@web-auto/webauto 0.1.4 → 0.1.6

package/modules/camo-runtime/src/autoscript/xhs-unified-template.mjs

@@ -19,6 +19,13 @@ function toPositiveInt(value, fallback, min = 1) {
   return Math.max(min, Math.floor(num));
 }
 
+function toNonNegativeInt(value, fallback = 0) {
+  if (value === null || value === undefined || value === '') return fallback;
+  const num = Number(value);
+  if (!Number.isFinite(num)) return fallback;
+  return Math.max(0, Math.floor(num));
+}
+
 function splitCsv(value) {
   if (Array.isArray(value)) {
     return value
@@ -32,9 +39,6 @@ function splitCsv(value) {
 }
 
 function pickCloseDependency(options) {
-  if (options.doReply || options.doLikes) return 'comment_match_gate';
-  if (options.matchGateEnabled) return 'comment_match_gate';
-  if (options.commentsHarvestEnabled) return 'comments_harvest';
   if (options.detailHarvestEnabled) return 'detail_harvest';
   return 'open_first_detail';
 }
@@ -73,7 +77,9 @@ function buildOpenFirstDetailScript(maxNotes, keyword) {
       const cover = item.querySelector('a.cover');
       if (!cover) return null;
       const href = String(cover.getAttribute('href') || '').trim();
-      const noteId = href.split('/').filter(Boolean).pop() || ('idx_' + index);
+      const lastSegment = href.split('/').filter(Boolean).pop() || '';
+      const normalized = lastSegment.split('?')[0].split('#')[0];
+      const noteId = normalized || ('idx_' + index);
       return { item, cover, href, noteId };
     })
     .filter(Boolean);
@@ -137,7 +143,9 @@ function buildOpenNextDetailScript(maxNotes) {
       const cover = item.querySelector('a.cover');
       if (!cover) return null;
       const href = String(cover.getAttribute('href') || '').trim();
-      const noteId = href.split('/').filter(Boolean).pop() || ('idx_' + index);
+      const lastSegment = href.split('/').filter(Boolean).pop() || '';
+      const normalized = lastSegment.split('?')[0].split('#')[0];
+      const noteId = normalized || ('idx_' + index);
       return { item, cover, href, noteId };
     })
     .filter(Boolean);
@@ -320,6 +328,7 @@ function buildCommentLikeScript(likeKeywords, maxLikesPerRound) {
   const state = window.__camoXhsState || (window.__camoXhsState = {});
   const keywords = ${JSON.stringify(likeKeywords)};
   const maxLikes = Number(${maxLikesPerRound});
+  const maxLikesLimit = maxLikes > 0 ? maxLikes : Number.POSITIVE_INFINITY;
   const nodes = Array.from(document.querySelectorAll('.comment-item'));
   const matches = Array.isArray(state.matchedComments) ? state.matchedComments : [];
   const targetIndexes = new Set(matches.map((row) => Number(row.index)));
@@ -327,7 +336,7 @@ function buildCommentLikeScript(likeKeywords, maxLikesPerRound) {
   let likedCount = 0;
   let skippedActive = 0;
   for (let idx = 0; idx < nodes.length; idx += 1) {
-    if (likedCount >= maxLikes) break;
+    if (likedCount >= maxLikesLimit) break;
     if (targetIndexes.size > 0 && !targetIndexes.has(idx)) continue;
     const item = nodes[idx];
     const text = String((item.querySelector('.content, .comment-content, p') || {}).textContent || '').trim();
@@ -460,16 +469,23 @@ function buildAbortScript(code) {
 export function buildXhsUnifiedAutoscript(rawOptions = {}) {
   const profileId = toTrimmedString(rawOptions.profileId, 'xiaohongshu-batch-1');
   const keyword = toTrimmedString(rawOptions.keyword, '手机膜');
-  const env = toTrimmedString(rawOptions.env, 'debug');
+  const env = toTrimmedString(rawOptions.env, 'prod');
   const outputRoot = toTrimmedString(rawOptions.outputRoot, '');
   const throttle = toPositiveInt(rawOptions.throttle, 900, 100);
   const tabCount = toPositiveInt(rawOptions.tabCount, 4, 1);
   const noteIntervalMs = toPositiveInt(rawOptions.noteIntervalMs, 1200, 200);
   const maxNotes = toPositiveInt(rawOptions.maxNotes, 30, 1);
-  const maxLikesPerRound = toPositiveInt(rawOptions.maxLikesPerRound, 2, 1);
+  const maxComments = toNonNegativeInt(rawOptions.maxComments, 0);
+  const resume = toBoolean(rawOptions.resume, false);
+  const incrementalMax = toBoolean(rawOptions.incrementalMax, true);
+  const maxLikesPerRound = toNonNegativeInt(rawOptions.maxLikesPerRound ?? rawOptions.maxLikes, 0);
   const matchMode = toTrimmedString(rawOptions.matchMode, 'any');
   const matchMinHits = toPositiveInt(rawOptions.matchMinHits, 1, 1);
   const replyText = toTrimmedString(rawOptions.replyText, '感谢分享，已关注');
+  const sharedHarvestPath = toTrimmedString(rawOptions.sharedHarvestPath, '');
+  const searchSerialKey = toTrimmedString(rawOptions.searchSerialKey, `${env}:${keyword}`);
+  const seedCollectCount = toNonNegativeInt(rawOptions.seedCollectCount, 0);
+  const seedCollectMaxRounds = toNonNegativeInt(rawOptions.seedCollectMaxRounds, 0);
 
   const doHomepage = toBoolean(rawOptions.doHomepage, true);
   const doImages = toBoolean(rawOptions.doImages, false);
@@ -506,6 +522,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
     profileId,
     throttle,
     defaults: {
+      disableTimeout: true,
       retry: { attempts: 2, backoffMs: 500 },
       impact: 'subscription',
       onFailure: 'chain_stop',
@@ -516,9 +533,9 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         eventCooldownMs: 300,
         jitterMs: 220,
         navigationMinIntervalMs: 1800,
-        timeoutMs: 180000,
+        timeoutMs: 0,
       },
-      timeoutMs: 180000,
+      timeoutMs: 0,
     },
     metadata: {
       keyword,
@@ -527,6 +544,10 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
       tabCount,
       noteIntervalMs,
       maxNotes,
+      maxComments,
+      maxLikesPerRound,
+      resume,
+      incrementalMax,
       doHomepage,
       doImages,
       doComments,
@@ -539,9 +560,14 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
       matchKeywords,
       likeKeywords,
       replyText,
+      sharedHarvestPath,
+      searchSerialKey,
+      seedCollectCount,
+      seedCollectMaxRounds,
       notes: [
         'open_next_detail intentionally stops script by throwing AUTOSCRIPT_DONE_* when exhausted.',
         'dev mode uses deterministic no-recovery policy (checkpoint recovery disabled).',
+        'resume=true keeps visited note history for断点续传; incrementalMax=true treats maxNotes as增量配额.',
         'when persistComments=true, xhs_comments_harvest emits full comments in operation result for downstream jsonl/file persistence.',
       ],
     },
@@ -608,7 +634,11 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
       {
         id: 'submit_search',
         action: 'xhs_submit_search',
-        params: { keyword },
+        params: {
+          keyword,
+          searchSerialKey,
+          sharedHarvestPath,
+        },
         trigger: 'home_search_input.exist',
         dependsOn: ['fill_keyword'],
         once: true,
@@ -626,15 +656,22 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
       {
         id: 'open_first_detail',
         action: 'xhs_open_detail',
-        params: { mode: 'first', maxNotes, keyword },
+        params: {
+          mode: 'first',
+          maxNotes,
+          keyword,
+          resume,
+          incrementalMax,
+          sharedHarvestPath,
+          seedCollectCount,
+          seedCollectMaxRounds,
+        },
         trigger: 'search_result_item.exist',
         dependsOn: ['submit_search'],
         once: true,
         timeoutMs: 90000,
-        validation: {
-          mode: 'post',
-          post: { page: { hostIncludes: ['xiaohongshu.com'], checkpointIn: ['detail_ready', 'comments_ready', 'search_ready'] } },
-        },
+        onFailure: 'continue',
+        validation: { mode: 'none' },
         checkpoint: {
           containerId: 'xiaohongshu_search.search_result_item',
           targetCheckpoint: 'detail_ready',
@@ -655,17 +692,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         impact: 'op',
         onFailure: 'continue',
         pacing: { operationMinIntervalMs: 2000, eventCooldownMs: 1200, jitterMs: 260 },
-        validation: {
-          mode: 'both',
-          pre: {
-            page: { hostIncludes: ['xiaohongshu.com'], checkpointIn: ['detail_ready', 'comments_ready'] },
-            container: { selector: '.note-detail-mask, .note-detail-page, .note-detail-dialog', mustExist: true, minCount: 1 },
-          },
-          post: {
-            page: { hostIncludes: ['xiaohongshu.com'], checkpointIn: ['detail_ready', 'comments_ready'] },
-            container: { selector: '.note-content, .note-scroller, .media-container', mustExist: true, minCount: 1 },
-          },
-        },
+        validation: { mode: 'none' },
         checkpoint: {
           containerId: 'xiaohongshu_detail.modal_shell',
           targetCheckpoint: 'detail_ready',
@@ -696,6 +723,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
           keyword,
           outputRoot,
           persistComments,
+          commentsLimit: maxComments,
           maxRounds: 48,
           scrollStep: 360,
           settleMs: 260,
@@ -717,9 +745,9 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         dependsOn: [detailHarvestEnabled ? 'detail_harvest' : 'open_first_detail'],
         once: false,
         oncePerAppear: true,
-        timeoutMs: 90000,
+        timeoutMs: 180000,
         retry: { attempts: 1, backoffMs: 0 },
-        impact: 'op',
+        impact: 'script',
         onFailure: 'continue',
         pacing: { operationMinIntervalMs: 2400, eventCooldownMs: 1500, jitterMs: 280 },
         validation: {
@@ -745,7 +773,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         action: 'xhs_comment_match',
         params: { keywords: matchKeywords, mode: matchMode, minHits: matchMinHits },
         trigger: 'detail_modal.exist',
-        dependsOn: [commentsHarvestEnabled ? 'comments_harvest' : (detailHarvestEnabled ? 'detail_harvest' : 'open_first_detail')],
+        dependsOn: [detailHarvestEnabled ? 'detail_harvest' : 'open_first_detail'],
         once: false,
         oncePerAppear: true,
         pacing: { operationMinIntervalMs: 2400, eventCooldownMs: 1200, jitterMs: 160 },
@@ -797,10 +825,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         once: false,
         oncePerAppear: true,
         pacing: { operationMinIntervalMs: 2500, eventCooldownMs: 1300, jitterMs: 180 },
-        validation: {
-          mode: 'post',
-          post: { page: { hostIncludes: ['xiaohongshu.com'], checkpointIn: ['home_ready', 'search_ready'] } },
-        },
+        validation: { mode: 'none' },
         checkpoint: {
           containerId: 'xiaohongshu_detail.discover_button',
           targetCheckpoint: 'search_ready',
@@ -814,7 +839,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         trigger: 'search_result_item.exist',
         dependsOn: ['close_detail'],
         once: false,
-        oncePerAppear: true,
+        oncePerAppear: false,
         retry: { attempts: 1, backoffMs: 0 },
         impact: 'op',
         onFailure: 'continue',
@@ -827,7 +852,7 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         trigger: 'search_result_item.exist',
         dependsOn: ['wait_between_notes', 'ensure_tab_pool'],
         once: false,
-        oncePerAppear: true,
+        oncePerAppear: false,
         timeoutMs: 180000,
         retry: { attempts: 2, backoffMs: 500 },
         impact: 'op',
@@ -836,15 +861,21 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
       {
         id: 'open_next_detail',
         action: 'xhs_open_detail',
-        params: { mode: 'next', maxNotes },
+        params: {
+          mode: 'next',
+          maxNotes,
+          resume,
+          incrementalMax,
+          sharedHarvestPath,
+        },
         trigger: 'search_result_item.exist',
         dependsOn: ['switch_tab_round_robin'],
         once: false,
-        oncePerAppear: true,
+        oncePerAppear: false,
         timeoutMs: 90000,
-        retry: { attempts: 1, backoffMs: 0 },
-        impact: 'script',
-        onFailure: 'stop_all',
+        retry: { attempts: 3, backoffMs: 1000 },
+        impact: 'op',
+        onFailure: 'continue',
         checkpoint: {
           containerId: 'xiaohongshu_search.search_result_item',
           targetCheckpoint: 'detail_ready',
@@ -914,6 +945,8 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         params: {
           acrossPages: true,
           settleMs: 320,
+          pageUrlIncludes: ['/search_result'],
+          requireMatchedPages: true,
           selectors: [
             { id: 'home_search_input', selector: '#search-input, input.search-input' },
             { id: 'search_result_item', selector: '.note-item', visible: false, minCount: 1 },
@@ -923,8 +956,8 @@ export function buildXhsUnifiedAutoscript(rawOptions = {}) {
         dependsOn: ['ensure_tab_pool'],
         once: true,
         timeoutMs: 90000,
-        impact: 'script',
-        onFailure: 'stop_all',
+        impact: 'op',
+        onFailure: 'continue',
       },
     ],
   };

package/modules/camo-runtime/src/container/runtime-core/operations/index.mjs

@@ -137,6 +137,56 @@ async function executeVerifySubscriptions({ profileId, params }) {
 
   const acrossPages = params.acrossPages === true;
   const settleMs = Math.max(0, Number(params.settleMs ?? 280) || 280);
+  const pageUrlIncludes = normalizeArray(params.pageUrlIncludes)
+    .map((item) => String(item || '').trim())
+    .filter(Boolean);
+  const pageUrlExcludes = normalizeArray(params.pageUrlExcludes)
+    .map((item) => String(item || '').trim())
+    .filter(Boolean);
+  const pageUrlRegex = String(params.pageUrlRegex || '').trim();
+  const pageUrlNotRegex = String(params.pageUrlNotRegex || '').trim();
+  const requireMatchedPages = params.requireMatchedPages !== false;
+
+  let includeRegex = null;
+  if (pageUrlRegex) {
+    try {
+      includeRegex = new RegExp(pageUrlRegex);
+    } catch {
+      return asErrorPayload('OPERATION_FAILED', `invalid pageUrlRegex: ${pageUrlRegex}`);
+    }
+  }
+  let excludeRegex = null;
+  if (pageUrlNotRegex) {
+    try {
+      excludeRegex = new RegExp(pageUrlNotRegex);
+    } catch {
+      return asErrorPayload('OPERATION_FAILED', `invalid pageUrlNotRegex: ${pageUrlNotRegex}`);
+    }
+  }
+
+  const hasPageFilter = (
+    pageUrlIncludes.length > 0
+    || pageUrlExcludes.length > 0
+    || Boolean(includeRegex)
+    || Boolean(excludeRegex)
+  );
+
+  const shouldVerifyPage = (rawUrl) => {
+    const url = String(rawUrl || '').trim();
+    if (pageUrlIncludes.length > 0 && !pageUrlIncludes.some((part) => url.includes(part))) {
+      return false;
+    }
+    if (pageUrlExcludes.length > 0 && pageUrlExcludes.some((part) => url.includes(part))) {
+      return false;
+    }
+    if (includeRegex && !includeRegex.test(url)) {
+      return false;
+    }
+    if (excludeRegex && excludeRegex.test(url)) {
+      return false;
+    }
+    return true;
+  };
 
   const collectForCurrentPage = async () => {
     const snapshot = await getDomSnapshotByProfile(profileId);
@@ -156,6 +206,7 @@ async function executeVerifySubscriptions({ profileId, params }) {
 
   let pagesResult = [];
   let overallOk = true;
+  let matchedPageCount = 0;
   if (!acrossPages) {
     const current = await collectForCurrentPage();
     overallOk = current.matches.every((item) => item.count >= item.minCount);
@@ -165,6 +216,16 @@ async function executeVerifySubscriptions({ profileId, params }) {
     const { pages, activeIndex } = extractPageList(listed);
     for (const page of pages) {
       const pageIndex = Number(page.index);
+      const listedUrl = String(page.url || '');
+      if (!shouldVerifyPage(listedUrl)) {
+        pagesResult.push({
+          index: pageIndex,
+          url: listedUrl,
+          skipped: true,
+          ok: true,
+        });
+        continue;
+      }
       if (Number.isFinite(activeIndex) && activeIndex !== pageIndex) {
         await callAPI('page:switch', { profileId, index: pageIndex });
         if (settleMs > 0) await new Promise((resolve) => setTimeout(resolve, settleMs));
@@ -173,15 +234,28 @@ async function executeVerifySubscriptions({ profileId, params }) {
       const pageOk = current.matches.every((item) => item.count >= item.minCount);
       overallOk = overallOk && pageOk;
       pagesResult.push({ index: pageIndex, ...current, ok: pageOk });
+      matchedPageCount += 1;
     }
     if (Number.isFinite(activeIndex)) {
       await callAPI('page:switch', { profileId, index: activeIndex });
     }
   }
 
+  if (acrossPages && hasPageFilter && requireMatchedPages && matchedPageCount === 0) {
+    return asErrorPayload('SUBSCRIPTION_MISMATCH', 'no page matched verify_subscriptions pageUrl filter', {
+      acrossPages,
+      pageUrlIncludes,
+      pageUrlExcludes,
+      pageUrlRegex: pageUrlRegex || null,
+      pageUrlNotRegex: pageUrlNotRegex || null,
+      pages: pagesResult,
+    });
+  }
+
   if (!overallOk) {
     return asErrorPayload('SUBSCRIPTION_MISMATCH', 'subscription selectors missing on one or more pages', {
       acrossPages,
+      matchedPageCount,
       pages: pagesResult,
     });
   }
@@ -190,7 +264,7 @@ async function executeVerifySubscriptions({ profileId, params }) {
     ok: true,
     code: 'OPERATION_DONE',
     message: 'verify_subscriptions done',
-    data: { acrossPages, pages: pagesResult },
+    data: { acrossPages, matchedPageCount, pages: pagesResult },
   };
 }
 

package/modules/camo-runtime/src/container/runtime-core/operations/selector-scripts.mjs

@@ -33,7 +33,20 @@ export function buildSelectorClickScript({ selector, highlight }) {
     }
     el.scrollIntoView({ behavior: 'smooth', block: 'center' });
     await new Promise((r) => setTimeout(r, 150));
-    el.click();
+    if (el instanceof HTMLElement) {
+      try { el.focus({ preventScroll: true }); } catch {}
+      const common = { bubbles: true, cancelable: true, view: window };
+      try {
+        if (typeof PointerEvent === 'function') {
+          el.dispatchEvent(new PointerEvent('pointerdown', { ...common, pointerType: 'mouse', button: 0 }));
+          el.dispatchEvent(new PointerEvent('pointerup', { ...common, pointerType: 'mouse', button: 0 }));
+        }
+      } catch {}
+      try { el.dispatchEvent(new MouseEvent('mousedown', { ...common, button: 0 })); } catch {}
+      try { el.dispatchEvent(new MouseEvent('mouseup', { ...common, button: 0 })); } catch {}
+    }
+    if (typeof el.click === 'function') el.click();
+    else el.dispatchEvent(new MouseEvent('click', { bubbles: true, cancelable: true, view: window, button: 0 }));
     if (${highlightLiteral} && el instanceof HTMLElement) {
       setTimeout(() => { el.style.outline = restoreOutline; }, 260);
     }
@@ -56,9 +69,63 @@ export function buildSelectorTypeScript({ selector, highlight, text }) {
     }
     el.scrollIntoView({ behavior: 'smooth', block: 'center' });
     await new Promise((r) => setTimeout(r, 150));
-    el.focus();
-    el.value = ${textLiteral};
-    el.dispatchEvent(new Event('input', { bubbles: true }));
+    if (el instanceof HTMLElement) {
+      try { el.focus({ preventScroll: true }); } catch {}
+      if (typeof el.click === 'function') el.click();
+    }
+    const value = ${textLiteral};
+    const fireInputEvent = (target, name, init) => {
+      try {
+        if (typeof InputEvent === 'function') {
+          target.dispatchEvent(new InputEvent(name, init));
+          return;
+        }
+      } catch {}
+      target.dispatchEvent(new Event(name, { bubbles: true, cancelable: init?.cancelable === true }));
+    };
+    const assignControlValue = (target, next) => {
+      if (target instanceof HTMLInputElement) {
+        const setter = Object.getOwnPropertyDescriptor(HTMLInputElement.prototype, 'value')?.set;
+        if (setter) setter.call(target, next);
+        else target.value = next;
+        if (typeof target.setSelectionRange === 'function') {
+          const cursor = String(next).length;
+          try { target.setSelectionRange(cursor, cursor); } catch {}
+        }
+        return true;
+      }
+      if (target instanceof HTMLTextAreaElement) {
+        const setter = Object.getOwnPropertyDescriptor(HTMLTextAreaElement.prototype, 'value')?.set;
+        if (setter) setter.call(target, next);
+        else target.value = next;
+        if (typeof target.setSelectionRange === 'function') {
+          const cursor = String(next).length;
+          try { target.setSelectionRange(cursor, cursor); } catch {}
+        }
+        return true;
+      }
+      return false;
+    };
+    const editableAssigned = assignControlValue(el, value);
+    if (!editableAssigned) {
+      if (el instanceof HTMLElement && el.isContentEditable) {
+        el.textContent = value;
+      } else {
+        throw new Error('Element not editable: ' + ${selectorLiteral});
+      }
+    }
+    fireInputEvent(el, 'beforeinput', {
+      bubbles: true,
+      cancelable: true,
+      data: value,
+      inputType: 'insertText',
+    });
+    fireInputEvent(el, 'input', {
+      bubbles: true,
+      cancelable: false,
+      data: value,
+      inputType: 'insertText',
+    });
     el.dispatchEvent(new Event('change', { bubbles: true }));
     if (${highlightLiteral} && el instanceof HTMLElement) {
       setTimeout(() => { el.style.outline = restoreOutline; }, 260);