reserved identifier violation

[elfring]

I would like to point out that an identifier like “__ARM_2D_H__” does not fit to the expected naming convention of the C++ language standard. Would you like to adjust your selection for unique names?

[GorgonMeducer]

Thank you for your feedback.

We knew this naming convention, and similar to cmsis and cmsis-dsp, arm-2d is part of the software infrastructure libraries provided by the arm for use with the arm toolchains (for most of the time I guess).

https://github.com/search?q=repo%3AARM-software%2FCMSIS-DSP%20__&type=code

https://github.com/ARM-software/CMSIS-DSP/blob/99e15d3d95084b6868c5579b6791aff5cbeefcd6/Include/dsp/none.h

And you are right. For the application code and example code, we should avoid using __ as the prefix. We will apply the change in the v1.1.6.

Thank you very much.

I hope this answers your questions.

[GorgonMeducer]

By the way, Arm-2D is a C library but not C++, hence by checking.

https://wiki.sei.cmu.edu/confluence/display/c/DCL37-C.+Do+not+declare+or+define+a+reserved+identifier

We are open to further discussion.

[elfring]

:thought_balloon: Would you become interested to use a development tool like “clang-tidy” for corresponding source code adjustments?

[GorgonMeducer]

Thank you. Using which tools isn't a problem. The key is following which software standard. Frankly speaking, this project has no plan to follow the SEI CERT C++ Coding standard in the short term.

The excellence of the SEI CERT C++ Coding standard does not mean it is the only good one that everyone must comply with. At the same time, we are not saying that we will not follow it in the future; we simply don't know about the future, and we have indeed not seen problems caused by failure to comply with this standard yet. We appreciate your concern about this issue. We have put compliance with some specifications, such as MISRA and SEI CERT C++ Coding standards, into our future plans.

I am looking forward to learning more in this kind of discussion.

[elfring]

:thought_balloon: I hope that you would also like to avoid that this software depends on undefined behaviour.

[GorgonMeducer]

:thought_balloon: I hope that you would also like to avoid that this software depends on undefined behaviour.

For this part, if you could point out detailed problem in a dedicated issue that would help.

Thank you very much.

[elfring]

:eyes: I got the impression that some ARM software components tamper with the reserved name space.

[GorgonMeducer]

Your impression is correct for arm-2d, nothing to hide, for the other components, it depends.

[elfring]

Which changes did improve the affected software?

[GorgonMeducer]

Please give a more specific question. Most of the commits are used to improve the arm-2d; otherwise, why are there any commits in the first place?

[elfring]

:crystal_ball: Would any ARM developers like to rename affected include guards (for example)?

[GorgonMeducer]

This is a question for arm developers; I know many who use the same manner, and I am definitely sure you know someone who doesn't like to use reserved identifiers.

I cannot represent any Arm developers, so I am not able to answer this question. Sorry.

[GorgonMeducer]

I want to make this clear. You can blame me for using reserved identifiers. I follow a different naming convention, which I have put it clear in arm-2d documents already, that is: "__" means private in this library.

Unless I see solid evidence that the existing arm-2d identifier with "__" as their prefix conflicts with the reserved ones in arm c compilers that we claim to support, i.e. IAR, GCC, LLVM and Arm Compiler 5/6, I will not change my mind.

Even if some arm-2d identifiers are addressed to conflicts with some existing reserved ones, I will treat them in a case-by-case manner.

Thank you for your time and suggestions.

[elfring]

Unless I see solid evidence that the existing arm-2d identifier with "__" as their prefix conflicts with the reserved ones in arm c compilers that we claim to support, i.e. IAR, GCC, LLVM and Arm Compiler 5/6, I will not change my mind.

:thought_balloon: It seems that some source code reviewers were also too careless according to concerns around undefined behaviour so far. :crystal_ball: Will development interests grow for better compliance with known programming rules?

[GorgonMeducer]

I am open to talking with you about this.

Would you mind providing me with some solid cases that:

existing arm-2d identifier with "__" as their prefix conflicts with the reserved ones in arm c compilers that we claim to support, i.e. IAR, GCC, LLVM and Arm Compiler 5/6,

This will help us to improve the project. Thank you.

[GorgonMeducer]

It seems that some source code reviewers were also too careless according to concerns around undefined behaviour so far.

We really welcome open-source community to help us with finding the undefined behaviours.

Please raise dedicated issues for each undefined behaviour.

[elfring]

:eyes: Did you overlook any warnings about undefined behaviour from linked information sources so far?

[GorgonMeducer]

It looks like a personal question for me. Please allow me to keep the answers to myself.

For the arm-2d project, our team reviews all the warnings, evaluates the risks and treats them with different policies. End users should see 0 warnings when using the arm compilers we claim to support.

if you address any warnings, please feel free to report them.

Thank you.

[elfring]

End users should see 0 warnings when using the arm compilers we claim to support.

if you address any warnings, please feel free to report them.

:thought_balloon: I find such a feedback questionable according to discussed programming language compliance concerns. Further source code analysis tools can help to reconsider remaining open issues, can't they? :thinking:

[GorgonMeducer]

We have yet to claim to follow the standard you mentioned. So I would put your suggestion as a new feature recommendation.

Thank you for your time.

[elfring]

We have yet to claim to follow the standard you mentioned.

Does this software belong to the implementation for the programming language “C” (or “C++”)?

[GorgonMeducer]

Arm-2D is under Apache 2.0. It is an open-source project. Anyone can check the source code and evaluate the risks based on their own judgements. We have clearly mentioned in our document that "__" is used as an indicator of private symbols.

Those who use arm-2d should follow the Apache 2.0 license and manage their own risk.

[elfring]

We have clearly mentioned in our document that "__" is used as an indicator of private symbols.

I interpret such information as a recurring programming mistake.

[GorgonMeducer]

I respect your judgement. Thank you.

[elfring]

I respect your judgement.

:crystal_ball: What does hinder you then to adhere to known programming language standard specifications?

[GorgonMeducer]

Please allow me to keep this personal question to myself.