Re: Port tree linter

On 2024-02-27 09:50, Hubert Tournier wrote:
> Hello there!
> 
> I made a small program that perhaps may be of interest to some port 
> maintainers.
> 
> It's called portlint2 (https://github.com/HubTou/portlint2), and it checks 
> the
> ports Index file and the port's makefiles, for the whole port tree, or for
> selected categories / maintainers / ports.
> 
> On a freshly updated port Index and port tree, it will produce a summary of
> findings like this:
> 
> Selected 34434 ports out of 34434 in the FreeBSD port tree, and found:
>   4 ports with unusual installation-prefix (warning)
>   339 ports with a comment string exceeding 70 characters (warning)
>   286 ports with an uncapitalized comment
>   11 ports comment ending with a dot
>   108 ports with a comment different between the Index and Makefile
>   2 ports with non existent description-file
>   4 ports with a maintainer different between the Index and Makefile
>   34 ports referring to unofficial categories (warning)
>   262 ports with categories different between the Index and Makefile
>   1251 ports with no www-site
>   300 ports with an unresolvable www-site hostname
>   658 ports with an unaccessible www-site
>   2 ports with a www-site different betwwen the Index and makefile
> 
> There are other checks, but it only prints summary lines for those with 1+ 
> occurrences.
> 
> Before this summary, it prints a list of affected ports per maintainer, for
> example like this:
> 
>   yuri@FreeBSD.org:
>     Diverging comments:
>       jamulus-server-3.10.0 RStudio-2022.12.0+353_6
>       RStudio-server-2022.12.0+353_6 qbittorrent-nox-4.6.3
>     Too long comments:
>       py39-pytest4-flakes-4.0.1 py39-spectral-0.22.4_1
>     Uncapitalized comments:
>       shunit2-2.1.8.93 libmicrodns-0.2.0 py39-mmcif-0.84 hq-1.0.1_9
>       ibus-m17n-1.4.28 jaq-1.3.0_1
>     Diverging categories:
>       obs-studio-30.0.2_1
>     HTTP Error 404 (Not found) on www-site:
>       eteroj-lv2-0.10.0_1 geonkick-lv2-2.10.0 lv2lint-0.16.2_2
>       midi-matrix-lv2-0.28.0_1 moony-lv2-0.36.0_1 orbit-lv2-0.1.661
>       py39-hsaudiotag3k-1.1.3.p1 sherlock-lv2-0.28.0_2 
> timely-lv2-g20190412_1
>       vm-lv2-0.14.0_2 graphlan-1.1.3_1 GroopM-0.3.4_4 thrust-1.9.5_1
>       py39-ta-lib-0.4.28 cmh-1.1.1_3 FlintQS-1.0
>       coin-or-flopc++-1.2.5.20200527_1 moab-5.3.1_5 mpfrcx-0.6.3_1
>       paritwine-0.1_3 vinci-1.0.5 ironscanner-1.1.0.20180828 
> sdformat-8.0.0_6
>       clash-1.18.0_2 dftd3-3.2.0.3_1 dftd4-3.5.0_1 octopus-13.0_1
>       openbabel-3.1.1.178 opsin-3.0.20190223_1 py39-dftd4-3.5.0
>       py39-openbabel-3.1.1.1 py39-phono3py-1.22.3_2 py39-pyked-0.4.1.16_1
>       xdrawchem-1.11.0.2_2 ntk-1.3.1001_1 redkite-1.3.1
>     Unresolvable www-site:
>       libpcl-1.12 geogram-1.7.9 daggy-2.1.3_1 silicon-0.1.124 
> ztoolkit-0.1.2_2
> 
> It also produces more detailed error logs on stderr.
> 
> I hope it will be useful to others.
> 
> I needed this because I have another program that mass checks Python ports 
> for
> unreported vulnerabilities,
> which needs an up-to-date and correct information in the ports Index to be 
> relevant...
> 
> Best regards,
> 
> Hubert
While I haven't (yet) tried it out. I'm grateful for your work. It'll 
potentially save a
bunch of work, Thanks!
Shouldn't this make it to ports-mgmt/ as portlinter?
Thanks again.

-- 
--Chris Hutchinson