Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1 | # |
| 2 | # partparse.py: parse a by-Guido-written-and-by-Jan-Hein-edited LaTeX file, |
| 3 | # and generate texinfo source. |
| 4 | # |
| 5 | # This is *not* a good example of good programming practices. In fact, this |
| 6 | # file could use a complete rewrite, in order to become faster, more |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 7 | # easily extensible and maintainable. |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 8 | # |
| 9 | # However, I added some comments on a few places for the pityful person who |
| 10 | # would ever need to take a look into this file. |
| 11 | # |
| 12 | # Have I been clear enough?? |
| 13 | # |
| 14 | # -jh |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 15 | # |
| 16 | # Yup. I made some performance improvements and hope this lasts a while; |
| 17 | # I don't want to be the schmuck who ends up re-writting it! |
| 18 | # |
| 19 | # -fld |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 20 | |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 21 | import sys, string, regex, getopt, os |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 22 | |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 23 | from types import IntType, ListType, StringType, TupleType |
| 24 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 25 | # Different parse modes for phase 1 |
| 26 | MODE_REGULAR = 0 |
| 27 | MODE_VERBATIM = 1 |
| 28 | MODE_CS_SCAN = 2 |
| 29 | MODE_COMMENT = 3 |
| 30 | MODE_MATH = 4 |
| 31 | MODE_DMATH = 5 |
| 32 | MODE_GOBBLEWHITE = 6 |
| 33 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 34 | the_modes = (MODE_REGULAR, MODE_VERBATIM, MODE_CS_SCAN, MODE_COMMENT, |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 35 | MODE_MATH, MODE_DMATH, MODE_GOBBLEWHITE) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 36 | |
| 37 | # Show the neighbourhood of the scanned buffer |
| 38 | def epsilon(buf, where): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 39 | wmt, wpt = where - 10, where + 10 |
| 40 | if wmt < 0: |
| 41 | wmt = 0 |
| 42 | if wpt > len(buf): |
| 43 | wpt = len(buf) |
| 44 | return ' Context ' + `buf[wmt:where]` + '.' + `buf[where:wpt]` + '.' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 45 | |
| 46 | # Should return the line number. never worked |
| 47 | def lin(): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 48 | global lineno |
| 49 | return ' Line ' + `lineno` + '.' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 50 | |
| 51 | # Displays the recursion level. |
| 52 | def lv(lvl): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 53 | return ' Level ' + `lvl` + '.' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 54 | |
| 55 | # Combine the three previous functions. Used often. |
| 56 | def lle(lvl, buf, where): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 57 | return lv(lvl) + lin() + epsilon(buf, where) |
| 58 | |
| 59 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 60 | # This class is only needed for _symbolic_ representation of the parse mode. |
| 61 | class Mode: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 62 | def __init__(self, arg): |
| 63 | if arg not in the_modes: |
| 64 | raise ValueError, 'mode not in the_modes' |
| 65 | self.mode = arg |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 66 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 67 | def __cmp__(self, other): |
| 68 | if type(self) != type(other): |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 69 | other = mode[other] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 70 | return cmp(self.mode, other.mode) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 71 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 72 | def __repr__(self): |
| 73 | if self.mode == MODE_REGULAR: |
| 74 | return 'MODE_REGULAR' |
| 75 | elif self.mode == MODE_VERBATIM: |
| 76 | return 'MODE_VERBATIM' |
| 77 | elif self.mode == MODE_CS_SCAN: |
| 78 | return 'MODE_CS_SCAN' |
| 79 | elif self.mode == MODE_COMMENT: |
| 80 | return 'MODE_COMMENT' |
| 81 | elif self.mode == MODE_MATH: |
| 82 | return 'MODE_MATH' |
| 83 | elif self.mode == MODE_DMATH: |
| 84 | return 'MODE_DMATH' |
| 85 | elif self.mode == MODE_GOBBLEWHITE: |
| 86 | return 'MODE_GOBBLEWHITE' |
| 87 | else: |
| 88 | raise ValueError, 'mode not in the_modes' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 89 | |
| 90 | # just a wrapper around a class initialisation |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 91 | mode = {} |
| 92 | for t in the_modes: |
| 93 | mode[t] = Mode(t) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 94 | |
| 95 | |
| 96 | # After phase 1, the text consists of chunks, with a certain type |
| 97 | # this type will be assigned to the chtype member of the chunk |
| 98 | # the where-field contains the file position where this is found |
| 99 | # and the data field contains (1): a tuple describing start- end end |
| 100 | # positions of the substring (can be used as slice for the buf-variable), |
| 101 | # (2) just a string, mostly generated by the changeit routine, |
| 102 | # or (3) a list, describing a (recursive) subgroup of chunks |
| 103 | PLAIN = 0 # ASSUME PLAINTEXT, data = the text |
| 104 | GROUP = 1 # GROUP ({}), data = [chunk, chunk,..] |
| 105 | CSNAME = 2 # CONTROL SEQ TOKEN, data = the command |
| 106 | COMMENT = 3 # data is the actual comment |
| 107 | DMATH = 4 # DISPLAYMATH, data = [chunk, chunk,..] |
| 108 | MATH = 5 # MATH, see DISPLAYMATH |
| 109 | OTHER = 6 # CHAR WITH CATCODE OTHER, data = char |
| 110 | ACTIVE = 7 # ACTIVE CHAR |
| 111 | GOBBLEDWHITE = 8 # Gobbled LWSP, after CSNAME |
| 112 | ENDLINE = 9 # END-OF-LINE, data = '\n' |
| 113 | DENDLINE = 10 # DOUBLE EOL, data='\n', indicates \par |
| 114 | ENV = 11 # LaTeX-environment |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 115 | # data =(envname,[ch,ch,ch,.]) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 116 | CSLINE = 12 # for texi: next chunk will be one group |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 117 | # of args. Will be set all on 1 line |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 118 | IGNORE = 13 # IGNORE this data |
| 119 | ENDENV = 14 # TEMP END OF GROUP INDICATOR |
| 120 | IF = 15 # IF-directive |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 121 | # data = (flag,negate,[ch, ch, ch,...]) |
| 122 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 123 | the_types = (PLAIN, GROUP, CSNAME, COMMENT, DMATH, MATH, OTHER, ACTIVE, |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 124 | GOBBLEDWHITE, ENDLINE, DENDLINE, ENV, CSLINE, IGNORE, ENDENV, IF) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 125 | |
| 126 | # class, just to display symbolic name |
| 127 | class ChunkType: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 128 | def __init__(self, chunk_type): |
| 129 | if chunk_type not in the_types: |
| 130 | raise ValueError, 'chunk_type not in the_types' |
| 131 | self.chunk_type = chunk_type |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 132 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 133 | def __cmp__(self, other): |
| 134 | if type(self) != type(other): |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 135 | other = chunk_type[other] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 136 | return cmp(self.chunk_type, other.chunk_type) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 137 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 138 | def __repr__(self): |
| 139 | if self.chunk_type == PLAIN: |
| 140 | return 'PLAIN' |
| 141 | elif self.chunk_type == GROUP: |
| 142 | return 'GROUP' |
| 143 | elif self.chunk_type == CSNAME: |
| 144 | return 'CSNAME' |
| 145 | elif self.chunk_type == COMMENT: |
| 146 | return 'COMMENT' |
| 147 | elif self.chunk_type == DMATH: |
| 148 | return 'DMATH' |
| 149 | elif self.chunk_type == MATH: |
| 150 | return 'MATH' |
| 151 | elif self.chunk_type == OTHER: |
| 152 | return 'OTHER' |
| 153 | elif self.chunk_type == ACTIVE: |
| 154 | return 'ACTIVE' |
| 155 | elif self.chunk_type == GOBBLEDWHITE: |
| 156 | return 'GOBBLEDWHITE' |
| 157 | elif self.chunk_type == DENDLINE: |
| 158 | return 'DENDLINE' |
| 159 | elif self.chunk_type == ENDLINE: |
| 160 | return 'ENDLINE' |
| 161 | elif self.chunk_type == ENV: |
| 162 | return 'ENV' |
| 163 | elif self.chunk_type == CSLINE: |
| 164 | return 'CSLINE' |
| 165 | elif self.chunk_type == IGNORE: |
| 166 | return 'IGNORE' |
| 167 | elif self.chunk_type == ENDENV: |
| 168 | return 'ENDENV' |
| 169 | elif self.chunk_type == IF: |
| 170 | return 'IF' |
| 171 | else: |
| 172 | raise ValueError, 'chunk_type not in the_types' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 173 | |
| 174 | # ...and the wrapper |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 175 | chunk_type = {} |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 176 | for t in the_types: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 177 | chunk_type[t] = ChunkType(t) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 178 | |
| 179 | # store a type object of the ChunkType-class-instance... |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 180 | chunk_type_type = type(chunk_type[PLAIN]) |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 181 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 182 | # this class contains a part of the parsed buffer |
| 183 | class Chunk: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 184 | def __init__(self, chtype, where, data): |
| 185 | if type(chtype) != chunk_type_type: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 186 | chtype = chunk_type[chtype] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 187 | self.chtype = chtype |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 188 | self.where = where |
| 189 | self.data = data |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 190 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 191 | def __repr__(self): |
| 192 | return 'chunk' + `self.chtype, self.where, self.data` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 193 | |
| 194 | # and the wrapper |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 195 | chunk = Chunk |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 196 | |
| 197 | |
| 198 | error = 'partparse.error' |
| 199 | |
| 200 | # |
| 201 | # TeX's catcodes... |
| 202 | # |
| 203 | CC_ESCAPE = 0 |
| 204 | CC_LBRACE = 1 |
| 205 | CC_RBRACE = 2 |
| 206 | CC_MATHSHIFT = 3 |
| 207 | CC_ALIGNMENT = 4 |
| 208 | CC_ENDLINE = 5 |
| 209 | CC_PARAMETER = 6 |
| 210 | CC_SUPERSCRIPT = 7 |
| 211 | CC_SUBSCRIPT = 8 |
| 212 | CC_IGNORE = 9 |
| 213 | CC_WHITE = 10 |
| 214 | CC_LETTER = 11 |
| 215 | CC_OTHER = 12 |
| 216 | CC_ACTIVE = 13 |
| 217 | CC_COMMENT = 14 |
| 218 | CC_INVALID = 15 |
| 219 | |
| 220 | # and the names |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 221 | cc_names = [ |
| 222 | 'CC_ESCAPE', |
| 223 | 'CC_LBRACE', |
| 224 | 'CC_RBRACE', |
| 225 | 'CC_MATHSHIFT', |
| 226 | 'CC_ALIGNMENT', |
| 227 | 'CC_ENDLINE', |
| 228 | 'CC_PARAMETER', |
| 229 | 'CC_SUPERSCRIPT', |
| 230 | 'CC_SUBSCRIPT', |
| 231 | 'CC_IGNORE', |
| 232 | 'CC_WHITE', |
| 233 | 'CC_LETTER', |
| 234 | 'CC_OTHER', |
| 235 | 'CC_ACTIVE', |
| 236 | 'CC_COMMENT', |
| 237 | 'CC_INVALID', |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 238 | ] |
| 239 | |
| 240 | # Show a list of catcode-name-symbols |
| 241 | def pcl(codelist): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 242 | result = '' |
| 243 | for i in codelist: |
| 244 | result = result + cc_names[i] + ', ' |
| 245 | return '[' + result[:-2] + ']' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 246 | |
| 247 | # the name of the catcode (ACTIVE, OTHER, etc.) |
| 248 | def pc(code): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 249 | return cc_names[code] |
| 250 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 251 | |
| 252 | # Which catcodes make the parser stop parsing regular plaintext |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 253 | regular_stopcodes = [CC_ESCAPE, CC_LBRACE, CC_RBRACE, CC_MATHSHIFT, |
| 254 | CC_ALIGNMENT, CC_PARAMETER, CC_SUPERSCRIPT, CC_SUBSCRIPT, |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 255 | CC_IGNORE, CC_ACTIVE, CC_COMMENT, CC_INVALID, CC_ENDLINE] |
| 256 | |
| 257 | # same for scanning a control sequence name |
| 258 | csname_scancodes = [CC_LETTER] |
| 259 | |
| 260 | # same for gobbling LWSP |
| 261 | white_scancodes = [CC_WHITE] |
| 262 | ##white_scancodes = [CC_WHITE, CC_ENDLINE] |
| 263 | |
| 264 | # make a list of all catcode id's, except for catcode ``other'' |
| 265 | all_but_other_codes = range(16) |
| 266 | del all_but_other_codes[CC_OTHER] |
| 267 | ##print all_but_other_codes |
| 268 | |
| 269 | # when does a comment end |
| 270 | comment_stopcodes = [CC_ENDLINE] |
| 271 | |
| 272 | # gather all characters together, specified by a list of catcodes |
| 273 | def code2string(cc, codelist): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 274 | ##print 'code2string: codelist = ' + pcl(codelist), |
| 275 | result = '' |
| 276 | for category in codelist: |
| 277 | if cc[category]: |
| 278 | result = result + cc[category] |
| 279 | ##print 'result = ' + `result` |
| 280 | return result |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 281 | |
| 282 | # automatically generate all characters of catcode other, being the |
| 283 | # complement set in the ASCII range (128 characters) |
| 284 | def make_other_codes(cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 285 | otherchars = range(256) # could be made 256, no problem |
| 286 | for category in all_but_other_codes: |
| 287 | if cc[category]: |
| 288 | for c in cc[category]: |
| 289 | otherchars[ord(c)] = None |
| 290 | result = '' |
| 291 | for i in otherchars: |
| 292 | if i != None: |
| 293 | result = result + chr(i) |
| 294 | return result |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 295 | |
| 296 | # catcode dump (which characters have which catcodes). |
| 297 | def dump_cc(name, cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 298 | ##print '\t' + name |
| 299 | ##print '=' * (8+len(name)) |
| 300 | if len(cc) != 16: |
| 301 | raise TypeError, 'cc not good cat class' |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 302 | ## for i in range(16): |
| 303 | ## print pc(i) + '\t' + `cc[i]` |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 304 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 305 | |
| 306 | # In the beginning,.... |
| 307 | epoch_cc = [None] * 16 |
| 308 | ##dump_cc('epoch_cc', epoch_cc) |
| 309 | |
| 310 | |
| 311 | # INITEX |
| 312 | initex_cc = epoch_cc[:] |
| 313 | initex_cc[CC_ESCAPE] = '\\' |
| 314 | initex_cc[CC_ENDLINE], initex_cc[CC_IGNORE], initex_cc[CC_WHITE] = \ |
| 315 | '\n', '\0', ' ' |
| 316 | initex_cc[CC_LETTER] = string.uppercase + string.lowercase |
| 317 | initex_cc[CC_COMMENT], initex_cc[CC_INVALID] = '%', '\x7F' |
| 318 | #initex_cc[CC_OTHER] = make_other_codes(initex_cc) I don't need them, anyway |
| 319 | ##dump_cc('initex_cc', initex_cc) |
| 320 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 321 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 322 | # LPLAIN: LaTeX catcode setting (see lplain.tex) |
| 323 | lplain_cc = initex_cc[:] |
| 324 | lplain_cc[CC_LBRACE], lplain_cc[CC_RBRACE] = '{', '}' |
| 325 | lplain_cc[CC_MATHSHIFT] = '$' |
| 326 | lplain_cc[CC_ALIGNMENT] = '&' |
| 327 | lplain_cc[CC_PARAMETER] = '#' |
| 328 | lplain_cc[CC_SUPERSCRIPT] = '^\x0B' # '^' and C-k |
| 329 | lplain_cc[CC_SUBSCRIPT] = '_\x01' # '_' and C-a |
| 330 | lplain_cc[CC_WHITE] = lplain_cc[CC_WHITE] + '\t' |
| 331 | lplain_cc[CC_ACTIVE] = '~\x0C' # '~' and C-l |
| 332 | lplain_cc[CC_OTHER] = make_other_codes(lplain_cc) |
| 333 | ##dump_cc('lplain_cc', lplain_cc) |
| 334 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 335 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 336 | # Guido's LaTeX environment catcoded '_' as ``other'' |
| 337 | # my own purpose catlist |
| 338 | my_cc = lplain_cc[:] |
| 339 | my_cc[CC_SUBSCRIPT] = my_cc[CC_SUBSCRIPT][1:] # remove '_' here |
| 340 | my_cc[CC_OTHER] = my_cc[CC_OTHER] + '_' # add it to OTHER list |
| 341 | dump_cc('my_cc', my_cc) |
| 342 | |
| 343 | |
| 344 | |
| 345 | # needed for un_re, my equivalent for regexp-quote in Emacs |
| 346 | re_meaning = '\\[]^$' |
| 347 | |
| 348 | def un_re(str): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 349 | result = '' |
| 350 | for i in str: |
| 351 | if i in re_meaning: |
| 352 | result = result + '\\' |
| 353 | result = result + i |
| 354 | return result |
| 355 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 356 | # NOTE the negate ('^') operator in *some* of the regexps below |
| 357 | def make_rc_regular(cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 358 | # problems here if '[]' are included!! |
| 359 | return regex.compile('[' + code2string(cc, regular_stopcodes) + ']') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 360 | |
| 361 | def make_rc_cs_scan(cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 362 | return regex.compile('[^' + code2string(cc, csname_scancodes) + ']') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 363 | |
| 364 | def make_rc_comment(cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 365 | return regex.compile('[' + code2string(cc, comment_stopcodes) + ']') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 366 | |
| 367 | def make_rc_endwhite(cc): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 368 | return regex.compile('[^' + code2string(cc, white_scancodes) + ']') |
| 369 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 370 | |
| 371 | |
| 372 | # regular: normal mode: |
| 373 | rc_regular = make_rc_regular(my_cc) |
| 374 | |
| 375 | # scan: scan a command sequence e.g. `newlength' or `mbox' or `;', `,' or `$' |
| 376 | rc_cs_scan = make_rc_cs_scan(my_cc) |
| 377 | rc_comment = make_rc_comment(my_cc) |
| 378 | rc_endwhite = make_rc_endwhite(my_cc) |
| 379 | |
| 380 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 381 | # parseit (BUF, PARSEMODE=mode[MODE_REGULAR], START=0, RECURSION-LEVEL=0) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 382 | # RECURSION-LEVEL will is incremented on entry. |
| 383 | # result contains the list of chunks returned |
| 384 | # together with this list, the buffer position is returned |
| 385 | |
| 386 | # RECURSION-LEVEL will be set to zero *again*, when recursively a |
| 387 | # {,D}MATH-mode scan has been enetered. |
| 388 | # This has been done in order to better check for environment-mismatches |
| 389 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 390 | def parseit(buf, parsemode=mode[MODE_REGULAR], start=0, lvl=0): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 391 | global lineno |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 392 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 393 | result = [] |
| 394 | end = len(buf) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 395 | if lvl == 0 and parsemode == mode[MODE_REGULAR]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 396 | lineno = 1 |
| 397 | lvl = lvl + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 398 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 399 | ##print 'parseit(' + epsilon(buf, start) + ', ' + `parsemode` + ', ' + `start` + ', ' + `lvl` + ')' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 400 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 401 | # |
| 402 | # some of the more regular modes... |
| 403 | # |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 404 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 405 | if parsemode in (mode[MODE_REGULAR], mode[MODE_DMATH], mode[MODE_MATH]): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 406 | cstate = [] |
| 407 | newpos = start |
| 408 | curpmode = parsemode |
| 409 | while 1: |
| 410 | where = newpos |
| 411 | #print '\tnew round: ' + epsilon(buf, where) |
| 412 | if where == end: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 413 | if lvl > 1 or curpmode != mode[MODE_REGULAR]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 414 | # not the way we started... |
| 415 | raise EOFError, 'premature end of file.' + lle(lvl, buf, where) |
| 416 | # the real ending of lvl-1 parse |
| 417 | return end, result |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 418 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 419 | pos = rc_regular.search(buf, where) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 420 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 421 | if pos < 0: |
| 422 | pos = end |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 423 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 424 | if pos != where: |
| 425 | newpos, c = pos, chunk(PLAIN, where, (where, pos)) |
| 426 | result.append(c) |
| 427 | continue |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 428 | |
| 429 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 430 | # |
| 431 | # ok, pos == where and pos != end |
| 432 | # |
| 433 | foundchar = buf[where] |
| 434 | if foundchar in my_cc[CC_LBRACE]: |
| 435 | # recursive subgroup parse... |
| 436 | newpos, data = parseit(buf, curpmode, where+1, lvl) |
| 437 | result.append(chunk(GROUP, where, data)) |
| 438 | |
| 439 | elif foundchar in my_cc[CC_RBRACE]: |
| 440 | if lvl <= 1: |
| 441 | raise error, 'ENDGROUP while in base level.' + lle(lvl, buf, where) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 442 | if lvl == 1 and mode != mode[MODE_REGULAR]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 443 | raise error, 'endgroup while in math mode. +lin() + epsilon(buf, where)' |
| 444 | return where + 1, result |
| 445 | |
| 446 | elif foundchar in my_cc[CC_ESCAPE]: |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 447 | # |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 448 | # call the routine that actually deals with |
| 449 | # this problem. If do_ret is None, than |
| 450 | # return the value of do_ret |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 451 | # |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 452 | # Note that handle_cs might call this routine |
| 453 | # recursively again... |
| 454 | # |
| 455 | do_ret, newpos = handlecs(buf, where, |
| 456 | curpmode, lvl, result, end) |
| 457 | if do_ret != None: |
| 458 | return do_ret |
| 459 | |
| 460 | elif foundchar in my_cc[CC_COMMENT]: |
| 461 | newpos, data = parseit(buf, |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 462 | mode[MODE_COMMENT], where+1, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 463 | result.append(chunk(COMMENT, where, data)) |
| 464 | |
| 465 | elif foundchar in my_cc[CC_MATHSHIFT]: |
| 466 | # note that recursive calls to math-mode |
| 467 | # scanning are called with recursion-level 0 |
| 468 | # again, in order to check for bad mathend |
| 469 | # |
| 470 | if where + 1 != end and buf[where + 1] in my_cc[CC_MATHSHIFT]: |
| 471 | # |
| 472 | # double mathshift, e.g. '$$' |
| 473 | # |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 474 | if curpmode == mode[MODE_REGULAR]: |
| 475 | newpos, data = parseit(buf, mode[MODE_DMATH], |
| 476 | where + 2, 0) |
| 477 | result.append(chunk(DMATH, where, data)) |
| 478 | elif curpmode == mode[MODE_MATH]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 479 | raise error, 'wrong math delimiiter' + lin() + epsilon(buf, where) |
| 480 | elif lvl != 1: |
| 481 | raise error, 'bad mathend.' + lle(lvl, buf, where) |
| 482 | else: |
| 483 | return where + 2, result |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 484 | else: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 485 | # |
| 486 | # single math shift, e.g. '$' |
| 487 | # |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 488 | if curpmode == mode[MODE_REGULAR]: |
| 489 | newpos, data = parseit(buf, mode[MODE_MATH], |
| 490 | where + 1, 0) |
| 491 | result.append(chunk(MATH, where, data)) |
| 492 | elif curpmode == mode[MODE_DMATH]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 493 | raise error, 'wrong math delimiiter' + lin() + epsilon(buf, where) |
| 494 | elif lvl != 1: |
| 495 | raise error, 'bad mathend.' + lv(lvl, buf, where) |
| 496 | else: |
| 497 | return where + 1, result |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 498 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 499 | elif foundchar in my_cc[CC_IGNORE]: |
| 500 | print 'warning: ignored char', `foundchar` |
| 501 | newpos = where + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 502 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 503 | elif foundchar in my_cc[CC_ACTIVE]: |
| 504 | result.append(chunk(ACTIVE, where, foundchar)) |
| 505 | newpos = where + 1 |
| 506 | |
| 507 | elif foundchar in my_cc[CC_INVALID]: |
| 508 | raise error, 'invalid char ' + `foundchar` |
| 509 | newpos = where + 1 |
| 510 | |
| 511 | elif foundchar in my_cc[CC_ENDLINE]: |
| 512 | # |
| 513 | # after an end of line, eat the rest of |
| 514 | # whitespace on the beginning of the next line |
| 515 | # this is what LaTeX more or less does |
| 516 | # |
| 517 | # also, try to indicate double newlines (\par) |
| 518 | # |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 519 | lineno = lineno + 1 |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 520 | savedwhere = where |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 521 | newpos, dummy = parseit(buf, mode[MODE_GOBBLEWHITE], where + 1, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 522 | if newpos != end and buf[newpos] in my_cc[CC_ENDLINE]: |
| 523 | result.append(chunk(DENDLINE, savedwhere, foundchar)) |
| 524 | else: |
| 525 | result.append(chunk(ENDLINE, savedwhere, foundchar)) |
| 526 | else: |
| 527 | result.append(chunk(OTHER, where, foundchar)) |
| 528 | newpos = where + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 529 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 530 | elif parsemode == mode[MODE_CS_SCAN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 531 | # |
| 532 | # scan for a control sequence token. `\ape', `\nut' or `\%' |
| 533 | # |
| 534 | if start == end: |
| 535 | raise EOFError, 'can\'t find end of csname' |
| 536 | pos = rc_cs_scan.search(buf, start) |
| 537 | if pos < 0: |
| 538 | pos = end |
| 539 | if pos == start: |
| 540 | # first non-letter right where we started the search |
| 541 | # ---> the control sequence name consists of one single |
| 542 | # character. Also: don't eat white space... |
| 543 | if buf[pos] in my_cc[CC_ENDLINE]: |
| 544 | lineno = lineno + 1 |
| 545 | pos = pos + 1 |
| 546 | return pos, (start, pos) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 547 | else: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 548 | spos = pos |
| 549 | if buf[pos] == '\n': |
| 550 | lineno = lineno + 1 |
| 551 | spos = pos + 1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 552 | pos2, dummy = parseit(buf, mode[MODE_GOBBLEWHITE], spos, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 553 | return pos2, (start, pos) |
| 554 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 555 | elif parsemode == mode[MODE_GOBBLEWHITE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 556 | if start == end: |
| 557 | return start, '' |
| 558 | pos = rc_endwhite.search(buf, start) |
| 559 | if pos < 0: |
| 560 | pos = start |
| 561 | return pos, (start, pos) |
| 562 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 563 | elif parsemode == mode[MODE_COMMENT]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 564 | pos = rc_comment.search(buf, start) |
| 565 | lineno = lineno + 1 |
| 566 | if pos < 0: |
| 567 | print 'no newline perhaps?' |
| 568 | raise EOFError, 'can\'t find end of comment' |
| 569 | pos = pos + 1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 570 | pos2, dummy = parseit(buf, mode[MODE_GOBBLEWHITE], pos, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 571 | return pos2, (start, pos) |
| 572 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 573 | else: |
| 574 | raise error, 'Unknown mode (' + `parsemode` + ')' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 575 | |
| 576 | |
| 577 | #moreresult = cswitch(buf[x1:x2], buf, newpos, parsemode, lvl) |
| 578 | |
| 579 | #boxcommands = 'mbox', 'fbox' |
| 580 | #defcommands = 'def', 'newcommand' |
| 581 | |
| 582 | endverbstr = '\\end{verbatim}' |
| 583 | |
| 584 | re_endverb = regex.compile(un_re(endverbstr)) |
| 585 | |
| 586 | # |
| 587 | # handlecs: helper function for parseit, for the special thing we might |
| 588 | # wanna do after certain command control sequences |
| 589 | # returns: None or return_data, newpos |
| 590 | # |
| 591 | # in the latter case, the calling function is instructed to immediately |
| 592 | # return with the data in return_data |
| 593 | # |
| 594 | def handlecs(buf, where, curpmode, lvl, result, end): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 595 | global lineno |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 596 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 597 | # get the control sequence name... |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 598 | newpos, data = parseit(buf, mode[MODE_CS_SCAN], where+1, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 599 | saveddata = data |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 600 | s_buf_data = s(buf, data) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 601 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 602 | if s_buf_data in ('begin', 'end'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 603 | # skip the expected '{' and get the LaTeX-envname '}' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 604 | newpos, data = parseit(buf, mode[MODE_REGULAR], newpos+1, lvl) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 605 | if len(data) != 1: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 606 | raise error, 'expected 1 chunk of data.' + lle(lvl, buf, where) |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 607 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 608 | # yucky, we've got an environment |
| 609 | envname = s(buf, data[0].data) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 610 | s_buf_saveddata = s(buf, saveddata) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 611 | ##print 'FOUND ' + s(buf, saveddata) + '. Name ' + `envname` + '.' + lv(lvl) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 612 | if s_buf_saveddata == 'begin' and envname == 'verbatim': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 613 | # verbatim deserves special treatment |
| 614 | pos = re_endverb.search(buf, newpos) |
| 615 | if pos < 0: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 616 | raise error, "%s not found.%s" \ |
| 617 | % (`endverbstr`, lle(lvl, buf, where)) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 618 | result.append(chunk(ENV, where, (envname, [chunk(PLAIN, newpos, (newpos, pos))]))) |
| 619 | newpos = pos + len(endverbstr) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 620 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 621 | elif s_buf_saveddata == 'begin': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 622 | # start parsing recursively... If that parse returns |
| 623 | # from an '\end{...}', then should the last item of |
| 624 | # the returned data be a string containing the ended |
| 625 | # environment |
| 626 | newpos, data = parseit(buf, curpmode, newpos, lvl) |
| 627 | if not data or type(data[-1]) is not StringType: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 628 | raise error, "missing 'end'" + lle(lvl, buf, where) \ |
| 629 | + epsilon(buf, newpos) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 630 | retenv = data[-1] |
| 631 | del data[-1] |
| 632 | if retenv != envname: |
| 633 | #[`retenv`, `envname`] |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 634 | raise error, 'environments do not match.%s%s' \ |
| 635 | % (lle(lvl, buf, where), epsilon(buf, newpos)) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 636 | result.append(chunk(ENV, where, (retenv, data))) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 637 | else: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 638 | # 'end'... append the environment name, as just |
| 639 | # pointed out, and order parsit to return... |
| 640 | result.append(envname) |
| 641 | ##print 'POINT of return: ' + epsilon(buf, newpos) |
| 642 | # the tuple will be returned by parseit |
| 643 | return (newpos, result), newpos |
| 644 | |
| 645 | # end of \begin ... \end handling |
| 646 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 647 | elif s_buf_data[0:2] == 'if': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 648 | # another scary monster: the 'if' directive |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 649 | flag = s_buf_data[2:] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 650 | |
| 651 | # recursively call parseit, just like environment above.. |
| 652 | # the last item of data should contain the if-termination |
| 653 | # e.g., 'else' of 'fi' |
| 654 | newpos, data = parseit(buf, curpmode, newpos, lvl) |
| 655 | if not data or data[-1] not in ('else', 'fi'): |
| 656 | raise error, 'wrong if... termination' + \ |
| 657 | lle(lvl, buf, where) + epsilon(buf, newpos) |
| 658 | |
| 659 | ifterm = data[-1] |
| 660 | del data[-1] |
| 661 | # 0 means dont_negate flag |
| 662 | result.append(chunk(IF, where, (flag, 0, data))) |
| 663 | if ifterm == 'else': |
| 664 | # do the whole thing again, there is only one way |
| 665 | # to end this one, by 'fi' |
| 666 | newpos, data = parseit(buf, curpmode, newpos, lvl) |
| 667 | if not data or data[-1] not in ('fi', ): |
| 668 | raise error, 'wrong if...else... termination' \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 669 | + lle(lvl, buf, where) \ |
| 670 | + epsilon(buf, newpos) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 671 | |
| 672 | ifterm = data[-1] |
| 673 | del data[-1] |
| 674 | result.append(chunk(IF, where, (flag, 1, data))) |
| 675 | #done implicitely: return None, newpos |
| 676 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 677 | elif s_buf_data in ('else', 'fi'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 678 | result.append(s(buf, data)) |
| 679 | # order calling party to return tuple |
| 680 | return (newpos, result), newpos |
| 681 | |
| 682 | # end of \if, \else, ... \fi handling |
| 683 | |
| 684 | elif s(buf, saveddata) == 'verb': |
| 685 | x2 = saveddata[1] |
| 686 | result.append(chunk(CSNAME, where, data)) |
| 687 | if x2 == end: |
| 688 | raise error, 'premature end of command.' + lle(lvl, buf, where) |
| 689 | delimchar = buf[x2] |
| 690 | ##print 'VERB: delimchar ' + `delimchar` |
| 691 | pos = regex.compile(un_re(delimchar)).search(buf, x2 + 1) |
| 692 | if pos < 0: |
| 693 | raise error, 'end of \'verb\' argument (' + \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 694 | `delimchar` + ') not found.' + \ |
| 695 | lle(lvl, buf, where) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 696 | result.append(chunk(GROUP, x2, [chunk(PLAIN, x2+1, (x2+1, pos))])) |
| 697 | newpos = pos + 1 |
| 698 | else: |
| 699 | result.append(chunk(CSNAME, where, data)) |
| 700 | return None, newpos |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 701 | |
| 702 | # this is just a function to get the string value if the possible data-tuple |
| 703 | def s(buf, data): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 704 | if type(data) is StringType: |
| 705 | return data |
| 706 | if len(data) != 2 or not (type(data[0]) is type(data[1]) is IntType): |
| 707 | raise TypeError, 'expected tuple of 2 integers' |
| 708 | x1, x2 = data |
| 709 | return buf[x1:x2] |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 710 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 711 | |
| 712 | ##length, data1, i = getnextarg(length, buf, pp, i + 1) |
| 713 | |
| 714 | # make a deep-copy of some chunks |
| 715 | def crcopy(r): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 716 | return map(chunkcopy, r) |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 717 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 718 | |
| 719 | # copy a chunk, would better be a method of class Chunk... |
| 720 | def chunkcopy(ch): |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 721 | if ch.chtype == chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 722 | return chunk(GROUP, ch.where, map(chunkcopy, ch.data)) |
| 723 | else: |
| 724 | return chunk(ch.chtype, ch.where, ch.data) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 725 | |
| 726 | |
| 727 | # get next argument for TeX-macro, flatten a group (insert between) |
| 728 | # or return Command Sequence token, or give back one character |
| 729 | def getnextarg(length, buf, pp, item): |
| 730 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 731 | ##wobj = Wobj() |
| 732 | ##dumpit(buf, wobj.write, pp[item:min(length, item + 5)]) |
| 733 | ##print 'GETNEXTARG, (len, item) =', `length, item` + ' ---> ' + wobj.data + ' <---' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 734 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 735 | while item < length and pp[item].chtype == chunk_type[ENDLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 736 | del pp[item] |
| 737 | length = length - 1 |
| 738 | if item >= length: |
| 739 | raise error, 'no next arg.' + epsilon(buf, pp[-1].where) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 740 | if pp[item].chtype == chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 741 | newpp = pp[item].data |
| 742 | del pp[item] |
| 743 | length = length - 1 |
| 744 | changeit(buf, newpp) |
| 745 | length = length + len(newpp) |
| 746 | pp[item:item] = newpp |
| 747 | item = item + len(newpp) |
| 748 | if len(newpp) < 10: |
| 749 | wobj = Wobj() |
| 750 | dumpit(buf, wobj.write, newpp) |
| 751 | ##print 'GETNEXTARG: inserted ' + `wobj.data` |
| 752 | return length, item |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 753 | elif pp[item].chtype == chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 754 | #grab one char |
| 755 | print 'WARNING: grabbing one char' |
| 756 | if len(s(buf, pp[item].data)) > 1: |
| 757 | pp.insert(item, chunk(PLAIN, pp[item].where, s(buf, pp[item].data)[:1])) |
| 758 | item, length = item+1, length+1 |
| 759 | pp[item].data = s(buf, pp[item].data)[1:] |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 760 | else: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 761 | item = item+1 |
| 762 | return length, item |
| 763 | else: |
| 764 | ch = pp[item] |
| 765 | try: |
| 766 | str = `s(buf, ch.data)` |
| 767 | except TypeError: |
| 768 | str = `ch.data` |
| 769 | if len(str) > 400: |
| 770 | str = str[:400] + '...' |
| 771 | print 'GETNEXTARG:', ch.chtype, 'not handled, data ' + str |
| 772 | return length, item |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 773 | |
| 774 | |
| 775 | # this one is needed to find the end of LaTeX's optional argument, like |
| 776 | # item[...] |
| 777 | re_endopt = regex.compile(']') |
| 778 | |
| 779 | # get a LaTeX-optional argument, you know, the square braces '[' and ']' |
| 780 | def getoptarg(length, buf, pp, item): |
| 781 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 782 | wobj = Wobj() |
| 783 | dumpit(buf, wobj.write, pp[item:min(length, item + 5)]) |
| 784 | ##print 'GETOPTARG, (len, item) =', `length, item` + ' ---> ' + wobj.data + ' <---' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 785 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 786 | if item >= length or \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 787 | pp[item].chtype != chunk_type[PLAIN] or \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 788 | s(buf, pp[item].data)[0] != '[': |
| 789 | return length, item |
| 790 | |
| 791 | pp[item].data = s(buf, pp[item].data)[1:] |
| 792 | if len(pp[item].data) == 0: |
| 793 | del pp[item] |
| 794 | length = length-1 |
| 795 | |
| 796 | while 1: |
| 797 | if item == length: |
| 798 | raise error, 'No end of optional arg found' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 799 | if pp[item].chtype == chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 800 | text = s(buf, pp[item].data) |
| 801 | pos = re_endopt.search(text) |
| 802 | if pos >= 0: |
| 803 | pp[item].data = text[:pos] |
| 804 | if pos == 0: |
| 805 | del pp[item] |
| 806 | length = length-1 |
| 807 | else: |
| 808 | item=item+1 |
| 809 | text = text[pos+1:] |
| 810 | |
| 811 | while text and text[0] in ' \t': |
| 812 | text = text[1:] |
| 813 | |
| 814 | if text: |
| 815 | pp.insert(item, chunk(PLAIN, 0, text)) |
| 816 | length = length + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 817 | return length, item |
| 818 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 819 | item = item+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 820 | |
| 821 | |
| 822 | # Wobj just add write-requests to the ``data'' attribute |
| 823 | class Wobj: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 824 | data = '' |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 825 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 826 | def write(self, data): |
| 827 | self.data = self.data + data |
Guido van Rossum | b819bdf | 1995-03-15 11:26:26 +0000 | [diff] [blame] | 828 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 829 | # ignore these commands |
Guido van Rossum | b819bdf | 1995-03-15 11:26:26 +0000 | [diff] [blame] | 830 | ignoredcommands = ('bcode', 'ecode') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 831 | # map commands like these to themselves as plaintext |
Guido van Rossum | 7760cde | 1995-03-17 16:03:11 +0000 | [diff] [blame] | 832 | wordsselves = ('UNIX', 'ABC', 'C', 'ASCII', 'EOF', 'LaTeX') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 833 | # \{ --> {, \} --> }, etc |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 834 | themselves = ('{', '}', ',', '.', '@', ' ', '\n') + wordsselves |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 835 | # these ones also themselves (see argargs macro in myformat.sty) |
| 836 | inargsselves = (',', '[', ']', '(', ')') |
| 837 | # this is how *I* would show the difference between emph and strong |
| 838 | # code 1 means: fold to uppercase |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 839 | markcmds = {'code': ('', ''), 'var': 1, 'emph': ('_', '_'), |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 840 | 'strong': ('*', '*')} |
| 841 | |
| 842 | # recognise patter {\FONTCHANGE-CMD TEXT} to \MAPPED-FC-CMD{TEXT} |
| 843 | fontchanges = {'rm': 'r', 'it': 'i', 'em': 'emph', 'bf': 'b', 'tt': 't'} |
| 844 | |
| 845 | # transparent for these commands |
Guido van Rossum | 7760cde | 1995-03-17 16:03:11 +0000 | [diff] [blame] | 846 | for_texi = ('emph', 'var', 'strong', 'code', 'kbd', 'key', 'dfn', 'samp', |
| 847 | 'file', 'r', 'i', 't') |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 848 | |
| 849 | |
| 850 | # try to remove macros and return flat text |
| 851 | def flattext(buf, pp): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 852 | pp = crcopy(pp) |
| 853 | ##print '---> FLATTEXT ' + `pp` |
| 854 | wobj = Wobj() |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 855 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 856 | i, length = 0, len(pp) |
| 857 | while 1: |
| 858 | if len(pp) != length: |
| 859 | raise 'FATAL', 'inconsistent length' |
| 860 | if i >= length: |
| 861 | break |
| 862 | ch = pp[i] |
| 863 | i = i+1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 864 | if ch.chtype == chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 865 | pass |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 866 | elif ch.chtype == chunk_type[CSNAME]: |
| 867 | s_buf_data = s(buf, ch.data) |
| 868 | if s_buf_data in themselves or hist.inargs and s_buf_data in inargsselves: |
| 869 | ch.chtype = chunk_type[PLAIN] |
| 870 | elif s_buf_data == 'e': |
| 871 | ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 872 | ch.data = '\\' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 873 | elif len(s_buf_data) == 1 \ |
| 874 | and s_buf_data in onlylatexspecial: |
| 875 | ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 876 | # if it is followed by an empty group, |
| 877 | # remove that group, it was needed for |
| 878 | # a true space |
| 879 | if i < length \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 880 | and pp[i].chtype==chunk_type[GROUP] \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 881 | and len(pp[i].data) == 0: |
| 882 | del pp[i] |
| 883 | length = length-1 |
| 884 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 885 | elif s_buf_data in markcmds.keys(): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 886 | length, newi = getnextarg(length, buf, pp, i) |
| 887 | str = flattext(buf, pp[i:newi]) |
| 888 | del pp[i:newi] |
| 889 | length = length - (newi - i) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 890 | ch.chtype = chunk_type[PLAIN] |
| 891 | markcmd = s_buf_data |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 892 | x = markcmds[markcmd] |
| 893 | if type(x) == TupleType: |
| 894 | pre, after = x |
| 895 | str = pre+str+after |
| 896 | elif x == 1: |
| 897 | str = string.upper(str) |
| 898 | else: |
| 899 | raise 'FATAL', 'corrupt markcmds' |
| 900 | ch.data = str |
| 901 | else: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 902 | if s_buf_data not in ignoredcommands: |
| 903 | print 'WARNING: deleting command ' + s_buf_data |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 904 | print 'PP' + `pp[i-1]` |
| 905 | del pp[i-1] |
| 906 | i, length = i-1, length-1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 907 | elif ch.chtype == chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 908 | length, newi = getnextarg(length, buf, pp, i-1) |
| 909 | i = i-1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 910 | ## str = flattext(buf, crcopy(pp[i-1:newi])) |
| 911 | ## del pp[i:newi] |
| 912 | ## length = length - (newi - i) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 913 | ## ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 914 | ## ch.data = str |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 915 | else: |
| 916 | pass |
| 917 | |
| 918 | dumpit(buf, wobj.write, pp) |
| 919 | ##print 'FLATTEXT: RETURNING ' + `wobj.data` |
| 920 | return wobj.data |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 921 | |
| 922 | # try to generate node names (a bit shorter than the chapter title) |
| 923 | # note that the \nodename command (see elsewhere) overules these efforts |
| 924 | def invent_node_names(text): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 925 | words = string.split(text) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 926 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 927 | ##print 'WORDS ' + `words` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 928 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 929 | if len(words) == 2 \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 930 | and string.lower(words[0]) == 'built-in' \ |
| 931 | and string.lower(words[1]) not in ('modules', 'functions'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 932 | return words[1] |
| 933 | if len(words) == 3 and string.lower(words[1]) == 'module': |
| 934 | return words[2] |
| 935 | if len(words) == 3 and string.lower(words[1]) == 'object': |
| 936 | return string.join(words[0:2]) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 937 | if len(words) > 4 \ |
| 938 | and (string.lower(string.join(words[-4:])) \ |
| 939 | == 'methods and data attributes'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 940 | return string.join(words[:2]) |
| 941 | return text |
| 942 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 943 | re_commas_etc = regex.compile('[,`\'@{}]') |
| 944 | |
| 945 | re_whitespace = regex.compile('[ \t]*') |
| 946 | |
| 947 | |
| 948 | ##nodenamecmd = next_command_p(length, buf, pp, newi, 'nodename') |
| 949 | |
| 950 | # look if the next non-white stuff is also a command, resulting in skipping |
| 951 | # double endlines (DENDLINE) too, and thus omitting \par's |
| 952 | # Sometimes this is too much, maybe consider DENDLINE's as stop |
| 953 | def next_command_p(length, buf, pp, i, cmdname): |
| 954 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 955 | while 1: |
| 956 | if i >= len(pp): |
| 957 | break |
| 958 | ch = pp[i] |
| 959 | i = i+1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 960 | if ch.chtype == chunk_type[ENDLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 961 | continue |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 962 | if ch.chtype == chunk_type[DENDLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 963 | continue |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 964 | if ch.chtype == chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 965 | if re_whitespace.search(s(buf, ch.data)) == 0 and \ |
| 966 | re_whitespace.match(s(buf, ch.data)) == len(s(buf, ch.data)): |
| 967 | continue |
| 968 | return -1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 969 | if ch.chtype == chunk_type[CSNAME]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 970 | if s(buf, ch.data) == cmdname: |
| 971 | return i # _after_ the command |
| 972 | return -1 |
| 973 | return -1 |
| 974 | |
| 975 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 976 | # things that are special to LaTeX, but not to texi.. |
| 977 | onlylatexspecial = '_~^$#&%' |
| 978 | |
Guido van Rossum | 23301a9 | 1993-05-24 14:19:37 +0000 | [diff] [blame] | 979 | class Struct: pass |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 980 | |
| 981 | hist = Struct() |
| 982 | out = Struct() |
| 983 | |
| 984 | def startchange(): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 985 | global hist, out |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 986 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 987 | hist.inenv = [] |
| 988 | hist.nodenames = [] |
| 989 | hist.cindex = [] |
| 990 | hist.inargs = 0 |
| 991 | hist.enumeratenesting, hist.itemizenesting = 0, 0 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 992 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 993 | out.doublenodes = [] |
| 994 | out.doublecindeces = [] |
| 995 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 996 | |
| 997 | spacech = [chunk(PLAIN, 0, ' ')] |
| 998 | commach = [chunk(PLAIN, 0, ', ')] |
| 999 | cindexch = [chunk(CSLINE, 0, 'cindex')] |
| 1000 | |
| 1001 | # the standard variation in symbols for itemize |
| 1002 | itemizesymbols = ['bullet', 'minus', 'dots'] |
| 1003 | |
| 1004 | # same for enumerate |
| 1005 | enumeratesymbols = ['1', 'A', 'a'] |
| 1006 | |
| 1007 | ## |
| 1008 | ## \begin{ {func,data,exc}desc }{name}... |
| 1009 | ## the resulting texi-code is dependent on the contents of indexsubitem |
| 1010 | ## |
| 1011 | |
| 1012 | # indexsubitem: `['XXX', 'function'] |
| 1013 | # funcdesc: |
| 1014 | # deffn {`idxsi`} NAME (FUNCARGS) |
| 1015 | |
| 1016 | # indexsubitem: `['XXX', 'method']` |
| 1017 | # funcdesc: |
| 1018 | # defmethod {`idxsi[0]`} NAME (FUNCARGS) |
| 1019 | |
| 1020 | # indexsubitem: `['in', 'module', 'MODNAME']' |
| 1021 | # datadesc: |
| 1022 | # defcv data {`idxsi[1:]`} NAME |
| 1023 | # excdesc: |
| 1024 | # defcv exception {`idxsi[1:]`} NAME |
| 1025 | # funcdesc: |
| 1026 | # deffn {function of `idxsi[1:]`} NAME (FUNCARGS) |
| 1027 | |
| 1028 | # indexsubitem: `['OBJECT', 'attribute']' |
| 1029 | # datadesc |
| 1030 | # defcv attribute {`OBJECT`} NAME |
| 1031 | |
| 1032 | |
| 1033 | ## this routine will be called on \begin{funcdesc}{NAME}{ARGS} |
| 1034 | ## or \funcline{NAME}{ARGS} |
| 1035 | ## |
| 1036 | def do_funcdesc(length, buf, pp, i): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1037 | startpoint = i-1 |
| 1038 | ch = pp[startpoint] |
| 1039 | wh = ch.where |
| 1040 | length, newi = getnextarg(length, buf, pp, i) |
| 1041 | funcname = chunk(GROUP, wh, pp[i:newi]) |
| 1042 | del pp[i:newi] |
| 1043 | length = length - (newi-i) |
| 1044 | save = hist.inargs |
| 1045 | hist.inargs = 1 |
| 1046 | length, newi = getnextarg(length, buf, pp, i) |
| 1047 | hist.inargs = save |
| 1048 | del save |
| 1049 | the_args = [chunk(PLAIN, wh, '()'[0])] + pp[i:newi] + \ |
Fred Drake | 7edd8d3 | 1996-10-09 16:11:26 +0000 | [diff] [blame] | 1050 | [chunk(PLAIN, wh, '()'[1])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1051 | del pp[i:newi] |
| 1052 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1053 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1054 | idxsi = hist.indexsubitem # words |
| 1055 | command = '' |
| 1056 | cat_class = '' |
Fred Drake | acc8754 | 1996-10-14 16:20:42 +0000 | [diff] [blame] | 1057 | if idxsi and idxsi[-1] in ('method', 'protocol', 'attribute'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1058 | command = 'defmethod' |
| 1059 | cat_class = string.join(idxsi[:-1]) |
| 1060 | elif len(idxsi) == 2 and idxsi[1] == 'function': |
| 1061 | command = 'deffn' |
| 1062 | cat_class = string.join(idxsi) |
| 1063 | elif len(idxsi) == 3 and idxsi[:2] == ['in', 'module']: |
| 1064 | command = 'deffn' |
| 1065 | cat_class = 'function of ' + string.join(idxsi[1:]) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1066 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1067 | if not command: |
| 1068 | raise error, 'don\'t know what to do with indexsubitem ' + `idxsi` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1069 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1070 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1071 | ch.data = command |
| 1072 | |
| 1073 | cslinearg = [chunk(GROUP, wh, [chunk(PLAIN, wh, cat_class)])] |
| 1074 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
| 1075 | cslinearg.append(funcname) |
| 1076 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
| 1077 | l = len(cslinearg) |
| 1078 | cslinearg[l:l] = the_args |
| 1079 | |
| 1080 | pp.insert(i, chunk(GROUP, wh, cslinearg)) |
| 1081 | i, length = i+1, length+1 |
| 1082 | hist.command = command |
| 1083 | return length, i |
| 1084 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1085 | |
| 1086 | ## this routine will be called on \begin{excdesc}{NAME} |
| 1087 | ## or \excline{NAME} |
| 1088 | ## |
| 1089 | def do_excdesc(length, buf, pp, i): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1090 | startpoint = i-1 |
| 1091 | ch = pp[startpoint] |
| 1092 | wh = ch.where |
| 1093 | length, newi = getnextarg(length, buf, pp, i) |
| 1094 | excname = chunk(GROUP, wh, pp[i:newi]) |
| 1095 | del pp[i:newi] |
| 1096 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1097 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1098 | idxsi = hist.indexsubitem # words |
| 1099 | command = '' |
| 1100 | cat_class = '' |
| 1101 | class_class = '' |
| 1102 | if len(idxsi) == 2 and idxsi[1] == 'exception': |
| 1103 | command = 'defvr' |
| 1104 | cat_class = string.join(idxsi) |
| 1105 | elif len(idxsi) == 3 and idxsi[:2] == ['in', 'module']: |
| 1106 | command = 'defcv' |
| 1107 | cat_class = 'exception' |
| 1108 | class_class = string.join(idxsi[1:]) |
| 1109 | elif len(idxsi) == 4 and idxsi[:3] == ['exception', 'in', 'module']: |
| 1110 | command = 'defcv' |
| 1111 | cat_class = 'exception' |
| 1112 | class_class = string.join(idxsi[2:]) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1113 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1114 | |
| 1115 | if not command: |
| 1116 | raise error, 'don\'t know what to do with indexsubitem ' + `idxsi` |
| 1117 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1118 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1119 | ch.data = command |
| 1120 | |
| 1121 | cslinearg = [chunk(GROUP, wh, [chunk(PLAIN, wh, cat_class)])] |
| 1122 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
| 1123 | if class_class: |
| 1124 | cslinearg.append(chunk(GROUP, wh, [chunk(PLAIN, wh, class_class)])) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1125 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1126 | cslinearg.append(excname) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1127 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1128 | pp.insert(i, chunk(GROUP, wh, cslinearg)) |
| 1129 | i, length = i+1, length+1 |
| 1130 | hist.command = command |
| 1131 | return length, i |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1132 | |
| 1133 | ## same for datadesc or dataline... |
| 1134 | def do_datadesc(length, buf, pp, i): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1135 | startpoint = i-1 |
| 1136 | ch = pp[startpoint] |
| 1137 | wh = ch.where |
| 1138 | length, newi = getnextarg(length, buf, pp, i) |
| 1139 | dataname = chunk(GROUP, wh, pp[i:newi]) |
| 1140 | del pp[i:newi] |
| 1141 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1142 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1143 | idxsi = hist.indexsubitem # words |
| 1144 | command = '' |
| 1145 | cat_class = '' |
| 1146 | class_class = '' |
| 1147 | if idxsi[-1] in ('attribute', 'option'): |
| 1148 | command = 'defcv' |
| 1149 | cat_class = idxsi[-1] |
| 1150 | class_class = string.join(idxsi[:-1]) |
| 1151 | elif len(idxsi) == 3 and idxsi[:2] == ['in', 'module']: |
| 1152 | command = 'defcv' |
| 1153 | cat_class = 'data' |
| 1154 | class_class = string.join(idxsi[1:]) |
| 1155 | elif len(idxsi) == 4 and idxsi[:3] == ['data', 'in', 'module']: |
| 1156 | command = 'defcv' |
| 1157 | cat_class = 'data' |
| 1158 | class_class = string.join(idxsi[2:]) |
Fred Drake | 11b6d24 | 1996-10-10 20:09:56 +0000 | [diff] [blame] | 1159 | else: |
| 1160 | command = 'defcv' |
| 1161 | cat_class = 'data' |
| 1162 | class_class = string.join(idxsi) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1163 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1164 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1165 | ch.data = command |
| 1166 | |
| 1167 | cslinearg = [chunk(GROUP, wh, [chunk(PLAIN, wh, cat_class)])] |
| 1168 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
| 1169 | if class_class: |
| 1170 | cslinearg.append(chunk(GROUP, wh, [chunk(PLAIN, wh, class_class)])) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1171 | cslinearg.append(chunk(PLAIN, wh, ' ')) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1172 | cslinearg.append(dataname) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1173 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1174 | pp.insert(i, chunk(GROUP, wh, cslinearg)) |
| 1175 | i, length = i+1, length+1 |
| 1176 | hist.command = command |
| 1177 | return length, i |
| 1178 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1179 | |
| 1180 | # regular indices: those that are not set in tt font by default.... |
| 1181 | regindices = ('cindex', ) |
| 1182 | |
| 1183 | # remove illegal characters from node names |
| 1184 | def rm_commas_etc(text): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1185 | result = '' |
| 1186 | changed = 0 |
| 1187 | while 1: |
| 1188 | pos = re_commas_etc.search(text) |
| 1189 | if pos >= 0: |
| 1190 | changed = 1 |
| 1191 | result = result + text[:pos] |
| 1192 | text = text[pos+1:] |
| 1193 | else: |
| 1194 | result = result + text |
| 1195 | break |
| 1196 | if changed: |
| 1197 | print 'Warning: nodename changhed to ' + `result` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1198 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1199 | return result |
| 1200 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1201 | # boolean flags |
| 1202 | flags = {'texi': 1} |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1203 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1204 | |
| 1205 | ## |
| 1206 | ## changeit: the actual routine, that changes the contents of the parsed |
| 1207 | ## chunks |
| 1208 | ## |
| 1209 | |
| 1210 | def changeit(buf, pp): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1211 | global onlylatexspecial, hist, out |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1212 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1213 | i, length = 0, len(pp) |
| 1214 | while 1: |
| 1215 | # sanity check: length should always equal len(pp) |
| 1216 | if len(pp) != length: |
| 1217 | raise 'FATAL', 'inconsistent length. thought ' + `length` + ', but should really be ' + `len(pp)` |
| 1218 | if i >= length: |
| 1219 | break |
| 1220 | ch = pp[i] |
| 1221 | i = i + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1222 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1223 | if type(ch) is StringType: |
| 1224 | #normally, only chunks are present in pp, |
| 1225 | # but in some cases, some extra info |
| 1226 | # has been inserted, e.g., the \end{...} clauses |
| 1227 | raise 'FATAL', 'got string, probably too many ' + `end` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1228 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1229 | if ch.chtype == chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1230 | # check for {\em ...} constructs |
| 1231 | if ch.data and \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1232 | ch.data[0].chtype == chunk_type[CSNAME] and \ |
| 1233 | s(buf, ch.data[0].data) in fontchanges.keys(): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1234 | k = s(buf, ch.data[0].data) |
| 1235 | del ch.data[0] |
| 1236 | pp.insert(i-1, chunk(CSNAME, ch.where, fontchanges[k])) |
| 1237 | length, i = length+1, i+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1238 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1239 | # recursively parse the contents of the group |
| 1240 | changeit(buf, ch.data) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1241 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1242 | elif ch.chtype == chunk_type[IF]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1243 | # \if... |
| 1244 | flag, negate, data = ch.data |
| 1245 | ##print 'IF: flag, negate = ' + `flag, negate` |
| 1246 | if flag not in flags.keys(): |
| 1247 | raise error, 'unknown flag ' + `flag` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1248 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1249 | value = flags[flag] |
| 1250 | if negate: |
| 1251 | value = (not value) |
| 1252 | del pp[i-1] |
| 1253 | length, i = length-1, i-1 |
| 1254 | if value: |
| 1255 | pp[i:i] = data |
| 1256 | length = length + len(data) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1257 | |
| 1258 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1259 | elif ch.chtype == chunk_type[ENV]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1260 | # \begin{...} .... |
| 1261 | envname, data = ch.data |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1262 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1263 | #push this environment name on stack |
| 1264 | hist.inenv.insert(0, envname) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1265 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1266 | #append an endenv chunk after grouped data |
| 1267 | data.append(chunk(ENDENV, ch.where, envname)) |
| 1268 | ##[`data`] |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1269 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1270 | #delete this object |
| 1271 | del pp[i-1] |
| 1272 | i, length = i-1, length-1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1273 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1274 | #insert found data |
| 1275 | pp[i:i] = data |
| 1276 | length = length + len(data) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1277 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1278 | if envname == 'verbatim': |
| 1279 | pp[i:i] = [chunk(CSLINE, ch.where, 'example'), |
| 1280 | chunk(GROUP, ch.where, [])] |
| 1281 | length, i = length+2, i+2 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1282 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1283 | elif envname == 'itemize': |
| 1284 | if hist.itemizenesting > len(itemizesymbols): |
| 1285 | raise error, 'too deep itemize nesting' |
| 1286 | ingroupch = [chunk(CSNAME, ch.where, |
| 1287 | itemizesymbols[hist.itemizenesting])] |
| 1288 | hist.itemizenesting = hist.itemizenesting + 1 |
| 1289 | pp[i:i] = [chunk(CSLINE, ch.where, 'itemize'), |
| 1290 | chunk(GROUP, ch.where, ingroupch)] |
| 1291 | length, i = length+2, i+2 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1292 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1293 | elif envname == 'enumerate': |
| 1294 | if hist.enumeratenesting > len(enumeratesymbols): |
| 1295 | raise error, 'too deep enumerate nesting' |
| 1296 | ingroupch = [chunk(PLAIN, ch.where, |
| 1297 | enumeratesymbols[hist.enumeratenesting])] |
| 1298 | hist.enumeratenesting = hist.enumeratenesting + 1 |
| 1299 | pp[i:i] = [chunk(CSLINE, ch.where, 'enumerate'), |
| 1300 | chunk(GROUP, ch.where, ingroupch)] |
| 1301 | length, i = length+2, i+2 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1302 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1303 | elif envname == 'description': |
| 1304 | ingroupch = [chunk(CSNAME, ch.where, 'b')] |
| 1305 | pp[i:i] = [chunk(CSLINE, ch.where, 'table'), |
| 1306 | chunk(GROUP, ch.where, ingroupch)] |
| 1307 | length, i = length+2, i+2 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1308 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1309 | elif (envname == 'tableiii') or (envname == 'tableii'): |
| 1310 | if (envname == 'tableii'): |
| 1311 | ltable = 2 |
| 1312 | else: |
| 1313 | ltable = 3 |
| 1314 | wh = ch.where |
| 1315 | newcode = [] |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1316 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1317 | #delete tabular format description |
| 1318 | # e.g., {|l|c|l|} |
| 1319 | length, newi = getnextarg(length, buf, pp, i) |
| 1320 | del pp[i:newi] |
| 1321 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1322 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1323 | newcode.append(chunk(CSLINE, wh, 'table')) |
| 1324 | ingroupch = [chunk(CSNAME, wh, 'asis')] |
| 1325 | newcode.append(chunk(GROUP, wh, ingroupch)) |
| 1326 | newcode.append(chunk(CSLINE, wh, 'item')) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1327 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1328 | #get the name of macro for @item |
| 1329 | # e.g., {code} |
| 1330 | length, newi = getnextarg(length, buf, pp, i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1331 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1332 | if newi-i != 1: |
| 1333 | raise error, 'Sorry, expected 1 chunk argument' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1334 | if pp[i].chtype != chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1335 | raise error, 'Sorry, expected plain text argument' |
| 1336 | hist.itemargmacro = s(buf, pp[i].data) |
| 1337 | del pp[i:newi] |
| 1338 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1339 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1340 | itembody = [] |
| 1341 | for count in range(ltable): |
| 1342 | length, newi = getnextarg(length, buf, pp, i) |
| 1343 | emphgroup = [ |
| 1344 | chunk(CSNAME, wh, 'emph'), |
| 1345 | chunk(GROUP, 0, pp[i:newi])] |
| 1346 | del pp[i:newi] |
| 1347 | length = length - (newi-i) |
| 1348 | if count == 0: |
| 1349 | itemarg = emphgroup |
| 1350 | elif count == ltable-1: |
| 1351 | itembody = itembody + \ |
| 1352 | [chunk(PLAIN, wh, ' --- ')] + emphgroup |
| 1353 | else: |
| 1354 | itembody = emphgroup |
| 1355 | newcode.append(chunk(GROUP, wh, itemarg)) |
| 1356 | newcode = newcode + itembody + [chunk(DENDLINE, wh, '\n')] |
| 1357 | pp[i:i] = newcode |
| 1358 | l = len(newcode) |
| 1359 | length, i = length+l, i+l |
| 1360 | del newcode, l |
| 1361 | |
| 1362 | if length != len(pp): |
| 1363 | raise 'STILL, SOMETHING wrong', `i` |
| 1364 | |
| 1365 | |
| 1366 | elif envname == 'funcdesc': |
| 1367 | pp.insert(i, chunk(PLAIN, ch.where, '')) |
| 1368 | i, length = i+1, length+1 |
| 1369 | length, i = do_funcdesc(length, buf, pp, i) |
| 1370 | |
| 1371 | elif envname == 'excdesc': |
| 1372 | pp.insert(i, chunk(PLAIN, ch.where, '')) |
| 1373 | i, length = i+1, length+1 |
| 1374 | length, i = do_excdesc(length, buf, pp, i) |
| 1375 | |
| 1376 | elif envname == 'datadesc': |
| 1377 | pp.insert(i, chunk(PLAIN, ch.where, '')) |
| 1378 | i, length = i+1, length+1 |
| 1379 | length, i = do_datadesc(length, buf, pp, i) |
| 1380 | |
| 1381 | else: |
| 1382 | print 'WARNING: don\'t know what to do with env ' + `envname` |
| 1383 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1384 | elif ch.chtype == chunk_type[ENDENV]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1385 | envname = ch.data |
| 1386 | if envname != hist.inenv[0]: |
| 1387 | raise error, '\'end\' does not match. Name ' + `envname` + ', expected ' + `hist.inenv[0]` |
| 1388 | del hist.inenv[0] |
| 1389 | del pp[i-1] |
| 1390 | i, length = i-1, length-1 |
| 1391 | |
| 1392 | if envname == 'verbatim': |
| 1393 | pp[i:i] = [ |
| 1394 | chunk(CSLINE, ch.where, 'end'), |
| 1395 | chunk(GROUP, ch.where, [ |
| 1396 | chunk(PLAIN, ch.where, 'example')])] |
| 1397 | i, length = i+2, length+2 |
| 1398 | elif envname == 'itemize': |
| 1399 | hist.itemizenesting = hist.itemizenesting - 1 |
| 1400 | pp[i:i] = [ |
| 1401 | chunk(CSLINE, ch.where, 'end'), |
| 1402 | chunk(GROUP, ch.where, [ |
| 1403 | chunk(PLAIN, ch.where, 'itemize')])] |
| 1404 | i, length = i+2, length+2 |
| 1405 | elif envname == 'enumerate': |
| 1406 | hist.enumeratenesting = hist.enumeratenesting-1 |
| 1407 | pp[i:i] = [ |
| 1408 | chunk(CSLINE, ch.where, 'end'), |
| 1409 | chunk(GROUP, ch.where, [ |
| 1410 | chunk(PLAIN, ch.where, 'enumerate')])] |
| 1411 | i, length = i+2, length+2 |
| 1412 | elif envname == 'description': |
| 1413 | pp[i:i] = [ |
| 1414 | chunk(CSLINE, ch.where, 'end'), |
| 1415 | chunk(GROUP, ch.where, [ |
| 1416 | chunk(PLAIN, ch.where, 'table')])] |
| 1417 | i, length = i+2, length+2 |
| 1418 | elif (envname == 'tableiii') or (envname == 'tableii'): |
| 1419 | pp[i:i] = [ |
| 1420 | chunk(CSLINE, ch.where, 'end'), |
| 1421 | chunk(GROUP, ch.where, [ |
| 1422 | chunk(PLAIN, ch.where, 'table')])] |
| 1423 | i, length = i+2, length + 2 |
| 1424 | pp.insert(i, chunk(DENDLINE, ch.where, '\n')) |
| 1425 | i, length = i+1, length+1 |
| 1426 | |
| 1427 | elif envname in ('funcdesc', 'excdesc', 'datadesc'): |
| 1428 | pp[i:i] = [ |
| 1429 | chunk(CSLINE, ch.where, 'end'), |
| 1430 | chunk(GROUP, ch.where, [ |
| 1431 | chunk(PLAIN, ch.where, hist.command)])] |
| 1432 | i, length = i+2, length+2 |
| 1433 | else: |
| 1434 | print 'WARNING: ending env ' + `envname` + 'has no actions' |
| 1435 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1436 | elif ch.chtype == chunk_type[CSNAME]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1437 | # control name transformations |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1438 | s_buf_data = s(buf, ch.data) |
| 1439 | if s_buf_data == 'optional': |
| 1440 | pp[i-1].chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1441 | pp[i-1].data = '[' |
| 1442 | if (i < length) and \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1443 | (pp[i].chtype == chunk_type[GROUP]): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1444 | cp=pp[i].data |
| 1445 | pp[i:i+1]=cp + [ |
| 1446 | chunk(PLAIN, ch.where, ']')] |
| 1447 | length = length+len(cp) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1448 | elif s_buf_data in ignoredcommands: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1449 | del pp[i-1] |
| 1450 | i, length = i-1, length-1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1451 | elif s_buf_data == '@' and \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1452 | i != length and \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1453 | pp[i].chtype == chunk_type[PLAIN] and \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1454 | s(buf, pp[i].data)[0] == '.': |
| 1455 | # \@. --> \. --> @. |
| 1456 | ch.data = '.' |
| 1457 | del pp[i] |
| 1458 | length = length-1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1459 | elif s_buf_data == '\\': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1460 | # \\ --> \* --> @* |
| 1461 | ch.data = '*' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1462 | elif len(s_buf_data) == 1 and \ |
| 1463 | s_buf_data in onlylatexspecial: |
| 1464 | ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1465 | # check if such a command is followed by |
| 1466 | # an empty group: e.g., `\%{}'. If so, remove |
| 1467 | # this empty group too |
| 1468 | if i < length and \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1469 | pp[i].chtype == chunk_type[GROUP] \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1470 | and len(pp[i].data) == 0: |
| 1471 | del pp[i] |
| 1472 | length = length-1 |
| 1473 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1474 | elif hist.inargs and s_buf_data in inargsselves: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1475 | # This is the special processing of the |
| 1476 | # arguments of the \begin{funcdesc}... or |
| 1477 | # \funcline... arguments |
| 1478 | # \, --> , \[ --> [, \] --> ] |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1479 | ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1480 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1481 | elif s_buf_data == 'renewcommand': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1482 | # \renewcommand{\indexsubitem}.... |
| 1483 | i, length = i-1, length-1 |
| 1484 | del pp[i] |
| 1485 | length, newi = getnextarg(length, buf, pp, i) |
| 1486 | if newi-i == 1 \ |
| 1487 | and i < length \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1488 | and pp[i].chtype == chunk_type[CSNAME] \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1489 | and s(buf, pp[i].data) == 'indexsubitem': |
| 1490 | del pp[i:newi] |
| 1491 | length = length - (newi-i) |
| 1492 | length, newi = getnextarg(length, buf, pp, i) |
| 1493 | text = flattext(buf, pp[i:newi]) |
| 1494 | if text[:1] != '(' or text[-1:] != ')': |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1495 | raise error, \ |
| 1496 | 'expected indexsubitem enclosed in parenteses' |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1497 | words = string.split(text[1:-1]) |
| 1498 | hist.indexsubitem = words |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1499 | ## print 'set hist.indexsubitem =', words |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1500 | del text, words |
| 1501 | else: |
| 1502 | print 'WARNING: renewcommand with unsupported arg removed' |
| 1503 | del pp[i:newi] |
| 1504 | length = length - (newi-i) |
| 1505 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1506 | elif s_buf_data == 'item': |
| 1507 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1508 | length, newi = getoptarg(length, buf, pp, i) |
| 1509 | ingroupch = pp[i:newi] |
| 1510 | del pp[i:newi] |
| 1511 | length = length - (newi-i) |
Fred Drake | 893e5e0 | 1996-10-25 22:13:10 +0000 | [diff] [blame] | 1512 | changeit(buf, ingroupch) # catch stuff inside the optional arg |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1513 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1514 | i, length = i+1, length+1 |
| 1515 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1516 | elif s_buf_data == 'ttindex': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1517 | idxsi = hist.indexsubitem |
| 1518 | |
| 1519 | cat_class = '' |
| 1520 | if len(idxsi) >= 2 and idxsi[1] in \ |
| 1521 | ('method', 'function', 'protocol'): |
| 1522 | command = 'findex' |
| 1523 | elif len(idxsi) >= 2 and idxsi[1] in \ |
| 1524 | ('exception', 'object'): |
| 1525 | command = 'vindex' |
Fred Drake | 7edd8d3 | 1996-10-09 16:11:26 +0000 | [diff] [blame] | 1526 | elif len(idxsi) == 3 and idxsi[:2] == ['in', 'module']: |
| 1527 | command = 'cindex' |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1528 | else: |
Fred Drake | 7edd8d3 | 1996-10-09 16:11:26 +0000 | [diff] [blame] | 1529 | print 'WARNING: can\'t categorize ' + `idxsi` \ |
| 1530 | + ' for \'ttindex\' command' |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1531 | command = 'cindex' |
| 1532 | |
| 1533 | if not cat_class: |
| 1534 | cat_class = '('+string.join(idxsi)+')' |
| 1535 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1536 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1537 | ch.data = command |
| 1538 | |
| 1539 | length, newi = getnextarg(length, buf, pp, i) |
| 1540 | arg = pp[i:newi] |
| 1541 | del pp[i:newi] |
| 1542 | length = length - (newi-i) |
| 1543 | |
| 1544 | cat_arg = [chunk(PLAIN, ch.where, cat_class)] |
| 1545 | |
| 1546 | # determine what should be set in roman, and |
| 1547 | # what in tt-font |
| 1548 | if command in regindices: |
| 1549 | |
| 1550 | arg = [chunk(CSNAME, ch.where, 't'), |
| 1551 | chunk(GROUP, ch.where, arg)] |
| 1552 | else: |
| 1553 | cat_arg = [chunk(CSNAME, ch.where, 'r'), |
| 1554 | chunk(GROUP, ch.where, cat_arg)] |
| 1555 | |
| 1556 | ingroupch = arg + \ |
| 1557 | [chunk(PLAIN, ch.where, ' ')] + \ |
| 1558 | cat_arg |
| 1559 | |
| 1560 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1561 | length, i = length+1, i+1 |
| 1562 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1563 | elif s_buf_data == 'ldots': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1564 | # \ldots --> \dots{} --> @dots{} |
| 1565 | ch.data = 'dots' |
| 1566 | if i == length \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1567 | or pp[i].chtype != chunk_type[GROUP] \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1568 | or pp[i].data != []: |
| 1569 | pp.insert(i, chunk(GROUP, ch.where, [])) |
| 1570 | i, length = i+1, length+1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1571 | elif s_buf_data in themselves: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1572 | # \UNIX --> UNIX |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1573 | ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1574 | if i != length \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1575 | and pp[i].chtype == chunk_type[GROUP] \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1576 | and pp[i].data == []: |
| 1577 | del pp[i] |
| 1578 | length = length-1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1579 | elif s_buf_data in for_texi: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1580 | pass |
| 1581 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1582 | elif s_buf_data == 'e': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1583 | # "\e" --> "\" |
| 1584 | ch.data = '\\' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1585 | ch.chtype = chunk_type[PLAIN] |
| 1586 | elif s_buf_data in ('lineiii', 'lineii'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1587 | # This is the most tricky one |
| 1588 | # \lineiii{a1}{a2}[{a3}] --> |
| 1589 | # @item @<cts. of itemargmacro>{a1} |
| 1590 | # a2 [ -- a3] |
| 1591 | # |
| 1592 | ##print 'LINEIIIIII!!!!!!!' |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 1593 | ## wobj = Wobj() |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1594 | ## dumpit(buf, wobj.write, pp[i-1:i+5]) |
| 1595 | ## print '--->' + wobj.data + '<----' |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1596 | if not hist.inenv: |
| 1597 | raise error, 'no environment for lineiii' |
| 1598 | if (hist.inenv[0] != 'tableiii') and \ |
| 1599 | (hist.inenv[0] != 'tableii'): |
| 1600 | raise error, \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1601 | 'wrong command (%s) in wrong environment (%s)' \ |
| 1602 | % (s_buf_data, `hist.inenv[0]`) |
| 1603 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1604 | ch.data = 'item' |
| 1605 | length, newi = getnextarg(length, buf, pp, i) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1606 | ingroupch = [chunk(CSNAME, 0, hist.itemargmacro), |
| 1607 | chunk(GROUP, 0, pp[i:newi])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1608 | del pp[i:newi] |
| 1609 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1610 | ## print 'ITEM ARG: --->', |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 1611 | ## wobj = Wobj() |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1612 | ## dumpit(buf, wobj.write, ingroupch) |
| 1613 | ## print wobj.data, '<---' |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1614 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1615 | grouppos = i |
| 1616 | i, length = i+1, length+1 |
| 1617 | length, i = getnextarg(length, buf, pp, i) |
| 1618 | length, newi = getnextarg(length, buf, pp, i) |
| 1619 | if newi > i: |
| 1620 | # we have a 3rd arg |
| 1621 | pp.insert(i, chunk(PLAIN, ch.where, ' --- ')) |
| 1622 | i = newi + 1 |
| 1623 | length = length + 1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1624 | ## pp[grouppos].data = pp[grouppos].data \ |
| 1625 | ## + [chunk(PLAIN, ch.where, ' ')] \ |
| 1626 | ## + pp[i:newi] |
| 1627 | ## del pp[i:newi] |
| 1628 | ## length = length - (newi-i) |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1629 | if length != len(pp): |
| 1630 | raise 'IN LINEIII IS THE ERR', `i` |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1631 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1632 | elif s_buf_data in ('chapter', 'section', 'subsection', 'subsubsection'): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1633 | #\xxxsection{A} ----> |
| 1634 | # @node A, , , |
| 1635 | # @xxxsection A |
| 1636 | ## also: remove commas and quotes |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1637 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1638 | length, newi = getnextarg(length, buf, pp, i) |
| 1639 | afternodenamecmd = next_command_p(length, buf, pp, newi, 'nodename') |
| 1640 | if afternodenamecmd < 0: |
| 1641 | cp1 = crcopy(pp[i:newi]) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1642 | pp[i:newi] = [chunk(GROUP, ch.where, pp[i:newi])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1643 | length, newi = length - (newi-i) + 1, i+1 |
| 1644 | text = flattext(buf, cp1) |
| 1645 | text = invent_node_names(text) |
| 1646 | else: |
| 1647 | length, endarg = getnextarg(length, buf, pp, afternodenamecmd) |
| 1648 | cp1 = crcopy(pp[afternodenamecmd:endarg]) |
| 1649 | del pp[newi:endarg] |
| 1650 | length = length - (endarg-newi) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1651 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1652 | pp[i:newi] = [chunk(GROUP, ch.where, pp[i:newi])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1653 | length, newi = length - (newi-i) + 1, i + 1 |
| 1654 | text = flattext(buf, cp1) |
| 1655 | if text[-1] == '.': |
| 1656 | text = text[:-1] |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 1657 | ## print 'FLATTEXT:', `text` |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1658 | if text in hist.nodenames: |
| 1659 | print 'WARNING: node name ' + `text` + ' already used' |
| 1660 | out.doublenodes.append(text) |
| 1661 | else: |
| 1662 | hist.nodenames.append(text) |
| 1663 | text = rm_commas_etc(text) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1664 | pp[i-1:i-1] = [chunk(CSLINE, ch.where, 'node'), |
| 1665 | chunk(GROUP, ch.where, [ |
| 1666 | chunk(PLAIN, ch.where, text+', , ,') |
| 1667 | ])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1668 | i, length = newi+2, length+2 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1669 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1670 | elif s_buf_data == 'funcline': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1671 | # fold it to a very short environment |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1672 | pp[i-1:i-1] = [chunk(CSLINE, ch.where, 'end'), |
| 1673 | chunk(GROUP, ch.where, [ |
| 1674 | chunk(PLAIN, ch.where, hist.command)])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1675 | i, length = i+2, length+2 |
| 1676 | length, i = do_funcdesc(length, buf, pp, i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1677 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1678 | elif s_buf_data == 'dataline': |
| 1679 | pp[i-1:i-1] = [chunk(CSLINE, ch.where, 'end'), |
| 1680 | chunk(GROUP, ch.where, [ |
| 1681 | chunk(PLAIN, ch.where, hist.command)])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1682 | i, length = i+2, length+2 |
| 1683 | length, i = do_datadesc(length, buf, pp, i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1684 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1685 | elif s_buf_data == 'excline': |
| 1686 | pp[i-1:i-1] = [chunk(CSLINE, ch.where, 'end'), |
| 1687 | chunk(GROUP, ch.where, [ |
| 1688 | chunk(PLAIN, ch.where, hist.command)])] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1689 | i, length = i+2, length+2 |
| 1690 | length, i = do_excdesc(length, buf, pp, i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1691 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1692 | elif s_buf_data == 'index': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1693 | #\index{A} ---> |
| 1694 | # @cindex A |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1695 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1696 | ch.data = 'cindex' |
| 1697 | length, newi = getnextarg(length, buf, pp, i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1698 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1699 | ingroupch = pp[i:newi] |
| 1700 | del pp[i:newi] |
| 1701 | length = length - (newi-i) |
| 1702 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1703 | length, i = length+1, i+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1704 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1705 | elif s_buf_data == 'bifuncindex': |
| 1706 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1707 | ch.data = 'findex' |
| 1708 | length, newi = getnextarg(length, buf, pp, i) |
| 1709 | ingroupch = pp[i:newi] |
| 1710 | del pp[i:newi] |
| 1711 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1712 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1713 | ingroupch.append(chunk(PLAIN, ch.where, ' ')) |
| 1714 | ingroupch.append(chunk(CSNAME, ch.where, 'r')) |
| 1715 | ingroupch.append(chunk(GROUP, ch.where, [ |
| 1716 | chunk(PLAIN, ch.where, |
| 1717 | '(built-in function)')])) |
| 1718 | |
| 1719 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1720 | length, i = length+1, i+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1721 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1722 | elif s_buf_data == 'obindex': |
| 1723 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1724 | ch.data = 'findex' |
| 1725 | length, newi = getnextarg(length, buf, pp, i) |
| 1726 | ingroupch = pp[i:newi] |
| 1727 | del pp[i:newi] |
| 1728 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1729 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1730 | ingroupch.append(chunk(PLAIN, ch.where, ' ')) |
| 1731 | ingroupch.append(chunk(CSNAME, ch.where, 'r')) |
| 1732 | ingroupch.append(chunk(GROUP, ch.where, [ |
| 1733 | chunk(PLAIN, ch.where, |
| 1734 | '(object)')])) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1735 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1736 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1737 | length, i = length+1, i+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1738 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1739 | elif s_buf_data == 'opindex': |
| 1740 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1741 | ch.data = 'findex' |
| 1742 | length, newi = getnextarg(length, buf, pp, i) |
| 1743 | ingroupch = pp[i:newi] |
| 1744 | del pp[i:newi] |
| 1745 | length = length - (newi-i) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1746 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1747 | ingroupch.append(chunk(PLAIN, ch.where, ' ')) |
| 1748 | ingroupch.append(chunk(CSNAME, ch.where, 'r')) |
| 1749 | ingroupch.append(chunk(GROUP, ch.where, [ |
| 1750 | chunk(PLAIN, ch.where, |
| 1751 | '(operator)')])) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1752 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1753 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1754 | length, i = length+1, i+1 |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1755 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1756 | elif s_buf_data == 'bimodindex': |
| 1757 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1758 | ch.data = 'pindex' |
| 1759 | length, newi = getnextarg(length, buf, pp, i) |
| 1760 | ingroupch = pp[i:newi] |
| 1761 | del pp[i:newi] |
| 1762 | length = length - (newi-i) |
| 1763 | |
| 1764 | ingroupch.append(chunk(PLAIN, ch.where, ' ')) |
| 1765 | ingroupch.append(chunk(CSNAME, ch.where, 'r')) |
| 1766 | ingroupch.append(chunk(GROUP, ch.where, [ |
| 1767 | chunk(PLAIN, ch.where, |
| 1768 | '(built-in)')])) |
| 1769 | |
| 1770 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1771 | length, i = length+1, i+1 |
| 1772 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1773 | elif s_buf_data == 'sectcode': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1774 | ch.data = 'code' |
| 1775 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1776 | elif s_buf_data == 'stmodindex': |
| 1777 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1778 | # use the program index as module index |
| 1779 | ch.data = 'pindex' |
| 1780 | length, newi = getnextarg(length, buf, pp, i) |
| 1781 | ingroupch = pp[i:newi] |
| 1782 | del pp[i:newi] |
| 1783 | length = length - (newi-i) |
| 1784 | |
| 1785 | ingroupch.append(chunk(PLAIN, ch.where, ' ')) |
| 1786 | ingroupch.append(chunk(CSNAME, ch.where, 'r')) |
| 1787 | ingroupch.append(chunk(GROUP, ch.where, [ |
| 1788 | chunk(PLAIN, ch.where, |
| 1789 | '(standard)')])) |
| 1790 | |
| 1791 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1792 | length, i = length+1, i+1 |
| 1793 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1794 | elif s_buf_data == 'stindex': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1795 | # XXX must actually go to newindex st |
| 1796 | wh = ch.where |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1797 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1798 | ch.data = 'cindex' |
| 1799 | length, newi = getnextarg(length, buf, pp, i) |
| 1800 | ingroupch = [chunk(CSNAME, wh, 'code'), |
| 1801 | chunk(GROUP, wh, pp[i:newi])] |
| 1802 | |
| 1803 | del pp[i:newi] |
| 1804 | length = length - (newi-i) |
| 1805 | |
| 1806 | t = ingroupch[:] |
| 1807 | t.append(chunk(PLAIN, wh, ' statement')) |
| 1808 | |
| 1809 | pp.insert(i, chunk(GROUP, wh, t)) |
| 1810 | i, length = i+1, length+1 |
| 1811 | |
| 1812 | pp.insert(i, chunk(CSLINE, wh, 'cindex')) |
| 1813 | i, length = i+1, length+1 |
| 1814 | |
| 1815 | t = ingroupch[:] |
| 1816 | t.insert(0, chunk(PLAIN, wh, 'statement, ')) |
| 1817 | |
| 1818 | pp.insert(i, chunk(GROUP, wh, t)) |
| 1819 | i, length = i+1, length+1 |
| 1820 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1821 | elif s_buf_data == 'indexii': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1822 | #\indexii{A}{B} ---> |
| 1823 | # @cindex A B |
| 1824 | # @cindex B, A |
| 1825 | length, newi = getnextarg(length, buf, pp, i) |
| 1826 | cp11 = pp[i:newi] |
| 1827 | cp21 = crcopy(pp[i:newi]) |
| 1828 | del pp[i:newi] |
| 1829 | length = length - (newi-i) |
| 1830 | length, newi = getnextarg(length, buf, pp, i) |
| 1831 | cp12 = pp[i:newi] |
| 1832 | cp22 = crcopy(pp[i:newi]) |
| 1833 | del pp[i:newi] |
| 1834 | length = length - (newi-i) |
| 1835 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1836 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1837 | ch.data = 'cindex' |
| 1838 | pp.insert(i, chunk(GROUP, ch.where, cp11 + [ |
| 1839 | chunk(PLAIN, ch.where, ' ')] + cp12)) |
| 1840 | i, length = i+1, length+1 |
| 1841 | pp[i:i] = [chunk(CSLINE, ch.where, 'cindex'), |
| 1842 | chunk(GROUP, ch.where, cp22 + [ |
| 1843 | chunk(PLAIN, ch.where, ', ')]+ cp21)] |
| 1844 | i, length = i+2, length+2 |
| 1845 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1846 | elif s_buf_data == 'indexiii': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1847 | length, newi = getnextarg(length, buf, pp, i) |
| 1848 | cp11 = pp[i:newi] |
| 1849 | cp21 = crcopy(pp[i:newi]) |
| 1850 | cp31 = crcopy(pp[i:newi]) |
| 1851 | del pp[i:newi] |
| 1852 | length = length - (newi-i) |
| 1853 | length, newi = getnextarg(length, buf, pp, i) |
| 1854 | cp12 = pp[i:newi] |
| 1855 | cp22 = crcopy(pp[i:newi]) |
| 1856 | cp32 = crcopy(pp[i:newi]) |
| 1857 | del pp[i:newi] |
| 1858 | length = length - (newi-i) |
| 1859 | length, newi = getnextarg(length, buf, pp, i) |
| 1860 | cp13 = pp[i:newi] |
| 1861 | cp23 = crcopy(pp[i:newi]) |
| 1862 | cp33 = crcopy(pp[i:newi]) |
| 1863 | del pp[i:newi] |
| 1864 | length = length - (newi-i) |
| 1865 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1866 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1867 | ch.data = 'cindex' |
| 1868 | pp.insert(i, chunk(GROUP, ch.where, cp11 + [ |
| 1869 | chunk(PLAIN, ch.where, ' ')] + cp12 |
| 1870 | + [chunk(PLAIN, ch.where, ' ')] |
| 1871 | + cp13)) |
| 1872 | i, length = i+1, length+1 |
| 1873 | pp[i:i] = [chunk(CSLINE, ch.where, 'cindex'), |
| 1874 | chunk(GROUP, ch.where, cp22 + [ |
| 1875 | chunk(PLAIN, ch.where, ' ')]+ cp23 |
| 1876 | + [chunk(PLAIN, ch.where, ', ')] + |
| 1877 | cp21)] |
| 1878 | i, length = i+2, length+2 |
| 1879 | pp[i:i] = [chunk(CSLINE, ch.where, 'cindex'), |
| 1880 | chunk(GROUP, ch.where, cp33 + [ |
| 1881 | chunk(PLAIN, ch.where, ', ')]+ cp31 |
| 1882 | + [chunk(PLAIN, ch.where, ' ')] + |
| 1883 | cp32)] |
| 1884 | i, length = i+2, length+2 |
| 1885 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1886 | elif s_buf_data == 'indexiv': |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1887 | length, newi = getnextarg(length, buf, pp, i) |
| 1888 | cp11 = pp[i:newi] |
| 1889 | cp21 = crcopy(pp[i:newi]) |
| 1890 | cp31 = crcopy(pp[i:newi]) |
| 1891 | cp41 = crcopy(pp[i:newi]) |
| 1892 | del pp[i:newi] |
| 1893 | length = length - (newi-i) |
| 1894 | length, newi = getnextarg(length, buf, pp, i) |
| 1895 | cp12 = pp[i:newi] |
| 1896 | cp22 = crcopy(pp[i:newi]) |
| 1897 | cp32 = crcopy(pp[i:newi]) |
| 1898 | cp42 = crcopy(pp[i:newi]) |
| 1899 | del pp[i:newi] |
| 1900 | length = length - (newi-i) |
| 1901 | length, newi = getnextarg(length, buf, pp, i) |
| 1902 | cp13 = pp[i:newi] |
| 1903 | cp23 = crcopy(pp[i:newi]) |
| 1904 | cp33 = crcopy(pp[i:newi]) |
| 1905 | cp43 = crcopy(pp[i:newi]) |
| 1906 | del pp[i:newi] |
| 1907 | length = length - (newi-i) |
| 1908 | length, newi = getnextarg(length, buf, pp, i) |
| 1909 | cp14 = pp[i:newi] |
| 1910 | cp24 = crcopy(pp[i:newi]) |
| 1911 | cp34 = crcopy(pp[i:newi]) |
| 1912 | cp44 = crcopy(pp[i:newi]) |
| 1913 | del pp[i:newi] |
| 1914 | length = length - (newi-i) |
| 1915 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1916 | ch.chtype = chunk_type[CSLINE] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1917 | ch.data = 'cindex' |
| 1918 | ingroupch = cp11 + \ |
| 1919 | spacech + cp12 + \ |
| 1920 | spacech + cp13 + \ |
| 1921 | spacech + cp14 |
| 1922 | pp.insert(i, chunk(GROUP, ch.where, ingroupch)) |
| 1923 | i, length = i+1, length+1 |
| 1924 | ingroupch = cp22 + \ |
| 1925 | spacech + cp23 + \ |
| 1926 | spacech + cp24 + \ |
| 1927 | commach + cp21 |
| 1928 | pp[i:i] = cindexch + [ |
| 1929 | chunk(GROUP, ch.where, ingroupch)] |
| 1930 | i, length = i+2, length+2 |
| 1931 | ingroupch = cp33 + \ |
| 1932 | spacech + cp34 + \ |
| 1933 | commach + cp31 + \ |
| 1934 | spacech + cp32 |
| 1935 | pp[i:i] = cindexch + [ |
| 1936 | chunk(GROUP, ch.where, ingroupch)] |
| 1937 | i, length = i+2, length+2 |
| 1938 | ingroupch = cp44 + \ |
| 1939 | commach + cp41 + \ |
| 1940 | spacech + cp42 + \ |
| 1941 | spacech + cp43 |
| 1942 | pp[i:i] = cindexch + [ |
| 1943 | chunk(GROUP, ch.where, ingroupch)] |
| 1944 | i, length = i+2, length+2 |
| 1945 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1946 | ## elif s_buf_data == 'indexsubitem': |
| 1947 | ## ch.data = flattext(buf, [ch]) |
| 1948 | ## ch.chtype = chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1949 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1950 | elif s_buf_data in ('noindent', 'indexsubitem'): |
| 1951 | pass |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1952 | |
| 1953 | else: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1954 | print "don't know what to do with keyword " + s_buf_data |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1955 | |
| 1956 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1957 | re_atsign = regex.compile('[@{}]') |
| 1958 | re_newline = regex.compile('\n') |
| 1959 | |
| 1960 | def dumpit(buf, wm, pp): |
| 1961 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1962 | global out |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 1963 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1964 | i, length = 0, len(pp) |
| 1965 | |
| 1966 | addspace = 0 |
| 1967 | |
| 1968 | while 1: |
| 1969 | if len(pp) != length: |
| 1970 | raise 'FATAL', 'inconsistent length' |
| 1971 | if i == length: |
| 1972 | break |
| 1973 | ch = pp[i] |
| 1974 | i = i + 1 |
| 1975 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1976 | dospace = addspace |
| 1977 | addspace = 0 |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1978 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1979 | if ch.chtype == chunk_type[CSNAME]: |
| 1980 | s_buf_data = s(buf, ch.data) |
| 1981 | wm('@' + s_buf_data) |
| 1982 | if s_buf_data == 'node' and \ |
| 1983 | pp[i].chtype == chunk_type[PLAIN] and \ |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1984 | s(buf, pp[i].data) in out.doublenodes: |
| 1985 | ##XXX doesnt work yet?? |
| 1986 | wm(' ZZZ-' + zfill(`i`, 4)) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1987 | if s_buf_data[0] in string.letters: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1988 | addspace = 1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 1989 | elif ch.chtype == chunk_type[PLAIN]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 1990 | if dospace and s(buf, ch.data) not in (' ', '\t'): |
| 1991 | wm(' ') |
| 1992 | text = s(buf, ch.data) |
| 1993 | while 1: |
| 1994 | pos = re_atsign.search(text) |
| 1995 | if pos < 0: |
| 1996 | break |
| 1997 | wm(text[:pos] + '@' + text[pos]) |
| 1998 | text = text[pos+1:] |
| 1999 | wm(text) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2000 | elif ch.chtype == chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2001 | wm('{') |
| 2002 | dumpit(buf, wm, ch.data) |
| 2003 | wm('}') |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2004 | elif ch.chtype == chunk_type[DENDLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2005 | wm('\n\n') |
| 2006 | while i != length and pp[i].chtype in \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2007 | (chunk_type[DENDLINE], chunk_type[ENDLINE]): |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2008 | i = i + 1 |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2009 | elif ch.chtype == chunk_type[OTHER]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2010 | wm(s(buf, ch.data)) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2011 | elif ch.chtype == chunk_type[ACTIVE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2012 | wm(s(buf, ch.data)) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2013 | elif ch.chtype == chunk_type[ENDLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2014 | wm('\n') |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2015 | elif ch.chtype == chunk_type[CSLINE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2016 | if i >= 2 and pp[i-2].chtype not in \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2017 | (chunk_type[ENDLINE], chunk_type[DENDLINE]) \ |
| 2018 | and (pp[i-2].chtype != chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2019 | or s(buf, pp[i-2].data)[-1] != '\n'): |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2020 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2021 | wm('\n') |
| 2022 | wm('@' + s(buf, ch.data)) |
| 2023 | if i == length: |
| 2024 | raise error, 'CSLINE expected another chunk' |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2025 | if pp[i].chtype != chunk_type[GROUP]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2026 | raise error, 'CSLINE expected GROUP' |
| 2027 | if type(pp[i].data) != ListType: |
| 2028 | raise error, 'GROUP chould contain []-data' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2029 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2030 | wobj = Wobj() |
| 2031 | dumpit(buf, wobj.write, pp[i].data) |
| 2032 | i = i + 1 |
| 2033 | text = wobj.data |
| 2034 | del wobj |
| 2035 | if text: |
| 2036 | wm(' ') |
| 2037 | while 1: |
| 2038 | pos = re_newline.search(text) |
| 2039 | if pos < 0: |
| 2040 | break |
| 2041 | print 'WARNING: found newline in csline arg' |
| 2042 | wm(text[:pos] + ' ') |
| 2043 | text = text[pos+1:] |
| 2044 | wm(text) |
| 2045 | if i >= length or \ |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2046 | pp[i].chtype not in (chunk_type[CSLINE], |
| 2047 | chunk_type[ENDLINE], chunk_type[DENDLINE]) \ |
| 2048 | and (pp[i].chtype != chunk_type[PLAIN] |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2049 | or s(buf, pp[i].data)[0] != '\n'): |
| 2050 | wm('\n') |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 2051 | |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2052 | elif ch.chtype == chunk_type[COMMENT]: |
| 2053 | ## print 'COMMENT: previous chunk =', pp[i-2] |
| 2054 | ## if pp[i-2].chtype == chunk_type[PLAIN]: |
| 2055 | ## print 'PLAINTEXT =', `s(buf, pp[i-2].data)` |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2056 | if s(buf, ch.data) and \ |
| 2057 | regex.match('^[ \t]*$', s(buf, ch.data)) < 0: |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2058 | if i >= 2 \ |
| 2059 | and pp[i-2].chtype not in (chunk_type[ENDLINE], chunk_type[DENDLINE]) \ |
| 2060 | and not (pp[i-2].chtype == chunk_type[PLAIN] |
| 2061 | and regex.match('\\(.\\|\n\\)*[ \t]*\n$', s(buf, pp[i-2].data)) >= 0): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2062 | wm('\n') |
| 2063 | wm('@c ' + s(buf, ch.data)) |
Guido van Rossum | 36f219d | 1996-09-11 21:30:40 +0000 | [diff] [blame] | 2064 | elif ch.chtype == chunk_type[IGNORE]: |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2065 | pass |
| 2066 | else: |
| 2067 | try: |
| 2068 | str = `s(buf, ch.data)` |
| 2069 | except TypeError: |
| 2070 | str = `ch.data` |
| 2071 | if len(str) > 400: |
| 2072 | str = str[:400] + '...' |
| 2073 | print 'warning:', ch.chtype, 'not handled, data ' + str |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2074 | |
| 2075 | |
| 2076 | |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2077 | def main(): |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2078 | outfile = None |
| 2079 | headerfile = 'texipre.dat' |
| 2080 | trailerfile = 'texipost.dat' |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2081 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2082 | try: |
| 2083 | opts, args = getopt.getopt(sys.argv[1:], 'o:h:t:') |
| 2084 | except getopt.error: |
| 2085 | args = [] |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2086 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2087 | if not args: |
| 2088 | print 'usage: partparse [-o outfile] [-h headerfile]', |
| 2089 | print '[-t trailerfile] file ...' |
| 2090 | sys.exit(2) |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 2091 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2092 | for opt, arg in opts: |
| 2093 | if opt == '-o': outfile = arg |
| 2094 | if opt == '-h': headerfile = arg |
| 2095 | if opt == '-t': trailerfile = arg |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 2096 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2097 | if not outfile: |
| 2098 | root, ext = os.path.splitext(args[0]) |
| 2099 | outfile = root + '.texi' |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 2100 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2101 | if outfile in args: |
| 2102 | print 'will not overwrite input file', outfile |
| 2103 | sys.exit(2) |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 2104 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2105 | outf = open(outfile, 'w') |
| 2106 | outf.write(open(headerfile, 'r').read()) |
Guido van Rossum | 7a2dba2 | 1993-11-05 14:45:11 +0000 | [diff] [blame] | 2107 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2108 | for file in args: |
| 2109 | if len(args) > 1: print '='*20, file, '='*20 |
| 2110 | buf = open(file, 'r').read() |
| 2111 | w, pp = parseit(buf) |
| 2112 | startchange() |
| 2113 | changeit(buf, pp) |
| 2114 | dumpit(buf, outf.write, pp) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2115 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2116 | outf.write(open(trailerfile, 'r').read()) |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2117 | |
Guido van Rossum | 5f18d6c | 1996-09-10 22:34:20 +0000 | [diff] [blame] | 2118 | outf.close() |
Guido van Rossum | 95cd2ef | 1992-12-08 14:37:55 +0000 | [diff] [blame] | 2119 | |
Guido van Rossum | 49604d3 | 1996-09-10 22:19:51 +0000 | [diff] [blame] | 2120 | if __name__ == "__main__": |
| 2121 | main() |