On Fri, Dec 25, 2015 at 12:13 PM, John Doty <jpd AT noqsi DOT com> = wrote:

On Dec 25, 2015, at 1:38 PM, Britton Kerin (britton DOT kerin AT gmail DOT com) = [via geda-user AT d= elorie.com] <geda-user AT delorie DOT com> wrote:

On Thu, Dec 24, 2015 at 1:39 PM, John Doty <jpd AT noqsi DOT com> = wrote:

On Dec 24, 2015, at 12:53 PM, Britton Kerin (britton DOT kerin AT gmail DOT c= om) [via ged= a-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:

Agreed.=C2=A0 = I like YAML for this reason.=C2=A0 You get a parser in every language for f= ree, without any other library material required.
=
AWK? sed? grep? cut? sort?

Why not?=C2=A0 Unless a newline in your regex really frightens you = that much...
=C2=A0
Records separated by newline with fields separa= ted by whitespace is *better* supported than YAML or any of the other candi= dates mentioned. The only things it lacks for our purposes is a spiffy name= and the need for extra layers of lasagna code.

=
Fine.=C2=A0 Put your money where your mouth is.=C2=A0 Send me a = full parser for pcb files with binding for perl, python, ruby, C, and virtu= ally every other extant language.
=
I=E2=80=99m not talking about pcb. I=E2=80=99m talking about geda= -gaf

=C2=A0

Geda-gaf is a muc= h cleaner design. The pcb format doesn=E2=80=99t even represent a proper mo= del of what a printed circuit board is.

Writing a = parser for geda-gaf files is simpler than penetrating an API wrapping such = a parser. That doesn=E2=80=99t mean you can do either in an instant.
<= /div>

Highly unlikely for .sch, = no chance whatsoever for pcb.=C2=A0 We're talking about one function ca= ll in each direction form YAML/JSON.=C2=A0 Not that I would advocate doing = anything to sch format.

It's easy so you s= hould be able to get it to me BEFORE your next bs email.
<= /div>
If having common parser makes things so easy, you should = be able to document API=E2=80=99s

<= div style=3D"">I thought you were the one who thought accessible file forma= ts were a good idea, and bundling big APIs around them not necessarily?=C2= =A0 So there's nothing to document.=C2=A0 Parsers are relatively easy b= ut as you admit not exactly instantaneous, so why not use a setup that obvi= ates the need for them? The major difference for people wanting to parse th= ings themselves is that the fields have names, so you just read the name ra= ther than digging through the spec or running a little experiment to figure= out which positional field is which.=C2=A0 In other words its easier for t= hat purpose too.

Using an ac= cessible format is entirely in keeping with the toolbox philosophy which yo= u normally advocate so strongly for in gEDA.=C2=A0 Why should pcb be the on= ly program that can easily fully parse pcb files?

Britton