add representation methods and encoders #70
Conversation
|
The encoders are useful to store objects on disk between multiple calls to a program. This is useful in some cases where you need to recover the previous state of your run. |
arfio
force-pushed
the
repr-encoders
branch
2 times, most recently
from
e934207
to
cfa5eea
Compare
There was a problem hiding this comment.
Thanks for the contribution. When testing I encountered some issues which I ask you to address.
I had to add some code to test the additions. Is there a way to provide some tests?
(please remember when running pytest, the trace server workspace is wiped out, so run it on a new workspace)
| series.print(array_print) | ||
|
|
||
| def __repr__(self): | ||
| return f'XYModel(title={self.title}, series={self.series})' |
There was a problem hiding this comment.
the print method was removd and hence calling ./tsp_cli_client --get-xy OUTPUT_ID --uuid UUID --items ITEMS --time-range START END NUM_TIMES fails:
Traceback (most recent call last):
File "/home/user/git/tsp-python-client/./tsp_cli_client", line 423, in <module>
xyModel.print(array_print=True)
AttributeError: 'XYModel' object has no attribute 'print'
There was a problem hiding this comment.
I replaced it using the repr instead. If a pretty print is needed for the CLI then it can be implemented but I feel like it is not needed here.
arfio
force-pushed
the
repr-encoders
branch
2 times, most recently
from
3e85218
to
5760d00
Compare
|
Thank you for the review, there were indeed quite a few bugs to fix. I wonder if there is a good way of validating these functionalities, should we just validate that no exceptions are raised ? or should we validate that the strings generated are equal to test strings ? This might require a lot of additional tests and code that might change in the future. What do you think ? |
|
The CI failed because on an issue with the trace server, it should be fixed when eclipse-tracecompass-incubator/org.eclipse.tracecompass.incubator#89 is merged. |
The encoders are useful to store objects on disk between multiple calls to a program. This is useful in some cases where you need to recover the previous state of your run. The representations implementation is very useful to print the results of the requests to know what you got. It is very useful for debugging and testing out things when developping. Signed-off-by: Arnaud Fiorini <[email protected]>
It's not easy to verify print-outs. They can change easily. You could verify some content that it's known to be there, but don't compare IDs or order that can change during multiple executions. Ideally there would be some test cases which could be used to test and helps to know how to use it. It doesn't need to be exhausting tests, but it would help. I had to read the code and google search to know how to try it. I'm ok to merge it as is, but it would be good to have an "hello world" test case for example purposes. |
|
@MatthewKhouzam are you still ok with this PR after the updates? |
This method adds encoders to be able to easily store the responses in json format.
It also adds repr() method to used models to be able to print and easily see what is contained in a response.