AnalyKeys.py 一个简单的脚本程序。GetCateKeys 分析关键字的分类，并按分类保存。DecodeUrl 将URL中带有中文的转码成%形式。 PostJson.py 将jekyll中_post目录下的文章到处json格式，满足阿里云搜索导入的数据格式。 GetKeys.py 解析url中的关键字，参数一是解析的文件，参数二是输出的中文关键字，最后打印行数

AnalyKeys.py

# -*- coding: cp936 -*-
'''
Created on 2013-6-19
@author: gdh
'''

import sys
import urllib

def GetCateKeys(fileName):
'''获取分类关键字'''
    f = open(fileName, "r")
    try:
        lines = f.readlines()
    finally:
        f.close()
    
    maps = {}
    for l in lines:
        fs = l.split()
        try:
            if(fs[11] and maps.has_key(fs[11])):
                if fs[5] not in maps[fs[11]]:
                    maps[fs[11]].append(fs[5])
            else:
                maps[fs[11]] = [fs[5]]
        except Exception as e:
            print e, fs
            
            
    metas = ("bing","jike","sogou","soso","panguso","youdao")
    path = "d:\\meta\\"
    for k in maps.keys():
        if k in metas:
            write = open(path + k + ".txt", 'w')
            newmap = maps[k][-100:]
            for l in newmap:
                write.write(l + "\n")
                
def DecodeUrl(fileName):
'''解析url中的汉字'''
    f = open(fileName, "r")
    try:
    lines = f.readlines()
    finally:
        f.close()
    
    write = open("d:/urls.txt", 'w')
    for l in lines:
        str = urllib.quote(l) 
        write.write(str + "\n")
    print 'count：%d' % len(lines)


if __name__ == '__main__':
    if sys.argv[1:]:
        GetCateKeys(sys.argv[1])
    else:
        print "need fileName"
    
    
    

GetKeys.py

# -*- coding: utf-8 -*-
'''
解析url中的关键字，参数一是解析的文件，参数二是输出的中文关键字，最后打印行数
@author: gdh
'''

import os
import sys
import urllib

shstr = "cat " + sys.argv[1]  + ''' | awk -F "key=" '{print $2}' | awk -F "&pageNo" '{print $1}' | awk  '{print $1}' | uniq '''
res = os.popen(shstr).read()

lines = res.split("\n")

write = open(sys.argv[2],"w")

for line in lines:
    write.write(urllib.unquote(line) + "\n")

print len(lines)

PostJson.py

#coding=utf-8
'''
将jekyll中_post目录下的文章到处json格式，用于阿里云搜索
@author: gdh
'''

import os
import re
import json
import time

def object2dict(obj):
    #convert object to a dict
    d = {}
    d.update(obj.__dict__)
    return d

class PostInfo():
    def __init__(self, title, tags, content, rooturl, f):
        self.title = title
        self.tag = {}
        for t in tags.split(","):
            self.tag[t.strip()] = 1
        self.body = content
        self.id = f.replace("-", "").replace(".html", "")
        self.display_text = "open search"
        self.hit_num = "88888"
        t = re.match('\\d{4}-\\d{2}-\\d{2}', f).group()
        self.url = rooturl + f.replace(t + "-", (t + "-").replace("-", "/"))
        self.update_timestamp = "%d" % time.mktime(time.strptime(t,'%Y-%m-%d'))
        self.create_timestamp = "%d" % time.mktime(time.strptime(t,'%Y-%m-%d'))
        self.type_id = "1"
        self.cat_id= [1,2]
        self.author="yhzhtk"
        self.grade="10"
        self.source="yhzhtk"
        self.boost = "1"

    
def getPostInfo(path, rooturl="http://yhzhtk.info/"):
    os.chdir(path)
    posts = [[open(f, "r").read(), f]
             for f in os.listdir(path) if f.endswith(".html")]
    pattern = re.compile(r"---\nlayout: post\ntitle: ([^\n]*)\ntags: \[([^\]]*)\]\n---\n(.*)", re.DOTALL)
    infos = []
    for post, f in posts:
        match = pattern.match(post)
        if match:
            title = match.group(1)
            tags =  match.group(2)
            content = match.group(3)
            content = re.sub("<[^>]*>","",content)
            content = re.sub("{%[^%]*%}","",content)
            content = content.replace("\n", "")
            infos.append(PostInfo(title, tags, content, rooturl, f))
    return infos

def genPostJson(infos):
    jsonstr = ""
    for info in infos:
        temp = json.dumps(info, ensure_ascii=False, default=object2dict)
        temp = '''{"fields":''' + temp + ''', "cmd": "ADD"}'''
        jsonstr += "," + temp
    if json:
        jsonstr = jsonstr[1:]
    jsonstr = "[" + jsonstr + "]"
    print jsonstr

    
if __name__ == '__main__':
    path = r"C:\Documents and Settings\yhzhtk\blog\_posts"
    infos = getPostInfo(path)
    genPostJson(infos);