dsh-lcx-codex 0.3.3 → 0.4.0-rc.13

package/lib/web-search-hosted.js

@@ -3,381 +3,31 @@ import { outputDomains, outputLineRange, parseWebRunOutput } from './web-run-out
 export const HOSTED_SEARCH_PARAMETERS = {
   type: 'object',
   properties: {
-    query: {
-      type: 'string',
-      description: 'Search the web for the requested topic. Include date or freshness requirements in the query when needed.',
-    },
+    query: { type: 'string', description: 'Advanced Responses Hosted Web Search query. Use DSH web_search for ordinary searches.' },
     searchContextSize: { type: 'string', enum: ['low', 'medium', 'high'] },
     allowedDomains: { type: 'array', items: { type: 'string' } },
     blockedDomains: { type: 'array', items: { type: 'string' } },
-    userLocation: {
-      type: 'object',
-      properties: {
-        country: { type: 'string' },
-        city: { type: 'string' },
-        region: { type: 'string' },
-        timezone: { type: 'string' },
-      },
-      additionalProperties: false,
-    },
+    userLocation: { type: 'object', properties: { country: { type: 'string' }, city: { type: 'string' }, region: { type: 'string' }, timezone: { type: 'string' } }, additionalProperties: false },
     externalWebAccess: { type: 'boolean' },
     returnTokenBudget: { type: 'string', enum: ['default', 'unlimited'] },
-    searchContentTypes: {
-      type: 'array',
-      items: { type: 'string', enum: ['text', 'image'] },
-    },
-    imageSettings: {
-      type: 'object',
-      properties: {
-        maxResults: { type: 'integer' },
-        caption: { type: 'boolean' },
-      },
-      additionalProperties: false,
-    },
+    searchContentTypes: { type: 'array', items: { type: 'string', enum: ['text', 'image'] } },
+    imageSettings: { type: 'object', properties: { maxResults: { type: 'integer' }, caption: { type: 'boolean' } }, additionalProperties: false },
   },
-  required: ['query'],
-  additionalProperties: false,
-}
-
-export const HOSTED_SEARCH_OUTPUT = {
-  type: 'object',
-  properties: {
-    mode: { type: 'string', enum: ['hosted'] },
-    action: { type: 'string' },
-    emulation: { type: 'string', enum: ['native'] },
-    content: { type: 'string' },
-    sources: { type: 'array', items: { type: 'object' } },
-    citations: { type: 'array', items: { type: 'object' } },
-    images: { type: 'array', items: { type: 'object' } },
-    warnings: { type: 'array', items: { type: 'string' } },
-    outputBlocks: { type: 'array', items: { type: 'object' } },
-    domains: { type: 'array', items: { type: 'string' } },
-    lineRange: { type: 'object' },
-    requestId: { type: 'string' },
-    responseId: { type: 'string' },
-    retrievedAt: { type: 'string' },
-    truncated: { type: 'boolean' },
-  },
-  required: ['mode', 'action', 'emulation', 'content', 'sources', 'citations', 'images', 'warnings', 'requestId', 'retrievedAt', 'truncated'],
-  additionalProperties: false,
-}
-
-function failure(message, code) {
-  const error = new Error(message)
-  error.code = code
-  return error
-}
-
-export function normalizeHostedSearchArgs(args) {
-  if (args === null || typeof args !== 'object' || Array.isArray(args)) throw failure('websearch_gpt arguments must be an object', 'WEB_INVALID_REQUEST')
-  if (typeof args.query !== 'string' || args.query.trim().length === 0) throw failure('websearch_gpt.query must be a non-empty string', 'WEB_INVALID_REQUEST')
-  const known = new Set([
-    'query',
-    'searchContextSize',
-    'allowedDomains',
-    'blockedDomains',
-    'userLocation',
-    'externalWebAccess',
-    'returnTokenBudget',
-    'searchContentTypes',
-    'imageSettings',
-  ])
-  if (Object.keys(args).some((key) => !known.has(key))) throw failure('websearch_gpt received an unknown field', 'WEB_INVALID_REQUEST')
-  const normalized = { query: args.query.trim() }
-  if (normalized.query.length > 16_000) throw failure('websearch_gpt.query is too long', 'WEB_INVALID_REQUEST')
-  if (args.searchContextSize !== undefined) {
-    if (!['low', 'medium', 'high'].includes(args.searchContextSize)) throw failure('websearch_gpt.searchContextSize is invalid', 'WEB_INVALID_REQUEST')
-    normalized.searchContextSize = args.searchContextSize
-  }
-  const allowedDomains = normalizeDomains(args.allowedDomains, 'allowedDomains')
-  const blockedDomains = normalizeDomains(args.blockedDomains, 'blockedDomains')
-  if (allowedDomains) normalized.allowedDomains = allowedDomains
-  if (blockedDomains) normalized.blockedDomains = blockedDomains
-  if (allowedDomains && blockedDomains) {
-    const blocked = new Set(blockedDomains)
-    if (allowedDomains.some((domain) => blocked.has(domain))) throw failure('websearch_gpt domain filters conflict', 'WEB_INVALID_REQUEST')
-  }
-  if (args.userLocation !== undefined) normalized.userLocation = normalizeUserLocation(args.userLocation)
-  if (args.externalWebAccess !== undefined) {
-    if (typeof args.externalWebAccess !== 'boolean') throw failure('websearch_gpt.externalWebAccess must be boolean', 'WEB_INVALID_REQUEST')
-    normalized.externalWebAccess = args.externalWebAccess
-  }
-  if (args.returnTokenBudget !== undefined) {
-    if (!['default', 'unlimited'].includes(args.returnTokenBudget)) throw failure('websearch_gpt.returnTokenBudget is invalid', 'WEB_INVALID_REQUEST')
-    normalized.returnTokenBudget = args.returnTokenBudget
-  }
-  if (args.searchContentTypes !== undefined) normalized.searchContentTypes = normalizeContentTypes(args.searchContentTypes)
-  if (args.imageSettings !== undefined) {
-    if (!normalized.searchContentTypes?.includes('image')) throw failure('websearch_gpt.imageSettings requires image search content', 'WEB_INVALID_REQUEST')
-    normalized.imageSettings = normalizeImageSettings(args.imageSettings)
-  }
-  return normalized
-}
-
-function normalizeDomains(value, field) {
-  if (value === undefined) return undefined
-  if (!Array.isArray(value) || value.length === 0 || value.length > 100) throw failure(`websearch_gpt.${field} must contain 1 to 100 domains`, 'WEB_INVALID_REQUEST')
-  const domains = value.map((item) => {
-    if (typeof item !== 'string') throw failure(`websearch_gpt.${field} contains an invalid domain`, 'WEB_INVALID_REQUEST')
-    const domain = item.trim().toLowerCase()
-    if (domain.length === 0 || domain.length > 253 || domain.includes('/') || domain.includes(':') || domain.endsWith('.')) {
-      throw failure(`websearch_gpt.${field} contains an invalid domain`, 'WEB_INVALID_REQUEST')
-    }
-    const labels = domain.split('.')
-    if (labels.length < 2 || labels.some((label) => !/^[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?$/u.test(label))) {
-      throw failure(`websearch_gpt.${field} contains an invalid domain`, 'WEB_INVALID_REQUEST')
-    }
-    return domain
-  })
-  if (new Set(domains).size !== domains.length) throw failure(`websearch_gpt.${field} contains duplicate domains`, 'WEB_INVALID_REQUEST')
-  return domains
-}
-
-function normalizeUserLocation(value) {
-  if (!isRecord(value)) throw failure('websearch_gpt.userLocation must be an object', 'WEB_INVALID_REQUEST')
-  const known = new Set(['country', 'city', 'region', 'timezone'])
-  if (Object.keys(value).some((key) => !known.has(key))) throw failure('websearch_gpt.userLocation contains an unknown field', 'WEB_INVALID_REQUEST')
-  const result = {}
-  if (value.country !== undefined) {
-    if (typeof value.country !== 'string' || !/^[a-z]{2}$/iu.test(value.country.trim())) throw failure('websearch_gpt.userLocation.country must be an ISO alpha-2 code', 'WEB_INVALID_REQUEST')
-    result.country = value.country.trim().toUpperCase()
-  }
-  for (const field of ['city', 'region']) {
-    if (value[field] === undefined) continue
-    if (typeof value[field] !== 'string' || value[field].trim().length === 0 || value[field].trim().length > 200) throw failure(`websearch_gpt.userLocation.${field} is invalid`, 'WEB_INVALID_REQUEST')
-    result[field] = value[field].trim()
-  }
-  if (value.timezone !== undefined) {
-    if (typeof value.timezone !== 'string' || value.timezone.trim().length === 0) throw failure('websearch_gpt.userLocation.timezone is invalid', 'WEB_INVALID_REQUEST')
-    const timezone = value.timezone.trim()
-    try {
-      new Intl.DateTimeFormat('en-US', { timeZone: timezone }).format()
-    } catch {
-      throw failure('websearch_gpt.userLocation.timezone must be an IANA timezone', 'WEB_INVALID_REQUEST')
-    }
-    result.timezone = timezone
-  }
-  if (Object.keys(result).length === 0) throw failure('websearch_gpt.userLocation must contain a location field', 'WEB_INVALID_REQUEST')
-  return result
-}
-
-function normalizeContentTypes(value) {
-  if (!Array.isArray(value) || value.length === 0 || value.length > 2 || value.some((item) => !['text', 'image'].includes(item))) {
-    throw failure('websearch_gpt.searchContentTypes is invalid', 'WEB_INVALID_REQUEST')
-  }
-  if (new Set(value).size !== value.length) throw failure('websearch_gpt.searchContentTypes contains duplicates', 'WEB_INVALID_REQUEST')
-  return [...value]
-}
-
-function normalizeImageSettings(value) {
-  if (!isRecord(value) || Object.keys(value).some((key) => !['maxResults', 'caption'].includes(key))) throw failure('websearch_gpt.imageSettings is invalid', 'WEB_INVALID_REQUEST')
-  const result = {}
-  if (value.maxResults !== undefined) {
-    if (!Number.isInteger(value.maxResults) || value.maxResults < 1 || value.maxResults > 100) throw failure('websearch_gpt.imageSettings.maxResults is invalid', 'WEB_INVALID_REQUEST')
-    result.maxResults = value.maxResults
-  }
-  if (value.caption !== undefined) {
-    if (typeof value.caption !== 'boolean') throw failure('websearch_gpt.imageSettings.caption must be boolean', 'WEB_INVALID_REQUEST')
-    result.caption = value.caption
-  }
-  if (Object.keys(result).length === 0) throw failure('websearch_gpt.imageSettings must contain a setting', 'WEB_INVALID_REQUEST')
-  return result
-}
-
-export function buildHostedSearchBody(args, model) {
-  const tool = {
-    type: 'web_search',
-    ...(args.searchContextSize ? { search_context_size: args.searchContextSize } : {}),
-    ...(args.allowedDomains || args.blockedDomains ? {
-      filters: {
-        ...(args.allowedDomains ? { allowed_domains: args.allowedDomains } : {}),
-        ...(args.blockedDomains ? { blocked_domains: args.blockedDomains } : {}),
-      },
-    } : {}),
-    ...(args.userLocation ? { user_location: { type: 'approximate', ...args.userLocation } } : {}),
-    ...(args.externalWebAccess !== undefined ? { external_web_access: args.externalWebAccess } : {}),
-    ...(args.returnTokenBudget ? { return_token_budget: args.returnTokenBudget } : {}),
-    ...(args.searchContentTypes ? { search_content_types: args.searchContentTypes } : {}),
-    ...(args.imageSettings ? {
-      image_settings: {
-        ...(args.imageSettings.maxResults !== undefined ? { max_results: args.imageSettings.maxResults } : {}),
-        ...(args.imageSettings.caption !== undefined ? { caption: args.imageSettings.caption } : {}),
-      },
-    } : {}),
-  }
-  return {
-    model,
-    input: [{ role: 'user', content: [{ type: 'input_text', text: args.query }] }],
-    tools: [tool],
-    tool_choice: 'required',
-    include: [
-      'web_search_call.action.sources',
-      ...(args.searchContentTypes?.includes('image') ? ['web_search_call.results'] : []),
-    ],
-    stream: false,
-    store: false,
-  }
-}
-
-function isRecord(value) {
-  return value !== null && typeof value === 'object' && !Array.isArray(value)
-}
-
-function textFrom(value, seen = new Set()) {
-  if (typeof value === 'string') return value
-  if (!value || typeof value !== 'object' || seen.has(value)) return ''
-  seen.add(value)
-  if (Array.isArray(value)) return value.map((item) => textFrom(item, seen)).filter(Boolean).join('\n')
-  if (typeof value.text === 'string' && ['output_text', 'text', 'input_text'].includes(value.type)) return value.text
-  if (Array.isArray(value.content)) return textFrom(value.content, seen)
-  return ''
-}
-
-function sourceFrom(value) {
-  if (!isRecord(value) || typeof value.url !== 'string' || value.url.trim().length === 0) return undefined
-  return {
-    url: value.url,
-    ...(typeof value.title === 'string' && value.title ? { title: value.title } : {}),
-    ...(typeof value.snippet === 'string' && value.snippet ? { snippet: value.snippet } : {}),
-    ...(typeof value.publishedAt === 'string' && value.publishedAt ? { publishedAt: value.publishedAt } : {}),
-    ...(typeof value.published_at === 'string' && value.published_at ? { publishedAt: value.published_at } : {}),
-    ...(typeof value.ref_id === 'string' && value.ref_id ? { refId: value.ref_id } : {}),
-    ...(typeof value.domain === 'string' && value.domain ? { domain: value.domain } : {}),
-  }
-}
-
-function responseArtifacts(response) {
-  const sources = []
-  const citations = []
-  const images = []
-  const actions = []
-  for (const item of Array.isArray(response?.output) ? response.output : []) {
-    if (item?.type === 'web_search_call') {
-      actions.push(typeof item.action?.type === 'string' ? item.action.type : 'search')
-      if (Array.isArray(item.action?.sources)) {
-        for (const value of item.action.sources) {
-          const source = sourceFrom(value)
-          if (source) sources.push(source)
-        }
-      }
-      for (const value of Array.isArray(item.results) ? item.results : []) {
-        const image = imageFrom(value)
-        if (image) images.push(image)
-      }
-    }
-    if (item?.type !== 'message') continue
-    for (const part of Array.isArray(item.content) ? item.content : []) {
-      if (part?.type !== 'output_text') continue
-      for (const annotation of Array.isArray(part.annotations) ? part.annotations : []) {
-        if (annotation?.type !== 'url_citation') continue
-        const source = sourceFrom(annotation)
-        if (source) {
-          citations.push(source)
-          sources.push(source)
-        }
-      }
-    }
-  }
-  return { sources, citations, images, actions }
-}
-
-function httpUrl(value) {
-  try {
-    const url = new URL(value)
-    if (!['http:', 'https:'].includes(url.protocol)) return undefined
-    return url
-  } catch {
-    return undefined
-  }
-}
-
-function imageFrom(value) {
-  if (!isRecord(value) || value.type !== 'image_result') return undefined
-  const imageUrl = httpUrl(value.image_url)?.toString()
-  if (!imageUrl) return undefined
-  const thumbnailUrl = httpUrl(value.thumbnail_url)?.toString()
-  const sourceWebsiteUrl = httpUrl(value.source_website_url)?.toString()
-  return {
-    imageUrl,
-    ...(thumbnailUrl ? { thumbnailUrl } : {}),
-    ...(sourceWebsiteUrl ? { sourceWebsiteUrl } : {}),
-    ...(typeof value.caption === 'string' && value.caption ? { caption: value.caption } : {}),
-  }
-}
-
-function canonicalUrl(value) {
-  const url = httpUrl(value)
-  if (url) {
-    url.hash = ''
-    for (const key of [...url.searchParams.keys()]) if (/^(utm_|gclid$|fbclid$)/iu.test(key)) url.searchParams.delete(key)
-    return url.toString()
-  }
-  return undefined
-}
-
-function uniqueByUrl(values) {
-  const result = []
-  const seen = new Set()
-  for (const value of values) {
-    const key = canonicalUrl(value.url)
-    if (!key || seen.has(key)) continue
-    seen.add(key)
-    result.push(value)
-  }
-  return result
-}
-
-export function parseHostedSearchResponse(response, requestId, maxResults = 8, retrievedAt = new Date().toISOString()) {
-  if (response?.error) {
-    const error = failure(response.error.message ?? 'LCX hosted Web Search failed', 'LCX_WEB_PROVIDER_ERROR')
-    throw error
-  }
-  if (response?.status !== 'completed') {
-    throw failure(`LCX hosted Web Search returned response status: ${String(response?.status ?? 'missing')}`, 'WEB_RESPONSE_INCOMPLETE')
-  }
-  const output = typeof response?.output_text === 'string' ? response.output_text : textFrom(response?.output ?? response?.content)
-  const outputBlocks = parseWebRunOutput(output)
-  const artifacts = responseArtifacts(response)
-  if (artifacts.actions.length === 0) throw failure('LCX hosted Web Search completed without executing web_search', 'WEB_SEARCH_NOT_EXECUTED')
-  const allSources = [...artifacts.sources]
-  for (const block of outputBlocks) {
-    if (block.url) allSources.push({ url: block.url, ...(block.title ? { title: block.title } : {}) })
-  }
-  const sources = uniqueByUrl(allSources)
-  const citations = uniqueByUrl(artifacts.citations)
-  const limited = sources.slice(0, Math.max(1, maxResults))
-  const images = artifacts.images.slice(0, Math.max(1, maxResults))
-  if (!output && limited.length === 0 && images.length === 0) throw failure('LCX hosted Web Search returned no text, sources, or images', 'WEB_NO_SOURCES')
-  return {
-    mode: 'hosted',
-    action: artifacts.actions[0],
-    emulation: 'native',
-    content: output,
-    sources: limited,
-    citations,
-    images,
-    warnings: artifacts.actions.length > 1 ? [`Multiple hosted search actions were returned: ${artifacts.actions.join(', ')}`] : [],
-    outputBlocks,
-    domains: outputDomains(outputBlocks),
-    ...(outputLineRange(outputBlocks) ? { lineRange: outputLineRange(outputBlocks) } : {}),
-    requestId,
-    ...(typeof response?.id === 'string' && response.id ? { responseId: response.id } : {}),
-    retrievedAt,
-    truncated: sources.length > limited.length || artifacts.images.length > images.length,
-  }
-}
-
-export function renderHostedSearchResult(value) {
-  const parts = []
-  if (value.content) parts.push(value.content)
-  if (value.sources?.length) {
-    parts.push(`来源：\n${value.sources.map((source) => `- [${source.title ?? source.url}](${source.url})${source.publishedAt ? `（${source.publishedAt}）` : ''}${source.snippet ? ` — ${source.snippet}` : ''}`).join('\n')}`)
-  }
-  if (value.images?.length) {
-    parts.push(`图片：\n${value.images.map((image) => `- [${image.caption ?? image.imageUrl}](${image.imageUrl})${image.sourceWebsiteUrl ? `（[来源页面](${image.sourceWebsiteUrl})）` : ''}`).join('\n')}`)
-  }
-  if (value.warnings?.length) parts.push(value.warnings.map((warning) => `警告：${warning}`).join('\n'))
-  if (value.retrievedAt) parts.push(`检索时间：${value.retrievedAt}`)
-  if (value.truncated) parts.push('来源已截断；如需更多结果，请缩小查询范围。')
-  return [{ type: 'text', text: parts.join('\n\n') || '没有找到结果。' }]
-}
+  required: ['query'], additionalProperties: false,
+}
+export const HOSTED_SEARCH_OUTPUT = { type: 'object', properties: { mode: { type: 'string', enum: ['hosted'] }, action: { type: 'string' }, emulation: { type: 'string', enum: ['native'] }, content: { type: 'string' }, sources: { type: 'array', items: { type: 'object' } }, citations: { type: 'array', items: { type: 'object' } }, images: { type: 'array', items: { type: 'object' } }, warnings: { type: 'array', items: { type: 'string' } }, outputBlocks: { type: 'array', items: { type: 'object' } }, domains: { type: 'array', items: { type: 'string' } }, lineRange: { type: 'object' }, requestId: { type: 'string' }, responseId: { type: 'string' }, retrievedAt: { type: 'string' }, truncated: { type: 'boolean' } }, required: ['mode','action','emulation','content','sources','citations','images','warnings','requestId','retrievedAt','truncated'], additionalProperties: false }
+function failure(message, code = 'WEB_INVALID_REQUEST') { const e = new Error(message); e.code = code; return e }
+function isRecord(value) { return value !== null && typeof value === 'object' && !Array.isArray(value) }
+function normalizeDomains(value, field) { if (value === undefined) return undefined; if (!Array.isArray(value) || value.length < 1 || value.length > 100) throw failure(`websearch_gpt_advanced.${field} must contain 1 to 100 domains`); const result = value.map((item) => { if (typeof item !== 'string') throw failure(`websearch_gpt_advanced.${field} contains an invalid domain`); const domain = item.trim().toLowerCase(); if (!domain || domain.length > 253 || domain.includes('/') || domain.includes(':') || domain.endsWith('.') || domain.split('.').length < 2) throw failure(`websearch_gpt_advanced.${field} contains an invalid domain`); return domain }); if (new Set(result).size !== result.length) throw failure(`websearch_gpt_advanced.${field} contains duplicates`); return result }
+function normalizeLocation(value) { if (!isRecord(value)) throw failure('websearch_gpt_advanced.userLocation must be an object'); const result = {}; if (value.country !== undefined) { if (typeof value.country !== 'string' || !/^[a-z]{2}$/iu.test(value.country.trim())) throw failure('userLocation.country must be ISO alpha-2'); result.country = value.country.trim().toUpperCase() } for (const field of ['city','region']) if (value[field] !== undefined) { if (typeof value[field] !== 'string' || !value[field].trim() || value[field].length > 200) throw failure(`userLocation.${field} is invalid`); result[field] = value[field].trim() } if (value.timezone !== undefined) { try { new Intl.DateTimeFormat('en-US', { timeZone: value.timezone }).format() } catch { throw failure('userLocation.timezone must be an IANA timezone') }; result.timezone = value.timezone } if (!Object.keys(result).length) throw failure('userLocation is empty'); return result }
+export function normalizeHostedSearchArgs(args) { if (!isRecord(args) || typeof args.query !== 'string' || !args.query.trim()) throw failure('websearch_gpt_advanced.query must be non-empty'); const query = args.query.trim(); if (query.length > 16_000) throw failure('query is too long'); const result = { query }; if (args.searchContextSize !== undefined) { if (!['low','medium','high'].includes(args.searchContextSize)) throw failure('searchContextSize is invalid'); result.searchContextSize = args.searchContextSize } const allowedDomains = normalizeDomains(args.allowedDomains, 'allowedDomains'); const blockedDomains = normalizeDomains(args.blockedDomains, 'blockedDomains'); if (allowedDomains) result.allowedDomains = allowedDomains; if (blockedDomains) result.blockedDomains = blockedDomains; if (allowedDomains && blockedDomains && allowedDomains.some((d) => new Set(blockedDomains).has(d))) throw failure('domain filters conflict'); if (args.userLocation !== undefined) result.userLocation = normalizeLocation(args.userLocation); if (args.externalWebAccess !== undefined) { if (typeof args.externalWebAccess !== 'boolean') throw failure('externalWebAccess must be boolean'); result.externalWebAccess = args.externalWebAccess } if (args.returnTokenBudget !== undefined) { if (!['default','unlimited'].includes(args.returnTokenBudget)) throw failure('returnTokenBudget is invalid'); result.returnTokenBudget = args.returnTokenBudget } if (args.searchContentTypes !== undefined) { if (!Array.isArray(args.searchContentTypes) || !args.searchContentTypes.length || args.searchContentTypes.length > 2 || args.searchContentTypes.some((v) => !['text','image'].includes(v))) throw failure('searchContentTypes is invalid'); result.searchContentTypes = [...new Set(args.searchContentTypes)] } if (args.imageSettings !== undefined) { if (!result.searchContentTypes?.includes('image') || !isRecord(args.imageSettings)) throw failure('imageSettings requires image search'); result.imageSettings = {}; if (args.imageSettings.maxResults !== undefined) { if (!Number.isInteger(args.imageSettings.maxResults) || args.imageSettings.maxResults < 1 || args.imageSettings.maxResults > 100) throw failure('imageSettings.maxResults is invalid'); result.imageSettings.maxResults = args.imageSettings.maxResults } if (args.imageSettings.caption !== undefined) { if (typeof args.imageSettings.caption !== 'boolean') throw failure('imageSettings.caption must be boolean'); result.imageSettings.caption = args.imageSettings.caption } } return result }
+export function buildHostedSearchBody(args, model, options = {}) { const tool = { type: 'web_search', ...(args.searchContextSize ? { search_context_size: args.searchContextSize } : {}), ...(args.allowedDomains || args.blockedDomains ? { filters: { ...(args.allowedDomains ? { allowed_domains: args.allowedDomains } : {}), ...(args.blockedDomains ? { blocked_domains: args.blockedDomains } : {}) } } : {}), ...(args.userLocation ? { user_location: { type: 'approximate', ...args.userLocation } } : {}), ...(args.externalWebAccess !== undefined ? { external_web_access: args.externalWebAccess } : {}), ...(args.returnTokenBudget ? { return_token_budget: args.returnTokenBudget } : {}), ...(args.searchContentTypes ? { search_content_types: args.searchContentTypes } : {}), ...(args.imageSettings ? { image_settings: { ...(args.imageSettings.maxResults !== undefined ? { max_results: args.imageSettings.maxResults } : {}), ...(args.imageSettings.caption !== undefined ? { caption: args.imageSettings.caption } : {}) } } : {}) }; return { model, input: [{ role: 'user', content: [{ type: 'input_text', text: args.query }] }], tools: [tool], tool_choice: 'required', include: ['web_search_call.action.sources', ...(args.searchContentTypes?.includes('image') ? ['web_search_call.results'] : [])], stream: false, store: false, ...(options.promptCacheKey ? { prompt_cache_key: options.promptCacheKey } : {}) } }
+function textFrom(value, seen = new Set()) { if (typeof value === 'string') return value; if (!value || typeof value !== 'object' || seen.has(value)) return ''; seen.add(value); if (Array.isArray(value)) return value.map((item) => textFrom(item, seen)).filter(Boolean).join('\n'); if (typeof value.text === 'string' && ['output_text','text','input_text'].includes(value.type)) return value.text; if (Array.isArray(value.content)) return textFrom(value.content, seen); return '' }
+function httpUrl(value) { try { const url = new URL(value); return ['http:','https:'].includes(url.protocol) ? url : undefined } catch { return undefined } }
+function sourceFrom(value) { if (!isRecord(value) || !httpUrl(value.url)) return undefined; return { url: httpUrl(value.url).toString(), ...(typeof value.title === 'string' && value.title ? { title: value.title } : {}), ...(typeof value.snippet === 'string' && value.snippet ? { snippet: value.snippet } : {}), ...(typeof value.publishedAt === 'string' ? { publishedAt: value.publishedAt } : typeof value.published_at === 'string' ? { publishedAt: value.published_at } : {}), ...(typeof value.ref_id === 'string' ? { refId: value.ref_id } : {}) } }
+function imageFrom(value) { if (!isRecord(value) || value.type !== 'image_result' || !httpUrl(value.image_url)) return undefined; return { imageUrl: httpUrl(value.image_url).toString(), ...(httpUrl(value.thumbnail_url) ? { thumbnailUrl: httpUrl(value.thumbnail_url).toString() } : {}), ...(httpUrl(value.source_website_url) ? { sourceWebsiteUrl: httpUrl(value.source_website_url).toString() } : {}), ...(typeof value.caption === 'string' ? { caption: value.caption } : {}) } }
+function responseArtifacts(response) { const sources = [], citations = [], images = [], actions = []; for (const item of Array.isArray(response?.output) ? response.output : []) { if (item?.type === 'web_search_call') { actions.push(typeof item.action?.type === 'string' ? item.action.type : 'search'); for (const value of item.action?.sources ?? []) { const source = sourceFrom(value); if (source) sources.push(source) } for (const value of item.results ?? []) { const image = imageFrom(value); if (image) images.push(image) } } if (item?.type === 'message') for (const part of item.content ?? []) if (part?.type === 'output_text') for (const annotation of part.annotations ?? []) if (annotation?.type === 'url_citation') { const source = sourceFrom(annotation); if (source) { citations.push(source); sources.push(source) } } } return { sources, citations, images, actions } }
+function canonicalUrl(value) { const url = httpUrl(value); if (!url) return undefined; url.hash = ''; for (const key of [...url.searchParams.keys()]) if (/^(utm_|gclid$|fbclid$)/iu.test(key)) url.searchParams.delete(key); return url.toString() }
+function uniqueByUrl(values) { const result = [], seen = new Set(); for (const value of values) { const key = canonicalUrl(value.url); if (!key || seen.has(key)) continue; seen.add(key); result.push(value) } return result }
+export function parseHostedSearchResponse(response, requestId, maxResults = 8, retrievedAt = new Date().toISOString()) { if (response?.error) throw failure(response.error.message ?? 'Hosted Web Search failed', 'LCX_WEB_PROVIDER_ERROR'); if (response?.status !== 'completed') throw failure(`Hosted Web Search status: ${String(response?.status ?? 'missing')}`, 'WEB_RESPONSE_INCOMPLETE'); const output = typeof response.output_text === 'string' ? response.output_text : textFrom(response?.output ?? response?.content); const outputBlocks = parseWebRunOutput(output); const artifacts = responseArtifacts(response); if (!artifacts.actions.length) throw failure('Hosted Web Search completed without web_search', 'WEB_SEARCH_NOT_EXECUTED'); const allSources = [...artifacts.sources, ...outputBlocks.flatMap((block) => block.url ? [{ url: block.url, ...(block.title ? { title: block.title } : {}) }] : [])]; const sources = uniqueByUrl(allSources); const citations = uniqueByUrl(artifacts.citations); const limited = sources.slice(0, Math.max(1, maxResults)); const images = artifacts.images.slice(0, Math.max(1, maxResults)); if (!output && !limited.length && !images.length) throw failure('Hosted Web Search returned no output', 'WEB_NO_SOURCES'); return { mode: 'hosted', action: artifacts.actions[0], emulation: 'native', content: output, sources: limited, citations, images, warnings: artifacts.actions.length > 1 ? [`Multiple hosted search actions were returned: ${artifacts.actions.join(', ')}`] : [], outputBlocks, domains: outputDomains(outputBlocks), ...(outputLineRange(outputBlocks) ? { lineRange: outputLineRange(outputBlocks) } : {}), requestId, ...(typeof response.id === 'string' ? { responseId: response.id } : {}), retrievedAt, truncated: sources.length > limited.length || artifacts.images.length > images.length } }
+export function renderHostedSearchResult(value) { const parts = []; if (value.content) parts.push(value.content); if (value.sources?.length) parts.push(`来源：\n${value.sources.map((s) => `- [${s.title ?? s.url}](${s.url})${s.snippet ? ` — ${s.snippet}` : ''}`).join('\n')}`); if (value.images?.length) parts.push(`图片：\n${value.images.map((i) => `- [${i.caption ?? i.imageUrl}](${i.imageUrl})`).join('\n')}`); if (value.warnings?.length) parts.push(value.warnings.map((w) => `警告：${w}`).join('\n')); parts.push(`检索时间：${value.retrievedAt}`); return [{ type: 'text', text: parts.filter(Boolean).join('\n\n') }] }

package/lib/web-search-ref-store.js

@@ -1,4 +1,4 @@
-import { AtomicWebSearchStore } from './web-search-store.js'
+import { JsonStore } from './json-store.js'
 
 const VERSION = 1
 
@@ -31,12 +31,12 @@ function unavailable(refId) {
 
 export class AlphaRefStore {
   constructor(file) {
-    this.store = new AtomicWebSearchStore(file, {
-      empty: () => ({ version: VERSION, sessions: {} }),
-      validate: validData,
-      corruptCode: 'LCX_ALPHA_REF_STORE_CORRUPT',
-      writeCode: 'LCX_ALPHA_REF_STORE_WRITE_FAILED',
-    })
+    this.store = new JsonStore(
+      file,
+      () => ({ version: VERSION, sessions: {} }),
+      validData,
+      'LCX_ALPHA_REF_STORE_CORRUPT',
+    )
   }
 
   record(sessionId, routeFingerprint, refs) {

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "dsh-lcx-codex",
-  "version": "0.3.3",
-  "description": "DSH web search and Responses native V2 compaction for Sub2API-proxied or NewAPI-relayed GPT models",
+  "version": "0.4.0-rc.13",
+  "description": "DSH Codex/Responses capabilities for existing openai-responses routes: hosted/alpha search and session-native V2 compaction",
   "keywords": [
     "deepseek-harness",
     "dsh",
@@ -35,11 +35,13 @@
   },
   "files": [
     "lib/*.js",
-    "scripts/probe-alpha.mjs",
+    "scripts/*.mjs",
+    "assets/*.jpg",
     "cordis.patch.yml",
     "LICENSE",
     "README.md",
     "README_EN.md",
+    "ARCHITECTURE.md",
     "CHANGELOG.md"
   ],
   "dsh": {
@@ -54,26 +56,30 @@
     "@earendil-works/pi-ai": "0.82.1"
   },
   "devDependencies": {
-    "@deepseek-ai/dsh-settings": "0.1.0-rc.8",
+    "@deepseek-ai/dsh-llm": "0.1.1-rc.2",
+    "@deepseek-ai/dsh-settings": "0.1.1-rc.2",
+    "@deepseek-ai/dsh-tools": "0.1.1-rc.2",
     "@deepseek-ai/schemastery": "^3.18.1"
   },
   "peerDependencies": {
     "@deepseek-ai/cordis": "^4.0.1",
-    "@deepseek-ai/dsh-client-runtime": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-client-ui-settings": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-client-ui-slots": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-credentials": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-llm": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-settings": "^0.1.0-rc.8",
-    "@deepseek-ai/dsh-tools": "^0.1.0-rc.8",
+    "@deepseek-ai/dsh-attachment": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-client-runtime": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-client-ui-settings": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-client-ui-slots": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-credentials": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-llm": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-session": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-settings": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-tools": "^0.1.1-rc.2",
     "@deepseek-ai/schemastery": "^3.18.1",
     "react": "^18.2.0"
   },
   "scripts": {
     "test": "node --test tests/*.test.mjs",
-    "test:schema": "node scripts/validate-dsh-schema.mjs",
-    "test:e2e": "node --test tests/sub2api.e2e.test.mjs",
-    "probe:alpha": "node scripts/probe-alpha.mjs"
+    "probe:alpha": "node scripts/probe-alpha.mjs",
+    "pack:local": "npm pack --ignore-scripts",
+    "test:schema": "node scripts/validate-dsh-schema.mjs"
   },
   "engines": {
     "node": ">=20"

package/scripts/probe-alpha.mjs

@@ -2,90 +2,42 @@ import { readFileSync } from 'node:fs'
 import { homedir } from 'node:os'
 import { join } from 'node:path'
 import { randomUUID } from 'node:crypto'
-import {
-  ALPHA_SCHEMA_FINGERPRINT,
-  buildAlphaSearchBody,
-  normalizeAlphaSearchArgs,
-  parseAlphaSearchResponse,
-  probeAlphaCapabilities,
-} from '../lib/web-search-alpha.js'
+import { ALPHA_SCHEMA_FINGERPRINT, buildAlphaSearchBody, normalizeAlphaSearchArgs, parseAlphaSearchResponse, probeAlphaCapabilities } from '../lib/web-search-alpha.js'
 import { AlphaCapabilityStore, alphaCapabilityFingerprint } from '../lib/web-search-capability.js'
-import { fetchJsonWithRetry } from '../lib/transport.js'
 
-function required(name, value) {
-  const normalized = String(value ?? '').trim()
-  if (!normalized) throw new Error(`${name} is required`)
-  return normalized
-}
-
-function apiKey() {
-  const file = String(process.env.LCX_API_KEY_FILE ?? '').trim()
-  if (file) return required('LCX_API_KEY_FILE content', readFileSync(file, 'utf8'))
-  return required('LCX_API_KEY or LCX_API_KEY_FILE', process.env.LCX_API_KEY)
-}
-
-const key = apiKey()
 const baseURL = String(process.env.LCX_BASE_URL ?? 'https://api.lcxbot.com/v1').replace(/\/+$/u, '')
-const model = required('LCX_MODEL', process.env.LCX_MODEL)
-const provider = String(process.env.LCX_PROVIDER ?? 'lcx')
+const model = String(process.env.LCX_MODEL ?? '').trim()
+const provider = String(process.env.LCX_PROVIDER ?? 'lcx').trim()
 const profile = String(process.env.LCX_ALPHA_PROFILE ?? '')
 const group = String(process.env.LCX_ALPHA_GROUP ?? '')
-const sessionId = `dsh-lcx-codex-alpha-probe-${Date.now()}`
-const timeoutMs = Number(process.env.LCX_E2E_TIMEOUT_MS ?? 120000)
-const capabilityPath = String(process.env.LCX_ALPHA_CAPABILITY_PATH ?? join(process.env.DSH_HOME ?? join(homedir(), '.dsh'), 'storages', 'lcx-codex', 'web-alpha-capabilities.json'))
-const trustedNativeProvenance = process.env.LCX_ALPHA_TRUSTED_NATIVE === '1'
-const clickProbeRef = String(process.env.LCX_ALPHA_CLICK_PROBE_REF ?? 'https://en.wikipedia.org/wiki/OpenAI').trim()
-const screenshotProbeRef = String(process.env.LCX_ALPHA_SCREENSHOT_PROBE_REF ?? 'https://arxiv.org/pdf/1706.03762').trim()
+const keyFile = String(process.env.LCX_API_KEY_FILE ?? '').trim()
+if (!model) throw new Error('Set LCX_MODEL to the exact GPT model id')
+if (!keyFile) throw new Error('Set LCX_API_KEY_FILE to a local file containing the bearer key')
+const apiKey = readFileSync(keyFile, 'utf8').trim()
+if (!apiKey) throw new Error('LCX_API_KEY_FILE is empty')
+const storeFile = process.env.LCX_ALPHA_CAPABILITY_PATH ?? join(process.env.DSH_HOME ?? join(homedir(), '.dsh'), 'storages', 'lcx-codex', 'web-alpha-capabilities.json')
+const sessionId = `lcx-probe-${randomUUID()}`
 
-const invoke = async (input) => {
-  const args = normalizeAlphaSearchArgs(input)
+async function invoke(rawArgs) {
+  const args = normalizeAlphaSearchArgs(rawArgs)
   const requestId = randomUUID()
-  const response = await fetchJsonWithRetry(
-    `${baseURL}/alpha/search`,
-    buildAlphaSearchBody(args, model, sessionId, true),
-    {
-      authorization: `Bearer ${key}`,
-      'content-type': 'application/json',
-      'x-client-request-id': requestId,
-      'session-id': sessionId,
-    },
-    AbortSignal.timeout(timeoutMs),
-    timeoutMs,
-    { maxAttempts: 1, maxResponseBytes: 4 * 1024 * 1024 },
-  )
-  return parseAlphaSearchResponse(response, { action: args.action, capability: 'unknown', requestId })
+  const response = await fetch(`${baseURL}/alpha/search`, {
+    method: 'POST',
+    headers: { 'content-type': 'application/json', authorization: `Bearer ${apiKey}`, 'x-client-request-id': requestId, 'session-id': sessionId },
+    body: JSON.stringify(buildAlphaSearchBody(args, model, sessionId, true, 2500)),
+  })
+  const text = await response.text()
+  let body
+  try { body = text ? JSON.parse(text) : {} } catch { body = { output: text, results: [] } }
+  if (!response.ok) { const e = new Error(body?.error?.message ?? body?.message ?? `HTTP ${response.status}`); e.status = response.status; throw e }
+  return parseAlphaSearchResponse(body, { action: args.action, capability: 'command-capable', requestId })
 }
 
-const actionProbes = process.env.LCX_ALPHA_PROBE_STRUCTURED === '1' ? {
-  image_query: { query: 'OpenAI logo' },
-  finance: { ticker: 'AAPL', assetType: 'equity', market: 'USA' },
-  weather: { location: 'United States, California, San Francisco', duration: 1 },
-  sports: { fn: 'schedule', league: 'nba', numberOfGames: 1 },
-  time: { utcOffset: '+00:00' },
+const structured = process.env.LCX_ALPHA_PROBE_STRUCTURED === '1'
+const actionProbes = structured ? {
+  image_query: { query: 'OpenAI logo' }, finance: { ticker: 'MSFT', assetType: 'equity', market: 'USA' }, weather: { location: 'San Francisco, CA' }, sports: { fn: 'standings', league: 'nba' }, time: { utcOffset: '+00:00' },
 } : {}
-
-const record = await probeAlphaCapabilities({
-  invoke,
-  schemaFingerprint: ALPHA_SCHEMA_FINGERPRINT,
-  trustedNativeProvenance,
-  actionProbes,
-  clickProbeRef,
-  screenshotProbeRef,
-})
+const result = await probeAlphaCapabilities({ invoke, schemaFingerprint: ALPHA_SCHEMA_FINGERPRINT, trustedNativeProvenance: process.env.LCX_ALPHA_TRUST_NATIVE === '1', actionProbes, clickProbeRef: process.env.LCX_ALPHA_CLICK_PROBE_REF, screenshotProbeRef: process.env.LCX_ALPHA_SCREENSHOT_PROBE_REF })
 const fingerprint = alphaCapabilityFingerprint({ baseURL, provider, model, profile, group, schemaFingerprint: ALPHA_SCHEMA_FINGERPRINT })
-new AlphaCapabilityStore(capabilityPath).put(fingerprint, record)
-
-const endpoint = new URL(baseURL)
-process.stdout.write(`${JSON.stringify({
-  endpoint: endpoint.host,
-  provider,
-  model,
-  profile,
-  group,
-  fingerprint: fingerprint.slice(0, 12),
-  classification: record.classification,
-  provenance: record.provenance,
-  actions: record.actions,
-  probedAt: record.probedAt,
-  capabilityPath,
-}, null, 2)}\n`)
+new AlphaCapabilityStore(storeFile).put(fingerprint, result)
+console.log(JSON.stringify({ fingerprint, storeFile, classification: result.classification, actions: result.actions, probedAt: result.probedAt, provenance: result.provenance }, null, 2))