enrich(ctrip): expand the adapter across Ctrip's travel verticals

clis/ctrip/attraction.js

@@ -0,0 +1,73 @@
+/**
+ * 携程门票 / 景点: a destination's top attractions by city id.
+ *
+ * you.ctrip.com's place page renders the top-rated attractions for a city as
+ * `/sight/<city><cityId>/<id>.html` links carrying the name, rating, and review
+ * count, so this navigates the place page (routed by the numeric city id, discover
+ * it via `ctrip search`) and reads those links, scoped to the requested city, by
+ * their stable pattern (see `buildAttractionExtractJs` in utils). Per-sight ticket
+ * prices sit on each attraction's own detail page and are out of scope here.
+ *
+ * There is no `EmptyResultError` path: a valid you.ctrip city always lists
+ * attractions, so a zero city-scoped result means a stale / invalid / unrenderable
+ * id, which surfaces as `CommandExecutionError` ("check the city id") rather than a
+ * genuine-empty result.
+ */
+import { AuthRequiredError, CommandExecutionError } from '@jackwener/opencli/errors';
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import {
+    buildAttractionExtractJs,
+    buildAttractionPlaceUrl,
+    buildWaitForAttractionsJs,
+    parseCityId,
+    parseListLimit,
+} from './utils.js';
+
+cli({
+    site: 'ctrip',
+    name: 'attraction',
+    access: 'read',
+    description: '列出携程某城市的热门景点（评分/点评数，城市 id 经 ctrip search 获取）',
+    domain: 'you.ctrip.com',
+    strategy: Strategy.COOKIE,
+    browser: true,
+    navigateBefore: false,
+    args: [
+        { name: 'city', required: true, positional: true, help: 'Numeric Ctrip city id (from `ctrip search`, e.g. 1 for 北京)' },
+        { name: 'limit', default: 20, help: 'Number of attractions (1-50)' },
+    ],
+    columns: [
+        'rank',
+        'name',
+        'rating', 'reviews',
+        'url',
+    ],
+    func: async (page, kwargs) => {
+        const cityId = parseCityId(kwargs.city);
+        const limit = parseListLimit(kwargs.limit);
+
+        const placeUrl = buildAttractionPlaceUrl(cityId);
+        await page.goto(placeUrl);
+        const waitResult = await page.evaluate(buildWaitForAttractionsJs(cityId));
+        if (waitResult === 'captcha') {
+            throw new AuthRequiredError('you.ctrip.com', 'Ctrip is asking for a captcha; complete it in your browser session and retry');
+        }
+        if (waitResult !== 'content') {
+            throw new CommandExecutionError(`Ctrip place page did not render attraction links for city id ${cityId} (state=${String(waitResult)}); check the city id`);
+        }
+        const raw = await page.evaluate(buildAttractionExtractJs(cityId));
+        if (!Array.isArray(raw)) {
+            throw new CommandExecutionError('Ctrip attraction DOM extraction returned malformed rows');
+        }
+        if (raw.length === 0) {
+            throw new CommandExecutionError('Ctrip attraction links rendered but parser did not find required sight anchors');
+        }
+        return raw.slice(0, limit).map((r, i) => ({
+            rank: i + 1,
+            name: r.name,
+            rating: r.rating,
+            reviews: r.reviews,
+            url: r.url,
+        }));
+    },
+});

clis/ctrip/bus.js

@@ -0,0 +1,85 @@
+/**
+ * 携程汽车票 search: intercity coach tickets by city name + date.
+ *
+ * The newbus SPA landing (`bus.ctrip.com/`) does not hydrate under the browser
+ * bridge, but the results route renders directly from a `?param=<json>` deep
+ * link (the payload the app's own search handler posts to `/list`). Schedules
+ * arrive via the `busListV2` XHR and render into `.list-item-parent` rows read
+ * by stable utility-class fields (see `buildBusExtractJs` in utils), so this
+ * reads by selector rather than positional innerText.
+ */
+import { ArgumentError, AuthRequiredError, CommandExecutionError, EmptyResultError } from '@jackwener/opencli/errors';
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import {
+    WAIT_FOR_BUS_JS,
+    buildBusExtractJs,
+    buildBusListUrl,
+    buildScrollUntilJs,
+    parseIsoDate,
+    parseListLimit,
+    parsePlaceName,
+} from './utils.js';
+
+cli({
+    site: 'ctrip',
+    name: 'bus',
+    access: 'read',
+    description: '搜索携程汽车票（按出发/到达城市名 + 日期）',
+    domain: 'bus.ctrip.com',
+    strategy: Strategy.COOKIE,
+    browser: true,
+    navigateBefore: false,
+    args: [
+        { name: 'from', required: true, positional: true, help: 'Departure city name (e.g. 北京 / 上海)' },
+        { name: 'to', required: true, positional: true, help: 'Arrival city name (e.g. 天津 / 杭州)' },
+        { name: 'date', required: true, help: 'Departure date (YYYY-MM-DD)' },
+        { name: 'limit', default: 20, help: 'Number of departures (1-50)' },
+    ],
+    columns: [
+        'rank',
+        'departureTime',
+        'fromStation', 'toStation',
+        'duration', 'price', 'status',
+        'url',
+    ],
+    func: async (page, kwargs) => {
+        const fromCity = parsePlaceName('from', kwargs.from);
+        const toCity = parsePlaceName('to', kwargs.to);
+        if (fromCity === toCity) {
+            throw new ArgumentError(`--from and --to must differ (got ${fromCity})`);
+        }
+        const date = parseIsoDate('date', kwargs.date);
+        const limit = parseListLimit(kwargs.limit);
+
+        const searchUrl = buildBusListUrl(fromCity, toCity, date);
+        await page.goto(searchUrl);
+        const waitResult = await page.evaluate(WAIT_FOR_BUS_JS);
+        if (waitResult === 'captcha') {
+            throw new AuthRequiredError('bus.ctrip.com', 'Ctrip is asking for a captcha; complete it in your browser session and retry');
+        }
+        if (waitResult !== 'content') {
+            throw new CommandExecutionError(`Ctrip bus page did not render schedule rows (state=${String(waitResult)})`);
+        }
+        const renderedCardCount = await page.evaluate(buildScrollUntilJs('.list-item-parent', limit));
+        const raw = await page.evaluate(buildBusExtractJs());
+        if (!Array.isArray(raw)) {
+            throw new CommandExecutionError('Ctrip bus DOM extraction returned malformed rows');
+        }
+        if (raw.length === 0) {
+            if (Number(renderedCardCount) > 0) {
+                throw new CommandExecutionError('Ctrip bus rows rendered but parser did not find required schedule anchors');
+            }
+            throw new EmptyResultError('ctrip bus', `No coaches for ${fromCity} to ${toCity} on ${date}`);
+        }
+        return raw.slice(0, limit).map((r, i) => ({
+            rank: i + 1,
+            departureTime: r.departureTime,
+            fromStation: r.fromStation,
+            toStation: r.toStation,
+            duration: r.duration,
+            price: r.price,
+            status: r.status,
+            url: searchUrl,
+        }));
+    },
+});

clis/ctrip/cruise.js

@@ -0,0 +1,96 @@
+/**
+ * 携程邮轮 search: cruise packages by departure port name.
+ *
+ * Ctrip's cruise search results live on the legacy `newpackage/search/sN.html`
+ * pages, keyed by an opaque per-port code `sN`. Those pages list every port as a
+ * link, so the command first navigates a stable port page (上海港, `s2`) to
+ * resolve the requested port name to its `sN` code, then loads that port's
+ * results and reads the `.route_info` cards (see helpers in utils).
+ */
+import { AuthRequiredError, CommandExecutionError, EmptyResultError } from '@jackwener/opencli/errors';
+import { cli, Strategy } from '@jackwener/opencli/registry';
+import {
+    WAIT_FOR_CRUISE_JS,
+    buildCruiseExtractJs,
+    buildCruisePortLookupJs,
+    buildCruiseSearchUrl,
+    parseListLimit,
+    parsePlaceName,
+} from './utils.js';
+
+// 上海港 lists every departure port as a link, so it doubles as the port-code index.
+const PORT_INDEX_CODE = '2';
+
+cli({
+    site: 'ctrip',
+    name: 'cruise',
+    access: 'read',
+    description: '搜索携程邮轮线路（按出发港名）',
+    domain: 'cruise.ctrip.com',
+    strategy: Strategy.COOKIE,
+    browser: true,
+    navigateBefore: false,
+    args: [
+        { name: 'port', required: true, positional: true, help: 'Departure cruise port name (e.g. 上海 / 威尼斯 / 罗马)' },
+        { name: 'limit', default: 20, help: 'Number of cruises (1-50)' },
+    ],
+    columns: [
+        'rank',
+        'title', 'star',
+        'boarding', 'sailingDate',
+        'tags', 'price',
+        'url',
+    ],
+    func: async (page, kwargs) => {
+        const port = parsePlaceName('port', kwargs.port);
+        const limit = parseListLimit(kwargs.limit);
+
+        const indexUrl = buildCruiseSearchUrl(PORT_INDEX_CODE);
+        await page.goto(indexUrl);
+        const indexWait = await page.evaluate(WAIT_FOR_CRUISE_JS);
+        if (indexWait === 'captcha') {
+            throw new AuthRequiredError('cruise.ctrip.com', 'Ctrip is asking for a captcha; complete it in your browser session and retry');
+        }
+        if (indexWait !== 'content') {
+            throw new CommandExecutionError(`Ctrip cruise page did not render (state=${String(indexWait)})`);
+        }
+        const portCode = await page.evaluate(buildCruisePortLookupJs(port));
+        if (!portCode) {
+            throw new EmptyResultError('ctrip cruise', `No cruise departure port matching "${port}" (try a listed sea-cruise port like 上海 / 威尼斯 / 罗马)`);
+        }
+
+        let searchUrl = indexUrl;
+        if (portCode !== PORT_INDEX_CODE) {
+            searchUrl = buildCruiseSearchUrl(portCode);
+            await page.goto(searchUrl);
+            const portWait = await page.evaluate(WAIT_FOR_CRUISE_JS);
+            if (portWait === 'captcha') {
+                throw new AuthRequiredError('cruise.ctrip.com', 'Ctrip is asking for a captcha; complete it in your browser session and retry');
+            }
+            if (portWait === 'empty') {
+                throw new EmptyResultError('ctrip cruise', `No cruises currently departing "${port}"`);
+            }
+            if (portWait !== 'content') {
+                throw new CommandExecutionError(`Ctrip cruise port page did not render (state=${String(portWait)})`);
+            }
+        }
+
+        const raw = await page.evaluate(buildCruiseExtractJs());
+        if (!Array.isArray(raw)) {
+            throw new CommandExecutionError('Ctrip cruise DOM extraction returned malformed rows');
+        }
+        if (raw.length === 0) {
+            throw new CommandExecutionError('Ctrip cruise cards rendered but parser did not find required itinerary anchors');
+        }
+        return raw.slice(0, limit).map((r, i) => ({
+            rank: i + 1,
+            title: r.title,
+            star: r.star,
+            boarding: r.boarding,
+            sailingDate: r.sailingDate,
+            tags: r.tags,
+            price: r.price,
+            url: searchUrl,
+        }));
+    },
+});