剝離字符串Python化的方式

Question

我想實現一個python等效以下的bash命令：

VERSION=$(curl --silent "http://nexus:8080/nexus/service/local/lucene/search?g=com.xxx.yyy&a=zzz" | sed -n 's|<latestRelease>\(.*\)</latestRelease>|\1|p' | sed -e 's/^[ \t]*//' | tail -1) 

我想出的代碼下面的代碼片段，其也部分地工作，並得到<latestRelease>1.0.11</latestRelease>輸出的一堆，這是完全可以預料的。但是，我被卡住了，只想得到1.0.11版本作爲python腳本的輸出。再次1.0.11可能根據在Nexus的最新版本不同，所以將是很好，如果專家能提出一個動態的解決方案在一個Python化的方式來剔除在bash的sed和尾部做了部分

#!/usr/bin/env python 

import os; 
import subprocess; 
import re 
import string; 

proc = subprocess.Popen(["curl", "--silent", "http://nexus:8080/nexus/service/local/lucene/search?g=com.xxx.yyy&a=zzz"], stdout=subprocess.PIPE) 
out = proc.communicate()[0] 
search_string = "<latestRelease>" 
for line in out.splitlines(): 
if search_string in line: 
    re.sub(r'\s*latestRelease\s*', '', line) 
    print line 

OUTPUT：

<latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 
    <latestRelease>1.0.11</latestRelease> 

所需的輸出：1.0.11

Answer 1

Python有相關的資料庫，幫助你用它：

• urllib2 - 要對URL的請求並獲得XML數據
• xml.etree.ElementTree解析XML結果，並獲得latestRelease節點值

例（應該在你的情況下工作，但我也沒辦法以測試它的工作）：

import xml.etree.ElementTree as ET 
from urllib2 import urlopen 

url = 'http://nexus:8080/nexus/service/local/lucene/search?g=com.xxx.yyy&a=zzz' 
tree = ET.parse(urlopen(url)) 
print tree.findtext('.//latestRelease')