Python3自动化_文件批量处理(文本、PDF；读取、筛选、导出)

利用Python3脚本语言的简练语法，高级语言的丰富类库，快速写了几个文件读取、筛选、导出的“脚本”。

这里简单总结一下关键功能。

读取ini配置文件

检查ini文件是否存在；检查输入的key在ini文件里是否有定义。

 1 import configparser 
 2

 4 def getConfigInfo(_ini_nm):
 5 
 6     # Open Ini File
 7     config = configparser.ConfigParser()
 8     if not config.read(os.path.join(os.getcwd(), _ini_nm + r'.ini')):
 9         printLog('E', 'Read Ini file fail.')
10 
11     while True:
12         sysCode = input(r'Please input the system code : (Press [Enter] to quit):').strip()
13         if 0 == len(sysCode.strip()):
14             exit()
15 
16         # Init ConnectionSettings
17         if sysCode in config.sections():
18             return dict(config[sysCode])
19         else:
20             print('Ini info of System [%s] is blank.\n' % sysCode)

多参数输入的获取

检查参数个数；检查参数合法性（长度，是否目录）；检查参数是否整个都是汉字。

 1 def _main():
 2 
 3     path = ''
 4     keyWord = ''
 5 
 6     while True:
 7         para = input(r'Please input the PDF directory and Key Word: (Press [Enter] to quit):').strip().split()
 8 
 9         if 2 != len(para): continue
10 
11         path = para[0]
12         keyWord = para[1]
13 
14         if 0 == len(path.strip()):
15             exit()
16 
17         if not os.path.exists(path):
18             print('input path is not a exists path.' + '\n')
19             continue
20 
21         flg = True
22         for char in keyWord.strip():
23             if char <= u'\u4e00' or char >= u'\u9fa5':
24                 flg = False
25                 break
26         if not flg:
27             print('Please input the Chinese Key Word for search.(Such as \'物流\').' + '\n')
28             continue
29 
30         break

PostgreSQL数据库处理

根据ini文件定义的数据库连接信息，尝试连库；执行SQL文。

 1 import psycopg2 
 4
 import traceback
 5 
 6 def connDB(_cfg):
 7     try:
 8         conn = psycopg2.connect(database=_cfg['servicename'],
 9                                 user=_cfg['dbuser'],
10                                 password=_cfg['dbpw'],
11                                 host=_cfg['host'],
12                                 port=_cfg['port'])
13         return conn
14     except Exception:
15         printLog('E', 'Exception occur at DB Connection.' + '\n' + traceback.format_exc())
16 
17 def executeSql(_cfg, _sql):
18     try:
19         conn = connDB(_cfg)
20         cur = conn.cursor()
21         cur.execute(_sql)
22 
23         results = cur.fetchall()
24         return list(map(lambda x: x[0], results))
25     except Exception:
26         printLog('E', 'Exception occur at Execute SQL.' + '\n' + traceback.format_exc())
27     finally:
28         cur.close()
29         conn.rollback()
30         conn.close()

日志处理

定义输出日志的级别；异常级别时，处理结束。

 1 logging.basicConfig(filename='log_' + datetime.now().strftime('%Y%m%d') + '.txt',
 2                     level=logging.INFO,
 3                     format=' %(asctime)s - %(levelname)s - %(message)s')
 4 
 5 logLevel = {'D': logging.DEBUG,
 6             'I': logging.INFO,
 7             'W': logging.WARNING,
 8             'E': logging.ERROR,
 9             'C': logging.CRITICAL}
10 
11 def printLog(_lvl, _msg):
12     logging.log(logLevel[_lvl], _msg)
13     if logging.ERROR == logLevel[_lvl]:
14         print(_msg)
15         exit()
16 
17 
18 printLog('E', 'srcpath is not a exists path.')
19 printLog('I', 'Get Src Path : %s' % srcPath)

MAP函数运用

列表元素批量处理，按第二个下划线字符截取字符串。

1 def getPreOfNm(x):
2     if 1 < x.count('_'):
3         return x[0:x.find('_', x.find('_') + 1)]
4     else:
5         return x
6 
7 # Get prefix of CRUD object name
8 prefixObjNm = list(set(map(getPreOfNm, lstTb)))
9 prefixObjNm.sort()

目录处理

目录/文件判断；目录的路径分割；完整路径的文件名取得；

 1 # Check the srcPath
 2 fullFilePaths = []
 3 if os.path.isdir(srcPath):
 4     for folderName, subFolders, fileNames in os.walk(srcPath):
 5         if os.path.split(folderName)[1] in ['tcs', 'doc']: continue
 6         for fn in fileNames:
 7             # Get src file
 8             mObj = fileNmReg.search(fn)
 9             if mObj:
10                 fullFilePaths.append(os.path.join(folderName, fn))
11 elif os.path.isfile(srcPath):
12     # Get src file
13     fn = os.path.basename(os.path.realpath(srcPath))
14     mObj = fileNmReg.search(fn)
15     if mObj:
16         fullFilePaths.append(srcPath)

PDF文件读取

来源：https://www.cnblogs.com/alexzhang92/p/11488949.html

 1 from pdfminer.converter import TextConverter
 2 from pdfminer.layout import LAParams
 3 from pdfminer.pdfinterp import PDFResourceManager, process_pdf
 4 import os
 5 
 6 
 7 def read_pdf(pdf):
 8     # resource manager
 9     rsrcmgr = PDFResourceManager()
10     retstr = StringIO()
11     laparams = LAParams()
12     # device
13     device = TextConverter(rsrcmgr, retstr, laparams=laparams)
14     process_pdf(rsrcmgr, device, pdf)
15     device.close()
16     content = retstr.getvalue()
17     retstr.close()
18     # 获取所有行
19     contents = str(content).split("\n")
20 
21     return contents

CSV文件导出

1 # Init result file
2 rstFile = open(os.path.join(srcPath, '[CRUD]' + datetime.now().strftime('%Y%m%d%H%M%S') + '.csv'), 'w', newline='')
3 rstWtr = csv.writer(rstFile, delimiter='\t', lineterminator='\n')
4 # Write head
5 rstWtr.writerow(['TYPE', 'CI', 'ENCODE', 'LINE NUM', 'CRUD', 'TABLE NM', 'FULL PATH'])

转载请注明原文链接，谢谢。

来源：oschina

链接：https://my.oschina.net/u/4418711/blog/4271937

标签

def

strip

pdfminer