In D21259#473832, @ngie wrote:

@fernape: could you please take this revision so I can sign off on the change?

Hi there,

• Remove shebanfix and SHEBANG_FILES (@ngie)
• Fix remaining config file paths

In D21259#465145, @ngie wrote:

In D21259#464853, @fernape wrote:

I have a new patch that addresses some of the things pointed by @mat and updates the port to 3.3.3 (recently released).

Since I'm not the creator of the review I don't know if it is OK to just arc diff --update this?

Feel free. I've been a bit busy with non-[F]OSS work lately.

I was also trying to add "do-test" to the Makefile, but it seems that "make check" with the project is broken when run from ports (I got past some of the issues, but ran into other ones that people won't see unless they enable PUMP).

• USES localbase
• Remove pycache files
• STRIP shared object
• Revert patch-Makefile.in so we deal with {bin,sbin}/distccd
• Remove linux specific examples
• Add config files
• Add missing DOC files
• README.pump include_server.1 and pump.1 are installed unconditionally

I have a new patch that addresses some of the things pointed by @mat and updates the port to 3.3.3 (recently released).

Hi Enjie,

Hi there,

Do not copy values into local variables

Address last of kib's comments.

In D19902#438246, @jilles wrote:

In D19902#437991, @hrs wrote:

In D19902#437962, @jilles wrote:

Perhaps it is cleaner to pass the int age to be searched for by reference. However, the asymmetry in the compare function is intentional as this allows finding an object with more data given the key, while having only one storage array.

Asymmetry of the two arguments gives more capability of data lookup? I do not see a big difference between passing an int value 22 by casting it to (void *) as in the original patch and passing a pointer to a struct person which contains age = 22, in terms of the lookup operation. Can you elaborate a case in which the symmetry arguments do not work?

The problem is not that it does not work, but that it creates a bogus value of type struct person which only contains an age. Although C's type system is unsound, I still prefer to avoid subverting it like this.

• Don't initialize locals in declaration
• Remove spaces after casts
• Remove excesive blank lines
• Revert to "asymmetric" compare function
• Add blank line before multi-line comment
• Remove return in main()

Addressing most of the issues with the revision:

In D19902#437776, @bcr wrote:

Reopen this based on comments by Konstantin and Bruce after the MFC to stable/12. @fernape: Can you look into their comments and address them in a followup patch? Thank you!

Is there anything else that needs to be done?

Properly escape \n sequences

• Eliminate cast to different size warning
• Replace comments with assertions

Fix typo

In D19481#417839, @tcberner wrote:

Yea, looks good to me.

Thanks!

If there are no other objections, I'll go ahead and commit this one.

• Remove CMAKE_ARGS since they are the default
• Remove LDFLAGS and modify src/CMakeLists.txt instead
• Update pkg-plist to match actual installation paths

Add actual port to review.