2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2009,2010,2011,2012,2014, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
36 * \brief Helper functions for the selection tokenizer.
38 * This file implements the functions in the headers scanner.h and
41 * \author Teemu Murtola <teemu.murtola@gmail.com>
42 * \ingroup module_selection
45 * \internal \file scanner_flex.h
46 * \brief Generated (from scanner.l) header file by Flex.
48 * This file contains definitions of functions that are needed in
49 * scanner_internal.cpp.
51 * \ingroup module_selection
56 #include "scanner_internal.h"
63 #include "gromacs/utility/cstringutil.h"
64 #include "gromacs/utility/exceptions.h"
65 #include "gromacs/utility/gmxassert.h"
66 #include "gromacs/utility/messagestringcollector.h"
67 #include "gromacs/utility/smalloc.h"
68 #include "gromacs/utility/stringutil.h"
71 #include "parsetree.h"
73 #include "selectioncollection-impl.h"
75 #include "selmethod.h"
79 * Step in which the allocated memory for pretty-printed input is incremented.
81 #define STRSTORE_ALLOCSTEP 1000
83 /* These are defined as macros in the generated scanner_flex.h.
84 * We undefine them here to have them as variable names in the subroutines.
85 * There are other ways of doing this, but this is probably the easiest. */
91 * Handles initialization of method parameter token.
94 init_param_token(YYSTYPE *yylval, gmx_ana_selparam_t *param, bool bBoolNo)
98 GMX_RELEASE_ASSERT(param->name != NULL,
99 "bBoolNo should only be set for a parameters with a name");
100 snew(yylval->str, strlen(param->name) + 3);
101 yylval->str[0] = 'n';
102 yylval->str[1] = 'o';
103 strcpy(yylval->str+2, param->name);
107 yylval->str = param->name ? gmx_strdup(param->name) : NULL;
113 * Processes a selection method token.
116 init_method_token(YYSTYPE *yylval, YYLTYPE *yylloc,
117 const gmx::SelectionParserSymbol *symbol,
118 bool bPosMod, gmx_sel_lexer_t *state)
120 gmx_ana_selmethod_t *method = symbol->methodValue();
121 /* If the previous token was not KEYWORD_POS, return EMPTY_POSMOD
122 * before the actual method to work around a limitation in Bison. */
123 if (!bPosMod && method->type != POS_VALUE)
125 state->nextMethodSymbol = symbol;
126 _gmx_sel_lexer_add_token(yylloc, NULL, 0, state);
129 _gmx_sel_lexer_add_token(yylloc, symbol->name().c_str(), -1, state);
130 yylval->meth = method;
131 if (!(method->flags & SMETH_MODIFIER) && method->nparams == 0)
134 switch (method->type)
138 state->bMatchOf = true;
139 return KEYWORD_NUMERIC;
140 case STR_VALUE: return KEYWORD_STR;
141 case GROUP_VALUE: return KEYWORD_GROUP;
143 GMX_THROW(gmx::InternalError("Unsupported keyword type"));
148 /* Method with parameters or a modifier */
149 if (method->flags & SMETH_MODIFIER)
151 /* Remove all methods from the stack */
153 if (method->param[1].name == NULL)
155 state->nextparam = &method->param[1];
160 if (method->param[0].name == NULL)
162 state->nextparam = &method->param[0];
166 if (state->msp >= state->mstack_alloc)
168 state->mstack_alloc += 10;
169 srenew(state->mstack, state->mstack_alloc);
171 state->mstack[state->msp] = method;
172 if (method->flags & SMETH_MODIFIER)
176 switch (method->type)
178 case INT_VALUE: return METHOD_NUMERIC;
179 case REAL_VALUE: return METHOD_NUMERIC;
180 case POS_VALUE: return METHOD_POS;
181 case GROUP_VALUE: return METHOD_GROUP;
184 GMX_THROW(gmx::InternalError("Unsupported method type"));
187 return INVALID; /* Should not be reached */
191 _gmx_sel_lexer_process_pending(YYSTYPE *yylval, YYLTYPE *yylloc,
192 gmx_sel_lexer_t *state)
194 if (state->nextparam)
196 gmx_ana_selparam_t *param = state->nextparam;
197 bool bBoolNo = state->bBoolNo;
202 _gmx_sel_lexer_add_token(yylloc, NULL, 0, state);
203 return END_OF_METHOD;
205 state->nextparam = NULL;
206 state->bBoolNo = false;
207 _gmx_sel_lexer_add_token(yylloc, param->name, -1, state);
208 return init_param_token(yylval, param, bBoolNo);
210 if (state->prev_pos_kw > 0)
212 --state->prev_pos_kw;
214 if (state->nextMethodSymbol)
216 const gmx::SelectionParserSymbol *symbol = state->nextMethodSymbol;
217 state->nextMethodSymbol = NULL;
218 return init_method_token(yylval, yylloc, symbol, true, state);
224 _gmx_sel_lexer_process_identifier(YYSTYPE *yylval, YYLTYPE *yylloc,
225 char *yytext, size_t yyleng,
226 gmx_sel_lexer_t *state)
228 /* Check if the identifier matches with a parameter name */
231 gmx_ana_selparam_t *param = NULL;
232 bool bBoolNo = false;
234 while (!param && sp >= 0)
237 for (i = 0; i < state->mstack[sp]->nparams; ++i)
239 /* Skip NULL parameters and too long parameters */
240 if (state->mstack[sp]->param[i].name == NULL
241 || strlen(state->mstack[sp]->param[i].name) > yyleng)
245 if (!strncmp(state->mstack[sp]->param[i].name, yytext, yyleng))
247 param = &state->mstack[sp]->param[i];
250 /* Check separately for a 'no' prefix on boolean parameters */
251 if (state->mstack[sp]->param[i].val.type == NO_VALUE
252 && yyleng > 2 && yytext[0] == 'n' && yytext[1] == 'o'
253 && !strncmp(state->mstack[sp]->param[i].name, yytext+2, yyleng-2))
255 param = &state->mstack[sp]->param[i];
267 if (param->val.type == NO_VALUE && !bBoolNo)
269 state->bMatchBool = true;
273 state->neom = state->msp - sp - 1;
274 state->nextparam = param;
275 state->bBoolNo = bBoolNo;
276 return END_OF_METHOD;
278 _gmx_sel_lexer_add_token(yylloc, param->name, -1, state);
279 return init_param_token(yylval, param, bBoolNo);
283 /* Check if the identifier matches with a symbol */
284 const gmx::SelectionParserSymbol *symbol
285 = state->sc->symtab->findSymbol(std::string(yytext, yyleng), false);
286 /* If there is no match, return the token as a string */
289 yylval->str = gmx_strndup(yytext, yyleng);
290 _gmx_sel_lexer_add_token(yylloc, yytext, yyleng, state);
293 gmx::SelectionParserSymbol::SymbolType symtype = symbol->type();
294 /* For method symbols, we need some extra processing. */
295 if (symtype == gmx::SelectionParserSymbol::MethodSymbol)
297 return init_method_token(yylval, yylloc, symbol, state->prev_pos_kw > 0, state);
299 _gmx_sel_lexer_add_token(yylloc, symbol->name().c_str(), -1, state);
300 /* Reserved symbols should have been caught earlier */
301 if (symtype == gmx::SelectionParserSymbol::ReservedSymbol)
303 GMX_THROW(gmx::InternalError(gmx::formatString(
304 "Mismatch between tokenizer and reserved symbol table (for '%s')",
305 symbol->name().c_str())));
307 /* For variable symbols, return the type of the variable value */
308 if (symtype == gmx::SelectionParserSymbol::VariableSymbol)
310 gmx::SelectionTreeElementPointer var = symbol->variableValue();
311 /* Return simple tokens for constant variables */
312 if (var->type == SEL_CONST)
317 yylval->i = var->v.u.i[0];
320 yylval->r = var->v.u.r[0];
325 GMX_THROW(gmx::InternalError("Unsupported variable type"));
328 yylval->sel = new gmx::SelectionTreeElementPointer(var);
331 case INT_VALUE: return VARIABLE_NUMERIC;
332 case REAL_VALUE: return VARIABLE_NUMERIC;
333 case POS_VALUE: return VARIABLE_POS;
334 case GROUP_VALUE: return VARIABLE_GROUP;
337 GMX_THROW(gmx::InternalError("Unsupported variable type"));
340 /* This position should not be reached. */
342 /* For position symbols, we need to return KEYWORD_POS, but we also need
343 * some additional handling. */
344 if (symtype == gmx::SelectionParserSymbol::PositionSymbol)
346 state->bMatchOf = true;
347 yylval->str = gmx_strdup(symbol->name().c_str());
348 state->prev_pos_kw = 2;
351 /* Should not be reached */
356 _gmx_sel_lexer_add_token(YYLTYPE *yylloc, const char *str, int len,
357 gmx_sel_lexer_t *state)
359 yylloc->startIndex = yylloc->endIndex = state->pslen;
360 /* Do nothing if the string is empty, or if it is a space and there is
361 * no other text yet, or if there already is a space. */
362 if (!str || len == 0 || strlen(str) == 0
363 || (str[0] == ' ' && str[1] == 0
364 && (state->pslen == 0 || state->pselstr[state->pslen - 1] == ' ')))
372 /* Allocate more memory if necessary */
373 if (state->nalloc_psel - state->pslen < len)
375 int incr = STRSTORE_ALLOCSTEP < len ? len : STRSTORE_ALLOCSTEP;
376 state->nalloc_psel += incr;
377 srenew(state->pselstr, state->nalloc_psel);
379 /* Append the token to the stored string */
380 strncpy(state->pselstr + state->pslen, str, len);
382 state->pselstr[state->pslen] = 0;
383 yylloc->endIndex = state->pslen;
387 _gmx_sel_init_lexer(yyscan_t *scannerp, struct gmx_ana_selcollection_t *sc,
388 bool bInteractive, int maxnr, bool bGroups,
389 struct gmx_ana_indexgrps_t *grps)
391 int rc = _gmx_sel_yylex_init(scannerp);
394 // TODO: Throw a more representative exception.
395 GMX_THROW(gmx::InternalError("Lexer initialization failed"));
398 gmx_sel_lexer_t *state = new gmx_sel_lexer_t;
401 state->errors = NULL;
402 state->bGroups = bGroups;
404 state->nexpsel = (maxnr > 0 ? static_cast<int>(sc->sel.size()) + maxnr : -1);
406 state->bInteractive = bInteractive;
408 snew(state->pselstr, STRSTORE_ALLOCSTEP);
409 state->pselstr[0] = 0;
411 state->nalloc_psel = STRSTORE_ALLOCSTEP;
412 state->currentLocation.startIndex = 0;
413 state->currentLocation.endIndex = 0;
415 snew(state->mstack, 20);
416 state->mstack_alloc = 20;
419 state->nextparam = NULL;
420 state->nextMethodSymbol = NULL;
421 state->prev_pos_kw = 0;
422 state->bBoolNo = false;
423 state->bMatchOf = false;
424 state->bMatchBool = false;
425 state->bCmdStart = true;
426 state->bBuffer = false;
428 _gmx_sel_yyset_extra(state, *scannerp);
432 _gmx_sel_free_lexer(yyscan_t scanner)
434 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
436 sfree(state->pselstr);
437 sfree(state->mstack);
440 _gmx_sel_yy_delete_buffer(state->buffer, scanner);
443 _gmx_sel_yylex_destroy(scanner);
447 _gmx_sel_set_lexer_error_reporter(yyscan_t scanner,
448 gmx::MessageStringCollector *errors)
450 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
451 state->errors = errors;
455 _gmx_sel_lexer_set_exception(yyscan_t scanner,
456 const boost::exception_ptr &ex)
458 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
459 state->exception = ex;
463 _gmx_sel_lexer_rethrow_exception_if_occurred(yyscan_t scanner)
465 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
466 if (state->exception)
468 boost::exception_ptr ex = state->exception;
469 state->exception = boost::exception_ptr();
470 rethrow_exception(ex);
475 _gmx_sel_is_lexer_interactive(yyscan_t scanner)
477 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
478 return state->bInteractive;
481 struct gmx_ana_selcollection_t *
482 _gmx_sel_lexer_selcollection(yyscan_t scanner)
484 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
488 gmx::MessageStringCollector *
489 _gmx_sel_lexer_error_reporter(yyscan_t scanner)
491 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
492 GMX_RELEASE_ASSERT(state->errors != NULL, "Error reporter not set");
493 return state->errors;
497 _gmx_sel_lexer_has_groups_set(yyscan_t scanner)
499 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
500 return state->bGroups;
503 struct gmx_ana_indexgrps_t *
504 _gmx_sel_lexer_indexgrps(yyscan_t scanner)
506 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
511 _gmx_sel_lexer_exp_selcount(yyscan_t scanner)
513 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
514 return state->nexpsel;
518 _gmx_sel_lexer_pselstr(yyscan_t scanner)
520 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
521 return state->pselstr;
525 _gmx_sel_lexer_set_current_location(yyscan_t scanner,
526 const gmx::SelectionLocation &location)
528 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
529 state->currentLocation = location;
532 const gmx::SelectionLocation &
533 _gmx_sel_lexer_get_current_location(yyscan_t scanner)
535 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
536 return state->currentLocation;
540 _gmx_sel_lexer_get_current_text(yyscan_t scanner)
542 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
543 return _gmx_sel_lexer_get_text(scanner, state->currentLocation);
547 _gmx_sel_lexer_get_text(yyscan_t scanner,
548 const gmx::SelectionLocation &location)
550 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
551 const int startIndex = location.startIndex;
552 const int endIndex = location.endIndex;
553 if (startIndex >= endIndex)
555 return std::string();
557 return std::string(&state->pselstr[startIndex], endIndex - startIndex);
561 _gmx_sel_lexer_clear_pselstr(yyscan_t scanner)
563 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
564 state->pselstr[0] = 0;
569 _gmx_sel_lexer_clear_method_stack(yyscan_t scanner)
571 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
577 _gmx_sel_finish_method(yyscan_t scanner)
579 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
588 _gmx_sel_set_lex_input_file(yyscan_t scanner, FILE *fp)
590 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
592 state->bBuffer = true;
593 state->buffer = _gmx_sel_yy_create_buffer(fp, YY_BUF_SIZE, scanner);
594 _gmx_sel_yy_switch_to_buffer(state->buffer, scanner);
598 _gmx_sel_set_lex_input_str(yyscan_t scanner, const char *str)
600 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
604 _gmx_sel_yy_delete_buffer(state->buffer, scanner);
606 state->bBuffer = true;
607 state->buffer = _gmx_sel_yy_scan_string(str, scanner);