Closed zcutlip closed 1 year ago
I haven't touched this for quite some time and no longer use it myself, therefore I'm unable to test this :( I'd be willing to blindly merge it but how well tested is it?
Not terribly well tested. I tried it out on some fairly complex open source kernel code, but that's about it. Do you have a set of tests or some sort of testing process I can do?
Not particularly. There are some unit tests but I'm not sure how applicable they are. If you've been using this for some time, that's probably good enough testing
@zcutlip thank you for fixing the issue.
I see more things with C code and incorrect regexps. For example a bunch of functions in Linux tree are not properly CTagged. e.g. jumping to arch_local_irq_restore
does not work. Looking at tags file I see
arch_local_irq_restore arch/x86/include/asm/irqflags.h /^static inline notrace void arch_local_irq_restore(unsigned long flags)$/;" f typeref:typename:notrace void signature:(unsigned long flags)
arch_local_irq_restore arch/x86/include/asm/paravirt.h /^static inline notrace void arch_local_irq_restore(unsigned long f)$/;" f typeref:typename:notrace void signature:(unsigned long f)
arch_local_irq_restore include/asm-generic/irqflags.h /^void arch_local_irq_restore(unsigned long flags);$/;" p typeref:typename:void signature:(unsigned long flags)
I wonder if parethesis in the regexp should be escaped. But as it is harder to change CTgas behavior maybe the plugin should workaround the pproblem?