fix:

修复抓取知网网页条目时，新条目不能放到原文件夹(collection)中;过滤卷期字段前面的0
colourful-lufruoloC · Jan 19, 2022 · a176fea · a176fea
1 parent c73c3b6
commit a176fea
Showing 2 changed files with 32 additions and 6 deletions.
diff --git a/chrome/content/scripts/jasminum.js b/chrome/content/scripts/jasminum.js
@@ -148,6 +148,7 @@ Zotero.Jasminum = new function () {
         if (items.length == 0) return;
         var item = items.shift();
         var itemCollections = item.getCollections();
+        Zotero.debug(itemCollections);
         var libraryID = item.libraryID;
         // Retrive meta data for webpage item
         if (Zotero.ItemTypes.getName(item.itemTypeID) === "webpage") {
@@ -157,21 +158,29 @@ Zotero.Jasminum = new function () {
             let postData = this.Scrape.createRefPostData([articleId]);
             let data = await this.Scrape.getRefText(postData);
             // Zotero.debug("** Jasminum webpage data");
-            // Zotero.debug(data);
+
             // Some item will be updated after published
             if (data.length === 0 && articleId.dbname.includes("TEMP")) {
-                articleId.dbname = articleId
-                    .dbname
-                    .replace(/TEMP.*$/, 'LAST' + item.getField("dateAdded").slice(0, 4));
-                postData = this.Scrape.createRefPostData([articleId]);
-                data = await this.Scrape.getRefText(postData);
+                articleId = await this.Scrape.getIDFromPage(item.getField("url"));
+                Zotero.debug([articleId]);
             }
+            postData = this.Scrape.createRefPostData([articleId]);
+            data = await this.Scrape.getRefText(postData);
             var newItems = await this.Utils.trans2Items(data, libraryID);
             let targetData = {
                 targetUrls: [item.getField("url")],
                 citations: [null]
             };
             newItems = await this.Utils.fixItem(newItems, targetData);
+            // Keep the same collection in newItem.
+            if (itemCollections.length) {
+                for (let collectionID of itemCollections) {
+                    for (let i of newItems) {
+                        i.addToCollection(collectionID);
+                        await i.saveTx();
+                    };
+                }
+            }
             // Move notes and attachments to newItems
             let childIDs = item.getNotes().concat(item.getAttachments());
             if (childIDs.length > 0) {

diff --git a/chrome/content/scripts/scrape.js b/chrome/content/scripts/scrape.js
@@ -279,6 +279,21 @@ Zotero.Jasminum.Scrape = new function () {
         return { dbname: dbname[1], filename: filename[1], dbcode: dbcode[1] };
     }.bind(Zotero.Jasminum);
 
+    /**
+     * Sometimes CNKI URL contains a temporary dbname, 
+     * you need to find a valid dbname from page.
+     * @param {String} CNKI url string
+     * @return {Object} {dbname: ..., filename: ..., dbcode: ...}
+     */
+    this.getIDFromPage = async function (url) {
+        let pageResp = await Zotero.HTTP.request("GET", url);
+        let page = this.Utils.string2HTML(pageResp.responseText);
+        let dbcode = page.querySelector("input#paramdbcode").value;
+        let filename = page.querySelector("input#paramfilename").value;
+        let dbname = page.querySelector("input#paramdbname").value;
+        return { dbname: dbname, filename: filename, dbcode: dbcode };
+    }.bind(Zotero.Jasminum);
+
 
     this.search = async function (fileData) {
         Zotero.debug("**Jasminum start search");
@@ -406,6 +421,8 @@ Zotero.Jasminum.Scrape = new function () {
             .replace("</li><li>", "") // divide results
             .replace(/<br>|\r/g, "\n")
             .replace(/vo (\d+)\n/, "VO $1\n") // Divide VO and IS to different line
+            .replace(/IS 0(\d+)\n/g, "IS $1\n")  // Remove leading 0
+            .replace(/VO 0(\d+)\n/g, "VO $1\n")
             .replace(/\n+/g, "\n")
             .replace(/\n([A-Z][A-Z1-9]\s)/g, "<br>$1")
             .replace(/\n/g, "")