@socialgouv/fiches-travail-data-types 4.438.0 → 4.440.0

package/build/fetch-data/parseDom.js

@@ -7,7 +7,6 @@ exports.parseDom = void 0;
 const cdtn_slugify_1 = __importDefault(require("@socialgouv/cdtn-slugify"));
 const got_1 = require("got");
 const email_1 = require("../email");
-const postProcess_1 = require("./postProcess");
 const referenceExtractor_1 = require("./referenceExtractor");
 const referenceResolver_1 = require("./referenceResolver");
 const $$ = (node, selector) => Array.from(node.querySelectorAll(selector));
@@ -22,16 +21,8 @@ const formatEmail = (node) => {
     node.removeAttribute("data-cfemail");
     node.textContent = value;
 };
-const formatPicture = (node) => {
-    const comment = node.parentElement.childNodes[0];
-    if (comment.nodeName !== "#comment") {
-        //upper sibbling node is not a comment so it's not a case we handle
-        return;
-    }
-    const [, src = ""] = comment.data.match(/src=["']([^'"]*)["']/);
-    if (src.length === 0) {
-        return;
-    }
+const SRC_REGEX = /src=["']([^'"]*)["']/;
+function getCleanSrc(src) {
     let [srcClean] = src.split("?");
     if (!srcClean.match(/^https?:\/\//)) {
         if (srcClean.slice(0, 1) !== "/") {
@@ -39,14 +30,50 @@ const formatPicture = (node) => {
         }
         srcClean = `https://travail-emploi.gouv.fr${srcClean}`;
     }
-    // we remove the ie comment that have timestamp in the url
-    comment.remove();
-    // we add e
-    const sourceNode = node.ownerDocument.createElement("source");
-    sourceNode.setAttribute("srcset", srcClean);
-    sourceNode.setAttribute("media", "(min-width: 300px)");
-    node.appendChild(sourceNode);
-    return node;
+    return srcClean;
+}
+const formatPicture = (node) => {
+    let comment;
+    node.parentElement
+        .childNodes
+        .forEach(function (childNode) {
+        if (childNode.nodeName === "#comment" || childNode.nodeType === 8) {
+            if (childNode.data.match(SRC_REGEX)) {
+                comment = childNode;
+            }
+        }
+    });
+    if (comment) {
+        const [, src = ""] = comment.data.match(SRC_REGEX);
+        if (src.length) {
+            const srcClean = getCleanSrc(src);
+            node.parentNode.innerHTML = `<img src="${srcClean}" style="width:100%;height:auto;" />`;
+            return;
+        }
+    }
+    let image;
+    node
+        .childNodes
+        .forEach(function (childNode) {
+        if (childNode.nodeName === "IMG") {
+            image = childNode;
+        }
+    });
+    if (image) {
+        node.replaceWith(image);
+    }
+};
+const formatImage = (node) => {
+    node.removeAttribute("onmousedown");
+    if (node.getAttribute("src").indexOf("data:image") === -1) {
+        node.removeAttribute("srcset");
+        node.removeAttribute("sizes");
+        let src = node.getAttribute("src");
+        if (!src.match(/^https?:\/\//)) {
+            const srcClean = getCleanSrc(src);
+            node.setAttribute("src", srcClean);
+        }
+    }
 };
 const formatAnchor = (node) => {
     if (node.innerHTML.trim() === "") {
@@ -80,6 +107,9 @@ const formatAnchor = (node) => {
         node.setAttribute("rel", "nofollow, noopener");
     }
 };
+const removeNode = (node) => {
+    node.remove();
+};
 const flattenCsBlocs = (node) => {
     node.insertAdjacentHTML("afterend", node.innerHTML);
     node.parentNode.removeChild(node);
@@ -101,27 +131,17 @@ function parseDom(dom, id, url) {
     if (!article) {
         throw new got_1.ParseError("no <main>");
     }
+    if (!id) {
+        throw new got_1.ParseError(`No id`);
+    }
     $$(article, "a").forEach(formatAnchor);
     $$(article, "picture").forEach(formatPicture);
     $$(article, "[data-cfemail]").forEach(formatEmail);
     $$(article, ".cs_blocs").forEach(flattenCsBlocs);
-    const imgs = $$(article, "img");
-    imgs.forEach((node) => {
-        // remove adaptImgFix(this) on hero img
-        node.removeAttribute("onmousedown");
-    });
-    imgs
-        .filter((node) => node.getAttribute("src").indexOf("data:image") === -1)
-        .forEach((node) => {
-        let src = node.getAttribute("src");
-        if (!src.match(/^https?:\/\//)) {
-            if (src.slice(0, 1) !== "/") {
-                src = "/" + src;
-            }
-            src = `https://travail-emploi.gouv.fr${src}`;
-            node.setAttribute("src", src);
-        }
-    });
+    $$(article, "img").forEach(formatImage);
+    $$(article, "style").forEach(removeNode);
+    $$(article, "button").forEach(removeNode);
+    $$(article, ".oembed-source").forEach(removeNode);
     let titleElement = $(article, "h1");
     if (!titleElement) {
         titleElement = $(article, "h2");
@@ -130,40 +150,29 @@ function parseDom(dom, id, url) {
         }
     }
     const title = titleElement.textContent.trim();
-    if (!id) {
-        throw new got_1.ParseError(`No id`);
-    }
-    const dateRaw = $(dom.window.document, "meta[property*=modified_time]") ||
-        $(dom.window.document, "meta[property$=published_time]");
+    const dateRaw = $(dom.window.document, "meta[property*=modified_time]") || $(dom.window.document, "meta[property$=published_time]");
     const [year, month, day] = dateRaw.getAttribute("content").split("-");
     let intro = $(article, ".main-article__chapo") || "";
-    intro =
-        intro && intro.innerHTML.replace(/\n/g, "").replace(/\s+/g, " ").trim();
-    const description = $(dom.window.document, "meta[name=description]")?.getAttribute("content") ??
-        "";
+    intro = intro && intro.innerHTML.replace(/\n/g, "").replace(/\s+/g, " ").trim();
+    const description = $(dom.window.document, "meta[name=description]")?.getAttribute("content") ?? "";
     const sections = [];
     const sectionTag = getSectionTag(article);
     // First pass is only to get a potential untitled section at the top of the article
     // This section has neither anchor nor title
     let nextArticleElement = $(article, ".main-article__texte > *");
     const untitledSection = {
-        anchor: "",
-        html: "",
-        text: "",
-        title: title,
+        anchor: "", html: "", text: "", title: title,
     };
-    while (nextArticleElement &&
-        nextArticleElement.tagName.toLowerCase() !== sectionTag) {
+    while (nextArticleElement && nextArticleElement.tagName.toLowerCase() !== sectionTag) {
         if (nextArticleElement.textContent) {
             if (!untitledSection.description) {
                 untitledSection.description = "temp description";
             }
-            untitledSection.html += (0, postProcess_1.htmlPostParser)(nextArticleElement.outerHTML
+            untitledSection.html += nextArticleElement.outerHTML
                 .replace(/\n+/g, "")
                 .replace(/>\s+</g, "><")
-                .replace(/\s+/g, " "));
-            untitledSection.text +=
-                " " + nextArticleElement.textContent.replace(/\s+/g, " ").trim();
+                .replace(/\s+/g, " ");
+            untitledSection.text += " " + nextArticleElement.textContent.replace(/\s+/g, " ").trim();
         }
         nextArticleElement = nextArticleElement.nextElementSibling;
     }
@@ -189,7 +198,7 @@ function parseDom(dom, id, url) {
             sections.push({
                 anchor: el.getAttribute("id") || (0, cdtn_slugify_1.default)(el.textContent),
                 description: sectionText.slice(0, 200).trim(),
-                html: (0, postProcess_1.htmlPostParser)(html.replace(/\n+/g, "").replace(/>\s+</g, "><").replace(/\s+/g, " ")),
+                html: html.replace(/\n+/g, "").replace(/>\s+</g, "><").replace(/\s+/g, " "),
                 references: getReferences(sectionText),
                 text: sectionText,
                 title: el.textContent.trim(),
@@ -200,13 +209,7 @@ function parseDom(dom, id, url) {
         throw new got_1.ParseError(`No sections`);
     }
     return {
-        date: `${day}/${month}/${year}`,
-        description,
-        intro,
-        pubId: id,
-        sections,
-        title,
-        url,
+        date: `${day}/${month}/${year}`, description, intro, pubId: id, sections, title, url,
     };
 }
 exports.parseDom = parseDom;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@socialgouv/fiches-travail-data-types",
-  "version": "4.438.0",
+  "version": "4.440.0",
   "main": "build/index.js",
   "module": "build/index.js",
   "files": [
@@ -34,7 +34,7 @@
     "babel-jest": "^27.4.4",
     "eslint": "^8.28.0",
     "eslint-plugin-jest": "^27.1.6",
-    "husky": "^7.0.4",
+    "husky": "^8.0.0",
     "jest": "^29.3.1",
     "jsdom": "^17.0.0",
     "prettier": "^2.5.1",

package/build/fetch-data/postProcess.d.ts

@@ -1 +0,0 @@
-export declare const htmlPostParser: (html: string) => string;

package/build/fetch-data/postProcess.js

@@ -1,61 +0,0 @@
-"use strict";
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || function (mod) {
-    if (mod && mod.__esModule) return mod;
-    var result = {};
-    if (mod != null) for (var k in mod) if (k !== "default" && Object.prototype.hasOwnProperty.call(mod, k)) __createBinding(result, mod, k);
-    __setModuleDefault(result, mod);
-    return result;
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.htmlPostParser = void 0;
-const cheerio = __importStar(require("cheerio"));
-const htmlPostParser = (html) => {
-    const $ = cheerio.load(html, null, false);
-    const arrImgSrc = [];
-    let currentIndex = 0;
-    $("style").remove();
-    // https://travail-emploi.gouv.fr/le-ministere-en-action/coronavirus-covid-19/questions-reponses-par-theme/article/mesures-de-prevention-dans-l-entreprise-contre-la-covid-19
-    $("button").remove();
-    $(".oembed-source").remove();
-    // https://travail-emploi.gouv.fr/emploi-et-insertion/accompagnement-des-mutations-economiques/activite-partielle-chomage-partiel/article/activite-partielle-chomage-partiel
-    $("*")
-        .contents()
-        .each(function () {
-        if (this.nodeType === 8) {
-            const regex = /src=['"](.*?)['"]/;
-            const result = regex.exec(this.nodeValue);
-            if (result) {
-                const src = result[0].slice(5, -1);
-                if (src)
-                    arrImgSrc.push(src);
-            }
-        }
-    });
-    $("picture").replaceWith(() => {
-        const src = arrImgSrc[currentIndex];
-        if (src) {
-            currentIndex++;
-            return `<img src="https://travail-emploi.gouv.fr/${src}" style="width:100%;height:auto;"/>`;
-        }
-        return $.html()?.toString() ?? "";
-    });
-    $("picture.adapt-img-wrapper").remove();
-    return $.html();
-};
-exports.htmlPostParser = htmlPostParser;