Fix NULL check

[siiky]

Fix #1655

[coveralls]

Coverage remained the same at 84.414% when pulling 089e54d on SiIky:master into 796684b on universal-ctags:master.

[masatake]

[jet@localhost]~/var/ctags% cat /tmp/foo.cc 
cat /tmp/foo.cc 
namespace S {
  struct T {
    enum E {
      TOKYO, KYOTO,
    } elt;
  };
}

struct S::T s = { .elt = S::T::E::TOKYO };

[jet@localhost]~/var/ctags% diff -uN WITHOUT-PATCH.tags WITH-PATCH.tags 
diff -uN WITHOUT-PATCH.tags WITH-PATCH.tags 
--- WITHOUT-PATCH.tags	2018-01-11 11:54:21.299130364 +0900
+++ WITH-PATCH.tags	2018-01-11 11:54:33.683074701 +0900
@@ -11,9 +11,9 @@
 E	/tmp/foo.cc	/^    enum E {$/;"	g	struct:S::T	file:
 S::T::E	/tmp/foo.cc	/^    enum E {$/;"	g	struct:S::T	file:
 TOKYO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
-S::TOKYO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
+S::T::TOKYO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
 KYOTO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
-S::KYOTO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
+S::T::KYOTO	/tmp/foo.cc	/^      TOKYO, KYOTO,$/;"	e	enum:S::T::E	file:
 elt	/tmp/foo.cc	/^    } elt;$/;"	m	struct:S::T	typeref:enum:S::T::E	file:
 S::T::elt	/tmp/foo.cc	/^    } elt;$/;"	m	struct:S::T	typeref:enum:S::T::E	file:
 s	/tmp/foo.cc	/^struct S::T s = { .elt = S::T::E::TOKYO };$/;"	v	typeref:struct:S::T

The result is obviously improved.
S::T::E::TOKYO is not included as a tag but it seems that it is intended.

	if(eScopeType == CXXScopeTypeEnum)
	{
		// If the scope kind is enumeration then we need to remove the
		// last scope part. This is what old ctags did.
		if(cxxScopeGetSize() < 2)
			return -1; // toplevel enum

Actually, with your patch, u-ctags generates the same tags file as e-ctags generates.

I will a pull request to your branch. So could you push it here?

(
BTW, I think you should not use master branch on your side.
As far as I know, the master branch should be used for synchronizing upstream.
If you modify your master branch with a hacking activity, you cannot do another hack.
)

[masatake]

See https://github.com/SiIky/ctags/pull/1.

[masatake]

(Thank you for pushing my change.)

[siiky]

Merged

I think you should not use master branch on your side

Sorry about that, next time I'll do it right. (This was my first PR)

[coveralls]

Coverage increased (+0.08%) to 84.495% when pulling 63b24e2 on SiIky:master into 796684b on universal-ctags:master.

[masatake]

Let's wait for the response from the maintainer of C/C++ parsers.

(This was my first PR)

I'm very proud of u-ctags project attracting a new hacker.
Though the required knowledge for developing u-ctags is only C and standard C library, I think developing u-ctags is very ... attractive. I cannot find a good word representing how I feel.

I hope that you enjoy hacking u-ctags more. So I would like to add more comments here.
I would like you to add a prefix to the subject of commit log like:

C++: fix NULL check

Here C++ is the prefix. If a patch is about a parser, use the name of parser as prefix.
Use main for the change for the common part of code. Please, see the other commit logs about more examples.

Knowing how to write a test case is important in this project. Please look at my patch which I submitted to your branch. See also http://docs.ctags.io/en/latest/units.html .

Thank you.

[coveralls]

Coverage increased (+0.08%) to 84.495% when pulling a877beb on SiIky:master into 796684b on universal-ctags:master.

[pragmaware]

Yep, looks good!
It's interesting that this bug didn't pop out earlier... it's there since the beginning of the parser.

[siiky]

(Didn't have time to reply yesterday)

I'm very proud of u-ctags project attracting a new hacker.

I don't know if I'll stay for long or contribute much to the project, but I'd like to at least try.

Though the required knowledge for developing u-ctags is only C and
standard C library

Not sure about those requirements, it seems to me there's a lot to learn before I can even grasp the ctags big picture and contribute something more valuable... (this PR shows exactly that, a C bug, didn't make tests or anything)

I would like you to add a prefix to the subject of commit

Done, was that what you wanted? Tell me if there's anything missing

[masatake]

Thank you.