Expand occasions parsing for discussions, introductions, and conversations

cinemas/barbican.org.uk/tests/__snapshots__/index.test.js.snap

@@ -1826,6 +1826,12 @@ Outdoor Cinema",
         "format": {},
         "notes": "",
         "occasions": {
+          "inConversation": {
+            "guests": [
+              "Emma Matthews",
+              "Louis Petit",
+            ],
+          },
           "panel": true,
         },
         "screen": "Cinema 3",

cinemas/genesiscinema.co.uk/tests/__snapshots__/index.test.js.snap

@@ -480,7 +480,13 @@ exports[`Genesis Cinema retrieve and transform 1`] = `
         "bookingUrl": "https://genesis.admit-one.co.uk/seats/?perfCode=11711",
         "format": {},
         "notes": "",
-        "occasions": {},
+        "occasions": {
+          "inConversation": {
+            "guests": [
+              "Nick Walker",
+            ],
+          },
+        },
         "status": {
           "soldOut": false,
         },

cinemas/ibraaz.org/tests/transform.test.js

@@ -0,0 +1,74 @@
+const transform = require("../transform");
+
+// Turang's film page, cut down to what the transform reads, in Ibraaz's own
+// words from a November 2026 retrieve.
+const TURANG_URL = "https://ibraaz.org/whats-on/turang";
+const turangPage = `
+  <div class="hero__inner">
+    <h1 class="headline">Turang</h1>
+    <p class="line--lg">Bachtiar Siagian</p>
+  </div>
+  <div class="richtext">
+    <p>With an introduction from <strong>Bunga Siagian</strong>, cultural producer, researcher, and daughter of Bachtiar Siagian, the screening presents the restored film as an act of narrative reclamation and an archive of decolonisation, created at a time of radical political upheaval in the country of its creation. This screening of <i>Turang</i> is curated by <strong>May Adadol Ingawanij</strong> with Ibraaz in conversation with the final chapter of Taring Padi’s <i>Rakyat Pasti Menang</i> at Ibraaz London, which runs until 22 November, 2026.</p>
+  </div>
+  <div class="tt-widget-fallback">
+    <a href="https://tickets.ibraaz.org/events/ibraaz/2526567/select-date">Book</a>
+  </div>
+`;
+
+// The Ticket Tailor event says nothing of the introduction.
+const sourcedEvents = {
+  "tickettailor.com": [
+    {
+      showingId: "tickettailor.com-2526567",
+      title: "Turang",
+      url: "https://www.tickettailor.com/events/ibraaz/2526567",
+      overview: {
+        duration: 7200000,
+        categories: [],
+        directors: [],
+        actors: [],
+      },
+      performances: [
+        {
+          time: new Date("2026-11-22T15:00:00Z").getTime(),
+          notes: "",
+          bookingUrl: "https://www.tickettailor.com/events/ibraaz/2526567",
+          status: {},
+          accessibility: {},
+          format: {},
+          occasions: {},
+        },
+      ],
+      matchingHints: { overview: "Turang" },
+    },
+  ],
+};
+
+describe("Ibraaz transform", () => {
+  it("reads the occasions the film page describes", async () => {
+    const [movie] = await transform(
+      { moviePages: { [TURANG_URL]: turangPage } },
+      sourcedEvents,
+    );
+
+    expect(movie.performances[0].occasions).toEqual({
+      intro: { guests: ["Bunga Siagian"] },
+    });
+  });
+
+  it("keeps the occasions the listing found", async () => {
+    const listed = structuredClone(sourcedEvents);
+    listed["tickettailor.com"][0].performances[0].occasions = { qAndA: true };
+
+    const [movie] = await transform(
+      { moviePages: { [TURANG_URL]: turangPage } },
+      listed,
+    );
+
+    expect(movie.performances[0].occasions).toEqual({
+      intro: { guests: ["Bunga Siagian"] },
+      qAndA: true,
+    });
+  });
+});

cinemas/ibraaz.org/transform.js

@@ -2,6 +2,8 @@ const cheerio = require("cheerio");
 const {
   getText,
   createOverview,
+  createOccasions,
+  mergeOccasions,
   generateShowingId,
 } = require("../../common/utils");
 const attributes = require("./attributes");
@@ -103,11 +105,20 @@ async function transform({ moviePages }, sourcedEvents) {
         duration: sourcedEvent.overview.duration,
       },
       // The event's performances already carry the booking URL, status and the
-      // accessibility and format read off the listing; the room they screen in
-      // is the one thing only the venue's own page knows.
+      // accessibility, format and occasions read off the listing. The room
+      // they screen in is only on the venue's own page, and so is the copy
+      // that names who is introducing - "With an introduction from Bunga
+      // Siagian" is in Turang's page and not its Ticket Tailor event - so the
+      // page's occasions are added beneath the listing's.
       performances: sourcedEvent.performances.map((performance) => ({
         ...performance,
         screen: screen || undefined,
+        occasions: mergeOccasions(
+          createOccasions("", {}, overview, {
+            date: new Date(performance.time),
+          }),
+          performance.occasions,
+        ),
       })),
       matchingHints: { overview },
     });

cinemas/ica.art/tests/__snapshots__/index.test.js.snap

@@ -592,6 +592,12 @@ exports[`Institute of Contemporary Arts retrieve and transform 1`] = `
         "format": {},
         "notes": "",
         "occasions": {
+          "inConversation": {
+            "guests": [
+              "Michael Witt",
+            ],
+          },
+          "intro": true,
           "qAndA": true,
         },
         "screen": "Cinema 1",
@@ -1452,6 +1458,12 @@ exports[`Institute of Contemporary Arts retrieve and transform 1`] = `
         "format": {},
         "notes": "",
         "occasions": {
+          "inConversation": {
+            "guests": [
+              "Alexander Carey-Morgan",
+              "Laurence Hills",
+            ],
+          },
           "qAndA": true,
         },
         "screen": "Cinema 1",

cinemas/phoenixcinema.co.uk/tests/__snapshots__/index.test.js.snap

@@ -69,6 +69,7 @@ exports[`Phoenix Cinema retrieve and transform 1`] = `
         "format": {},
         "notes": "This screening will be followed by a Q&A",
         "occasions": {
+          "intro": true,
           "qAndA": true,
         },
         "screen": "1",
@@ -107,6 +108,7 @@ exports[`Phoenix Cinema retrieve and transform 1`] = `
         "format": {},
         "notes": "This screening will be followed by a Q&A",
         "occasions": {
+          "intro": true,
           "qAndA": true,
         },
         "screen": "1",
@@ -789,7 +791,11 @@ exports[`Phoenix Cinema retrieve and transform 1`] = `
         "format": {},
         "notes": "This screening will be followed by a Q&A",
         "occasions": {
-          "qAndA": true,
+          "qAndA": {
+            "guests": [
+              "Aengus Collins O'Malley",
+            ],
+          },
         },
         "screen": "1",
         "status": {

cinemas/picturehouses.com-central/tests/__snapshots__/index.test.js.snap

@@ -4023,7 +4023,9 @@ Q&A",
           "dimension": "2d",
         },
         "notes": "Ad and Trailer Free",
-        "occasions": {},
+        "occasions": {
+          "intro": true,
+        },
         "screen": "1",
         "status": {
           "soldOut": false,

cinemas/riocinema.org.uk/tests/__snapshots__/index.test.js.snap

@@ -395,7 +395,13 @@ No Ads or Trailers",
 Special Event
 No Ads or Trailers",
         "occasions": {
-          "qAndA": true,
+          "qAndA": {
+            "guests": [
+              "Rabbi Herschel Gluck",
+              "Winstan Whitter",
+              "Elin Moe",
+            ],
+          },
         },
         "screen": "1",
         "status": {

cinemas/thegardencinema.co.uk/tests/__snapshots__/index.test.js.snap

@@ -434,7 +434,14 @@ The screening will be introduced.",
         "bookingUrl": "https://bookings.thegardencinema.co.uk/TheGardenCinema.dll/TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_5958172.TcsSection_3060731",
         "format": {},
         "notes": "The screening will be followed by a discussion.",
-        "occasions": {},
+        "occasions": {
+          "inConversation": {
+            "guests": [
+              "Dan Dooner",
+              "Ben Mills",
+            ],
+          },
+        },
         "status": {
           "soldOut": false,
         },

cinemas/thelexicinema.co.uk/tests/__snapshots__/index.test.js.snap

@@ -1187,7 +1187,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991762.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -1201,7 +1201,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991779.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -1232,7 +1232,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991803.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -1246,7 +1246,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991814.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -1277,7 +1277,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991784.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -1291,7 +1291,7 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         },
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9991801.TcsSection_5209705",
         "format": {},
-        "notes": "",
+        "notes": "Presented in collaboration with Cinema Mentiré",
         "occasions": {},
         "screen": "1",
         "status": {
@@ -2210,7 +2210,9 @@ exports[`The Lexi Cinema retrieve and transform 1`] = `
         "bookingUrl": "https://thelexicinema.co.uk/TheLexiCinema.dll/Booking?Booking=TSelectItems.waSelectItemsPrompt.TcsWebMenuItem_0.TcsWebTab_0.TcsPerformance_9987157.TcsSection_5209807",
         "format": {},
         "notes": "Accessible screening",
-        "occasions": {},
+        "occasions": {
+          "intro": true,
+        },
         "screen": "2",
         "status": {
           "soldOut": false,

common/occasions.js

@@ -232,6 +232,23 @@ const routinePreview =
 
 const qAndA = String.raw`(?:q\s*&\s*a|q\s*and\s*a|screen\s?talk)`;
 
+// "Switches Screening + Director Conversation" - a conversation billed by who
+// is in it rather than with "in". Only the roles that are someone in the room
+// for the film, so a "Film Conversation" (a strand name) isn't read as one.
+const conversationRole = String.raw`(?:director|filmmaker|film-maker|writer|actor|actress|producer|artist|curator|composer|author|cast|creator)s?(?:['’]s?)?`;
+
+// Curators put works "in conversation with" one another: Ibraaz's Turang "is
+// curated by May Adadol Ingawanij with Ibraaz in conversation with the final
+// chapter of Taring Padi's Rakyat Pasti Menang", an exhibition. That is a
+// dialogue between artworks, not someone in the room.
+const conversationWithWork =
+  /\bin conversation with\s+(?:(?:the|an?|our|its|their|this|these)\s+)?(?:[\p{L}'’-]+\s+){0,3}?(?:exhibitions?|chapters?|installations?|works?|programmes?|seasons?|series|collections?|archives?|shows?)(?![\p{L}'’])/iu;
+
+const notInTheRoomOrAWork = new RegExp(
+  `${notInTheRoom.source}|${conversationWithWork.source}`,
+  "iu",
+);
+
 // Each rule finds one key. `names` names the detail its trailing phrase fills;
 // rules for the same key are tried in order and the first to match wins, so a
 // named rule is listed before the bare one it refines.
@@ -257,7 +274,8 @@ const rules = [
   {
     key: "qAndA",
     pattern: new RegExp(
-      String.raw`\b${qAndA}\s+(?:with|w\/|from|featuring|by)\s+`,
+      // "a Q&A discussion with Rabbi Herschel Gluck" (the Rio) is a Q&A
+      String.raw`\b${qAndA}(?:\s+(?:discussion|session))?\s+(?:with|w\/|from|featuring|by)\s+`,
       "i",
     ),
     names: "guests",
@@ -271,15 +289,36 @@ const rules = [
 
   {
     key: "inConversation",
-    pattern: /\bin conversation with\s+/i,
+    pattern: new RegExp(
+      String.raw`\b(?:in|${conversationRole})\s+conversation\s+with\s+`,
+      "i",
+    ),
     names: "guests",
-    unless: notInTheRoom,
+    unless: notInTheRoomOrAWork,
   },
+  // A discussion only counts with someone named in it: "lead a discussion
+  // following this screening with director Nadine White" (the Lexi). A bare
+  // one is as often the audience talking - "A post-film discussion where every
+  // voice is welcome", the Lexi's "friendly film discussion group for
+  // seniors" - and a panel or Q&A discussion is a panel or a Q&A, as is
+  // Birkbeck's "post-screening discussion panel with Michal Oleszczyk".
   {
     key: "inConversation",
-    pattern: /\bin conversation\b/i,
+    pattern: new RegExp(
+      String.raw`(?<!\b(?:panel|${qAndA})\s)\bdiscussion\s+(?:(?!panel\b)[\p{L}'’-]+\s+){0,4}?with\s+`,
+      "iu",
+    ),
+    names: "guests",
     unless: notInTheRoom,
   },
+  {
+    key: "inConversation",
+    pattern: new RegExp(
+      String.raw`\bin conversation\b|\b${conversationRole}\s+conversation\b|(?:^|\s)\+\s*conversation\b`,
+      "i",
+    ),
+    unless: notInTheRoomOrAWork,
+  },
 
   {
     key: "panel",
@@ -301,7 +340,14 @@ const rules = [
   },
   {
     key: "intro",
-    pattern: /\bintroduc(?:tion|tions|ed)\b|(?:^|[^\p{L}])intros?(?![\p{L}])/iu,
+    // "introduce" only with the film or the screening as its object -
+    // "Charmaine Simpson ... will introduce the films" (the Lexi) - since
+    // programme copy introduces audiences to things ("This season introduces
+    // audiences to ...", "The second selection of films will introduce films
+    // about the war"). Not "introducing", which is as often a title -
+    // "Introducing the Dwights".
+    pattern:
+      /\bintroduc(?:tion|tions|ed)\b|\bintroduces?\s+(?:the|this|these|each|tonight's|tonight’s)\s+(?:films?|screenings?|evening|night|programme|feature|double bill)\b|(?:^|[^\p{L}])intros?(?![\p{L}])/iu,
     unless: notInTheRoom,
   },
 
@@ -348,7 +394,7 @@ const rules = [
 // One cheap test every rule implies, so text that says nothing but a film's
 // name costs one pass rather than a dozen.
 const anySignal =
-  /q\s*&\s*a|q\s*and\s*a|screen\s?talk|intro|live|premi[eè]re|preview|conversation|in person|attendance|panel|masterclass|special guest/i;
+  /q\s*&\s*a|q\s*and\s*a|screen\s?talk|intro|live|premi[eè]re|preview|conversation|discussion|in person|attendance|panel|masterclass|special guest/i;
 
 // The occasions one piece of text describes, as an occasions object. A
 // sentence of a description is read with `fromDescription`, which leaves out