If you want a peek at something I've played with: https://github.com/marekr/kicad-sexpr/commit/3bf73b984eaeef3d8aede351592a2cc07677cb10
I avoid boost like the plague ^^ On Fri, Sep 9, 2016 at 10:55 AM, Wayne Stambaugh <[email protected]> wrote: > I understand completely. I've written enough of our sexpr code to know > how it works and where it's strengths and weaknesses lie. I just want > to make sure broken file formats are flagged as such by any parsers that > we write. If our output formatters are creating these files, that is an > all together different issue that needs to be addressed. > > At some point, if I ever get the time, I would like to review this. It > might be useful use a property tree like parser/formatter such as how > xml is handled or some other similar construct rather than the way we > are doing it now. I took a look at the boost::property_tree class it > looks like a possible candidate. I was hoping to do something like this > for the new schematic file format. It's primarily a issue of finding > the free time to learn something new at the moment. > > On 09/09/2016 10:38 AM, Mark Roszko wrote: >> ....the existing parser is "by the seat of our pants" style parsing >> where there is no centralized parsing logic and instead >> manual definition of expected tokens for every single parsable atom >> and well, this has led to places where expectRightParenthesis() isn't >> called and the like. But that won't break anything because it >> continues walking down the token list after the case for an atom is >> executed. >> >> Meh, I probably suck at explaining ^ >> >> Looks like I forgot to submit a patch for these files as I submitted >> one for another template file before that I think JP commited. Because >> my strict SEXPR parser barked :3 >> >> On Fri, Sep 9, 2016 at 10:19 AM, Wayne Stambaugh <[email protected]> >> wrote: >>> Hmm. I took another look at this patch and I'm concerned as to why the >>> page layout template parser did not choke on this. That should be fixed >>> as well. >>> >>> On 9/8/2016 2:56 PM, Werner Almesberger wrote: >>>> The default and logo page layout templates are missing some opening >>>> parentheses. Eeschema's parser accepts them anyway, but it tripped >>>> my s-expr parser. >>>> >>>> The gost templates and the built-in default in >>>> common/page_layout/page_layout_default_description.cpp >>>> are both correct. >>>> >>>> - Werner >>>> >>>> >>>> >>>> _______________________________________________ >>>> Mailing list: https://launchpad.net/~kicad-developers >>>> Post to : [email protected] >>>> Unsubscribe : https://launchpad.net/~kicad-developers >>>> More help : https://help.launchpad.net/ListHelp >>>> >>> >>> >>> _______________________________________________ >>> Mailing list: https://launchpad.net/~kicad-developers >>> Post to : [email protected] >>> Unsubscribe : https://launchpad.net/~kicad-developers >>> More help : https://help.launchpad.net/ListHelp >> >> >> -- Mark _______________________________________________ Mailing list: https://launchpad.net/~kicad-developers Post to : [email protected] Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
