Python crawling PubMed data: All 985 universities, 211 colleges and universities can swap Liu Cheng

He went to the annual study section points out the time it recently so I have a lot of friends to help them analyze how to do the swap in advance. Retest and swap always inseparable. Today, to share some of the important points of the swap, with the hope that the trend when you adjust the rules, the swap can understand.

Perhaps everyone to swap the impression that as long as the wait for the re-examination of the notification admissions unit can, in fact, this is wrong, this will go through a lot of processes, such as screen out, but the final decision lies in the hands of instructors, every year one-fifth of the candidates reached the score, but fell on the swap. So what are the specific reasons for it?

Author | small Kankan

Edit | Hu Weiwei

Source | CSDN (ID: CSDNnews)

PubMed opportunities and risks in swap

First, the acquisition swap information when there is a major delay, there is no timely access to first-hand information, contact the instructor in advance led to a lot of people, won a good impression.

Second, performance standards, access to the swap too little information, did not go to their favorite universities, and forced no choice.

Third, the swap too much information, we all know that from January to April is the outbreak of the swap information, staring at the large amount of content every day, it will be missed, not ready for rapid screening information, resulting in a hurry to deal with the swap, while ignoring its own matching.

More than three brush probability is greatest, so we must evaluate their own comprehensive ability, swap is a technology live, recognize their own, to choose the right is king. Here then are some tips swap.

To be successful swap, we must first grasp the first-hand information about the swap, and timely access to an insufficient number of specialized universities where admissions, and in line with your condition, it is necessary. Meanwhile, the toner is also a requirement, as follows,

First, have a re-examination of national eligibility rules;

Second, the entrance exam subjects of both families adjust to the professional examination subjects are identical, two subjects of the professional examination subjects to the same or similar (not professional counterparts, the chances of success of the swap is zero);

Third, give priority to internal transfers of this profession, then consider multi-disciplinary transfers, and finally inter-institution transfers;

Fourth, the college counseling Yanzhao Ban confirm whether enrollment is full, full do not consider (enrollment vacancies obtain first-hand information is very important);

Fifth, do not wait for the results of re-examination, re-examination of school time are different, easy to miss the chance to swap.

Selenium actual development reptiles data transfers

网上有很多网站都在公布调剂信息，其中大多数实验室的一手调剂信息都是公布在论坛的，再有很多网站小编转手Copy自己的网站里，那么今天就要说一说怎么去获取调剂数据并进行分析。

本文只获取数据并进行数据分析，以帮助大家更好的了解调剂形势。

选取中国考研网站上公布调剂查询页面，其页面效果大致如图所示：

先对页面结构进行分析，可以看出是【信息流】+【粗->细】+【翻页】的结构。

针对这种结构，采用Python+Selenium进行爬虫开发。

Selenium是一个用于Web应用程序测试的工具。Selenium测试直接运行在浏览器中，就像真正的用户在操作一样。支持的浏览器包括IE（7, 8, 9, 10, 11）、Mozilla 、Firefox、Safari、Google Chrome、Opera等。

这个工具的主要功能包括：测试与浏览器的兼容性——测试你的应用程序看是否能够很好地工作在不同浏览器和操作系统之上。测试系统功能——创建回归测试检验软件功能和用户需求。支持自动录制动作和自动生成 .Net、Java、Perl等不同语言的测试脚本。

大学名称、专业名称以及调剂标题名称，均在列表中。那么：

第一步：获取【信息流所在的HTML元素】

通过对HTML进行解析，可以发现所有的信息流的都是DIV标签包裹的，其共同特征都是class=”info-item font14” 。

由此想到Selenium中的一个接口find_elements_by_xpath，那么Xpath是什么呢？

XPath即为XML路径语言（XML Path Language），它是一种用来确定XML文档中某部分位置的语言。取节点 XPath使用路径表达式在XML文档中选取节点。节点是通过沿着路径或Step来选取的。

下面列出了最有用的路径表达式：

表达式	描述
nodename	选取此节点的所有子节点。
/	从根节点选取。
//	从匹配选择的当前节点选择文档中的节点，而不考虑它们的位置。
.	选取当前节点。
..	选取当前节点的父节点。
@	选取属性。

在下面的表格中，已列出了一些路径表达式以及表达式的结果：

路径表达式	结果
bookstore	选取 bookstore 元素的所有子节点。
/bookstore	选取根元素 bookstore。注释：假如路径起始于正斜杠( / )，则此路径始终代表到某元素的绝对路径！
bookstore/book	选取属于 bookstore 的子元素的所有 book 元素。
//book	选取所有 book 子元素，而不管它们在文档中的位置。
bookstore//book	选择属于 bookstore 元素的后代的所有 book 元素，而不管它们位于 bookstore 之下的什么位置。
//@lang	选取名为 lang 的所有属性。