I have a python script to download source code from a list of repositories, some of them are big.
Sometimes, svn hangs in the middle of a check out. Is there a way to watch over svn process, and 开发者_StackOverflow中文版so I know it is hang or not?
You can use PySVN, and register a callback for each "event" processes. PySVN can also poll a "cancel" callback. The first callback could start a timer, and if the timer expires, you can do tell the "cancel" callback to return False, thus cancelling the checkout.
#!/usr/bin/python
url = "svn://server/path/to/repo"
path = "/path/to/local/wc"
import pysvn
import threading
# Set to something reasonable
SVN_TIMEOUT = 1000
svn_timer = None
stop_svn = False
def timer_expired():
# Too long since last SVN event, so do something sensible...
print "SVN took too long!"
global stop_svn
stop_svn = True
def svn_cancel():
return stop_svn
def notify( event_dict ):
global svn_timer
if svn_timer:
svn_timer.cancel()
svn_timer = threading.Timer(SVN_TIMEOUT, timer_expired)
svn_timer.start()
svn_client = pysvn.Client()
svn_client.callback_notify = notify
svn_client.callback_cancel = svn_cancel
svn_timer = threading.Timer(SVN_TIMEOUT, timer_expired)
svn_timer.start()
revision = svn_client.checkout(url,path)
if svn_timer:
svn_timer.cancel()
You can keep polling the stdout of the svn process, and test how often you get new files. If you don't get a new file within x seconds, bounce the process.
Launch svn using subprocess from within your master script, and poll stdout while waiting for the process to complete.
精彩评论