diff options
Diffstat (limited to 'tnslc/parse')
-rw-r--r-- | tnslc/parse/parse.tnsl | 56 | ||||
-rw-r--r-- | tnslc/parse/token.tnsl | 56 | ||||
-rw-r--r-- | tnslc/parse/tokenizer.tnsl | 17 |
3 files changed, 70 insertions, 59 deletions
diff --git a/tnslc/parse/parse.tnsl b/tnslc/parse/parse.tnsl index 5d3bfa5..9927d56 100644 --- a/tnslc/parse/parse.tnsl +++ b/tnslc/parse/parse.tnsl @@ -14,12 +14,64 @@ EXPRESS OR IMPLIED #/ -/; module parse +/; export module parse :include "parse/token.tnsl" :include "parse/tokenizer.tnsl" ;/ -/; print_tokens(~{}parse.Token dat) +/# The various types of tokens #/ +; enum TOKEN_TYPE [int] { + LINESEP = 0, + INLNSEP = 1, + DELIMIT = 2, + AUGMENT = 3, + LITERAL = 4, + KEYTYPE = 5, + PREWORD = 6, + KEYWORD = 7, + DEFWORD = 8 +} + +/# Token struct definition #/ +; struct Token { + int + token_type, + line, + col, + + ~{}charp + data +} + +/; method Token + + /; print + ;tnsl.io.print("{ ") + ;tnsl.io.print(self.token_type) + ;tnsl.io.print(" ") + ;tnsl.io.print(self.data`) + ;tnsl.io.print(" ") + ;tnsl.io.print(self.line) + ;tnsl.io.print(" ") + ;tnsl.io.print(self.col) + ;tnsl.io.print(" } ") + ;/ + + /; operator delete + ;delete self.data + ;/ + + /; add_char (~{}charp part) + # ;uint l = len self.data` + # ;realloc self.data, l + len part + /;loop (int i = 0; i < len part`) [i++] + # ;self.data`{l + i} = part{i} + ;self.data`.append(part`{i}) + ;/ + ;/ +;/ + +/; print_tokens(~{}Token dat) /;loop (int i = 0; i < len dat`) [i++] ;dat`{i}.print() ;/ diff --git a/tnslc/parse/token.tnsl b/tnslc/parse/token.tnsl index ce2bcdb..92748d8 100644 --- a/tnslc/parse/token.tnsl +++ b/tnslc/parse/token.tnsl @@ -14,58 +14,6 @@ EXPRESS OR IMPLIED #/ -/# The various types of tokens #/ -; enum TOKEN_TYPE [int] { - LINESEP = 0, - INLNSEP = 1, - DELIMIT = 2, - AUGMENT = 3, - LITERAL = 4, - KEYTYPE = 5, - PREWORD = 6, - KEYWORD = 7, - DEFWORD = 8 -} - -/# Token struct definition #/ -; struct Token { - int - token_type, - line, - col, - - ~{}charp - data -} - -/; method Token - - /; print - ;tnsl.io.print("{ ") - ;tnsl.io.print(self.token_type) - ;tnsl.io.print(" ") - ;tnsl.io.print(self.data`) - ;tnsl.io.print(" ") - ;tnsl.io.print(self.line) - ;tnsl.io.print(" ") - ;tnsl.io.print(self.col) - ;tnsl.io.print(" } ") - ;/ - - /; operator delete - ;delete self.data - ;/ - - /; add_char (~{}charp part) - # ;uint l = len self.data` - # ;realloc self.data, l + len part - /;loop (int i = 0; i < len part`) [i++] - # ;self.data`{l + i} = part{i} - ;self.data`.append(part`{i}) - ;/ - ;/ -;/ - /# Reserved words and characters, as well as helper funcs for checking their token types. @@ -142,6 +90,7 @@ "raw", "asm", "inline", + "virtual", "delete", @@ -151,7 +100,8 @@ ;{}{}charp LITERALS = { "true", - "false" + "false", + "null" } ;{}charp RESERVED = "`~!#%^&*()-=+[]{}|;:,.<>/" diff --git a/tnslc/parse/tokenizer.tnsl b/tnslc/parse/tokenizer.tnsl index 139877c..3a66e24 100644 --- a/tnslc/parse/tokenizer.tnsl +++ b/tnslc/parse/tokenizer.tnsl @@ -21,41 +21,50 @@ /; break_token ({}charp dat, charp c) [bool] /; if (len dat == 0) ;return false + ;; else if (dat{0} == '"' || dat{0} == '\'') ;return string_closed(dat, c) + ;; else if (is_in_string(~RESERVED, dat{len dat - 1})) + /; if (is_in_string(~RESERVED, c)) ;dat.append(c) ;return get_token_type(~dat) == TOKEN_TYPE.DEFWORD + ;; else if (len dat == 1 && dat{0} == '.') ;return !is_digit(c) + ;/ + ;return true + ;; else if (is_in_string(~RESERVED, c)) + /; if (is_numeric_literal(~dat) && !is_float(~dat) && c == '.') ;return false + ;/ + ;return true ;/ + ;return is_whitespace(c) ;/ /; strip_and_expand (~{}Token dat) [{}Token] ;{}Token out = {} - ;{}charp cbst = "/#", cben = "#/" - ;bool cblk = false /; loop (int i = 0; i < len dat`) [i++] /; if (!cblk) - /; if (string_equate(dat`{i}.data, ~cbst)) + /; if (string_equate(dat`{i}.data`, "/#")) ;cblk = true ;; else ;out.append(dat`{i}) ;/ - ;; else if (string_equate(dat`{i}.data, ~cben)) + ;; else if (string_equate(dat`{i}.data`, "#/")) ;cblk = false ;/ ;/ |