Thanks! Yes, the syntax design for the core of ATS3 is mostly done at this stage. The comments may not be up-to-date but should not be far off. Here are a few notes:
Keywords for initiating a declaration usually start with the symbol # For instance, #infixl (50) + - #include "a-path-to-a-file" A function-like in ATS is something like a function but not a function. For example, $delay, $raise, $extcode, .... As you may have guessed correctly, the name for a function-like usually starts with the symbol $. Pointed brackets are now used to group parameters for templates. For instance, fun<a:type><b,c:tbox> foo<a><b.c>(...) = ... 'type' in ATS3 is for 't@ype' in ATS2; 'tbox' for 'type'. And annother name for 'type' is 'tflt' / 'tflat'. I assume that you will provide some documentation for language-xats :) This will be a BIG help for people interested in implementing things like emacs modes. Cheers! On Sat, Oct 20, 2018 at 10:35 AM Vanessa McHale <vanessa.mch...@iohk.io> wrote: > language-xats is being written by reading the source code - that of the > lexer and that of the parser. The parser has some comments which I > assume are a notation like BNF, e.g. > > https://github.com/githwxi/ATS-Xanadu/blob/master/srcgen/xats/SATS/parsing.sats#L349 > > On 10/20/18 1:02 AM, Kiwamu Okabe wrote: > > On Sat, Oct 20, 2018 at 1:42 PM Vanessa McHale <vanessa.mch...@iohk.io> > wrote: > >> Cross-compilation is in decent shape even with ATS2, at least on Debian. > > Thanks. I hope ATS3 also do that. > > > >> I have built binaries for many platforms you can imagine, including > >> Redox https://github.com/vmchale/illiterate/releases > > Wow! Literate programming for ATS!!! > > I believe you are the best architect of ATS ecosystem. Best balance. > > > >> As for package management - that I hope to handle (in part) myself. I > >> have a much better grasp of ATS now and so I can read the canonical > >> parser/lexer and should be able to write a better language library: > >> https://github.com/vmchale/language-xats > > Are you writing above library without any documentation? Just reading > > Hongwei's code? > > I think your parser is useful for me to understand ATS3 language. > > > >> The dependency resolution (and some of the cross-compiling/caching edge > >> cases) for atspkg are broken, but I think I can fix much of that when > >> writing xpkg :) > > Thanks. Great. > > > >> As for an assembler backend - that may be out of my area of expertise > >> but there are relatively few places where GCC/C is not expressive > >> enough. I'd love to see ATS genuinely beat C, but the opportunities for > >> doing so seem sparse - only FORTRAN and C++ (or maybe J) do this > reliably. > > I also like C backend. My motivation is based on waning debugable binary > > with ATS3 such using DWARF. Such assembler backend with LLVM may be > > useful than swapping C's DWARF with own ATS own DWARF. Maybe. > > > > Best, > > -- > > Kiwamu Okabe at METASEPI DESIGN > > > > -- > You received this message because you are subscribed to the Google Groups > "ats-lang-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to ats-lang-users+unsubscr...@googlegroups.com. > To post to this group, send email to ats-lang-users@googlegroups.com. > Visit this group at https://groups.google.com/group/ats-lang-users. > To view this discussion on the web visit > https://groups.google.com/d/msgid/ats-lang-users/b7750359-e809-7c58-5c5e-9c5f5c047ba6%40iohk.io > . > -- You received this message because you are subscribed to the Google Groups "ats-lang-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to ats-lang-users+unsubscr...@googlegroups.com. To post to this group, send email to ats-lang-users@googlegroups.com. Visit this group at https://groups.google.com/group/ats-lang-users. To view this discussion on the web visit https://groups.google.com/d/msgid/ats-lang-users/CAPPSPLoRykaR5WH%2BJ%3DRuZpFnw1mf2Hwucg%3Dgw7eCuaSAWRF1ZQ%40mail.gmail.com.
