package com.github.xsocket.job.parser;
import java.io.File;
import java.io.FileInputStream;
import java.io.InputStream;
import java.util.Calendar;
import javax.mail.BodyPart;
import javax.mail.Multipart;
import javax.mail.Session;
import javax.mail.internet.MimeMessage;
import org.apache.commons.io.FileUtils;
import org.apache.http.client.fluent.Request;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import com.github.xsocket.job.AbstractResumeParser;
import com.github.xsocket.job.Resume;
import com.github.xsocket.job.ResumeParser;
import com.github.xsocket.job.resume.ZhilianResume;
import com.github.xsocket.job.util.QuotedPrintableUtils;
/**
* 智联邮件简历解析工具。
* @author MWQ
*
*/
public class ZhilianEmailResumeParser extends AbstractResumeParser implements ResumeParser {
private static final String COMPANY_SPLIT = " ";
private static final String BASIC_INFO_SPLIT = "\\|";
@Override
public String getName() {
return "智联招聘(邮件)";
}
@Override
public boolean canParse(File file) {
if(file == null) {
return false;
} else {
String name = file.getName();
return (name.contains("Zhaopin.com") || name.startsWith("智联招聘")) && name.endsWith(".eml");
}
}
@Override
public Resume parse(File file) throws Exception {
ZhilianResume resume = new ZhilianResume();
String fileName = file.getName();
if(fileName.contains("Zhaopin.com")) {
resume.setJob(intercept(fileName, "应聘 ", "-"));
} else {
// 解析已经解析完成的文件
int start = fileName.lastIndexOf("-") + 1;
int end = fileName.lastIndexOf(".eml");
if(start > 0 && end > start) {
resume.setJob(fileName.substring(start, end));
}
}
Document doc = parse2Html(file);
/*
解析样例如下的数据：
董文朋
男|4年工作经验|1992年5月|未婚 现居住于北京-通州区|本科|湖北-孝感户口
..... 其他信息
2014/02 - 2015/05 辽河油田泰利达有限公司 （1年3个月）
。。。。。。 其他信息
2008/09 - 2012/07 北京国际经贸研修学院 信息管理与信息系统 统招 本科
*/
Elements sections = doc.select("table table table table tr");
// 下一行要解析的数据类型
String nextSection = "姓名";
for(Element section : sections) {
String text = section.text();
if(isNullOrEmpty(text)) {
continue;
}
if("姓名".equals(nextSection)) {
resume.setName(text.trim());
nextSection = "综合";
} else if("综合".equals(nextSection)) {
parseBasicInfo(resume, text);
nextSection = "上家公司";
} else if("上家公司".equals(nextSection)) {
// 判断是否是真是的公司信息
if(text.length() < 20) {
continue;
} else if(text.charAt(4) != '/') {
continue;
} else if(text.charAt(8) != '-') {
continue;
}
String[] infos = text.split(COMPANY_SPLIT);
if(infos.length>=4) {
resume.setCompany(infos[3]);
}
nextSection = "毕业院校";
} else if("毕业院校".equals(nextSection)) {
for(String edu : EDUCATIONS.keySet()) {
if(text.contains(edu)) {
resume.setEducation(edu);
String[] infos = text.split(COMPANY_SPLIT);
if(infos.length>=4) {
resume.setSchool(infos[3]);
}
nextSection = "结束";
break;
}
}
if("结束".equals(nextSection)) {
break;
}
}
}
tryFetchContact(resume, doc);
return resume;
}
protected void tryFetchContact(ZhilianResume resume, Document doc) {
final String SPLIT1 = "url=";
final String SPLIT2 = "ldparam=";
Elements as = doc.select("table table table table tr td a");
for(Element elem : as) {
String href = elem.attr("href");
if(href.contains(SPLIT2) && href.contains(SPLIT1)) {
String url = href.substring(href.lastIndexOf(SPLIT1) + SPLIT1.length(), href.length());
String content;
try {
content = Request.Get(url).execute().returnContent().asString();
Document doc2 = Jsoup.parse(content);
Elements infos = doc2.select("div.login_content p a");
resume.setName(infos.get(0).text());
resume.setPhone(infos.get(1).text());
resume.setMail(infos.get(2).text());
} catch (Exception e) {
e.printStackTrace(System.err);
}
return;
}
}
}
protected void parseBasicInfo(ZhilianResume resume, String text) {
String[] values = text.split(BASIC_INFO_SPLIT);
String next = null;
for(String value : values) {
String str = value.trim();
if(str.contains("男")) {
resume.setSex("男");
next = "工作经验";
} else if(str.contains("女")) {
resume.setSex("女");
next = "工作经验";
} else if(str.contains("工作经验")) {
int index = str.indexOf("年");
if(index > 0) {
resume.setWorkDuration(str.substring(0, index));
} else {
resume.setWorkDuration(str.substring(0, str.indexOf("工作经验")));
}
next = "生日";
} else if("生日".equals(next)){
int index = str.indexOf("年");
if(index > 0) {
resume.setBirthday(str);
try {
int birthYear = Integer.parseInt(str.substring(0, index));
Integer age = Calendar.getInstance().get(Calendar.YEAR) - birthYear;
resume.setAge(age.toString());
} catch(Exception e) {
// 解析出生年份失败，无法计算年龄
}
}
next = "城市";
} else if("城市".equals(next)){
int index = str.indexOf("现居住于");
if(index > 0) {
resume.setCity(str.substring(index + 1, str.length()));
}
next = "学历";
} else if("学历".equals(next)) {
if(EDUCATIONS.containsKey(str)) {
resume.setEducation(str);
}
}
}
}
protected Document parse2Html(File file) throws Exception {
InputStream in = new FileInputStream(file);
Session mailSession = Session.getDefaultInstance(System.getProperties(), null);
MimeMessage msg = new MimeMessage(mailSession, in);
Multipart part = (Multipart) msg.getContent();
String html = "";
for(int i = 0; i < part.getCount(); i++) {
BodyPart body = part.getBodyPart(i);
String type = body.getContentType();
if(type.startsWith("text/html")) {
html = body.getContent().toString();
break;
}
}
in.close();
if(html == null || html.length() == 0) {
String content = FileUtils.readFileToString(file);
final String endFlag = "</html>";
int start = content.indexOf("<html");
int end = content.indexOf(endFlag);
content = content.substring(start, end + endFlag.length());
html = QuotedPrintableUtils.decode(content.getBytes(), "gb2312");
System.err.println(html);
}
return Jsoup.parse(html);
}
}
