指令碼寫的好，下班下得早！程式設計師的日常工作除了編寫程式程式碼，還不可避免地需要處理相關的測試和驗證工作。

例如，訪問某個網站一直不通，需要確定此地址是否可訪問，伺服器返回什麼，進而確定問題在於什麼。完成這個任務，如果一味希望採用編譯型語言來編寫這樣的程式碼，實踐中的時間和精力是不夠的，這個時候就需要發揮指令碼的神奇作用！

好不誇張的說，能否寫出高效實用的指令碼程式碼，直接影響著一個程式設計師的幸福生活[下班時間]。下面整理 8 個實用的 Python 指令碼，需要的時候改改直接用，建議收藏！

1.解決 linux 下 unzip 亂碼的問題。

import os
import sys
import zipfile
import argparse

s = '\x1b[%d;%dm%s\x1b[0m'
 
       

def unzip(path):

    file = zipfile.ZipFile(path,"r")
    if args.secret:
        file.setpassword(args.secret)

    for name in file.namelist():
        try:
            utf8name=name.decode('gbk')
            pathname = os.path.dirname(utf8name)
        except:
            utf8name=name
            pathname = os.path.dirname(utf8name)

        #print s % (1,92,'  >> extracting:'),utf8name
 

        #pathname = os.path.dirname(utf8name)
        if not os.path.exists(pathname) and pathname != "":
            os.makedirs(pathname)
        data = file.read(name)
        if not os.path.exists(utf8name):
            try:
                fo = open(utf8name,"w")
                fo.write(data)
                fo.close
            except:
                pass
    file.close()

def main(argv):
    ######################################################
 

    # for argparse
    p = argparse.ArgumentParser(description='解決unzip亂碼')
    p.add_argument('xxx',type=str,nargs='*',\
        help='命令物件.')
    p.add_argument('-s','--secret',action='store',\
        default=None,help='密碼')
    global args
    args = p.parse_args(argv[1:])
    xxx = args.xxx

    for path in xxx:
        if path.endswith('.zip'):
            if os.path.exists(path):
                print s % (1,97,'  ++ unzip:'),path
                unzip(path)
            else:
                print s % (1,91,'  !! file doesn\'t exist.'),path
        else:
            print s % (1,'  !! file isn\'t a zip file.'),path

if __name__ == '__main__':
    argv = sys.argv
    main(argv)
複製程式碼

2.統計當前根目錄程式碼行數。

# coding=utf-8
import os
import time
# 設定根目錄
basedir = './'
filelists = []
# 指定想要統計的檔案型別
whitelist = ['cpp','h']
#遍歷檔案,遞迴遍歷資料夾中的所有
def getFile(basedir):
    global filelists
    for parent,dirnames,filenames in os.walk(basedir):
        for filename in filenames:
            ext = filename.split('.')[-1]
            #只統計指定的檔案型別，略過一些log和cache檔案
            if ext in whitelist:
                filelists.append(os.path.join(parent,filename))
#統計一個的行數
def countLine(fname):
    count = 0
    # 把檔案做二進位制看待,read.
    for file_line in open(fname,'rb').readlines():
        if file_line != '' and file_line != '\n': #過濾掉空行
            count += 1
    print (fname + '----',count)
    return count
if __name__ == '__main__' :
    startTime = time.clock()
    getFile(basedir)
    totalline = 0
    for filelist in filelists:
        totalline = totalline + countLine(filelist)
    print ('total lines:',totalline)
    print ('Done! Cost Time: %0.2f second' % (time.clock() - startTime))
複製程式碼

3.掃描當前目錄和所有子目錄並顯示大小。

import os
import sys      
try:
    directory = sys.argv[1]   
except IndexError:
    sys.exit("Must provide an argument.")

dir_size = 0   
fsizedicr = {'Bytes': 1,'Kilobytes': float(1) / 1024,'Megabytes': float(1) / (1024 * 1024),'Gigabytes': float(1) / (1024 * 1024 * 1024)}
for (path,dirs,files) in os.walk(directory):      
    for file in files:                              
        filename = os.path.join(path,file)
        dir_size += os.path.getsize(filename)       

fsizeList = [str(round(fsizedicr[key] * dir_size,2)) + " " + key for key in fsizedicr] 

if dir_size == 0: print ("File Empty") 
else:
  for units in sorted(fsizeList)[::-1]: 
      print ("Folder Size: " + units)
複製程式碼

4.將源目錄240天以上的所有檔案移動到目標目錄。

import shutil
import sys
import time
import os
import argparse

usage = 'python move_files_over_x_days.py -src [SRC] -dst [DST] -days [DAYS]'
description = 'Move files from src to dst if they are older than a certain number of days.  Default is 240 days'

args_parser = argparse.ArgumentParser(usage=usage,description=description)
args_parser.add_argument('-src','--src',nargs='?',default='.',help='(OPTIONAL) Directory where files will be moved from. Defaults to current directory')
args_parser.add_argument('-dst','--dst',required=True,help='(REQUIRED) Directory where files will be moved to.')
args_parser.add_argument('-days','--days',type=int,default=240,help='(OPTIONAL) Days value specifies the minimum age of files to be moved. Default is 240.')
args = args_parser.parse_args()

if args.days < 0:
	args.days = 0

src = args.src  # 設定源目錄
dst = args.dst  # 設定目標目錄
days = args.days # 設定天數
now = time.time()  # 獲得當前時間

if not os.path.exists(dst):
	os.mkdir(dst)

for f in os.listdir(src):  # 遍歷源目錄所有檔案
    if os.stat(f).st_mtime < now - days * 86400:  # 判斷是否超過240天
        if os.path.isfile(f):  # 檢查是否是檔案
            shutil.move(f,dst)  # 移動檔案
複製程式碼

5.掃描指令碼目錄，並給出不同型別指令碼的計數。

import os																	
import shutil																
from time import strftime												

logsdir="c:\logs\puttylogs"											
zipdir="c:\logs\puttylogs\zipped_logs"							
zip_program="zip.exe"												

for files in os.listdir(logsdir):										
	if files.endswith(".log"):										
		files1=files+"."+strftime("%Y-%m-%d")+".zip"		
		os.chdir(logsdir) 												
		os.system(zip_program + " " +  files1 +" "+ files)	
		shutil.move(files1,zipdir)									 
		os.remove(files)													
複製程式碼

6.下載Leetcode的演演算法題。

import sys
import re
import os
import argparse
import requests
from lxml import html as lxml_html

try:
    import html
except ImportError:
    import HTMLParser
    html = HTMLParser.HTMLParser()

try:
    import cPickle as pk
except ImportError:
    import pickle as pk

class LeetcodeProblems(object):
    def get_problems_info(self):
        leetcode_url = 'https://leetcode.com/problemset/algorithms'
        res = requests.get(leetcode_url)
        if not res.ok:
            print('request error')
            sys.exit()
        cm = res.text
        cmt = cm.split('tbody>')[-2]
        indexs = re.findall(r'<td>(\d+)</td>',cmt)
        problem_urls = ['https://leetcode.com' + url \
                        for url in re.findall(
                            r'<a href="(/problems/.+?)"',cmt)]
        levels = re.findall(r"<td value='\d*'>(.+?)</td>",cmt)
        tinfos = zip(indexs,levels,problem_urls)
        assert (len(indexs) == len(problem_urls) == len(levels))
        infos = []
        for info in tinfos:
            res = requests.get(info[-1])
            if not res.ok:
                print('request error')
                sys.exit()
            tree = lxml_html.fromstring(res.text)
            title = tree.xpath('//meta[@property="og:title"]/@content')[0]
            description = tree.xpath('//meta[@property="description"]/@content')
            if not description:
                description = tree.xpath('//meta[@property="og:description"]/@content')[0]
            else:
                description = description[0]
            description = html.unescape(description.strip())
            tags = tree.xpath('//div[@id="tags"]/following::a[@class="btn btn-xs btn-primary"]/text()')
            infos.append(
                {
                    'title': title,'level': info[1],'index': int(info[0]),'description': description,'tags': tags
                }
            )

        with open('leecode_problems.pk','wb') as g:
            pk.dump(infos,g)
        return infos

    def to_text(self,pm_infos):
        if self.args.index:
            key = 'index'
        elif self.args.title:
            key = 'title'
        elif self.args.tag:
            key = 'tags'
        elif self.args.level:
            key = 'level'
        else:
            key = 'index'

        infos = sorted(pm_infos,key=lambda i: i[key])

        text_template = '## {index} - {title}\n' \
            '~{level}~  {tags}\n' \
            '{description}\n' + '\n' * self.args.line
        text = ''
        for info in infos:
            if self.args.rm_blank:
                info['description'] = re.sub(r'[\n\r]+',r'\n',info['description'])
            text += text_template.format(**info)

        with open('leecode problems.txt','w') as g:
            g.write(text)

    def run(self):
        if os.path.exists('leecode_problems.pk') and not self.args.redownload:
            with open('leecode_problems.pk','rb') as f:
                pm_infos = pk.load(f)
        else:
            pm_infos = self.get_problems_info()

        print('find %s problems.' % len(pm_infos))
        self.to_text(pm_infos)

def handle_args(argv):
    p = argparse.ArgumentParser(description='extract all leecode problems to location')
    p.add_argument('--index',action='store_true',help='sort by index')
    p.add_argument('--level',help='sort by level')
    p.add_argument('--tag',help='sort by tag')
    p.add_argument('--title',help='sort by title')
    p.add_argument('--rm_blank',help='remove blank')
    p.add_argument('--line',default=10,help='blank of two problems')
    p.add_argument('-r','--redownload',help='redownload data')
    args = p.parse_args(argv[1:])
    return args

def main(argv):
    args = handle_args(argv)
    x = LeetcodeProblems()
    x.args = args
    x.run()

if __name__ == '__main__':
    argv = sys.argv
    main(argv)
複製程式碼

7.將 Markdown 轉換為 HTML。

import sys
import os

from bs4 import BeautifulSoup
import markdown

class MarkdownToHtml:

    headTag = '<head><meta charset="utf-8" /></head>'

    def __init__(self,cssFilePath = None):
        if cssFilePath != None:
            self.genStyle(cssFilePath)

    def genStyle(self,cssFilePath):
        with open(cssFilePath,'r') as f:
            cssString = f.read()
        self.headTag = self.headTag[:-7] + '<style type="text/css">{}</style>'.format(cssString) + self.headTag[-7:]

    def markdownToHtml(self,sourceFilePath,destinationDirectory = None,outputFileName = None):
        if not destinationDirectory:
            # 未定義輸出目錄則將原始檔目錄(注意要轉換為絕對路徑)作為輸出目錄
            destinationDirectory = os.path.dirname(os.path.abspath(sourceFilePath))
        if not outputFileName:
            # 未定義輸出檔名則沿用輸入檔名
            outputFileName = os.path.splitext(os.path.basename(sourceFilePath))[0] + '.html'
        if destinationDirectory[-1] != '/':
            destinationDirectory += '/'
        with open(sourceFilePath,'r',encoding='utf8') as f:
            markdownText = f.read()
        # 編譯出原始 HTML 文字
        rawHtml = self.headTag + markdown.markdown(markdownText,output_format='html5')
        # 格式化 HTML 文字為可讀性更強的格式
        beautifyHtml = BeautifulSoup(rawHtml,'html5lib').prettify()
        with open(destinationDirectory + outputFileName,'w',encoding='utf8') as f:
            f.write(beautifyHtml)

if __name__ == "__main__":
    mth = MarkdownToHtml()
    # 做一個命令列引數列表的淺拷貝，不包含指令碼檔名
    argv = sys.argv[1:]
    # 目前列表 argv 可能包含原始檔路徑之外的元素（即選項資訊）
    # 程式最後遍歷列表 argv 進行編譯 markdown 時，列表中的元素必須全部是原始檔路徑
    outputDirectory = None
    if '-s' in argv:
        cssArgIndex = argv.index('-s') +1
        cssFilePath = argv[cssArgIndex]
        # 檢測樣式表文件路徑是否有效
        if not os.path.isfile(cssFilePath):
            print('Invalid Path: '+cssFilePath)
            sys.exit()
        mth.genStyle(cssFilePath)
        # pop 順序不能隨意變化
        argv.pop(cssArgIndex)
        argv.pop(cssArgIndex-1)
    if '-o' in argv:
        dirArgIndex = argv.index('-o') +1
        outputDirectory = argv[dirArgIndex]
        # 檢測輸出目錄是否有效
        if not os.path.isdir(outputDirectory):
            print('Invalid Directory: ' + outputDirectory)
            sys.exit()
        # pop 順序不能隨意變化
        argv.pop(dirArgIndex)
        argv.pop(dirArgIndex-1)
    # 至此，列表 argv 中的元素均是原始檔路徑
    # 遍歷所有原始檔路徑
    for filePath in argv:
        # 判斷檔案路徑是否有效
        if os.path.isfile(filePath):
            mth.markdownToHtml(filePath,outputDirectory)
        else:
            print('Invalid Path: ' + filePath)
複製程式碼

8.文字檔案編碼檢測與轉換。

import sys
import os
import argparse
from chardet.universaldetector import UniversalDetector

parser = argparse.ArgumentParser(description = '文字檔案編碼檢測與轉換')
parser.add_argument('filePaths',nargs = '+',help = '檢測或轉換的檔案路徑')
parser.add_argument('-e','--encoding',nargs = '?',const = 'UTF-8',help = '''
目標編碼。支援的編碼有：
ASCII,(Default) UTF-8 (with or without a BOM),UTF-16 (with a BOM),UTF-32 (with a BOM),Big5,GB2312/GB18030,EUC-TW,HZ-GB-2312,ISO-2022-CN,EUC-JP,SHIFT_JIS,ISO-2022-JP,ISO-2022-KR,KOI8-R,MacCyrillic,IBM855,IBM866,ISO-8859-5,windows-1251,ISO-8859-2,windows-1250,EUC-KR,ISO-8859-1,windows-1252,ISO-8859-7,windows-1253,ISO-8859-8,windows-1255,TIS-620
''')
parser.add_argument('-o','--output',help = '輸出目錄')
# 解析引數，得到一個 Namespace 物件
args = parser.parse_args()
# 輸出目錄不為空即視為開啟轉換,若未指定轉換編碼，則預設為 UTF-8
if args.output != None:
    if not args.encoding:
        # 預設使用編碼 UTF-8
        args.encoding = 'UTF-8'
    # 檢測使用者提供的輸出目錄是否有效
    if not os.path.isdir(args.output):
        print('Invalid Directory: ' + args.output)
        sys.exit()
    else:
        if args.output[-1] != '/':
            args.output += '/'
# 例項化一個通用檢測器
detector = UniversalDetector()
print()
print('Encoding (Confidence)',':','File path')
for filePath in args.filePaths:
    # 檢測檔案路徑是否有效，無效則跳過
    if not os.path.isfile(filePath):
        print('Invalid Path: ' + filePath)
        continue
    # 重置檢測器
    detector.reset()
    # 以二進位制模式讀取檔案
    for each in open(filePath,'rb'):
        # 檢測器讀取資料
        detector.feed(each)
        # 若檢測完成則跳出迴圈
        if detector.done:
            break
    # 關閉檢測器
    detector.close()
    # 讀取結果
    charEncoding = detector.result['encoding']
    confidence = detector.result['confidence']
    # 列印資訊
    if charEncoding is None:
        charEncoding = 'Unknown'
        confidence = 0.99
    print('{} {:>12} : {}'.format(charEncoding.rjust(8),'('+str(confidence*100)+'%)',filePath))
    if args.encoding and charEncoding != 'Unknown' and confidence > 0.6:
        # 若未設定輸出目錄則覆蓋原始檔
        outputPath = args.output + os.path.basename(filePath) if args.output else filePath
        with open(filePath,encoding = charEncoding,errors = 'replace') as f:
            temp = f.read()
        with open(outputPath,encoding = args.encoding,errors = 'replace') as f:
            f.write(temp)
複製程式碼

最後兩個指令碼內容選至實驗樓的課程《使用 Python3 編寫系列實用指令碼》，課程對這兩個指令碼有詳細的實現過程講解，感興趣的同學可以直接前往實驗樓進行學習！

知乎專欄同步：zhuanlan.zhihu.com/p/85728888