improvements

build the dev environment and try to solve the mismatch query.

.gitignore

@@ -86,3 +86,6 @@ typings/
 # Custom Misc
 build/
 builds/
+.DS_Store
+package.json
+package-lock.json

README.md

@@ -5,8 +5,11 @@
 
 New features: 
 - Improve the process of dealing with reCAPTCHA, especially in China;
-- Fix url error when encounter special characters in the paper title;
 - When updating many items, a random time delay is added, which may avoid Google Scholar traffic detection and reduce the frequency of reCAPTCHA;
+- Fix url error when encounter special characters in the paper title;
+- Add a rough time range to the generated query url:
+  > For example, if a paper titled "Explaining and Exploiting Adversarial Examples" has a year of "2013" in Zotero, the original version will produce a URL that results in an incorrect query (https://scholar.google.com/scholar?q=Explaining+and+Harnessing+Adversarial+Examples&hl=en&as_sdt=0%2C5&as_ylo=2013&as_yhi=2013), while a looser range, such as 2012 - 2014, will hit the correct query (https://scholar.google.com/scholar?q=Explaining+and+Harnessing+Adversarial+Examples&hl=en&as_sdt=0%2C5&as_ylo=2012&as_yhi=2014).
+
 
 v2.0.5
 - Adaptation for Zotero 6.0;

chrome/content/zsc.js

@@ -12,7 +12,7 @@ let zsc = {
  _extraEntrySep: "\n",
 };
 
-zsc._extraRegex = new RegExp(zsc._extraPrefix + ".*");
+zsc._extraRegex = new RegExp(zsc._extraPrefix + ".{0,20}");
 
 let isDebug = function () {
  return (
@@ -184,20 +184,43 @@ zsc.retrieveCitationData = function (item, cb) {
  let citeCount;
  let xhr = new XMLHttpRequest();
  xhr.open("GET", url, true);
+ // xhr.responseType = "document";
  xhr.onreadystatechange = function () {
  if (this.readyState == 4 && this.status == 200) {
  // // debug on response text
  // if (isDebug()) Zotero.debug(this.responseText);
 
  // check if response includes meaningful content
  if (this.responseText.indexOf('class="gs_r gs_or gs_scl"') != -1) {
- if (isDebug())
+ if (isDebug()) {
+ Zotero.debug(
+ "[scholar-citations] received non-captcha scholar results!"
+ );
+
+ // check if the returned title match with the itme title
+ var title1 = item
+ .getField("title")
+ .trim()
+ .toLowerCase()
+ .replace(/ +/g, " ");
+ Zotero.debug("[scholar-citations] the item title: " + title1);
+ var parser = new DOMParser();
+ var htmlDoc = parser.parseFromString(this.responseText, "text/html");
+ var title2 = htmlDoc
+ .getElementsByClassName("gs_rt")[0]
+ .innerText.trim()
+ .toLowerCase()
+ .replace(/ +/g, " ");
+ Zotero.debug("[scholar-citations] the queried title: " + title2);
  Zotero.debug(
- "[scholar-citations] received non-captcha scholar results"
+ "[scholar-citations] will item title equals to queried title?" +
+ (title1 === title2)
  );
+ }
+
  cb(item, zsc.getCiteCount(this.responseText));
- // check if response includes captcha
  } else if (
+ // check if response includes captcha
  this.responseText.indexOf("www.google.com/recaptcha/api.js") != -1
  ) {
  if (isDebug())
@@ -225,6 +248,7 @@ zsc.retrieveCitationData = function (item, cb) {
  }
  } else if (this.readyState == 4 && this.status == 429) {
  if (this.responseText.indexOf("www.google.com/recaptcha/api.js") == -1) {
+ // failed without captcha
  if (isDebug())
  Zotero.debug(
  "[scholar-citations] could not retrieve the google scholar data. Server returned: [" +
@@ -237,6 +261,7 @@ zsc.retrieveCitationData = function (item, cb) {
  " seconds before sending further requests."
  );
  } else {
+ // failed with captcha
  if (isDebug())
  Zotero.debug(
  "[scholar-citations] received a captcha instead of a scholar result"

test/http/429.js

@@ -1,14 +1,16 @@
-let http = require('http');
+let http = require("http");
 
 let port = 8080;
 
-http.createServer(function(req, res) {
- console.log('incomming request!');
- console.log('method: ', req.method);
- console.log('url: ', req.url);
- console.log('header: ', req.headers);
- res.writeHead(429, {'Content-Type': 'text/plain', 'Retry-After': 3600});
- res.end('Yikes! Your\'re blocked!');
-}).listen(port);
+http
+ .createServer(function (req, res) {
+ console.log("incomming request!");
+ console.log("method: ", req.method);
+ console.log("url: ", req.url);
+ console.log("header: ", req.headers);
+ res.writeHead(429, { "Content-Type": "text/plain", "Retry-After": 3600 });
+ res.end("Yikes! Your're blocked!");
+ })
+ .listen(port);
 
-console.log('Starting super simple http server on localhost:' + port + '!');
+console.log("Starting super simple http server on localhost:" + port + "!");

test/http/captcha.js

@@ -1,14 +1,16 @@
-let http = require('http');
+let http = require("http");
 
 let port = 8080;
 
-http.createServer(function(req, res) {
- console.log('incomming request!');
- console.log('method: ', req.method);
- console.log('url: ', req.url);
- console.log('header: ', req.headers);
- res.writeHead(200, {'Content-Type': 'text/plain'});
- res.end('Such Captcha! Much Protec! www.google.com/recaptcha/api.js Wow!');
-}).listen(port);
+http
+ .createServer(function (req, res) {
+ console.log("incomming request!");
+ console.log("method: ", req.method);
+ console.log("url: ", req.url);
+ console.log("header: ", req.headers);
+ res.writeHead(200, { "Content-Type": "text/plain" });
+ res.end("Such Captcha! Much Protec! www.google.com/recaptcha/api.js Wow!");
+ })
+ .listen(port);
 
-console.log('Starting super simple http server on localhost:' + port + '!');
+console.log("Starting super simple http server on localhost:" + port + "!");