| ID: | 5732 |
| From: | Erich Heinzle <a1...@gmail.com> |
| Date: | Sat, 9 Jul 2022 14:28:52 +0930 |
| Subject: | Re: [pcb-rnd] file name policy change proposal |
| in-reply-to: | 5731 from rn...@igor2.repo.hu |
--00000000000095341705e3582df0 Content-Type: text/plain; charset="UTF-8" Of tgese, rch/rsc/rfp/rsy seem ok in the same way I will look at files in a design directory and know what foo.GBL foo.txt foo.sym foo.fp and foo.GTO all contain, it remains very helpful to have file endings which distinguish schematic and pcb layouts, even if they are both in lihata format, to avoid accidentally saving a layout on top of/overwriting a pcb layout, which is easy to do if .lht is the default file ending for both formats. I noticed this when jumping between sch-rnd and pcb-rnd tge other day, and being prompted to use .lht as a file ending or to load tjings ending in .lht so the argument for dedicated file endings is a human factors/safety argument as much as anything, which also applies to collaboration regards, Erich On Sat, 9 Jul 2022 14:13 , <rnd2@igor2.repo.hu> wrote: > > > On Fri, 8 Jul 2022, Bdale Garbee wrote: > > >rnd2@igor2.repo.hu writes: > > > >> In what situation would you name your initial file foo.pcb when running > >> pcb-rnd instead of foo.lht knowing your Makefile expects foo.lht for > >> lihata? > > > >I wouldn't. > > > >> Or are you trying to create a geda/pcb file using pcb-rnd? > > > >Nope. > > > >My issue is that I have a lot of existing .sch and .pcb files in my > >design tree, where those suffixes currently have meaning relative to the > >content. While I could of course customize things any way I want, if > >you change the default behavior of the ringdove suite tools to generate > > You wouldn't need to customize or change anything for your existing > projects. > > >.sch and .pcb files with lihata content, I'm confident I'll end up > >dealing with files that have those extensions and lihata content one way > >or another .. borrowing designs, trying to collaborate with others, > >whatever. > > Ok, collaborating is a good point, if you had exisitng Makefiles assuming > a certain naming convention and others use a different naming convention > you surely would need to rename those files on import. > > >I'm certainly not married to .lht as an extension for PCB designs .. the > >total number of boards I've created in pcb-rnd to date would take me > >only a few minutes to rename to match whatever your new default is, > >updating my "harness logic" as needed to match. I'd do that work > >because I like using tools as close to their "default" behavior as > >possible since a lot of what I work on is done in collaboration with > >others and any time we customize at this sort of level, it's just making > >more work in that context. > > > >So .. what I hope for is that the default file name extensions for > >ringdove are "unique" at least as far as avoiding conflict with any > >existing tools that I, or others already in your user community, have > built > >infrastructure dependencies around, so that I can happily use the > >ringdove default extensions for the designs I use ringdove tools for, > >while keeping my big tree of existing design data around... and never > >(as a human looking at files in directories) be confused about what is > >what. > > Ok, so we can't identify any specific technical detail that would break, > but you want to both stick to the new default file names and your existing > file names. > > That sounds reasonable as user preference. I wonder if we can find a > shorter variant, because my user preference is not to make file names > longer. Of coure this would sacrifice the "natural" part, but it seems we > need to sacrifice something. > > > What about these: > > 1. rcb/rch; symbols could be rym, footprints rfp. So r + second char + > last char of the original, sort of replacing first char instead of > inserting one. So basically you type the same thing, just replace first > char with r. Except for the footprint where we insert r before fp. > > 2. rpb/rsh; symbols could be rsm, footprints rfp. Same idea, but r + first > char + last char > > 3. rp/rs; symbols could be ry, footprints rf. So two chars only, r for > ringdove, then one for the purpose. If it's non-natural, "unreadable" > anyway, why not make it also short? > > 4. bor/sem for board/schematics, sbl/fpt for symbol and footprint (so > using totally different characters/terms only loosely connected - > guaranteed it won't match anything you ever used in this context) > > 5. pbd/sms for Printed (circuit) BoarD and ScheMaticS, sbl/fpt for symbol > and footprint (first char, a middle char and last char, consistently) > > > Best regards, > > Igor2 > > > > --00000000000095341705e3582df0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"auto">Of tgese, rch/rsc/rfp/rsy seem ok<div dir=3D"auto"><br></= div><div dir=3D"auto">in the same way I will look at files in a design dire= ctory and know what</div><div dir=3D"auto"><br></div><div dir=3D"auto">foo.= GBL</div><div dir=3D"auto">foo.txt</div><div dir=3D"auto">foo.sym</div><div= dir=3D"auto">foo.fp</div><div dir=3D"auto">and foo.GTO</div><div dir=3D"au= to"><br></div><div dir=3D"auto">all contain,=C2=A0 it remains very helpful = to have file endings which distinguish schematic and pcb layouts,=C2=A0 eve= n if they are both in lihata format,=C2=A0 to avoid accidentally saving a l= ayout on top of/overwriting a pcb layout,=C2=A0 which is easy to do if .lht= is the default file ending for both formats. I noticed this when jumping b= etween sch-rnd and pcb-rnd tge other day,=C2=A0 and being prompted to use .= lht as a file ending or to load tjings ending in .lht</div><div dir=3D"auto= "><br></div><div dir=3D"auto">so the argument for dedicated file endings is= a human factors/safety argument as much as anything, which also applies to= collaboration</div><div dir=3D"auto"><br></div><div dir=3D"auto">regards,= =C2=A0</div><div dir=3D"auto"><br></div><div dir=3D"auto">Erich</div><div d= ir=3D"auto"><br></div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr"= class=3D"gmail_attr">On Sat, 9 Jul 2022 14:13 , <<a href=3D"mailto:rnd2= @igor2.repo.hu">rnd2@igor2.repo.hu</a>> wrote:<br></div><blockquote clas= s=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;pad= ding-left:1ex"><br> <br> On Fri, 8 Jul 2022, Bdale Garbee wrote:<br> <br> ><a href=3D"mailto:rnd2@igor2.repo.hu" target=3D"_blank" rel=3D"noreferr= er">rnd2@igor2.repo.hu</a> writes:<br> ><br> >> In what situation would you name your initial file foo.pcb when ru= nning <br> >> pcb-rnd instead of foo.lht knowing your Makefile expects foo.lht f= or <br> >> lihata?<br> ><br> >I wouldn't.<br> ><br> >> Or are you trying to create a geda/pcb file using pcb-rnd?<br> ><br> >Nope. <br> ><br> >My issue is that I have a lot of existing .sch and .pcb files in my<br> >design tree, where those suffixes currently have meaning relative to th= e<br> >content.=C2=A0 While I could of course customize things any way I want,= if<br> >you change the default behavior of the ringdove suite tools to generate= <br> <br> You wouldn't need to customize or change anything for your existing <br= > projects.<br> <br> >.sch and .pcb files with lihata content, I'm confident I'll end= up<br> >dealing with files that have those extensions and lihata content one wa= y<br> >or another .. borrowing designs, trying to collaborate with others,<br> >whatever.<br> <br> Ok, collaborating is a good point, if you had exisitng Makefiles assuming <= br> a certain naming convention and others use a different naming convention <b= r> you surely would need to rename those files on import.<br> <br> >I'm certainly not married to .lht as an extension for PCB designs .= . the<br> >total number of boards I've created in pcb-rnd to date would take m= e<br> >only a few minutes to rename to match whatever your new default is,<br> >updating my "harness logic" as needed to match.=C2=A0 I'd= do that work<br> >because I like using tools as close to their "default" behavi= or as<br> >possible since a lot of what I work on is done in collaboration with<br= > >others and any time we customize at this sort of level, it's just m= aking<br> >more work in that context.<br> ><br> >So .. what I hope for is that the default file name extensions for<br> >ringdove are "unique" at least as far as avoiding conflict wi= th any<br> >existing tools that I, or others already in your user community, have b= uilt<br> >infrastructure dependencies around, so that I can happily use the<br> >ringdove default extensions for the designs I use ringdove tools for,<b= r> >while keeping my big tree of existing design data around... and never<b= r> >(as a human looking at files in directories) be confused about what is<= br> >what. <br> <br> Ok, so we can't identify any specific technical detail that would break= , <br> but you want to both stick to the new default file names and your existing = <br> file names. <br> <br> That sounds reasonable as user preference. I wonder if we can find a <br> shorter variant, because my user preference is not to make file names <br> longer. Of coure this would sacrifice the "natural" part, but it = seems we <br> need to sacrifice something.<br> <br> <br> What about these:<br> <br> 1. rcb/rch; symbols could be rym, footprints rfp. So r + second char + <br> last char of the original, sort of replacing first char instead of <br> inserting one. So basically you type the same thing, just replace first <br= > char with r. Except for the footprint where we insert r before fp.<br> <br> 2. rpb/rsh; symbols could be rsm, footprints rfp. Same idea, but r + first = <br> char + last char<br> <br> 3. rp/rs; symbols could be ry, footprints rf. So two chars only, r for <br> ringdove, then one for the purpose. If it's non-natural, "unreadab= le" <br> anyway, why not make it also short?<br> <br> 4. bor/sem for board/schematics, sbl/fpt for symbol and footprint (so <br> using totally different characters/terms only loosely connected - <br> guaranteed it won't match anything you ever used in this context)<br> <br> 5. pbd/sms for Printed (circuit) BoarD and ScheMaticS, sbl/fpt for symbol <= br> and footprint (first char, a middle char and last char, consistently)<br> <br> <br> Best regards,<br> <br> Igor2<br> <br> <br> <br> </blockquote></div> --00000000000095341705e3582df0--
Reply subtree:
5732 Re: [pcb-rnd] file name policy change proposal from Erich Heinzle <a1...@gmail.com>