Date: Sat, 13 Jan 2024 21:20:02 +0100 From: Daniel Engberg <daniel.engberg.lists@pyret.net> To: Charlie Li <vishwin@freebsd.org> Cc: Gleb Popov <arrowd@freebsd.org>, ports-committers@freebsd.org, dev-commits-ports-all@freebsd.org, dev-commits-ports-main@freebsd.org Subject: Re: git: 25e6f68a6661 - main - textproc/libxml2: Update to 2.11.6 Message-ID: <d6023e74db7025fcf6ff4e8110491ec3@mail.infomaniak.com> In-Reply-To: <84d0b8cc-3400-4eee-82ee-409120cdaba9@freebsd.org> References: <202401121705.40CH5JhG014492@gitrepo.freebsd.org> <ZaIcEvYZ2Yq_jW9Z@FreeBSD.org> <CALH631=meexWeEvPY4dBsVviAEms9OfcmKTk-YHGfw4fHJCzSw@mail.gmail.com> <a228d839c5a29d9991b72f1e0664da04@mail.infomaniak.com> <02ebab1b-a763-45e4-9380-0d82c83c22a4@freebsd.org> <CALH631keLuwEchvHA8OWuHy_LiWsodJi8VL5NN-L5p1WYsKGwA@mail.gmail.com> <d9314eaa-4629-49d4-8929-ba7b0551762a@freebsd.org> <30d803b8ab6713c88aa8fe2c0e516e51@mail.infomaniak.com> <84d0b8cc-3400-4eee-82ee-409120cdaba9@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--_=_swift_1705177202_df9c4f792d380b0d1dd7eb170b2cf2bb_=_ Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 2024-01-13T20:27:45.000+01:00, Charlie Li <vishwin@freebsd.org> wrote:= >=C2=A0Daniel=C2=A0Engberg=C2=A0wrote: >>=C2=A0=C2=A0On=C2=A02024-0= 1-13T18:46:34.000+01:00,=C2=A0Charlie=C2=A0Li >>=C2=A0=C2=A0<vishwin@free= bsd.org>=C2=A0wrote: >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0The=C2=A0Build= =C2=A0Instructions=C2=A0section=C2=A0of=C2=A0the=C2=A0README=C2=A0specifica= lly=C2=A0states >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0"Autotools=C2=A0(for= =C2=A0POSIX=C2=A0systems=C2=A0like=C2=A0Linux,=C2=A0BSD,=C2=A0macOS)",= =C2=A0even=C2=A0in >>=C2=A0=C2=A0the >>=C2=A0=C2=A0 >>=C2=A0=C2=A0= =C2=A0latest=C2=A0trunk.=C2=A0CMake=C2=A0was=C2=A0originally=C2=A0added= =C2=A0mainly=C2=A0for=C2=A0Windows >>=C2=A0=C2=A0support. >>=C2=A0= =C2=A0 >>=C2=A0=C2=A0=C2=A0Their=C2=A0CMake=C2=A0support=C2=A0on=C2=A0pla= tforms=C2=A0like=C2=A0ours=C2=A0still=C2=A0has=C2=A0an >>=C2=A0=C2=A0outs= tanding=C2=A0bug >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0pertaining=C2=A0to= =C2=A0dependency=C2=A0resolution,=C2=A0which=C2=A0at=C2=A0least=C2=A0I= =C2=A0consider=C2=A0a >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0showstopper.= =C2=A0While=C2=A0upstream=C2=A0have=C2=A0been=C2=A0accepting=C2=A0and=C2= =A0responsive=C2=A0to >>=C2=A0=C2=A0any >>=C2=A0=C2=A0 >>=C2=A0=C2= =A0=C2=A0and=C2=A0all=C2=A0improvements=C2=A0to=C2=A0their=C2=A0CMake=C2= =A0support,=C2=A0that=C2=A0is=C2=A0irrelevant >>=C2=A0=C2=A0until >>= =C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0they=C2=A0explicitly=C2=A0bless=C2=A0it= =C2=A0as=C2=A0an=C2=A0equal=C2=A0to=C2=A0autotools.=C2=A0(Not=C2=A0to=C2= =A0say >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0autotools >>=C2=A0=C2=A0 = >>=C2=A0=C2=A0=C2=A0is=C2=A0perfect=C2=A0either,=C2=A0far=C2=A0from=C2= =A0it) >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0Personally,=C2=A0between= =C2=A0the=C2=A0two=C2=A0choices=C2=A0here,=C2=A0I=C2=A0prefer=C2=A0CMake.= =C2=A0But >>=C2=A0=C2=A0personal >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2= =A0preferences=C2=A0are=C2=A0irrelevant=C2=A0wrt=C2=A0liability=C2=A0and= =C2=A0support=C2=A0issues. >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0Hi, >>= =C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0That's=C2=A0a=C2=A0bit=C2=A0contractiona= ry?=C2=A0We=C2=A0upstream=C2=A0patches,=C2=A0community=C2=A0and >>=C2= =A0=C2=A0others=C2=A0 >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0also=C2=A0do= =C2=A0and=C2=A0yet=C2=A0there's=C2=A0a=C2=A0"support=C2=A0issue"=C2=A0despi= te=C2=A0autotools=C2=A0is >>=C2=A0=C2=A0"far=C2=A0from=C2=A0 >>=C2= =A0=C2=A0 >>=C2=A0=C2=A0=C2=A0it"=C2=A0(perfect)?=C2=A0Why=C2=A0not=C2= =A0embrace=C2=A0instead=C2=A0of=C2=A0obstructing?=C2=A0Please >>=C2=A0= =C2=A0keep=C2=A0in=C2=A0 >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0mind=C2= =A0ports=C2=A0is=C2=A0a=C2=A0joint=C2=A0/=C2=A0community=C2=A0effort=C2= =A0which=C2=A0is=C2=A0why=C2=A0we=C2=A0have >>=C2=A0=C2=A0groups,=C2= =A0 >>=C2=A0=C2=A0 >>=C2=A0=C2=A0=C2=A0guidelines=C2=A0etc=C2=A0and= =C2=A0for=C2=A0custom=C2=A0trees=C2=A0there's=C2=A0overlay=C2=A0support= =C2=A0for >>=C2=A0=C2=A0who=C2=A0want=C2=A0 >>=C2=A0=C2=A0 >>=C2= =A0=C2=A0=C2=A0or=C2=A0need=C2=A0to=C2=A0diverge. >=C2=A0 >=C2=A0=C2= =A0Not=C2=A0contradictory=C2=A0at=C2=A0all.=C2=A0autotools=C2=A0has=C2= =A0and=C2=A0remains=C2=A0the=C2=A0default=C2=A0and >=C2=A0thus=C2=A0 >= =C2=A0 >=C2=A0the=C2=A0most=C2=A0supported=C2=A0option=C2=A0for=C2=A0our= =C2=A0case.=C2=A0If=C2=A0there=C2=A0is=C2=A0any=C2=A0perceived=C2=A0 >= =C2=A0 >=C2=A0obstruction,=C2=A0it=C2=A0is=C2=A0on=C2=A0upstream=C2=A0dec= laring=C2=A0that=C2=A0another=C2=A0option=C2=A0is >=C2=A0equally=C2=A0 = >=C2=A0 >=C2=A0viable=C2=A0and=C2=A0supported=C2=A0for=C2=A0our=C2=A0case= ,=C2=A0in=C2=A0a=C2=A0*release*. >=C2=A0 >=C2=A0Community=C2=A0effort= =C2=A0goes=C2=A0multiple=C2=A0ways,=C2=A0but=C2=A0the=C2=A0buck=C2=A0still= =C2=A0stops=C2=A0with >=C2=A0both=C2=A0 >=C2=A0 >=C2=A0us=C2=A0and= =C2=A0whatever=C2=A0upstream=C2=A0we=C2=A0port/deal=C2=A0with.=C2=A0Users= =C2=A0in=C2=A0particular >=C2=A0could=C2=A0be=C2=A0 >=C2=A0 >=C2= =A0reporting=C2=A0bugs=C2=A0and=C2=A0other=C2=A0issues=C2=A0here=C2=A0or= =C2=A0upstream.=C2=A0We=C2=A0may=C2=A0not=C2=A0have=C2=A0 >=C2=A0 >= =C2=A0upstream's=C2=A0full=C2=A0picture=C2=A0of=C2=A0warts=C2=A0and=C2= =A0considerations;=C2=A0they=C2=A0will >=C2=A0certainly=C2=A0 >=C2= =A0 >=C2=A0not=C2=A0have=C2=A0our=C2=A0(or=C2=A0any=C2=A0other=C2=A0opera= ting=C2=A0system=C2=A0distribution's)=C2=A0full=C2=A0 >=C2=A0 >=C2= =A0picture.=C2=A0Especially=C2=A0with=C2=A0something=C2=A0reported=C2=A0ups= tream=C2=A0that=C2=A0is >=C2=A0specific=C2=A0to=C2=A0 >=C2=A0 >=C2= =A0our=C2=A0implementation,=C2=A0if=C2=A0it=C2=A0is=C2=A0because=C2=A0the= =C2=A0port/package=C2=A0from=C2=A0main=C2=A0uses >=C2=A0a=C2=A0 >=C2= =A0 >=C2=A0method=C2=A0different=C2=A0than=C2=A0that=C2=A0prescribed= =C2=A0for=C2=A0us,=C2=A0it=C2=A0becomes=C2=A0our >=C2=A0problem!=C2=A0 = >=C2=A0 >=C2=A0This=C2=A0is=C2=A0the=C2=A0liability. >=C2=A0 >=C2= =A0You=C2=A0can=C2=A0repeat=C2=A0the=C2=A0word=C2=A0or=C2=A0meaning=C2= =A0of=C2=A0"collaboration"=C2=A0until=C2=A0blue=C2=A0in >=C2=A0the=C2= =A0 >=C2=A0 >=C2=A0face,=C2=A0but=C2=A0the=C2=A0reality=C2=A0in=C2= =A0the=C2=A0world=C2=A0of=C2=A0spare=C2=A0time/effort=C2=A0is=C2=A0limited.= >=C2=A0Yes,=C2=A0 >=C2=A0 >=C2=A0anyone=C2=A0can=C2=A0use=C2=A0the= =C2=A0other=C2=A0method=C2=A0in=C2=A0overlays=C2=A0and=C2=A0whatnot,=C2= =A0but=C2=A0the >=C2=A0main=C2=A0 >=C2=A0 >=C2=A0ports=C2=A0tree= =C2=A0needs=C2=A0to=C2=A0be=C2=A0kept=C2=A0production-ready=C2=A0as=C2= =A0much=C2=A0as=C2=A0possible, >=C2=A0and=C2=A0 >=C2=A0 >=C2=A0part= =C2=A0of=C2=A0that=C2=A0means=C2=A0respecting=C2=A0upstream's=C2=A0wishes= =C2=A0as=C2=A0much=C2=A0as >=C2=A0practicable=C2=A0 >=C2=A0 >=C2= =A0so=C2=A0that=C2=A0we=C2=A0as=C2=A0a=C2=A0community=C2=A0(especially= =C2=A0including=C2=A0users)=C2=A0stay=C2=A0on=C2=A0good=C2=A0 >=C2=A0 >= =C2=A0terms=C2=A0with=C2=A0upstream.=C2=A0We=C2=A0can=C2=A0upstream=C2= =A0stuff=C2=A0that=C2=A0they=C2=A0accept=C2=A0to=C2=A0our=C2=A0 >=C2= =A0 >=C2=A0hearts'=C2=A0content,=C2=A0but=C2=A0upstream=C2=A0still=C2= =A0needs=C2=A0to=C2=A0explicitly=C2=A0acknowledge >=C2=A0them=C2=A0 >= =C2=A0 >=C2=A0as=C2=A0production-ready=C2=A0before=C2=A0we=C2=A0use=C2= =A0them=C2=A0in=C2=A0our=C2=A0main=C2=A0tree.=C2=A0Using=C2=A0a=C2=A0 >= =C2=A0 >=C2=A0different=C2=A0method=C2=A0than=C2=A0what=C2=A0upstream= =C2=A0declared=C2=A0for=C2=A0us=C2=A0to=C2=A0use=C2=A0is=C2=A0not=C2=A0 >= =C2=A0 >=C2=A0exactly=C2=A0respecting=C2=A0their=C2=A0wishes. >=C2= =A0 >=C2=A0--=C2=A0 >=C2=A0 >=C2=A0Charlie=C2=A0Li >=C2=A0 >= =C2=A0...nope,=C2=A0still=C2=A0don't=C2=A0have=C2=A0an=C2=A0exit=C2=A0line.= Hi, It's a collaboration and seemingly multiple people seem to a= gree both within and outside the project on that given the engagement, yo= u seem to disagree on that which is a bit concerning. Everything is a l= iability like that including even updating and everything is best effort.= But again, if you or others want to start to upstream Autotools fixes et= c by all means go ahead but has yet to occur. If we can work with upstrea= m and reduce the amount of local patches, hacks and have clean Makefiles = that's a win for everyone irregardless of build system. Best regards= , Daniel --_=_swift_1705177202_df9c4f792d380b0d1dd7eb170b2cf2bb_=_ Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <div>On 2024-01-13T20:27:45.000+01:00, Charlie Li <vishwin@freebsd.org&g= t; wrote:<br></div><div class=3D"ik_mail_quote answerContentMessage"><block= quote class=3D"ws-ng-quote"><pre style=3D"white-space: normal;"><div>Daniel= Engberg wrote:<br></div><blockquote class=3D"ws-ng-quote"><div> On 2024-0= 1-13T18:46:34.000+01:00, Charlie Li <<a href=3D"mailto:vishwin@freebsd.o= rg" class=3D"defaultMailLink">vishwin@freebsd.org</a>> wrote:<br></div><= div> The Build Instructions section of the README specifically states<b= r></div><div> "Autotools (for POSIX systems like Linux, BSD, macOS)", e= ven in the<br></div><div> latest trunk. CMake was originally added main= ly for Windows support.<br></div><div> Their CMake support on platforms= like ours still has an outstanding bug<br></div><div> pertaining to de= pendency resolution, which at least I consider a<br></div><div> showsto= pper. While upstream have been accepting and responsive to any<br></div><di= v> and all improvements to their CMake support, that is irrelevant unti= l<br></div><div> they explicitly bless it as an equal to autotools. (No= t to say<br></div><div> autotools<br></div><div> is perfect either,= far from it)<br></div><div> <br></div><div> Personally, between the tw= o choices here, I prefer CMake. But personal<br></div><div> preferences= are irrelevant wrt liability and support issues.<br></div><div> <br></div>= <div> <br></div><div> Hi,<br></div><div> <br></div><div> That's a bit contr= actionary? We upstream patches, community and others <br></div><div> also d= o and yet there's a "support issue" despite autotools is "far from <br></di= v><div> it" (perfect)? Why not embrace instead of obstructing? Please keep = in <br></div><div> mind ports is a joint / community effort which is why we= have groups, <br></div><div> guidelines etc and for custom trees there's o= verlay support for who want <br></div><div> or need to diverge.<br></div><d= iv> <br></div></blockquote><div> Not contradictory at all. autotools has an= d remains the default and thus <br></div><div>the most supported option for= our case. If there is any perceived <br></div><div>obstruction, it is on u= pstream declaring that another option is equally <br></div><div>viable and = supported for our case, in a *release*.<br></div><div><br></div><div>Commun= ity effort goes multiple ways, but the buck still stops with both <br></div= ><div>us and whatever upstream we port/deal with. Users in particular could= be <br></div><div>reporting bugs and other issues here or upstream. We may= not have <br></div><div>upstream's full picture of warts and consideration= s; they will certainly <br></div><div>not have our (or any other operating = system distribution's) full <br></div><div>picture. Especially with somethi= ng reported upstream that is specific to <br></div><div>our implementation,= if it is because the port/package from main uses a <br></div><div>method d= ifferent than that prescribed for us, it becomes our problem! <br></div><di= v>This is the liability.<br></div><div><br></div><div>You can repeat the wo= rd or meaning of "collaboration" until blue in the <br></div><div>face, but= the reality in the world of spare time/effort is limited. Yes, <br></div><= div>anyone can use the other method in overlays and whatnot, but the main <= br></div><div>ports tree needs to be kept production-ready as much as possi= ble, and <br></div><div>part of that means respecting upstream's wishes as = much as practicable <br></div><div>so that we as a community (especially in= cluding users) stay on good <br></div><div>terms with upstream. We can upst= ream stuff that they accept to our <br></div><div>hearts' content, but upst= ream still needs to explicitly acknowledge them <br></div><div>as productio= n-ready before we use them in our main tree. Using a <br></div><div>differe= nt method than what upstream declared for us to use is not <br></div><div>e= xactly respecting their wishes.<br></div><div><br></div><div>-- <br></div><= div>Charlie Li<br></div><div>...nope, still don't have an exit line.<br></d= iv></pre></blockquote></div><div>Hi,<br></div><div><br></div><div>It's a co= llaboration and seemingly multiple people seem to agree both within and out= side the project on that given the engagement, you seem to disagree on that= which is a bit concerning. Everything is a liability like that including e= ven updating and everything is best effort. But again, if you or others wan= t to start to upstream Autotools fixes etc by all means go ahead but has ye= t to occur. If we can work with upstream and reduce the amount of local pat= ches, hacks and have clean Makefiles that's a win for everyone irregardless= of build system.<br></div><div><br></div><div>Best regards,<br></div><div>= Daniel<br></div> --_=_swift_1705177202_df9c4f792d380b0d1dd7eb170b2cf2bb_=_--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d6023e74db7025fcf6ff4e8110491ec3>