Date: Fri, 31 May 2024 21:59:37 +0800 From: Po-Chuan Hsieh <sunpoet@freebsd.org> To: Max Brazhnikov <makc@freebsd.org> Cc: Moin Rahman <bofh@freebsd.org>, "Jason E. Hale" <jhale@freebsd.org>, ports-committers <ports-committers@freebsd.org>, dev-commits-ports-all@freebsd.org, dev-commits-ports-main@freebsd.org Subject: Re: git: 7a740b7c908f - main - */tex*: Update version 20230313=>20240312 Message-ID: <CAMHz58Rox4_SUhX6fr0p9knX_gNse2gNc0fBY9hJabB0LVO4fA@mail.gmail.com> In-Reply-To: <CAJie3yoz5JUDkEyhDJ4n796Ku5DLbLNkg%2B3Zn0NJACpGd2Sp6A@mail.gmail.com> References: <202405291834.44TIYggq003718@gitrepo.freebsd.org> <CAJE75NH1VCtXE9iua1MZAc-XUX4t%2Bv04RQ%2BCbiOF2k=K%2BgykFA@mail.gmail.com> <CAF822A0-A1EC-40A2-9983-445DCBD95ACF@freebsd.org> <CAJE75NG1UAwkiMrCZ=TwcC3ppHXvf4jUBUJDu43Zqku2EcU8LQ@mail.gmail.com> <CAJE75NH6vYeraWr%2BXvUAKW8AviVK28h%2BJ6X2A%2BBB3MTBaZk%2BQw@mail.gmail.com> <5B96E765-95E2-4599-80FD-BE9DBA5263FC@freebsd.org> <CAMHz58Sjic3%2B=f9LdeAourYv9EptPjbG9hBLdxoNO6Gf3Pz-cQ@mail.gmail.com> <CAJie3yoz5JUDkEyhDJ4n796Ku5DLbLNkg%2B3Zn0NJACpGd2Sp6A@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--00000000000020c1cd0619c0671c Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, May 31, 2024 at 7:09=E2=80=AFPM Max Brazhnikov <makc@freebsd.org> w= rote: > On Fri, 31 May 2024 at 13:02, Po-Chuan Hsieh <sunpoet@freebsd.org> wrote: > > I always use "poudriere bulk -r -t" > > -r Recursively test all dependencies as well. This flag is > > automatically set when using -at. > > -t Add some testing to the specified ports. Add -r to > recursively > > test all port dependencies as well. When used with -a > then -rk > > are implied. > > > > The side effect is it will block the build if the check fails. > > For example, devel/doxygen has an fs_violation issue (caused by > fontconfig db). > > I have to build it manually without the -t flag. > > -k Do not consider failures to be fatal when using -t. Do not > skip > dependent ports on findings. This flag is automatically set > when using -at. > I'm aware of the -k flag (also mentioned in the help of -t flag) but decide not to use it. It's easier to find out the problematic ports since they are listed as failed ports. If I use -k flag, I have to dig into all logs to find out the issue. And for plist issue of orphaned files, the generated package is incorrect due to missing new files. It may result in incorrect build status of dependent ports. I have to manually remove the package. Therefore, maybe it's better for me to let it fail and skip the dependent ports (and save some time). --00000000000020c1cd0619c0671c Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div dir=3D"ltr">On Fri, May 31, 2024 at 7:09=E2=80=AFPM M= ax Brazhnikov <<a href=3D"mailto:makc@freebsd.org">makc@freebsd.org</a>&= gt; wrote:<br></div><div class=3D"gmail_quote"><blockquote class=3D"gmail_q= uote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,2= 04);padding-left:1ex">On Fri, 31 May 2024 at 13:02, Po-Chuan Hsieh <<a h= ref=3D"mailto:sunpoet@freebsd.org" target=3D"_blank">sunpoet@freebsd.org</a= >> wrote:<br> > I always use "poudriere bulk -r -t"<br> >=C2=A0 =C2=A0 =C2=A0 -r=C2=A0 =C2=A0 =C2=A0 =C2=A0Recursively test all = dependencies as well.=C2=A0 This flag is<br> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0automatically se= t when using -at.<br> >=C2=A0 =C2=A0 =C2=A0 -t=C2=A0 =C2=A0 =C2=A0 =C2=A0Add some testing to t= he specified ports.=C2=A0 Add -r to recursively<br> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0test all port de= pendencies as well.=C2=A0 When used with -a then -rk<br> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0are implied.<br> ><br> > The side effect is it will block the build if the check fails.<br> > For example, devel/doxygen has an fs_violation issue (caused by fontco= nfig db).<br> > I have to build it manually without the -t flag.<br> <br> =C2=A0 =C2=A0 -k=C2=A0 =C2=A0 =C2=A0 =C2=A0Do not consider failures to be f= atal when using -t.=C2=A0 Do not skip<br> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0dependent ports on findings= .=C2=A0 This flag is automatically set<br> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0when using -at.<br></blockq= uote><div><br></div><div>I'm aware of the -k flag (also mentioned in th= e help of -t flag) but decide not to use it.</div><div>It's easier to f= ind out the problematic ports since they are listed as failed ports.<br></d= iv><div>If I use -k flag, I have to dig into all=C2=A0logs=C2=A0to find out= the issue.</div><div>And for plist issue of orphaned files, the generated = package is incorrect due to missing new files.</div><div>It may result in i= ncorrect build status of dependent ports.</div><div>I have to manually remo= ve the package.</div><div>Therefore, maybe it's better for me to let it= fail and skip the dependent ports (and save some time).</div><div><br></di= v></div></div> --00000000000020c1cd0619c0671c--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAMHz58Rox4_SUhX6fr0p9knX_gNse2gNc0fBY9hJabB0LVO4fA>