fix(ui/import): read author, series and volume out of the download folder name

static/app.js

@@ -1010,8 +1010,8 @@ $("#import-deselect-all").addEventListener("click", () => {
 let namesTargets = [];
 
 function nameToRequest(item) {
-  const { number, rest } = splitEpisodeNumber(item.name);
-  return { title: cleanFileName(rest) || item.name, volume: number };
+  const { title, volume } = parseDownloadName(item.name);
+  return { title, volume };
 }
 
 $("#import-from-names").addEventListener("click", () => {
@@ -1023,8 +1023,10 @@ $("#import-from-names").addEventListener("click", () => {
   const type = namesTargets[0].media_type;
   namesTargets = namesTargets.filter((it) => it.media_type === type);
   $("#names-intro").textContent = t("names.intro", { n: namesTargets.length, type });
-  $("#names-series").value = lastSeries;
-  $("#names-authors").value = "";
+  // the folder names usually carry both, e.g. "Autor - Serie 03 - Titel"
+  const first = parseDownloadName(namesTargets[0].name);
+  $("#names-series").value = first.series || lastSeries;
+  $("#names-authors").value = first.authors;
   $("#names-library").innerHTML = libOptions(type) || noLibraryOption();
   renderNamesPreview();
   $("#names-dialog").showModal();
@@ -1176,6 +1178,7 @@ function splitItem(i) {
 // ---- quick request from a scanned file ----
 let quickItemIndex = null;
 let quickEpisode = null;
+let quickSeries = "";    // series read from the folder name, if the hit has none
 let quickPicked = null;   // metadata hit the manual fields were filled from
 let quickAutoSearch = false;  // the search fired on open, not by the user
 let lastSeries = "";
@@ -1200,29 +1203,115 @@ function splitEpisodeNumber(name) {
   return { number: null, rest: name };
 }
 
-function cleanFileName(name) {
+// release noise: Audible treats every word of a query as required, so "6CD",
+// "vorgelesen" or a year answers a perfectly good folder name with nothing
+const RELEASE_TAG_RE = /\b(?:mp3cd|mp3|m4b|flac|ogg|opus|epub|pdf|azw3|cbz|cbr|retail|unabridged|ungek(?:ue|ü)rzt|gek(?:ue|ü)rzt|vorgelesen|abook|audiobook|h(?:oe|ö)rbuch|\d+\s?cds?|\d+\s?mp|(?:19|20)\d{2}|kompl.*)\b/gi;
+// "Grant County 01", "Georgia-Serie 11", "017": a trailing number is the volume
+const SERIES_VOLUME_RE = /^(.*?)[\s._-]*(?:reihe|serie)?[\s._-]*(\d{1,3})$/i;
+// "Folge 124", "Band 3": a counter word in front of the number is not a series
+const COUNTER_WORD_RE = /^(?:folge|folgen|teil|band|nr\.?|nummer|episode|ep|vol|volume|track|cd|disc)$/i;
+// a language tag from the release name; uppercase only, so a Spanish "de" stays
+const LANG_TAG_RE = /\b(?:DE|EN|GER|ENG|MULTI)\b/g;
+// "Karin Slaughter", "J R R Tolkien" - every word capitalised, no digits, so a
+// series like "Die drei ???" is not mistaken for an author
+const PERSON_RE = /^\p{Lu}[\p{L}.'’-]*(?: \p{Lu}[\p{L}.'’-]*){1,3}$/u;
+
+function stripTags(name) {
   return name
-    .replace(/[._\-\[\]()]+/g, " ")
-    .replace(/\b(mp3|m4b|flac|epub|pdf|cbz|cbr|retail|unabridged|kompl.*)\b/gi, " ")
+    .replace(LANG_TAG_RE, " ")
+    .replace(RELEASE_TAG_RE, " ")
     .replace(/\s+/g, " ")
     .trim();
 }
 
+function splitSeriesVolume(part) {
+  const m = SERIES_VOLUME_RE.exec(part.trim());
+  if (!m) return null;
+  const head = m[1].trim();
+  return { series: COUNTER_WORD_RE.test(head) ? "" : head, volume: parseInt(m[2], 10) };
+}
+
+function cleanFileName(name) {
+  return stripTags(name.replace(/[._\-\[\]()]+/g, " "));
+}
+
+// Download folders follow a few scene conventions: "Autor - Serie 03 - Titel",
+// "Serie - 017 - Titel", "Autor - Titel (Serie 7)", "Autor_-_Titel-6CD-2013".
+// The series number and the tags must stay out of the search query, but the
+// leading segment must stay in it - for "Die drei ???" it is what Audible
+// titles the books after.
+function parseDownloadName(name) {
+  let series = "", volume = null, authors = "", lead = "";
+  let s = name.replace(/_/g, " ").replace(/(\p{L})\.(?=\p{L})/gu, "$1 ");
+  // "(Georgia 7)" names the series, "(ADAC Motorwelt Edition)" is noise
+  s = s.replace(/[[(]([^)\]]*)[)\]]/g, (_, inner) => {
+    const sv = splitSeriesVolume(stripTags(inner));
+    if (sv) {
+      series = sv.series || series;
+      volume = sv.volume;
+    }
+    return " ";
+  });
+  const parts = stripTags(s)
+    .split(/\s+[-–—]+\s+/)
+    .map((p) => p.replace(/^[\s.\-–—]+|[\s.\-–—]+$/g, ""))
+    .filter(Boolean);
+  if (parts.length >= 3) {
+    const sv = splitSeriesVolume(parts[1]);
+    if (sv) {
+      volume = sv.volume;
+      // a named middle part leaves the first one as the author, a bare number
+      // ("Serie - 017 - Titel") as the series
+      if (sv.series) {
+        series = sv.series;
+        authors = parts[0];
+      } else {
+        series = parts[0];
+      }
+      lead = parts.splice(0, 2)[0];
+    }
+  } else if (parts.length === 2) {
+    // "Die drei ??? 017 - Titel": series and number share the first part
+    const sv = splitSeriesVolume(parts[0]);
+    if (sv && sv.series) {
+      series = sv.series;
+      volume = sv.volume;
+      lead = sv.series;
+      parts.shift();
+    }
+  }
+  if (!lead && parts.length > 1 && PERSON_RE.test(parts[0])) {
+    authors = lead = parts.shift();
+  }
+  let rest = parts.join(" - ");
+  if (volume == null) {
+    const ep = splitEpisodeNumber(rest);
+    if (ep.number != null) {
+      volume = ep.number;
+      rest = ep.rest;
+    }
+  }
+  const title = cleanFileName(rest) || cleanFileName(name) || name;
+  return { title, volume, series, authors, query: cleanFileName(`${lead} ${rest}`) || title };
+}
+
 function openQuickDialog(i) {
   quickItemIndex = i;
   const item = scanItems[i];
   // the number wrecks the Audible search, so it becomes the volume instead
-  const { number, rest } = splitEpisodeNumber(item.name);
-  quickEpisode = number;
+  const parsed = parseDownloadName(item.name);
+  quickEpisode = parsed.volume;
+  quickSeries = parsed.series;
   $("#quick-file").textContent = (item.rel_dir ? item.rel_dir + "/" : "") + item.name;
-  $("#quick-episode").textContent = number != null ? t("quick.episodeNote", { n: number }) : "";
-  $("#quick-q").value = cleanFileName(rest);
+  $("#quick-episode").textContent =
+    parsed.volume != null ? t("quick.episodeNote", { n: parsed.volume }) : "";
+  $("#quick-q").value = parsed.query;
   quickPicked = null;
   $("#quick-req-filter").value = "";
-  $("#quick-m-title").value = cleanFileName(rest);
-  $("#quick-m-authors").value = "";
-  $("#quick-m-volume").value = number ?? "";
-  $("#quick-m-series").value = lastSeries;
+  $("#quick-m-title").value = parsed.title;
+  $("#quick-m-authors").value = parsed.authors;
+  $("#quick-m-volume").value = parsed.volume ?? "";
+  $("#quick-m-series").value = parsed.series || lastSeries;
   $("#quick-based-on").hidden = true;
   $("#quick-manual-summary").textContent = t("quick.manualSummary");
   $("#quick-manual").open = false;
@@ -1370,7 +1459,7 @@ function pickQuickResult(r) {
   return requestAndConnect({
     title: r.title, authors: r.authors, narrator: r.narrator || "",
     external_id: r.external_id, volume: r.volume ?? quickEpisode,
-    year: r.year, series: r.series, cover_url: r.cover_url,
+    year: r.year, series: r.series || quickSeries, cover_url: r.cover_url,
   });
 }
 
@@ -1380,7 +1469,7 @@ function editBeforeRequest(r) {
   quickPicked = r;
   $("#quick-m-title").value = r.title || "";
   $("#quick-m-authors").value = r.authors || "";
-  $("#quick-m-series").value = r.series || "";
+  $("#quick-m-series").value = r.series || quickSeries;
   $("#quick-m-volume").value = r.volume ?? quickEpisode ?? "";
   $("#quick-based-on").textContent = t("quick.basedOn", { title: r.title });
   $("#quick-based-on").hidden = false;