**我在下面解决了这个问题** ：我认为将来可能会对其他人有所帮助，所以我要保持质疑而不是将其删除。它是python与其他语言嵌套文件导入问题。但是，如果有人理解为什么在python中这样做的复杂性，那么解释性答案将会非常受欢迎。
我的代码运行正常，文件目录设置如下：
sniffer //folder
-__init__.py
-Sniffer.py
-database.py
I switched it to:
Main
-snifferLaunch.py
-flashy
--sniffer
---Sniffer.py
---database.py
理论上，如果我更改导入以查找文件夹，它仍然应该以相同的方式运行...
我的印象是，即使它是嵌套的，也可以导入python文件。例如
然而，我发现这是假的，我误解了吗？所以我试着看一个导入这样的文件的例子：导入Sniffer //在snifferLaunch中应该遍历每个文件并尝试查找Sniffer.py文件。
将flashy.sniffer.Sniffer导入为Sniffer
这确实导入了我认为的文件。当我运行它时，它会在启动时追踪错误：
Traceback (most recent call last):
File "snifferLaunch.py", line 19, in <module>
import flashy.sniffer.Sniffer
File "/Users/tai/Desktop/FlashY/flashy/sniffer/__init__.py", line 110, in <module>
File "/Users/tai/Desktop/FlashY/flashy/sniffer/__init__.py", line 107, in forInFile
File "/Users/tai/Desktop/FlashY/flashy/sniffer/__init__.py", line 98, in runFlashY
File "/Users/tai/Desktop/FlashY/flashy/sniffer/__init__.py", line 89, in db
AttributeError: 'module' object has no attribute 'getDecompiledFiles'
这通常会让我去寻找一个getDecompiledFiles函数。问题是代码中没有getDecompiledFiles。有一个get_Decompiled_Files函数。
我的代码看起来像这样（删除了非必要部分）。你看到我的错了吗？我搜索了整个项目，无法在任何地方找到getDecompiledFiles函数。我不知道为什么会有这个属性......
snifferLaunch：
import flashy.sniffer.Sniffer as Sniffer
import flashy.sniffer.database as database
import flashy.sniffer.cleaner as cleaner
def open_websites(line):
#opens a list of websites from local file "urlIn.txt" and runs the Sniffer on them.
#It retrieves the swfs from each url and storing them in the local out/"the modified url" /"hashed swf.swf" and the file contains the decompiled swf
print( "opening websites")
newSwfFiles = [];
# reads in all of the lines in urlIn.txt
#for line in urlsToRead:
if line[0] !="#":
newLine = cleaner.remove_front(line);
# note the line[:9] is done to avoid the http// which creates an additional file to go into. The remaining part of the url is still unique.
outFileDirectory = decSwfsFolder + "/" + newLine
cleaner.check_or_create_dir(outFileDirectory)
try:
newSwfFiles = Sniffer.open_url(line, []);
except:
print " Sniffer.openURL failed"
pass
# for all of the files there it runs jpex on them. (in the future this will delete the file after jpex runs so we don't run jpex more than necessary)
for location in newSwfFiles:
cleaner.check_or_create_dir(outFileDirectory + "/" + location)
#creates the command for jpex flash decompiler, the command + file to save into + location of the swf to decompile
newCommand = javaCommand + "/" + newLine + "/" + location +"/ " + swfLoc +"/"+ location
os.system(newCommand)
print ("+++this is the command: " + newCommand+"\n")
# move the swf into a new swf file for db storage
oldLocation = swfFolder + location;
newLocation = decSwfsFolder + "/" + newLine + "/" + location + "/" + "theSwf"+ "/"
cleaner.check_or_create_dir(newLocation )
if(os.path.exists(oldLocation)):
# if the file already exists at that location do not move it simply delete it (the) duplicate
if(os.path.exists(newLocation +"/"+ location)):
os.remove(oldLocation)
else:
shutil.move(swfFolder + location, newLocation)
if cleanup:
cleaner.cleanSwf();
# newSwfFiles has the directory file location of each new added file: "directory/fileHash.swf"
def db():
database.get_decompiled_files()
def run_flashY(line):
#Run FlashY a program that decompiles all of the swfs found at urls defined in urlIn.txt.
#Each decompiled file will be stored in the PaperG Amazon S3 bucket: decompiled_swfs.
#run the program for each line
#open all of the websites in the url file urlIn.txt
open_websites(line)
#store the decompiled swfs in the database
db()
#remove all files from local storage
cleaner.clean_out()
#kill all instances of firefox
def for_in_file():
#run sniffer for each line in the file
#for each url, run then kill firefox to prevent firefox buildup
for line in urlsToRead:
run_flashY(line)
cleaner.kill_firefox()
#Main Functionality
if __name__ == '__main__':
#initialize and run the program on launch
for_in_file()
嗅探器文件：
import urllib2
from urllib2 import Request, urlopen, URLError, HTTPError
import shutil
import sys
import re
import os
import hashlib
import time
import datetime
from selenium import webdriver
import glob
import thread
import httplib
from collections import defaultdict
import cleaner
a=[];
b=[];
newSwfFiles=[];
theURL='';
curPath = os.path.dirname(os.path.realpath(__file__))
#firebug gets all network data
fireBugPath = curPath +'/firebug-1.12.8b1.xpi';
#netExport exports firebug's http archive (network req/res) in the form of a har file
netExportPath = curPath +'/netExport.xpi';
harLoc = curPath +"/har/";
swfLoc = curPath +"/swfs";
cleanThis=True
#remove har file(s) after reading them out to gather swf files
profile = webdriver.firefox.firefox_profile.FirefoxProfile();
profile.add_extension( fireBugPath);
profile.add_extension(netExportPath);
hashLib = hashlib.md5()
#firefox preferences
profile.set_preference("app.update.enabled", False)
profile.native_events_enabled = True
profile.set_preference("webdriver.log.file", curPath +"webFile.txt")
profile.set_preference("extensions.firebug.DBG_STARTER", True);
profile.set_preference("extensions.firebug.currentVersion", "1.12.8");
profile.set_preference("extensions.firebug.addonBarOpened", True);
profile.set_preference('extensions.firebug.consoles.enableSite', True)
profile.set_preference("extensions.firebug.console.enableSites", True);
profile.set_preference("extensions.firebug.script.enableSites", True);
profile.set_preference("extensions.firebug.net.enableSites", True);
profile.set_preference("extensions.firebug.previousPlacement", 1);
profile.set_preference("extensions.firebug.allPagesActivation", "on");
profile.set_preference("extensions.firebug.onByDefault", True);
profile.set_preference("extensions.firebug.defaultPanelName", "net");
#set net export preferences
profile.set_preference("extensions.firebug.netexport.alwaysEnableAutoExport", True);
profile.set_preference("extensions.firebug.netexport.autoExportToFile", True);
profile.set_preference("extensions.firebug.netexport.saveFiles", True);
profile.set_preference("extensions.firebug.netexport.autoExportToServer", False);
profile.set_preference("extensions.firebug.netexport.Automation", True);
profile.set_preference("extensions.firebug.netexport.showPreview", False);
profile.set_preference("extensions.firebug.netexport.pageLoadedTimeout", 15000);
profile.set_preference("extensions.firebug.netexport.timeout", 10000);
profile.set_preference("extensions.firebug.netexport.defaultLogDir",harLoc);
profile.update_preferences();
browser = webdriver.Firefox(firefox_profile=profile);
def open_url(url,s):
#open each url, find all of the har files with them and get those files.
theURL = url;
time.sleep(6);
#browser = webdriver.Chrome();
browser.get(url); #load the url in firefox
browser.set_page_load_timeout(30)
time.sleep(3); #wait for the page to load
browser.execute_script("window.scrollTo(0, document.body.scrollHeight/5);")
time.sleep(1); #wait for the page to load
browser.execute_script("window.scrollTo(0, document.body.scrollHeight/4);")
time.sleep(1); #wait for the page to load
browser.execute_script("window.scrollTo(0, document.body.scrollHeight/3);")
time.sleep(1); #wait for the page to load
browser.execute_script("window.scrollTo(0, document.body.scrollHeight/2);")
time.sleep(1); #wait for the page to load
browser.execute_script("window.scrollTo(0, document.body.scrollHeight);")
searchText='';
time.sleep(20); #wait for the page to load
# print(browser.page_source);
#close the browser and get all the swfs from the created har file.
#uses the a & b arrays to find the swf files from generated har files
get_swfs_from_har()
#clean out the slashes
clean_f_slashes()
#get all files
get_all_files()
#ensure that some files were gained
assert a != []
assert b != []
assert newSwfFiles != []
#if the files (har, swf, out) should be cleaned out do so. This can be toggled for dubugging
if(cleanThis):
cleaner.clean_har()
return newSwfFiles;
def remove_non_url(t):
#remove matched urls that are not actually urls
a=[];
for b in t:
if(b.lower()[:4] !="http" and b.lower()[:4] != "www." ):
if(b[:2] == "//" and b.__len__() >10):
a.append(theURL+"/"+b[2:]);
else:
while((b.lower()[:4] !="http" or b.lower()[:4] !="www." or b.lower()[:1] !="//") and b.__len__() >10):
b=b[1:b.__len__()];
if( b.__len__() >10):
if(b[:1] == "//" ):
if not b in a:
a.append(theURL+b[2:b.__len__()]);
else:
if not b in a:
a.append(b);
else:
if not b in a:
a.append(b);
return a;
def get_swfs_from_har():
#validate that the files in the har are actual swf files
files = [f for f in os.listdir(harLoc) if re.match((theURL[7:]+ '.*.har'), f)]
for n in files:
with open (harLoc + n , "r") as theF:
textt = theF.read();
swfObjects= re.findall('\{[^\{]*(?:http:\/\/|https:\/\/|www\.|\/\/)[^}]*\.swf[^}]+', textt.lower())
#swfObjects = "".join(str(i) for i in swfObjects)
for obj in swfObjects:
l=[]
otherL=[]
links = re.findall('(?:http:\/\/|https:\/\/|www\.|\/\/)[^"]+', obj)
for url in links:
url=url[:url.__len__()-1]
ending = url[url.__len__()-6:];
if ".swf" in ending:
l.append(url);
elif "." not in ending:
otherL.append(url);
for c in l:
if not c in a and c.__len__() >20:
a.append(c);
if(otherL.__len__()>0):
theMostLikelyLink=otherL[0];
b.append(theMostLikelyLink);
##adds the 1st link after the swf
otherL.remove(theMostLikelyLink);
else:
b.append(None);
def clean_f_slashes():
#remove unrelated characters from swfs
for x in a:
newS='';
if(',' in x or ';' in x or '\\' in x):
for d in x:
if(d != '\\' and d != ',' and d != ';'):
newS+=d;
else:
newS=x;
if "http" not in newS.lower():
if "www" in newS:
newS= "http://" + newS;
else:
newS = "http://www."+newS
while(newS[:3]!="htt"):
newS=newS[1:];
a.remove(x);
if(newS.__len__() >15):
a.append(newS);
def get_all_files():
#get all of the files from the array of valid swfs
os.chdir(swfLoc);
for openUrl in a:
place = a.index(openUrl);
try:
req = Request(openUrl)
response = urlopen(req)
fData = urllib2.urlopen(openUrl)
iText = fData.read()
#get the hex hash of the file
hashLib.update(iText);
hashV =hashLib.hexdigest()+".swf";
outUrl= get_redirected_url(b[place]);
#check if file already exists, if it does do not add a duplicate
theFile = [f for f in os.listdir(swfLoc) if re.match((hashV), f)]
if hashV not in theFile:
lFile = open(outUrl+"," +hashV, "w")
lFile.write(iText)
lFile.close();
#except and then ignore are invalid urls.
except:
pass
#Remove all files less than 8kb, anything less than this size is unlikely to be an advertisement. Most flash ads seen so far are 25kb or larger
sFiles = [f for f in os.listdir(swfLoc)]
for filenames in sFiles:
sizeF = os.path.getsize(filenames);
#if the file is smaller remove it
if(sizeF<8000):
cleaner.remove_file(filenames)
else:
newSwfFiles.append(filenames);
def x_str(s):
#check if a unicode expression exists and convert it to a string
if s is None:
return ''
return str(s)
def get_redirected_url(s):
#get the url that another url will redirect to
if s is None:
return "";
if ".macromedia" in s:
return ""
browser.get(s);
time.sleep(20);
theredirectedurl=cleaner.removeFront(browser.current_url);
aUrl= re.findall("[^/]+",theredirectedurl)[0].encode('ascii','ignore')
return aUrl;
答案 0 :(得分：0)
有趣......所以我实际上意识到我错了。
我仍然不知道为什么它会期望一个不存在的功能，但我确实有猜测。
我已将__init__.py文件拉出来用作snifferLaunch文件。这是由于我最初对__init__.py的误解，并假设它与其他语言的主要内容相似。
我相信__init__.pyc文件中包含一个过时的旧函数。基本上我相信有一个文件永远不应该运行，它已经过时，并以某种方式被调用。它是唯一存在该函数的文件，我忽略了它，因为我认为不应该调用它。
解决方案如下，错误是由于我误用了__init__。
我更改了我的导入语句：
from flashy.sniffer import Sniffer
import flashy.sniffer.database as database
import flashy.sniffer.cleaner as cleaner
我在__init__.py中创建了新的空白__init__.pyc和flashy/sniffer/个文件。
这可以防止getDecompiledFiles的错误期望，并且还允许运行代码。我收到“无法找到此文件”错误，因为它未正确识别为模块。如果有人能够解释那里发生了什么，那么有关此的更多信息将不胜感激。我以为你可以在没有init语句的情况下运行python文件，但是当嵌套在其他文件夹中时，它似乎必须作为python模块打开。
我的文件结构现在看起来像这样：
Main
-snifferLaunch.py //with changed import statements
-flashy
--sniffer
---Sniffer.py
---database.py
---__init__.py //blank
---__init__.pyc // blank
它似乎是python与其他语言问题。还有其他人经历过这个吗？