Hi there,
I can't explain the Linux/Windows difference.
Fuseki was wrong (it could double decode because the servlet API does
decoding)
For Linux - looks like there is still a bug - it should work (I tried
using using curl and a Fuseki server in Jetty).
What does the Fuseki log file say? The POST or PUT should be logged. Is
it the same on Windows and Linux?
There seems to be missing text in your message - see below.
But.
Because "#" is fragment and there are several pieces of software that
may be involved in %-encoding handling, it is best to not use '#' for a
Graph Store Protocol graph name. You may find on other systems it
remains as %23, which is not entirely wrong.
Andy
On 06/09/2021 15:31, Sarbashis Das wrote:
Hi Users,
I have been testing apache-jena-fuseki version 4.1.0. and
found one issue with the graph name.
Previously I was using apache-jena-fuseki 3.14.0 and upload ttl files using
python post request using graph names. As an example
http://localhost:3030/eva-dev/data?graph=http://www.mysite.com/eva/graph%23backhaul
(Graph names has “#” which is encoded as %23)
Now with fuseki v4.1.0, my python request failed with error as follow in linux
and Mac
urllib3.exceptions.ProtocolError: ('Connection aborted.', BrokenPipeError(32,
'Broken pipe'))
During handling of the above exception, another exception occurred:
Is there something missing here?
However it works with windows system
I am wondering what is the reason for this difference and how can I fix that.
Below are some info to reproduce the issue
Apache-jena-fuseki-4.1.0
Java version:
java 12.0.1 2019-04-16
Java(TM) SE Runtime Environment (build 12.0.1+12)
Java HotSpot(TM) 64-Bit Server VM (build 12.0.1+12, mixed mode, sharing)
Python version: 3.8.2
Thanks in advance for your support
Regards,
Sarbashis
