ns-rss-spider 1.0.14 → 1.1.0

package/dist/cjs/parseContent/parseContent.js

@@ -38,6 +38,7 @@ var import_zx = require("zx");
 var import_html_entities = require("html-entities");
 var import_generateSsrContent = require("./generateSsrContent");
 var import_getBasicFromItem = require("./getBasicFromItem");
+var import_getArticleImage = require("../utils/getArticleImage");
 async function parseContent($, $element, item, strategy) {
   var _a;
   const srcs = [];
@@ -117,7 +118,8 @@ async function parseContent($, $element, item, strategy) {
     ...(0, import_getBasicFromItem.getBasicFromItem)(item),
     content,
     ssrContent: (0, import_generateSsrContent.generateSsrContent)(content, images),
-    images
+    images,
+    image: (0, import_getArticleImage.getImage)(images)
   };
 }
 // Annotate the CommonJS export names for ESM import in node:

package/dist/cjs/parsers/leiphone.d.ts

@@ -0,0 +1,2 @@
+import { ParseStrategy } from "../types";
+export declare const leiphone: ParseStrategy;

package/dist/cjs/parsers/leiphone.js

@@ -0,0 +1,81 @@
+var __create = Object.create;
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __getProtoOf = Object.getPrototypeOf;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toESM = (mod, isNodeMode, target) => (target = mod != null ? __create(__getProtoOf(mod)) : {}, __copyProps(
+  // If the importer is in node compatibility mode or this is not an ESM
+  // file that has been converted to a CommonJS file using a Babel-
+  // compatible transform (i.e. "__esModule" has not been set), then set
+  // "default" to the CommonJS "module.exports" for node compatibility.
+  isNodeMode || !mod || !mod.__esModule ? __defProp(target, "default", { value: mod, enumerable: true }) : target,
+  mod
+));
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/parsers/leiphone.ts
+var leiphone_exports = {};
+__export(leiphone_exports, {
+  leiphone: () => leiphone
+});
+module.exports = __toCommonJS(leiphone_exports);
+var import_axios = __toESM(require("axios"));
+var import_zx = require("zx");
+var import_cheerio = require("cheerio");
+var import_constants = require("../utils/constants");
+var leiphone = {
+  parse: true,
+  fetcher: "http",
+  getContentElementFromArticle: ($) => {
+    const el = $(".article-template .lph-article-comView");
+    return $(el);
+  },
+  getThumbs: async (articles) => {
+    const res = await import_axios.default.get("https://www.leiphone.com/", {
+      responseType: "text",
+      headers: {
+        "User-Agent": import_constants.chromeUA
+      }
+    }).catch((e) => {
+      console.error(import_zx.chalk.red("【get thumb】获取 html 错误"), e);
+    });
+    if (!res)
+      return void 0;
+    const html = res.data;
+    const $ = (0, import_cheerio.load)(html, null, false);
+    const kv = {};
+    $(".lph-pageList ul.clr > li .img > a:nth-child(2)").each((_, $a) => {
+      var _a;
+      const thumb = $($a).find("img").attr("data-original");
+      if (!thumb) {
+        return;
+      }
+      const href = $($a).attr("href") || "";
+      const guid = (_a = articles.find((a) => {
+        const r = a.guid == href;
+        return r;
+      })) == null ? void 0 : _a.guid;
+      if (guid && thumb) {
+        kv[guid] = thumb;
+      }
+    });
+    return kv;
+  }
+};
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  leiphone
+});

package/dist/cjs/parsers/oschina.d.ts

@@ -0,0 +1,2 @@
+import { ParseStrategy } from "../types";
+export declare const oschina: ParseStrategy;

package/dist/cjs/parsers/oschina.js

@@ -0,0 +1,37 @@
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/parsers/oschina.ts
+var oschina_exports = {};
+__export(oschina_exports, {
+  oschina: () => oschina
+});
+module.exports = __toCommonJS(oschina_exports);
+var oschina = {
+  parse: true,
+  fetcher: "http",
+  getContentElementFromArticle: ($) => {
+    const el = $(".article-detail > .content");
+    $(el).find(".ad-wrap").remove();
+    return $(el);
+  }
+};
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  oschina
+});

package/dist/cjs/strategy.js

@@ -30,6 +30,8 @@ var import_kr = require("./parsers/36kr");
 var import_techrunch = require("./parsers/techrunch");
 var import_mydrivers = require("./parsers/mydrivers");
 var import_tmtpost = require("./parsers/tmtpost");
+var import_leiphone = require("./parsers/leiphone");
+var import_oschina = require("./parsers/oschina");
 var strategies = {
   cnbeta: import_cnbeta.cnbeta,
   ifanr: import_ifanr.ifanr,
@@ -38,7 +40,9 @@ var strategies = {
   "36kr": import_kr._36kr,
   techrunch: import_techrunch.techrunch,
   mydrivers: import_mydrivers.mydrivers,
-  tmtpost: import_tmtpost.tmtpost
+  tmtpost: import_tmtpost.tmtpost,
+  leiphone: import_leiphone.leiphone,
+  oschina: import_oschina.oschina
 };
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {

package/dist/cjs/types.d.ts

@@ -25,6 +25,7 @@ export interface SimpleArticle {
     source?: string;
     categories?: string;
     thumb?: string;
+    image?: string;
 }
 export interface RichArticleImage {
     url: string;

package/dist/cjs/upload/index.d.ts

@@ -22,4 +22,5 @@ export declare function uploadContent({ app, getUploadApi, items, }: {
     source?: string | undefined;
     categories?: string | undefined;
     thumb?: string | undefined;
+    image?: string | undefined;
 }[]>;

package/dist/cjs/utils/getArticleImage.d.ts

@@ -0,0 +1,11 @@
+import { RichArticleImage } from "../types";
+/**
+ * 从 article images 中获取 image 数据。目前用作无 thumb 时的降级
+ * 若都没有尺寸，选第一张
+    对于有尺寸的图片，按如下优先级
+    第一个长宽比大于 1.5 且长大于 80 小于 1000 的图片
+    若没有，则第一个图片
+ * @param images
+ * @returns
+ */
+export declare const getImage: (images: RichArticleImage[]) => string | undefined;

package/dist/cjs/utils/getArticleImage.js

@@ -0,0 +1,37 @@
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/utils/getArticleImage.ts
+var getArticleImage_exports = {};
+__export(getArticleImage_exports, {
+  getImage: () => getImage
+});
+module.exports = __toCommonJS(getArticleImage_exports);
+var getImage = (images) => {
+  if (!(images == null ? void 0 : images.length)) {
+    return void 0;
+  }
+  const img = images.find((img2) => {
+    return img2.width && img2.height && img2.width / img2.height >= 1.5 && img2.width < 2e3;
+  });
+  return img == null ? void 0 : img.url;
+};
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  getImage
+});

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ns-rss-spider",
-  "version": "1.0.14",
+  "version": "1.1.0",
   "description": "",
   "main": "dist/cjs/index.js",
   "types": "dist/cjs/index.d.ts",