Related
I'm trying to get the list of files that are fully uploaded on the FTP server.
I have access to this FTP server where a 3rd party writes data and marker files every 15 minutes. Once the data file is completely uploaded then a marker file gets created. we know once this marker file is there that means data files are ready and we can download it. I'm looking for a way to efficiently approach this problem. I want to check every minute if there are any new stable files on FTP server, if there is then I'll download those files. one preferred way is see if the marker file is 2 minutes old then we are good to download marker file and corresponding data file.
I'm new with python and looking for help.
I have some code till I list out the files
import paramiko
from datetime import datetime, timedelta
FTP_HOST = 'host_address'
FTP_PORT = 21
FTP_USERNAME = 'username'
FTP_PASSWORD = 'password'
FTP_ROOT_PATH = 'path_to_dir'
def today():
return datetime.strftime(datetime.now(), '%Y%m%d')
def open_ftp_connection(ftp_host, ftp_port, ftp_username, ftp_password):
"""
Opens ftp connection and returns connection object
"""
client = paramiko.SSHClient()
client.load_system_host_keys()
try:
transport = paramiko.Transport(ftp_host, ftp_port)
except Exception as e:
return 'conn_error'
try:
transport.connect(username=ftp_username, password=ftp_password)
except Exception as identifier:
return 'auth_error'
ftp_connection = paramiko.SFTPClient.from_transport(transport)
return ftp_connection
def show_ftp_files_stat():
ftp_connection = open_ftp_connection(FTP_HOST, int(FTP_PORT), FTP_USERNAME, FTP_PASSWORD)
full_ftp_path = FTP_ROOT_PATH + "/" + today()
file_attr_list = ftp_connection.listdir_attr(full_ftp_path)
print(file_attr_list)
for file_attr in file_attr_list:
print(file_attr.filename, file_attr.st_size, file_attr.st_mtime)
if __name__ == '__main__':
show_ftp_files_stat()
Sample file name
org-reference-delta-quotes.REF.48C2.20200402.92.1.1.txt.gz
Sample corresponding marker file name
org-reference-delta-quotes.REF.48C2.20200402.92.note.txt.gz
I solved my use case with 2 min stable rule, if modified time is within 2 min of the current time, I consider them stable.
import logging
import time
from datetime import datetime, timezone
from ftplib import FTP
FTP_HOST = 'host_address'
FTP_PORT = 21
FTP_USERNAME = 'username'
FTP_PASSWORD = 'password'
FTP_ROOT_PATH = 'path_to_dir'
logger = logging.getLogger()
logger.setLevel(logging.ERROR)
def today():
return datetime.strftime(datetime.now(tz=timezone.utc), '%Y%m%d')
def current_utc_ts():
return datetime.utcnow().timestamp()
def current_utc_ts_minus_120():
return int(datetime.utcnow().timestamp()) - 120
def yyyymmddhhmmss_string_epoch_ts(dt_string):
return time.mktime(time.strptime(dt_string, '%Y%m%d%H%M%S'))
def get_ftp_connection(ftp_host, ftp_username, ftp_password):
try:
ftp = FTP(ftp_host, ftp_username, ftp_password)
except Exception as e:
print(e)
logger.error(e)
return 'conn_error'
return ftp
def get_list_of_files(ftp_connection, date_to_process):
full_ftp_path = FTP_ROOT_PATH + "/" + date_to_process + "/"
ftp_connection.cwd(full_ftp_path)
entries = list(ftp_connection.mlsd())
entry_list = [line for line in entries if line[0].endswith('.gz') | line[0].endswith('.zip')]
ftp_connection.quit()
print('Total file count', len(entry_list))
return entry_list
def parse_file_list_to_dict(entries):
try:
file_dict_list = []
for line in entries:
file_dict = dict({"file_name": line[0],
"server_timestamp": int(yyyymmddhhmmss_string_epoch_ts(line[1]['modify'])),
"server_date": line[0].split(".")[3])
file_dict_list.append(file_dict)
except IndexError as e:
# Output expected IndexErrors.
logging.exception(e)
except Exception as exception:
# Output unexpected Exceptions.
logging.exception(exception, False)
return file_dict_list
def get_stable_files_dict_list(dict_list):
stable_list = list(filter(lambda d: d['server_timestamp'] < current_utc_ts_minus_120(), dict_list))
print('stable file count: {}'.format(len(stable_list)))
return stable_list
if __name__ == '__main__':
ftp_connection = get_ftp_connection(FTP_HOST, FTP_USERNAME, FTP_PASSWORD)
if ftp_connection == 'conn_error':
logger.error('Failed to connect FTP Server!')
else:
file_list = get_list_of_files(ftp_connection, today())
parse_file_list = parse_file_list_to_dict(file_list)
stable_file_list = get_stable_files_dict_list(parse_file_list)
I am trying to develop a script which sends an email about checking ping regularly at one hour interval of time. I am using Python to program this script and I cannot create a log file to keep the ping logs which I need to mail. I'm new to using subprocess module and its functions.
import threading
import os
def check_ping():
threading.Timer(5.0, check_ping).start()
hostname = "www.google.com"
response = os.system("ping -c 4 " + hostname)
'''
def trace_route():
threading.Timer(5.0, trace_route).start()
hostname = "www.google.com"
response = os.system("traceroute" + hostname)
'''
check_ping()
#trace_route()
output = check_ping()
file = open("sample.txt","a")
file.write(output)
file.close()
import os, platform
import threading
def check_ping():
threading.Timer(10.0,check_ping).start()
hostname = "www.google.com"
response = os.system("ping " + ("-n 1 " if platform.system().lower()=="windows" else "-c 1 ") + hostname)
# and then check the response...
if response == 0:
pingstatus = "Network Active"
else:
pingstatus = "Network Error"
return pingstatus
pingstatus = check_ping()
This is what I came up with:
using subprocess instead of os.system
added timeout of 8 seconds
writing to csv file instead of txt file
added timestamps to csv file, without which I don't really see the point of logging in the first place
import os
import threading
import time
from subprocess import Popen, PIPE
def check_ping():
threading.Timer(10.0,check_ping).start()
# Get current time
timestamp = int(time.time())
# Build the command
hostname = "www.google.com"
if os.name == 'nt':
command = ['ping', '-n', '1', hostname]
else:
command = ['ping', '-c', '1', hostname]
# Create process
pingProcess = Popen(command, stdout=PIPE, stderr=PIPE)
try:
# Timeout 8 seconds, to avoid overlap with the next ping command
outs, errs = pingProcess.communicate(timeout=8)
except TimeoutExpired:
# If timed out, kill
pingProcess.kill()
outs, errs = pingProcess.communicate()
# Get the return code of the process
response = pingProcess.returncode
# and then check the response...
# These four lines can be removed, they are just to see if the system
# works.
if response == 0:
print("Network Active")
else:
print("Network Error")
# You most likely want a CSV file, as most programs accept this file type,
# including Microsoft Excel and LibreOffice Calc
# Further, I'm sure you want timestamps with the results.
file = open("ping.csv","a")
file.write(str(timestamp) + "," + str(response) + "\n")
file.close()
check_ping()
Here is another version without using the system's ping command, but instead using a python library for pinging. This ensures that the code works on all operating systems:
import threading
import time
from ping3 import ping
def check_ping():
threading.Timer(10.0,check_ping).start()
# Get current time
timestamp = int(time.time())
# Build the command
hostname = "www.google.com"
# Run ping
ping_result = ping(hostname, timeout=8)
ping_success = False if ping_result is None else True
# and then check the response...
# These four lines can be removed, they are just to see if the system
# works.
if ping_success:
print("Network Active (" + str(ping_result) + ")")
else:
print("Network Error")
# You most likely want a CSV file, as most programs accept this file type,
# including Microsoft Excel and LibreOffice Calc
# Further, I'm sure you want timestamps with the results.
file = open("ping.csv", "a")
ping_value_str = str(ping_result) if ping_success else "NaN"
file.write(str(timestamp) + "," + ("0" if ping_success else "1") + "," + ping_value_str + "\n")
file.close()
check_ping()
The following script is an extract from
https://github.com/RittmanMead/obi-metrics-agent/blob/master/obi-metrics-agent.py
The script is written in jython & it hits the weblogic admin console to extract metrics
The problem is it runs only once and does not loop infinitely
Here's the script that I've extracted from the original for my purpose:
import calendar, time
import sys
import getopt
print '---------------------------------------'
# Check the arguments to this script are as expected.
# argv[0] is script name.
argLen = len(sys.argv)
if argLen -1 < 2:
print "ERROR: got ", argLen -1, " args, must be at least two."
print '$FMW_HOME/oracle_common/common/bin/wlst.sh obi-metrics-agent.py <AdminUserName> <AdminPassword> [<AdminServer_t3_url>] [<Carbon|InfluxDB>] [<target host>] [<target port>] [targetDB influx db>'
exit()
outputFormat='CSV'
url='t3://localhost:7001'
targetHost='localhost'
targetDB='obi'
targetPort='8086'
try:
wls_user = sys.argv[1]
wls_pw = sys.argv[2]
url = sys.argv[3]
outputFormat=sys.argv[4]
targetHost=sys.argv[5]
targetPort=sys.argv[6]
targetDB=sys.argv[7]
except:
print ''
print wls_user, wls_pw,url, outputFormat,targetHost,targetPort,targetDB
now_epoch = calendar.timegm(time.gmtime())*1000
if outputFormat=='InfluxDB':
import httplib
influx_msgs=''
connect(wls_user,wls_pw,url)
results = displayMetricTables('Oracle_BI*','dms_cProcessInfo')
while True:
for table in results:
tableName = table.get('Table')
rows = table.get('Rows')
rowCollection = rows.values()
iter = rowCollection.iterator()
while iter.hasNext():
row = iter.next()
rowType = row.getCompositeType()
keys = rowType.keySet()
keyIter = keys.iterator()
inst_name= row.get('Name').replace(' ','-')
try:
server= row.get('Servername').replace(' ','-').replace('/','_')
except:
try:
server= row.get('ServerName').replace(' ','-').replace('/','_')
except:
server='unknown'
try:
host= row.get('Host').replace(' ','-')
except:
host=''
while keyIter.hasNext():
columnName = keyIter.next()
value = row.get(columnName )
if columnName.find('.value')>0:
metric_name=columnName.replace('.value','')
if value is not None:
if outputFormat=='InfluxDB':
influx_msg= ('%s,server=%s,host=%s,metric_group=%s,metric_instance=%s value=%s %s') % (metric_name,server,host,tableName,inst_name, value,now_epoch*1000000)
influx_msgs+='\n%s' % influx_msg
conn = httplib.HTTPConnection('%s:%s' % (targetHost,targetPort))
## TODO pretty sure should be urlencoding this ...
a=conn.request("POST", ("/write?db=%s" % targetDB), influx_msg)
r=conn.getresponse()
if r.status != 204:
print 'Failed to send to InfluxDB! Error %s Reason %s' % (r.status,r.reason)
print influx_msg
#sys.exit(2)
else:
print 'Skipping None value %s,server=%s,host=%s,metric_group=%s,metric_instance=%s value=%s %s' % (metric_name,server,host,tableName,inst_name, value,now_epoch*1000000)
I've tried to use the While loop, but that just stopped the code from exiting and not re-looping
What I want to achieve is to loop it infinitely post connection to weblogic
i.e. after this line
connect(wls_user,wls_pw,url)
and perhaps sleep for 5 seconds before re-running
Any and all help will be appreciated
Thanks
P
You can use this kind of condition for the loop :
mainLoop = 'true'
while mainLoop == 'true' :
and this for the pause between iterations :
java.lang.Thread.sleep(3 * 1000)
I am trying to SSH into a bunch of routers and grep some info from each. Some of the router IPs might not be valid as they are got from the loop. For the ones that do not exist, i am getting a socket.timeout error (obviously) and the script exits execution at that point. I want python to ignore this error and continue with the rest of the iterations. I have included the except/pass/continue commands, but they somehow do not seem to be doing their job and i am still getting the socket.timeout error and exiting the script :
import paramiko
import socket
import subprocess
import time
site = ['sao2','mia2','par2','scs2','sin4','sjl1','syd3','tok2','tor1','wdc1','akl1','chc1','chi2','cop2','dal1','fra2']
csr_hostname = 'csr_1'
csr_pass = str('*******')
username = str('test_user')
csrcmd = 'sh ver \n'
f = open ("version_output.txt", 'r+')
#*********************************************************************
for DC in site :
y = str(DC)
mcp = socket.gethostbyname(y)
mcpip = mcp.split('.')
mcpip2 = int(mcpip[0])
mcpip3 = int(mcpip[1])
mcpip4 = int(mcpip[2])
print (DC + ":")
f.write('\n')
f.write(DC)
f.write('\n=====\n')
for i in range(5,33):
x=37+(2*i)
mcpcsr = (str(mcpip2) + ('.') + str(mcpip3)+ ('.') + str(mcpip4)+ ('.') + str(x))
fqdn1=(str(i)+'-01.'+str(DC))
print i
ssh = paramiko.SSHClient()
ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
ssh.connect(mcpcsr, username=username, password=csr_pass, look_for_keys = False, allow_agent = False, timeout=10)
router_channel = ssh.invoke_shell()
router_channel.settimeout(2)
try :
buff = ''
router_channel.send(csrcmd)
time.sleep(1)
buff = router_channel.recv(99999)
for item in buff.split("\n"):
if "Cisco IOS XE Software, Version" in item :
csrver = (fqdn1 + " : " + item.strip())
if "Cisco IOS Software" in item :
csrver1 = (item.strip())
print csrver
print csrver1
f.write(csrver)
f.write('\n')
f.write(csrver1)
f.write('\n')
fqdn1 = ''
ssh.close()
#*********************************************************************
except socket.timeout:
print ("bla bla")
pass
continue
except socket.error :
print ("bla bla")
pass
continue
except paramiko.SSHException, socket.error :
print ("bla bla")
pass
continue
except Exception :
print ("bla bla")
pass
continue
#*********************************************************************
Output :
Traceback (most recent call last):
File "csrnum.py", line 36, in ?
ssh.connect(mcpcsr, username=username, password=csr_pass, look_for_keys = False, allow_agent = False, timeout=10)
File "/usr/lib/python2.4/site-packages/paramiko/client.py", line 291, in connect
sock.connect(addr)
File "<string>", line 1, in connect
socket.timeout: timed out
Any insight into this will be much appreciated and grateful for!
Figured it out! The exceptions were set one indent more than the one where the ssh connections were attempted. Made a small change to the code to accomodate this and the exceptions are working as expected!
I'm attempting to debug a Subversion post-commit hook that calls some python scripts. What I've been able to determine so far is that when I run post-commit.bat manually (I've created a wrapper for it to make it easier) everything succeeds, but when SVN runs it one particular step doesn't work.
We're using CollabNet SVNServe, which I know from the documentation removes all environment variables. This had caused some problems earlier, but shouldn't be an issue now.
Before Subversion calls a hook script, it removes all variables - including $PATH on Unix, and %PATH% on Windows - from the environment. Therefore, your script can only run another program if you spell out that program's absolute name.
The relevant portion of post-commit.bat is:
echo -------------------------- >> c:\svn-repos\company\hooks\svn2ftp.out.log
set SITENAME=staging
set SVNPATH=branches/staging/wwwroot/
"C:\Python3\python.exe" C:\svn-repos\company\hooks\svn2ftp.py ^
--svnUser="svnusername" ^
--svnPass="svnpassword" ^
--ftp-user=ftpuser ^
--ftp-password=ftppassword ^
--ftp-remote-dir=/ ^
--access-url=svn://10.0.100.6/company ^
--status-file="C:\svn-repos\company\hooks\svn2ftp-%SITENAME%.dat" ^
--project-directory=%SVNPATH% "staging.company.com" %1 %2 >> c:\svn-repos\company\hooks\svn2ftp.out.log
echo -------------------------- >> c:\svn-repos\company\hooks\svn2ftp.out.log
When I run post-commit.bat manually, for example: post-commit c:\svn-repos\company 12345, I see output like the following in svn2ftp.out.log:
--------------------------
args1: c:\svn-repos\company
args0: staging.company.com
abspath: c:\svn-repos\company
project_dir: branches/staging/wwwroot/
local_repos_path: c:\svn-repos\company
getting youngest revision...
done, up-to-date
--------------------------
However, when I commit something to the repo and it runs automatically, the output is:
--------------------------
--------------------------
svn2ftp.py is a bit long, so I apologize but here goes. I'll have some notes/disclaimers about its contents below it.
#!/usr/bin/env python
"""Usage: svn2ftp.py [OPTION...] FTP-HOST REPOS-PATH
Upload to FTP-HOST changes committed to the Subversion repository at
REPOS-PATH. Uses svn diff --summarize to only propagate the changed files
Options:
-?, --help Show this help message.
-u, --ftp-user=USER The username for the FTP server. Default: 'anonymous'
-p, --ftp-password=P The password for the FTP server. Default: '#'
-P, --ftp-port=X Port number for the FTP server. Default: 21
-r, --ftp-remote-dir=DIR The remote directory that is expected to resemble the
repository project directory
-a, --access-url=URL This is the URL that should be used when trying to SVN
export files so that they can be uploaded to the FTP
server
-s, --status-file=PATH Required. This script needs to store the last
successful revision that was transferred to the
server. PATH is the location of this file.
-d, --project-directory=DIR If the project you are interested in sending to
the FTP server is not under the root of the
repository (/), set this parameter.
Example: -d 'project1/trunk/'
This should NOT start with a '/'.
2008.5.2 CKS
Fixed possible Windows-related bug with tempfile, where the script didn't have
permission to write to the tempfile. Replaced this with a open()-created file
created in the CWD.
2008.5.13 CKS
Added error logging. Added exception for file-not-found errors when deleting files.
2008.5.14 CKS
Change file open to 'rb' mode, to prevent Python's universal newline support from
stripping CR characters, causing later comparisons between FTP and SVN to report changes.
"""
try:
import sys, os
import logging
logging.basicConfig(
level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(message)s',
filename='svn2ftp.debug.log',
filemode='a'
)
console = logging.StreamHandler()
console.setLevel(logging.ERROR)
logging.getLogger('').addHandler(console)
import getopt, tempfile, smtplib, traceback, subprocess
from io import StringIO
import pysvn
import ftplib
import inspect
except Exception as e:
logging.error(e)
#capture the location of the error
frame = inspect.currentframe()
stack_trace = traceback.format_stack(frame)
logging.debug(stack_trace)
print(stack_trace)
#end capture
sys.exit(1)
#defaults
host = ""
user = "anonymous"
password = "#"
port = 21
repo_path = ""
local_repos_path = ""
status_file = ""
project_directory = ""
remote_base_directory = ""
toAddrs = "developers#company.com"
youngest_revision = ""
def email(toAddrs, message, subject, fromAddr='autonote#company.com'):
headers = "From: %s\r\nTo: %s\r\nSubject: %s\r\n\r\n" % (fromAddr, toAddrs, subject)
message = headers + message
logging.info('sending email to %s...' % toAddrs)
server = smtplib.SMTP('smtp.company.com')
server.set_debuglevel(1)
server.sendmail(fromAddr, toAddrs, message)
server.quit()
logging.info('email sent')
def captureErrorMessage(e):
sout = StringIO()
traceback.print_exc(file=sout)
errorMessage = '\n'+('*'*80)+('\n%s'%e)+('\n%s\n'%sout.getvalue())+('*'*80)
return errorMessage
def usage_and_exit(errmsg):
"""Print a usage message, plus an ERRMSG (if provided), then exit.
If ERRMSG is provided, the usage message is printed to stderr and
the script exits with a non-zero error code. Otherwise, the usage
message goes to stdout, and the script exits with a zero
errorcode."""
if errmsg is None:
stream = sys.stdout
else:
stream = sys.stderr
print(__doc__, file=stream)
if errmsg:
print("\nError: %s" % (errmsg), file=stream)
sys.exit(2)
sys.exit(0)
def read_args():
global host
global user
global password
global port
global repo_path
global local_repos_path
global status_file
global project_directory
global remote_base_directory
global youngest_revision
try:
opts, args = getopt.gnu_getopt(sys.argv[1:], "?u:p:P:r:a:s:d:SU:SP:",
["help",
"ftp-user=",
"ftp-password=",
"ftp-port=",
"ftp-remote-dir=",
"access-url=",
"status-file=",
"project-directory=",
"svnUser=",
"svnPass="
])
except getopt.GetoptError as msg:
usage_and_exit(msg)
for opt, arg in opts:
if opt in ("-?", "--help"):
usage_and_exit()
elif opt in ("-u", "--ftp-user"):
user = arg
elif opt in ("-p", "--ftp-password"):
password = arg
elif opt in ("-SU", "--svnUser"):
svnUser = arg
elif opt in ("-SP", "--svnPass"):
svnPass = arg
elif opt in ("-P", "--ftp-port"):
try:
port = int(arg)
except ValueError as msg:
usage_and_exit("Invalid value '%s' for --ftp-port." % (arg))
if port < 1 or port > 65535:
usage_and_exit("Value for --ftp-port must be a positive integer less than 65536.")
elif opt in ("-r", "--ftp-remote-dir"):
remote_base_directory = arg
elif opt in ("-a", "--access-url"):
repo_path = arg
elif opt in ("-s", "--status-file"):
status_file = os.path.abspath(arg)
elif opt in ("-d", "--project-directory"):
project_directory = arg
if len(args) != 3:
print(str(args))
usage_and_exit("host and/or local_repos_path not specified (" + len(args) + ")")
host = args[0]
print("args1: " + args[1])
print("args0: " + args[0])
print("abspath: " + os.path.abspath(args[1]))
local_repos_path = os.path.abspath(args[1])
print('project_dir:',project_directory)
youngest_revision = int(args[2])
if status_file == "" : usage_and_exit("No status file specified")
def main():
global host
global user
global password
global port
global repo_path
global local_repos_path
global status_file
global project_directory
global remote_base_directory
global youngest_revision
read_args()
#repository,fs_ptr
#get youngest revision
print("local_repos_path: " + local_repos_path)
print('getting youngest revision...')
#youngest_revision = fs.youngest_rev(fs_ptr)
assert youngest_revision, "Unable to lookup youngest revision."
last_sent_revision = get_last_revision()
if youngest_revision == last_sent_revision:
# no need to continue. we should be up to date.
print('done, up-to-date')
return
if last_sent_revision or youngest_revision < 10:
# Only compare revisions if the DAT file contains a valid
# revision number. Otherwise we risk waiting forever while
# we parse and uploading every revision in the repo in the case
# where a repository is retroactively configured to sync with ftp.
pysvn_client = pysvn.Client()
pysvn_client.callback_get_login = get_login
rev1 = pysvn.Revision(pysvn.opt_revision_kind.number, last_sent_revision)
rev2 = pysvn.Revision(pysvn.opt_revision_kind.number, youngest_revision)
summary = pysvn_client.diff_summarize(repo_path, rev1, repo_path, rev2, True, False)
print('summary len:',len(summary))
if len(summary) > 0 :
print('connecting to %s...' % host)
ftp = FTPClient(host, user, password)
print('connected to %s' % host)
ftp.base_path = remote_base_directory
print('set remote base directory to %s' % remote_base_directory)
#iterate through all the differences between revisions
for change in summary :
#determine whether the path of the change is relevant to the path that is being sent, and modify the path as appropriate.
print('change path:',change.path)
ftp_relative_path = apply_basedir(change.path)
print('ftp rel path:',ftp_relative_path)
#only try to sync path if the path is in our project_directory
if ftp_relative_path != "" :
is_file = (change.node_kind == pysvn.node_kind.file)
if str(change.summarize_kind) == "delete" :
print("deleting: " + ftp_relative_path)
try:
ftp.delete_path("/" + ftp_relative_path, is_file)
except ftplib.error_perm as e:
if 'cannot find the' in str(e) or 'not found' in str(e):
# Log, but otherwise ignore path-not-found errors
# when deleting, since it's not a disaster if the file
# we want to delete is already gone.
logging.error(captureErrorMessage(e))
else:
raise
elif str(change.summarize_kind) == "added" or str(change.summarize_kind) == "modified" :
local_file = ""
if is_file :
local_file = svn_export_temp(pysvn_client, repo_path, rev2, change.path)
print("uploading file: " + ftp_relative_path)
ftp.upload_path("/" + ftp_relative_path, is_file, local_file)
if is_file :
os.remove(local_file)
elif str(change.summarize_kind) == "normal" :
print("skipping 'normal' element: " + ftp_relative_path)
else :
raise str("Unknown change summarize kind: " + str(change.summarize_kind) + ", path: " + ftp_relative_path)
ftp.close()
#write back the last revision that was synced
print("writing last revision: " + str(youngest_revision))
set_last_revision(youngest_revision) # todo: undo
def get_login(a,b,c,d):
#arguments don't matter, we're always going to return the same thing
try:
return True, "svnUsername", "svnPassword", True
except Exception as e:
logging.error(e)
#capture the location of the error
frame = inspect.currentframe()
stack_trace = traceback.format_stack(frame)
logging.debug(stack_trace)
#end capture
sys.exit(1)
#functions for persisting the last successfully synced revision
def get_last_revision():
if os.path.isfile(status_file) :
f=open(status_file, 'r')
line = f.readline()
f.close()
try: i = int(line)
except ValueError:
i = 0
else:
i = 0
f = open(status_file, 'w')
f.write(str(i))
f.close()
return i
def set_last_revision(rev) :
f = open(status_file, 'w')
f.write(str(rev))
f.close()
#augmented ftp client class that can work off a base directory
class FTPClient(ftplib.FTP) :
def __init__(self, host, username, password) :
self.base_path = ""
self.current_path = ""
ftplib.FTP.__init__(self, host, username, password)
def cwd(self, path) :
debug_path = path
if self.current_path == "" :
self.current_path = self.pwd()
print("pwd: " + self.current_path)
if not os.path.isabs(path) :
debug_path = self.base_path + "<" + path
path = os.path.join(self.current_path, path)
elif self.base_path != "" :
debug_path = self.base_path + ">" + path.lstrip("/")
path = os.path.join(self.base_path, path.lstrip("/"))
path = os.path.normpath(path)
#by this point the path should be absolute.
if path != self.current_path :
print("change from " + self.current_path + " to " + debug_path)
ftplib.FTP.cwd(self, path)
self.current_path = path
else :
print("staying put : " + self.current_path)
def cd_or_create(self, path) :
assert os.path.isabs(path), "absolute path expected (" + path + ")"
try: self.cwd(path)
except ftplib.error_perm as e:
for folder in path.split('/'):
if folder == "" :
self.cwd("/")
continue
try: self.cwd(folder)
except:
print("mkd: (" + path + "):" + folder)
self.mkd(folder)
self.cwd(folder)
def upload_path(self, path, is_file, local_path) :
if is_file:
(path, filename) = os.path.split(path)
self.cd_or_create(path)
# Use read-binary to avoid universal newline support from stripping CR characters.
f = open(local_path, 'rb')
self.storbinary("STOR " + filename, f)
f.close()
else:
self.cd_or_create(path)
def delete_path(self, path, is_file) :
(path, filename) = os.path.split(path)
print("trying to delete: " + path + ", " + filename)
self.cwd(path)
try:
if is_file :
self.delete(filename)
else:
self.delete_path_recursive(filename)
except ftplib.error_perm as e:
if 'The system cannot find the' in str(e) or '550 File not found' in str(e):
# Log, but otherwise ignore path-not-found errors
# when deleting, since it's not a disaster if the file
# we want to delete is already gone.
logging.error(captureErrorMessage(e))
else:
raise
def delete_path_recursive(self, path):
if path == "/" :
raise "WARNING: trying to delete '/'!"
for node in self.nlst(path) :
if node == path :
#it's a file. delete and return
self.delete(path)
return
if node != "." and node != ".." :
self.delete_path_recursive(os.path.join(path, node))
try: self.rmd(path)
except ftplib.error_perm as msg :
sys.stderr.write("Error deleting directory " + os.path.join(self.current_path, path) + " : " + str(msg))
# apply the project_directory setting
def apply_basedir(path) :
#remove any leading stuff (in this case, "trunk/") and decide whether file should be propagated
if not path.startswith(project_directory) :
return ""
return path.replace(project_directory, "", 1)
def svn_export_temp(pysvn_client, base_path, rev, path) :
# Causes access denied error. Couldn't deduce Windows-perm issue.
# It's possible Python isn't garbage-collecting the open file-handle in time for pysvn to re-open it.
# Regardless, just generating a simple filename seems to work.
#(fd, dest_path) = tempfile.mkstemp()
dest_path = tmpName = '%s.tmp' % __file__
exportPath = os.path.join(base_path, path).replace('\\','/')
print('exporting %s to %s' % (exportPath, dest_path))
pysvn_client.export( exportPath,
dest_path,
force=False,
revision=rev,
native_eol=None,
ignore_externals=False,
recurse=True,
peg_revision=rev )
return dest_path
if __name__ == "__main__":
logging.info('svnftp.start')
try:
main()
logging.info('svnftp.done')
except Exception as e:
# capture the location of the error for debug purposes
frame = inspect.currentframe()
stack_trace = traceback.format_stack(frame)
logging.debug(stack_trace[:-1])
print(stack_trace)
# end capture
error_text = '\nFATAL EXCEPTION!!!\n'+captureErrorMessage(e)
subject = "ALERT: SVN2FTP Error"
message = """An Error occurred while trying to FTP an SVN commit.
repo_path = %(repo_path)s\n
local_repos_path = %(local_repos_path)s\n
project_directory = %(project_directory)s\n
remote_base_directory = %(remote_base_directory)s\n
error_text = %(error_text)s
""" % globals()
email(toAddrs, message, subject)
logging.error(e)
Notes/Disclaimers:
I have basically no python training so I'm learning as I go and spending lots of time reading docs to figure stuff out.
The body of get_login is in a try block because I was getting strange errors saying there was an unhandled exception in callback_get_login. Never figured out why, but it seems fine now. Let sleeping dogs lie, right?
The username and password for get_login are currently hard-coded (but correct) just to eliminate variables and try to change as little as possible at once. (I added the svnuser and svnpass arguments to the existing argument parsing.)
So that's where I am. I can't figure out why on earth it's not printing anything into svn2ftp.out.log. If you're wondering, the output for one of these failed attempts in svn2ftp.debug.log is:
2012-09-06 15:18:12,496 INFO svnftp.start
2012-09-06 15:18:12,496 INFO svnftp.done
And it's no different on a successful run. So there's nothing useful being logged.
I'm lost. I've gone way down the rabbit hole on this one, and don't know where to go from here. Any ideas?
It looks as if you are overwriting your logging level. Try setting both to DEBUG and see what happens.
import sys, os
import logging
logging.basicConfig(
level=logging.DEBUG, # DEBUG here
format='%(asctime)s %(levelname)s %(message)s',
filename='svn2ftp.debug.log',
filemode='a'
)
console = logging.StreamHandler()
console.setLevel(logging.ERROR) # ERROR here
logging.getLogger('').addHandler(console)
Additionally you are printing in some places and logging in others. I am not sure that the logging library automatically redirects sys.stdout to the logging console. I would convert all print statements to logging statements to be consistent.