How to dump all wiki pages from Phabricator

参照的源代码在这里：https://gist.github.com/HackToHell/9f261728f4ceda1605c9b3b1f46addbc

install software

yum -y install \
	python-sqlalchemy \
	python-pandas \
	MySQL-python;

code

# -*- coding: utf-8 -*-
import pandas as pd
import os
import errno
from sqlalchemy import *
import sys
reload(sys)
sys.setdefaultencoding('utf8')

# Change me to the phriction database
engine = create_engine( 
        'mysql://user:[email protected]/phabricator_phriction?charset=utf8mb4')

df=pd.read_sql('select * from phriction_content',engine)

def get(x):
    x=x.iloc[0]
    strpath='data/' + x['slug'][:-1] + '.md'
    #print strpath
    strdir = os.path.dirname(strpath)
    #print strdir
    try:
        os.makedirs(strdir)
    except OSError as exc:
        if exc.errno == errno.EEXIST and os.path.isdir(strdir):
            pass
        else:
            raise

    with open(strpath, 'w') as fp:
        fp.write(x['content'])

for name,group in df.groupby('title'):
    if len(group) > 1:
        x=group[group['version'] == group['version'].max()] #Get only the latest version
        get(x)
    else:
        get(group)

NOTE: 这里有个小 bug：如果当某个页面 A 被 move 成 B 之后呢，这个程序只能 dump 出 A.md，而且里面是空的。