moving sfc over here

2026-06-01 16:44:33 -07:00
parent 177978d184
commit f548c7d88d
15 changed files with 3534 additions and 19 deletions
@@ -0,0 +1,161 @@
+% This is a header file that contains sfc's record types
+%
+% This is in order to
+%   1. share records across modules; and,
+%   2. allow external modules to just use the sfc records
+
+%-------------------------------------------------------
+% API Types: sfc internal token representation
+%
+% -export_type([
+%     sf_token_type/0
+%     sf_token/0
+% ]).
+%-------------------------------------------------------
+
+% 
+-type sfc_token_type()
+    :: bcom      % /* ... */
+     | lcom      % //
+     | ws        % whitespace
+     % literals
+     | char      % 'a'
+     | string    % "foo"
+     | int10     % 69_420
+     | int16     % 0xDEAD_BEEF
+     | bytes     % #DEAD_BEEF
+     | ak        % ak_ABC
+     | ct        % ct_ABC
+     | sg        % sg_ABC
+     % kwds/variables/etc
+     | id         % foo, foo_bar, foo_bar'baz'  _'foo'
+     | con        % Foo, Foo_Bar, FooBar
+     | qid        % Foo.Bar.baz
+     | qcon       % Foo.Bar.Baz
+     | tvar       % 'foo, 'foo_bar, '_'foo'_'bar'''
+     % kwds ops and punct are all collapsed by
+     % so_scan:scan down to eg {'contract', {420, 69}}
+     % where {420, 69} is the source location
+     % these are three different parsers
+     | kwd        % contract, interface, payable, etc
+     | op         % "=!<>+-*/:&|?~@^"
+     | punct      % ".." | oneof(",.;()[]{}")
+     % kwds and punct are kind of the same thing
+     % but i'll keep them separate now for my own sanity. ok
+     % i guess op or symbol or whatever is fine.
+     %
+     % not going to overthink. if having them separate
+     % becomes an issue it's easy enough to collapse. harder
+     % to separate afterward if collapsing is wrong.
+     .
+
+
+-type sfc_pos() :: {Line :: pos_integer(), Col :: pos_integer()}.
+
+-record(sfc_token,
+        {type   :: sfc_token_type(),
+         pos    :: sfc_pos(),
+         string :: string()}).
+
+-type sfc_token() :: #sfc_token{}.
+
+
+% tokens are in essence the "chunk boundaries" of
+% the file
+%
+% because we have semantic whitespace, we have to be
+% careful about block declarations, because we don't
+% have an explicit open/close block token. blocks can
+% be closed in one of two ways:
+%
+%   1. a new block at a previous indent level:
+%           switch(foo)
+%               // block starts here
+%               Bar => bar()
+%               Baz => baz()
+%           quux()      // ends because indent level
+%   2. it's part of some type of list:
+%
+%           [switch(foo)
+%               // block starts here
+%               Bar => bar()
+%               Baz => baz(),   // ends here
+%            switch(bizz)
+%               // block starts here
+%               Bar => bar()
+%               Baz => baz()]
+%
+% in order to avoid speculatively inserting virtual
+% close tokens, at least on first write-out, we're
+% going to disambiguate list notions right away
+
+% token groups
+% lists = (_, _, _)
+%       | [_, _, _]
+%       | {_, _, _}
+%-record(sfc_ast1_block,
+%        {indent = none :: none | pos_integer(),
+%         decls  = none :: [sfc_ast1_decl()]}).
+%
+%-type sfc_ast() :: 
+%
+%-type sfc_list_group() :: {'(', [sfc_token()], ')'}
+%                        | {'[', [sfc_token()], ']'}
+%                        | {'{', [sfc_token()], '}'}
+%                        | {proof, 
+%                        .
+
+
+
+% @doc
+% this one is very specific so it deserves its own
+% record type: unterminated block comments at the end
+% of files. these are ok in legacy sophia, so we have to
+% specifically account for this error
+-record(sfc_err_bcom_unterminated,
+        {prev_tokens :: [sfc_token()],
+         break_pos    :: sfc_pos(),
+         rest         :: string()}).
+
+-record(sfc_err_no_tokmatch,
+        {prev_tokens :: [sfc_token()],
+         break_pos    :: sfc_pos(),
+         rest         :: string()}).
+
+
+-record(sfc_err_delims,
+        {past       :: [sfc_token()],
+         open_stack :: [sfc_token()],
+         bad_close  :: sfc_token(),
+         future     :: [sfc_token()]}).
+% FIXME
+-record(sfc_err_nyi, {}).
+-record(sfc_err_empty_file, {}).
+
+%-record(src_parse_error,
+%        {atom   = none :: none | atom(),
+%         string = 
+
+%j-record(sfc_err_gulp_ct,
+%j        {gulped :: 
+
+% @doc
+% generic placeholder error for now
+-record(sfc_err,
+        {atom           :: atom(),
+         string = none  :: none | iolist(),
+         extra  = none  :: none | any()}).
+
+% @doc all errors SFC can return conveniently listed in
+% one place
+-type sfc_err() :: #sfc_err_bcom_unterminated{}
+                 | #sfc_err_no_tokmatch{}
+                 | #sfc_err_nyi{}
+                 | #sfc_err_empty_file{}
+                 | #sfc_err{}.
+
+
+%% FIXME
+-type sfc_ast() :: any().
+
+