Have you tried it on another website?
I have seen the timeout too, but once I it with some new variable names and a new website it worked everytime and pretty fast too. I think that an not properly closed socket is too blame for the timeouts. The resources need to be released, that’s what the close method is for.
(
total = 0
for i = 1 to 10 do
(
start = timeStamp()
request = (dotNetClass "system.Net.WebRequest").create "http://www.penproductions.ca"
request.Credentials = (dotnetClass "System.Net.CredentialCache").DefaultCredentials
response = request.GetResponse()
result_URI = response.ResponseUri.ToString()
result_HEADER = response.Headers.ToString()
response.Close()
end = timeStamp()
total += end - start
format "%
%
" (end-start) result_URI
)
format "
totalTime of 10 runs : %
" (total/10)
)
[color=SeaGreen]-- 844
-- http://www.penproductions.ca/
-- 359
-- http://www.penproductions.ca/
-- 266
-- http://www.penproductions.ca/
-- 266
-- http://www.penproductions.ca/
-- 265
-- http://www.penproductions.ca/
-- 250
-- http://www.penproductions.ca/
-- 250
-- http://www.penproductions.ca/
-- 266
-- http://www.penproductions.ca/
-- 250
-- http://www.penproductions.ca/
-- 265
-- http://www.penproductions.ca/
-- totalTime of 10 runs : 328
-- OK[/color]
Average of 328ms, that’s not bad.
-Johan
request.Timeout = 1000
Sets the max timeout to 1 sec. for example. Still you’d have to handle the timeout exception, could be in a simple try catch.
-Johan
To see if there’s an internet connection you can also use this example, but always using the try catch method…
(
---Check if internet connection is available
host = dotnetClass "System.Net.Dns"
try
(
hostInfo = host.GetHostByName("www.google.com")
)
catch
(
print (getCurrentException())
)
)
Also if you had this to JHN’s code after request.Credentials it’s faster because it only downloads the metainformation contained in the HEADERS and not the whole entity-body (as explained in HTTP/1.1 RFC2616 Protocol)
...
request.Credentials = (dotnetClass "System.Net.CredentialCache").DefaultCredentials
[B]request.Method = "HEAD"[/B]
response = request.GetResponse()
...
Without request.Method = “HEAD”
1265
http://www.penproductions.ca/
526
http://www.penproductions.ca/
370
http://www.penproductions.ca/
374
http://www.penproductions.ca/
363
http://www.penproductions.ca/
364
http://www.penproductions.ca/
362
http://www.penproductions.ca/
368
http://www.penproductions.ca/
378
http://www.penproductions.ca/
358
http://www.penproductions.ca/
totalTime of 10 runs : 472
With request.Method = “HEAD”
348
http://www.penproductions.ca/
173
http://www.penproductions.ca/
176
http://www.penproductions.ca/
176
http://www.penproductions.ca/
177
http://www.penproductions.ca/
178
http://www.penproductions.ca/
175
http://www.penproductions.ca/
177
http://www.penproductions.ca/
175
http://www.penproductions.ca/
176
http://www.penproductions.ca/
totalTime of 10 runs : 193
JHN, the problem isn’t with timing out I don’t think, I think it is with not having a connection in the first place. When I have a connection it works great.
Kameleon, thanks, I’m using a try catch now to get around the problem, I was just hoping that there was a cleaner way.
Maybe it’s good to mention that try catch is a perfectly legal way in c# to catch exceptions, I even think it’s the only way. I know it’s not considered great practice in max, but in this instance where you set a timeout like 1 second and a try-catch to capture the timeout exception, looks perfectly okay to me!
-Johan
I guess I’m a bit anal and want to keep it really clean. TryCatch is working fine.