "Spec-driven parser for music track strings: artists, feat credits, remixes/versions, junk and more from single track strings (store titles, YouTube titles, filenames).",
   "license": "MIT",
   "author": "Maksim Maksimov",