typedefs - TANGO Controls

# 5 years ago
EnricoSegre	nleclercq Ok. Let's go. Could you help me to factorize some code in the polymorphic VIs? For instance, let's take the _TangoReadAttritute.vi. What would you do to factorize the underlying code? If you wish to look at, I worked out the first 5 instances (alphabetically). It is not that I really need that now, it is more of an exercise and to show what could minimally be done to have a leaner and easier to maintain code. I haven't checked the results for real, and don't exclude that some mistake slipped in, for that you need systematic unit tests. Other incidental comments: if the exercise would be carried on for all the 35 instances, you would get rid of 35 appearances of a deprecated VI. I suppose that the typedef _DataType.ctl could be obsoleted and replaced everywhere by _AttributeDataType_t.ctl. _TangoExceptionToLabviewError.vi has an input argument "location" whose use, for what I saw in the codebase, seems pleonastic. I think its value could be derived from the call chain, except from minor variances which look completely unintentional. I speculate that perhaps more factorization could be done using variants or classes - but that is a more far fetched design decision, probably revolutionizing the API, and as of now I really miss the overview for proposing anything.

# 5 years ago

nleclercq
Ok. Let's go. Could you help me to factorize some code in the polymorphic VIs? For instance, let's take the _TangoReadAttritute.vi. What would you do to factorize the underlying code?

If you wish to look at, I worked out the first 5 instances (alphabetically). It is not that I really need that now, it is more of an exercise and to show what could minimally be done to have a leaner and easier to maintain code.
I haven't checked the results for real, and don't exclude that some mistake slipped in, for that you need systematic unit tests.

Other incidental comments:

if the exercise would be carried on for all the 35 instances, you would get rid of 35 appearances of a deprecated VI.
I suppose that the typedef _DataType.ctl could be obsoleted and replaced everywhere by _AttributeDataType_t.ctl.
_TangoExceptionToLabviewError.vi has an input argument "location" whose use, for what I saw in the codebase, seems pleonastic. I think its value could be derived from the call chain, except from minor variances which look completely unintentional.
I speculate that perhaps more factorization could be done using variants or classes - but that is a more far fetched design decision, probably revolutionizing the API, and as of now I really miss the overview for proposing anything.

# 5 years ago
nleclercq	I just gave you PR a try and I'm quite impress by the result in terms of side effects on the existing code. I don't see any regression - at least with the provided client and server examples. Congrats for that. My main concern is now focused on the "orphan" vis. Could we continue our discussion on github please? Thanks.