Mercurial > hg > index.cgi

comparison src/parse.s @ 124:8770e6f977c3

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Rework parser to use parse_wordtab for symbols too There's no reason not to use the parse_wordtab table thing to match the symbols with their token codes. It takes less space than the combined code and tables to do it separately.
author William Astle <lost@l-w.ca>
date Mon, 01 Jan 2024 15:57:59 -0700
parents 5681cdada362
children 0607e4e20702
comparison
equal deleted inserted replaced
123:5681cdada362 124:8770e6f977c3
80 bsr parse_nextchar ; eat the space 80 bsr parse_nextchar ; eat the space
81 bne parse_nexttok0 ; brif not end of input 81 bne parse_nexttok0 ; brif not end of input
82 parse_nexttok1 ldb #token_eot ; flag end of input 82 parse_nexttok1 ldb #token_eot ; flag end of input
83 bra parse_nexttok6 ; go return it 83 bra parse_nexttok6 ; go return it
84 parse_nexttok2 sty parse_tokenst ; save start of current token after skipping spaces 84 parse_nexttok2 sty parse_tokenst ; save start of current token after skipping spaces
85 bsr parse_toupper ; make sure we have upper case letters for matching
86 ldx #parse_wt ; point to keyword parsing table
87 bsr parse_wordtab ; go see if we have a match in the keyword table
88 bcc parse_nexttok6 ; brif we do - return it
89 ldy parse_tokenst ; return to the start of the token - pointer probably clobbered
90 bsr parse_curchar ; get back input character (may have been clobbered)
85 cmpa #'. ; leading decimal? 91 cmpa #'. ; leading decimal?
86 beq parse_nexttok3 ; brif so - parse number 92 beq parse_nexttok3 ; brif so - parse number
87 cmpa #'0 ; is it a digit 93 cmpa #'0 ; is it a digit
88 blo parse_nexttok4 ; brif not 94 blo parse_nexttok10 ; brif not
89 cmpa #'9 ; is it still a digit? 95 cmpa #'9 ; is it still a digit?
90 bhi parse_nexttok4 ; brif not 96 bhi parse_nexttok10 ; brif not
91 parse_nexttok3 jmp parse_number ; go parse a number 97 parse_nexttok3 jmp parse_number ; go parse a number
92 parse_nexttok4 ldx #parse_chartab ; point to list of single character tokens to recognize
93 parse_nexttok5 ldb 1,x ; get token value
94 cmpa ,x++ ; character match (and move to next entry)
95 bne parse_nexttok7 ; brif not
96 parse_nexttok6 stb parse_curtok ; save token type 98 parse_nexttok6 stb parse_curtok ; save token type
97 leay 1,y ; eat the input character 99 leay 1,y ; eat the input character
98 clra ; clear C to indicate no error (and clear Z also) 100 clra ; clear C to indicate no error (and clear Z also)
99 rts 101 rts
100 parse_nexttok7 cmpb #token_eot ; end of table? 102 parse_nexttok10 cmpa #'A ; is it alpha?
101 bne parse_nexttok5 ; brif not
102 clrb ; initialize relational flags
103 pshs d ; save input character and relational flags for later
104 parse_nexttok8 cmpa #'< ; less than?
105 blo parse_nexttok9 ; brif not <, =, or >
106 cmpa #'> ; still <, =, or >?
107 bhi parse_nexttok9 ; brif not
108 suba #'< ; adjust < to 0
109 cmpa #1 ; set C if <, clear if = or >
110 rola ; now 4 if >, 2 if =, or 1 if <
111 eora 1,s ; merge with previous relational characters
112 cmpa 1,s ; if it doesn't match, we have a dupe
113 bne parse_nexttok9 ; brif it's not valid - we won't recognize more in the token
114 sta 1,s ; save new relational flags
115 bsr parse_nextchar ; fetch next input
116 sta ,s ; save input character
117 bne parse_nexttok8 ; brif there was one - go handle it
118 parse_nexttok9 puls d ; get back input character and relational flag
119 tstb ; was it a relational operator?
120 beq parse_nexttok10 ; brif not
121 ldx #parse_reltab ; point to relational operator token table
122 ldb b,x ; get the token code
123 clra ; flag no error
124 rts ; return - but don't advance - we already did looking for multiples
125 parse_nexttok10 bsr parse_toupper ; convert to upper case
126 cmpa #'A ; is it alpha?
127 blo parse_nexttok11 ; brif not 103 blo parse_nexttok11 ; brif not
128 cmpa #'Z ; is it still alpha? 104 cmpa #'Z ; is it still alpha?
129 bls parse_nexttok12 ; brif so 105 bls parse_nexttok12 ; brif so
130 parse_nexttok11 comb ; flag error - unrecognized token 106 parse_nexttok11 comb ; flag error - unrecognized token
131 ldb #token_error 107 ldb #token_error
132 rts 108 rts
109 parse_nexttok12 bsr parse_nextcharu ; fetch next input character
110 cmpa #'0 ; is it alphanumeric?
111 blo parse_nexttok13 ; brif not
112 cmpa #'9 ; is it numeric?
113 bls parse_nexttok12 ; brif so - keep skipping it
114 cmpa #'A ; is it alpha?
115 blo parse_nexttok13 ; brif not
116 cmpa #'Z ; is it still alpha?
117 bls parse_nexttok12 ; brif so - keep skipping it
118 parse_nexttok13 tfr y,d ; calculate length of identifier
119 subd parse_tokenst
120 std val0+val.strlen ; save it for reference
121 ldb #token_ident ; indicate an identifier (variable name, etc.)
122 rts ; return result (C will be clear from SUBD above)
133 parse_nextcharu bsr parse_nextchar ; fetch next input character 123 parse_nextcharu bsr parse_nextchar ; fetch next input character
134 beq parse_toupper0 ; brif end of input 124 beq parse_toupper0 ; brif end of input
135 parse_toupper cmpa #'a ; is it lower case alpha? 125 parse_toupper cmpa #'a ; is it lower case alpha?
136 blo parse_toupper0 ; brif not 126 blo parse_toupper0 ; brif not
137 cmpa #'z ; is it still lower case alpha? 127 cmpa #'z ; is it still lower case alpha?
138 bhi parse_toupper0 ; brif not 128 bhi parse_toupper0 ; brif not
139 suba #0x20 ; adjust to upper case alpha 129 suba #0x20 ; adjust to upper case alpha
140 parse_toupper0 rts ; Z only set here if input was zero entering from parse_nextcharu 130 parse_toupper0 rts ; Z only set here if input was zero entering from parse_nextcharu
141 ; We parse alpha keywords and identifiers here, of the form [a-zA-Z][a-zA-Z0-9]* with a possible nonalpha characters
142 ; in actual keywords. We use a table to parse keywords. As soon as we find a character that doesn't match a keyword
143 ; table entry, we fall back to looking for the end of an identifier and then returning that.
144 parse_nexttok12 ldx #parse_wordtab ; point to keyword table
145 bsr parse_nexttok16 ; process this table entry
146 cmpb #token_ident ; did we match a token?
147 bne parse_nexttok6 ; brif so - go return it
148 parse_nexttok13 cmpa #'0 ; was it alphanumeric?
149 blo parse_nexttok15 ; brif not
150 cmpa #'9 ; was it numeric?
151 bls parse_nexttok14 ; brif so
152 cmpa #'A ; was it alpha?
153 blo parse_nexttok15 ; brif not
154 cmpa #'Z ; is it still alpha?
155 bhi parse_nexttok15 ; brif not
156 parse_nexttok14 bsr parse_nextcharu ; fetch next character and force upper case
157 bne parse_nexttok13 ; if not end of input, see if we have alphanumeric
158 parse_nexttok15 tfr y,d ; fetch input location
159 subd parse_tokenst ; calculate length of token
160 std val0+val.strlen ; save the length of the identifier
161 ldb #token_ident ; set token type to identifier (variable name, probably)
162 rts ; return token type, do not advance since we already did above
163 ; This routine parses tokens using the table at parse_wordtab. The table is structured as follows: 131 ; This routine parses tokens using the table at parse_wordtab. The table is structured as follows:
164 ; 132 ;
165 ; * two bytes which contain the length of the table less the two bytes for this length value 133 ; * two bytes which contain the length of the table less the two bytes for this length value
166 ; * a sequence of entries consisting of a single byte matching character and a token code followed 134 ; * a sequence of entries consisting of a single byte matching character and a token code followed
167 ; by an optional sub table, structured exactly the same way. 135 ; by an optional sub table, structured exactly the same way.
201 bra parse_wordtab3 ; otherwise just move to the next entry 169 bra parse_wordtab3 ; otherwise just move to the next entry
202 parse_wordtab5 leay a,y ; move back the specified number of characters 170 parse_wordtab5 leay a,y ; move back the specified number of characters
203 parse_wordtab6 clra ; clear C to indicate a match 171 parse_wordtab6 clra ; clear C to indicate a match
204 puls a,x,pc ; clean up stack and return 172 puls a,x,pc ; clean up stack and return
205 parse_number jmp parse_tokerr 173 parse_number jmp parse_tokerr
206 ; Relational token table, bits are > = <
207 parse_reltab fcb token_error
208 fcb token_lt
209 fcb token_eq
210 fcb token_le
211 fcb token_gt
212 fcb token_ne
213 fcb token_ge
214 fcb token_reltrue
215 ; Single character token lookup table
216 parse_chartab fcb 0x21,token_bang ; !
217 fcb 0x23,token_hash ; #
218 fcb 0x24,token_dollar ; $
219 fcb 0x25,token_percent ; %
220 fcb 0x26,token_amp ; &
221 fcb 0x27,token_apos ; '
222 fcb 0x28,token_oparen ; (
223 fcb 0x29,token_cparen ; )
224 fcb 0x2a,token_star ; *
225 fcb 0x2b,token_plus ; +
226 fcb 0x2c,token_comma ; ,
227 fcb 0x2d,token_minus ; -
228 fcb 0x2f,token_slash ; /
229 fcb 0x3a,token_stmtsep ; :
230 fcb 0x3b,token_semi ; ;
231 fcb 0x3f,token_print ; ? - print shortcut
232 fcb 0x40,token_at ; @
233 fcb 0x5e,token_exp ; ^ - exponentiation
234 fcb 0x00,token_eot ; end of table flag
235 ; Parse tokens - define them in order using the macro parse_tokdef 174 ; Parse tokens - define them in order using the macro parse_tokdef
236 *pragmapush list 175 *pragmapush list
237 *pragma nolist 176 *pragma nolist
238 parse_toknum set 0 177 parse_toknum set 0
239 parse_tokdef macro noexpand 178 parse_tokdef macro noexpand