Closed GoogleCodeExporter closed 9 years ago
I have the same problem. Any progress?
Original comment by andrewdi...@gmail.com
on 13 Oct 2011 at 9:55
No... feel free to implement a fix. I can pull it in and cut a new release.
Original comment by mosa...@gmail.com
on 13 Oct 2011 at 7:05
If I find the time I can look at that myself at some stage, but if somebody
that needs this could provide a fix that would probably be faster.
Original comment by mosa...@gmail.com
on 18 Nov 2011 at 5:37
Looks like I might not get around to this anytime soon so I am changing to
awaiting a contribution from someone else.
Original comment by mosa...@gmail.com
on 8 Dec 2011 at 5:31
The problem might be related with the method used to create the request data
(createRequestData on class Transport ). Currently this method is using a
default constructor call to build the ByteArrayOutputStream used to hold the
request information, The buffer capacity is initially 32 bytes by default as
per javadocs.
That means that it will need to increase the size of the array dinamically
(Very expensive proceess) to make space to hold the 2.5 Mg.
So i think we can alleviate this by calling the ByteArrayOutputStream
constructor passing as argument the size of the object, (Or at least and
aproximated value).
Will try to test this solution, the problem will be to determine the
approximate size of the object
Original comment by jose.cas...@gmail.com
on 16 May 2012 at 3:53
That sounds like a good improvement. In the short term it might even be fine to
be able to set an initial value as a static parameter on the transport so that
you can set it high when calling a service with a large load.
Original comment by mosa...@gmail.com
on 11 Jun 2012 at 5:40
I am working on that will send you a pull request shortly
Original comment by jose.cas...@gmail.com
on 12 Jun 2012 at 2:43
Thank you!
Original comment by rbrlnx
on 2 Aug 2012 at 10:14
Hello,
I am also having difficulties with KSOAP httpTransport.call() being too slow.
Our IPhone app on IOS is able to call our webservice superfast.
For Android, using KSOAP, the same webservice takes several seconds.
When will a version be available with the fix described above?
Thanks in advance!
Original comment by i...@pharox.nl
on 7 Aug 2012 at 9:12
[deleted comment]
Fix is in master and will be available with 3.0.0-RC.4
Original comment by mosa...@gmail.com
on 12 Nov 2012 at 6:24
Hello,
I have the same issue, but it occurs when using in a local network (e.g. the
device has the IP 192.168.1.3 and the server 192.168.1.3) and this network is
not connected to the Internet. Nevertheless it works well when the network is
connected to the Internet. In this case the HttpTransport.call() method returns
immediately, but when not connected to the Internet it takes up to 20 seconds
to return (what is a lot of time for the task that uses it).
I have tried with the 3.0.0-RC.4 but it did not work.
Could you check this out, please?
Thank you!
Original comment by j...@boleteriafacil.com
on 8 Jun 2013 at 11:12
[deleted comment]
[deleted comment]
Still no fix?
Ksoap2 on android is useless for 3G/Edge.
Where iOS app connects to our web service (https) instantly and returns in no
time, our android version is taking 5+ seconds just to open the output stream.
Once it's open, it will send the data in a couple of milliseconds.
We use KeepAliveHttpsTransportSE.
It's working fine on WiFi.
Original comment by dco...@gmail.com
on 21 Jun 2013 at 3:53
Forgot to mention, we are using 3.0.0 RC4
Original comment by dco...@gmail.com
on 21 Jun 2013 at 3:53
Hello,
I have the same issue, but it occurs when using in a local network (e.g. the
device has the IP 192.168.1.3 and the server 192.168.1.3) and this network is
not connected to the Internet. Nevertheless it works well when the network is
connected to the Internet. In this case the HttpTransport.call() method returns
immediately, but when not connected to the Internet it takes up to 20 seconds
to return (what is a lot of time for the task that uses it).
I have tried with the 3.0.0-RC.4 but it did not work.
Could you check this out, please?
Thank you!
ı have the same problem too. still no fix?
Original comment by bluerom...@gmail.com
on 11 Aug 2014 at 12:22
Original issue reported on code.google.com by
v...@itgorod.ru
on 14 Sep 2011 at 9:36