biglaw_splitter

splitterMTQwNTIxNDUzOQ==.py

#!/usr/bin/python
#coding=utf8
 
import os
import re
import sys
import pymysql
 
#===============================================
# normalize input string to prevent from errors
#===============================================
def str_normalize(str):
    try:
        res = ''
        if str is not None:
            strlist = str.split('\'')
            cnt = 0
            for value in strlist:
                if(cnt > 0):
                    res += '\\\''
                cnt += 1
                res += value
    except:
        print ("E001")
        sys.exit('E001')
        
    return res
 
#===============================================
# Put many regular expression in proper order
#===============================================
def get_regexlist():
    try:
        regexlist = []
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[\s　]+事[　\s]+實.*)\n(?P<Fact1Word>(.*\n)*)(?P<Fact2>[　\s]*理[　\s]+由.*)\n(?P<Fact2Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[　\s]*事[　\s]*實及理[　\s]*由.*)\n(?P<Fact1Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[\s　]+事[　\s]+實.*)\n(?P<Fact1Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[　\s]*犯罪事實[　\s]*及[　\s]*理由.*)\n(?P<Fact1Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[　\s]*犯罪事實.*)\n(?P<Fact1Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n(?P<Premain>(.*\n)*)[　\s]*主[　\s]+文[　\s]*\n+(?P<Main>(.*\n)*)(?P<Fact1>[　\s]*理[　\s]+由.*)\n(?P<Fact1Word>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
        regexlist.append('.*\n(?P<Location>\S+)-\S+-(?P<Caseyear>\d+)(?P<Character>\D+)(?P<Number>\d+)\n.*\n.*\s--\s(?P<Type>\S+)\n\S裁判字號\S\n.*\n\S裁判日期\S\n(?P<Judgedate>\d+).*\n\S裁判案由\S\n(?P<Judgereason>(.*\n)*)\S裁判全文\S\n((?:.*分院)|(?:.*法院))(?P<Judge>[^　\s]+)[　\s]+.*號\n*(?P<Premain>(.*\n)*?)(?P<Tail>中[　\s]+華[　\s]+民[　\s]+國.*\n(?:.*\n)*).*')
    except:
        print ("E001")
        sys.exit('E001')
 
    return regexlist
 
#===============================================
# Get name of columns lied in table dynamically
#===============================================
def get_all_column_name(db, cursor):
    try:
        sql = "SELECT column_name FROM INFORMATION_SCHEMA.COLUMNS WHERE table_name='refereebook'"
        cursor.execute(sql)
        mytuple = cursor.fetchall()
        db.commit()
    except:
        db.rollback()
        print ("E004")
        sys.exit('E004')
    try:
        columns = []
        for emt in mytuple:
            columns.append(emt[0])
    except:
        print ("E001")
        sys.exit('E001')
    return columns
 
#===============================================
# Try to decompose the file
#===============================================
def anylize(filepath,writeDB):
    # Check filepath
    try:
        os.path.isfile(filepath)
    except:
        print ("E001")
        sys.exit('E001')
    # Open database connectiion
    try:
        db = pymysql.connect(host="changethis", user="changethis", passwd="changethis", db="changethis", charset="utf8")
        cursor = db.cursor()
    except:
        print ("E002")
        sys.exit('E002')
 
    # Open the file and translate into a string
	
    try:
        fp = open(filepath, 'r', encoding="utf8")
        filecontent = ""
        for line in fp:
            filecontent += line
    except:
        print ("E001")
        sys.exit('E001')
 
    # Use regular expression
    try:
        columnlist = get_all_column_name(db, cursor)
        regexlist = get_regexlist()
        regex_ver = 0
        for regex in regexlist:
            results = re.search(regex, filecontent)
            if(results is not None):
                break;
            regex_ver += 1
        sqllist = []
        for column in columnlist:
            if(re.search(column, regex) is not None):
                if(column == "Judgedate"):
                    year = str(int(results.group(column)[0:3]) + 1911)
                    month = results.group(column)[3:5]
                    day = results.group(column)[5:7]
                    sqllist.append(str_normalize(year + '-' + month + '-' + day))
                else:
                    sqllist.append(str_normalize(results.group(column)))
            else:
                sqllist.append(None)
    except Exception as e:
        print ("E003")
        sys.exit('E003')
    if(not writeDB):
        print (sqllist)
        return 0
    # Construct SQL command and insert into table
    if(writeDB):
        try:
            sql = "INSERT INTO `refereebook`("
            for column in columnlist[1:]:
                sql += "`{0}`".format(column)
                if(column == columnlist[-1]):
                    sql += ")"
                    break;
                else:
                    sql += ","
            sql += "VALUES ("
            for emt in sqllist[1:]:
                if(emt is None):
                    sql += 'NULL'
                else:
                    sql += "\'{0}\'".format(emt)
                if(emt == sqllist[-1]):
                    sql += ")"
                    break;
                else:
                    sql += ","
            cursor.execute(sql)
            db.commit()
            StatusCode = cursor.lastrowid
        except Exception as e:
            db.rollback()
            print (e)
            print ("E004")
            sys.exit('E004')
    
    return StatusCode
 
 
#===============================================
# 'E001': Constant error
# 'E002': Database connection error
# 'E003': Regular expression error
# 'E004': SQL fail
# other: ID of successfully inserted record 
#===============================================
def main(argv = None):
    if argv is None:
        argv = sys.argv
    if(len(sys.argv) == 2):
        return anylize(sys.argv[1], True)
    if(len(sys.argv) == 3):
        return anylize(sys.argv[1], False)
    print ("E001")
    sys.exit('E001')
    
if __name__ == "__main__":
    scode = main()
    print(scode)
    sys.exit(scode)