Given your comments, I think it does makes sense to add new checkers. Thanks!
That's fair. As with all things open source, it has a long life as long as we have people contributing to it. ☺.
Cheers,
Doug.
On Thu, 2019-04-18 at 16:11 +0200, Mark B wrote:
Yep, but you know, this kind of changes are quite big, my customer for example is starting to use c++11 just now after ten years and we are taking about a medium company so I guess codan has still long life :)
I think that's fine for the short term.
A number of us are hoping to move to clangd as our language server. I think C++20 will force our hand as there are some significant changes coming. Clang has a much larger community than ours and it would be good to latch onto that.
But we are a long way from feature parity with our existing language services so I don't expect a switch before that becomes an issue.
That being said, it would be too bad if all this work going into codan and other work going into refactoring goes to waste in a couple of years. But we really need a plan before knowing how to manage our short term investments.
In other words, as the contributor, you get to choose what best meets your needs and we'll be happy to support you how we can. ☺
Cheers,
Doug.
On Thu, 2019-04-18 at 07:19 +0200, Mark B wrote:
I think we should add basic checkers avoiding to add too much specific ones. I think the checker I added would be enough. Codan framework provides to the developer a unique feature:
real-time check with full ide integration. No need other dependency. I'm performing a study payed by one of my customers and I'm pushing the customer to invest in Codan framework to gain on maintenance activity because it decrease dramatically the bug
life since they use Eclipse as basic environment. However I like to receive feedback about this point because if the conclusion is that is better to avoid to add new checkers I think I can definitely prefer the plug-in way removing the commits I did. Creating
a plug-in for cdt is ok from point of view.
Marco
Hi Macro,
Thank you for all the great work! One thing I have been wondering is if we should go ahead with adding a lot of new checkers or not in Codan. My worry is that we have a lot to maintain and false positives can be a recurring problem. There are also a lot of
tools that have mostly replaced the need of Codan like clang-tidy, clang-static-analyzer, pc-lint, Cppcheck etc. Clangd also covers a lot through warnings and errors that come directly from the compiler and does so as-you-type.
I am not saying that I am against adding more checkers and I don’t want to discourage you from improving CDT in this area. But this concern has been at the back of my mind and I’d love to hear your thoughts (or anyone’s) on the subject,
Otherwise, I’m happy to help reviewing where (when?) I can. (Well, except the formatter, I don’t know much about!).
Regards,
Marc-André
> On Apr 17, 2019, at 12:27 PM, marco <marco.stornelli@xxxxxxxxx> wrote:
>
> Hi guys,
>
> first of all thanks to everyone for comments received about merged commits. However there are a lot of them without feedback, the effort for the rebase each time is high, so please review the commits. Thanks!
>
> Current status:
>
> Formatter bug fixes with higher priority:
> Bug 405409 - Formatting namespaces with inactive sections
> Bug 376395 - Fix const methods with never join wrapped lines
> Bug 546221 - Fix exception for nested namespaces
> Bug 390324 - Fix cast operator containing func declarator
>
> Formatter improvements
> Bug 500000 - Added proper formatting for lambda expressions
> Bug 461435 - Added options to format pointers in methods (ready to be merged @Jonah can you do it?)
>
> Codan improvements
> Bug 545954 - Added checker for functions/methods blacklist
> Bug 545959 - Added checker for assignment operator
> Bug 545702 - Added new checker to verify copyright information
> Bug 545714 - Added a shallow copy checker
> Bug 545956 - Added checker for virtual methods in ctor/dtor
> Bug 546173 - Add a check for returning of local variable addresses
> Bug 545977 - Added checker for variable initialization
> Bug 545704 - Added checker for float comparison
> Bug 343274 - Added checker for const and static methods check
> Bug 545952 - Added checker to check for goto usage
> Bug 355174 - Added cheker for missing case in a switch
>
> New UI feature
> Bug 546395 - Added "copy qualified name" feature
>
> Refactoring
> Bug 303870 - Add override virtual methods functionality (ready to be merged after review)
>
> Codan fix
> Bug 545877 - Don't take into account init var values in CFG
> _______________________________________________
> cdt-dev mailing list
>
cdt-dev@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
>
https://www.eclipse.org/mailman/listinfo/cdt-dev
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/cdt-dev
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.eclipse.org_mailman_listinfo_cdt-2Ddev&d=DwICAg&c=yzoHOc_ZK-sxl-kfGNSEvlJYanssXN3q-lhj0sp26wE&r=NrrbvTHWa2Nbp_kAN0Hl1o3lM1WAwSes64uBjxjNhMc&m=x3qV_l8T9r1mki5viXtSUIeOO46ohB7H5cYyG14Ze9o&s=fHLMZJuJ6Ebg09V3N0EEAU2PkDkyr5dwJ4W0z54D1gs&e=
_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/cdt-dev
_______________________________________________ cdt-dev mailing list cdt-dev@xxxxxxxxxxxTo change your delivery options, retrieve your password, or unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/cdt-dev
|
