2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2009,2010,2011,2012,2014,2015,2016,2017,2018, 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/smalloc.h"
67 #include "gromacs/utility/stringutil.h"
70 #include "parsetree.h"
72 #include "selectioncollection-impl.h"
74 #include "selmethod.h"
77 /* These are defined as macros in the generated scanner_flex.h.
78 * We undefine them here to have them as variable names in the subroutines.
79 * There are other ways of doing this, but this is probably the easiest. */
85 * Handles initialization of method parameter token.
88 init_param_token(YYSTYPE *yylval, gmx_ana_selparam_t *param, bool bBoolNo)
92 GMX_RELEASE_ASSERT(param->name != nullptr,
93 "bBoolNo should only be set for a parameters with a name");
94 snew(yylval->str, strlen(param->name) + 3);
97 strcpy(yylval->str+2, param->name);
101 yylval->str = param->name ? gmx_strdup(param->name) : nullptr;
107 * Processes a selection method token.
110 init_method_token(YYSTYPE *yylval, YYLTYPE *yylloc,
111 const gmx::SelectionParserSymbol *symbol,
112 bool bPosMod, gmx_sel_lexer_t *state)
114 gmx_ana_selmethod_t *method = symbol->methodValue();
115 /* If the previous token was not KEYWORD_POS, return EMPTY_POSMOD
116 * before the actual method to work around a limitation in Bison. */
117 if (!bPosMod && method->type != POS_VALUE)
119 state->nextMethodSymbol = symbol;
120 _gmx_sel_lexer_add_token(yylloc, nullptr, 0, state);
123 _gmx_sel_lexer_add_token(yylloc, symbol->name().c_str(), -1, state);
124 yylval->meth = method;
125 if (!(method->flags & SMETH_MODIFIER) && method->nparams == 0)
128 switch (method->type)
132 state->bMatchOf = true;
133 return KEYWORD_NUMERIC;
134 case STR_VALUE: return KEYWORD_STR;
135 case GROUP_VALUE: return KEYWORD_GROUP;
137 GMX_THROW(gmx::InternalError("Unsupported keyword type"));
142 /* Method with parameters or a modifier */
143 if (method->flags & SMETH_MODIFIER)
145 /* Remove all methods from the stack */
147 if (method->param[1].name == nullptr)
149 state->nextparam = &method->param[1];
154 if (method->param[0].name == nullptr)
156 state->nextparam = &method->param[0];
160 if (state->msp >= state->mstack_alloc)
162 state->mstack_alloc += 10;
163 srenew(state->mstack, state->mstack_alloc);
165 state->mstack[state->msp] = method;
166 if (method->flags & SMETH_MODIFIER)
170 switch (method->type)
172 case INT_VALUE: return METHOD_NUMERIC;
173 case REAL_VALUE: return METHOD_NUMERIC;
174 case POS_VALUE: return METHOD_POS;
175 case GROUP_VALUE: return METHOD_GROUP;
178 GMX_THROW(gmx::InternalError("Unsupported method type"));
184 _gmx_sel_lexer_process_pending(YYSTYPE *yylval, YYLTYPE *yylloc,
185 gmx_sel_lexer_t *state)
187 if (state->nextparam)
189 gmx_ana_selparam_t *param = state->nextparam;
190 bool bBoolNo = state->bBoolNo;
195 _gmx_sel_lexer_add_token(yylloc, nullptr, 0, state);
196 return END_OF_METHOD;
198 state->nextparam = nullptr;
199 state->bBoolNo = false;
200 _gmx_sel_lexer_add_token(yylloc, param->name, -1, state);
201 return init_param_token(yylval, param, bBoolNo);
203 if (state->prev_pos_kw > 0)
205 --state->prev_pos_kw;
207 if (state->nextMethodSymbol)
209 const gmx::SelectionParserSymbol *symbol = state->nextMethodSymbol;
210 state->nextMethodSymbol = nullptr;
211 return init_method_token(yylval, yylloc, symbol, true, state);
217 _gmx_sel_lexer_process_identifier(YYSTYPE *yylval, YYLTYPE *yylloc,
218 char *yytext, size_t yyleng,
219 gmx_sel_lexer_t *state)
221 /* Check if the identifier matches with a parameter name */
224 gmx_ana_selparam_t *param = nullptr;
225 bool bBoolNo = false;
227 while (!param && sp >= 0)
230 for (i = 0; i < state->mstack[sp]->nparams; ++i)
232 /* Skip NULL parameters and too long parameters */
233 if (state->mstack[sp]->param[i].name == nullptr
234 || strlen(state->mstack[sp]->param[i].name) > yyleng)
238 if (!strncmp(state->mstack[sp]->param[i].name, yytext, yyleng))
240 param = &state->mstack[sp]->param[i];
243 /* Check separately for a 'no' prefix on boolean parameters */
244 if (state->mstack[sp]->param[i].val.type == NO_VALUE
245 && yyleng > 2 && yytext[0] == 'n' && yytext[1] == 'o'
246 && !strncmp(state->mstack[sp]->param[i].name, yytext+2, yyleng-2))
248 param = &state->mstack[sp]->param[i];
260 if (param->val.type == NO_VALUE && !bBoolNo)
262 state->bMatchBool = true;
266 state->neom = state->msp - sp - 1;
267 state->nextparam = param;
268 state->bBoolNo = bBoolNo;
269 return END_OF_METHOD;
271 _gmx_sel_lexer_add_token(yylloc, param->name, -1, state);
272 return init_param_token(yylval, param, bBoolNo);
276 /* Check if the identifier matches with a symbol */
277 const gmx::SelectionParserSymbol *symbol
278 = state->sc->symtab->findSymbol(std::string(yytext, yyleng));
279 /* If there is no match, return the token as a string */
282 yylval->str = gmx_strndup(yytext, yyleng);
283 _gmx_sel_lexer_add_token(yylloc, yytext, yyleng, state);
286 gmx::SelectionParserSymbol::SymbolType symtype = symbol->type();
287 /* For method symbols, we need some extra processing. */
288 if (symtype == gmx::SelectionParserSymbol::MethodSymbol)
290 return init_method_token(yylval, yylloc, symbol, state->prev_pos_kw > 0, state);
292 _gmx_sel_lexer_add_token(yylloc, symbol->name().c_str(), -1, state);
293 /* Reserved symbols should have been caught earlier */
294 if (symtype == gmx::SelectionParserSymbol::ReservedSymbol)
296 GMX_THROW(gmx::InternalError(gmx::formatString(
297 "Mismatch between tokenizer and reserved symbol table (for '%s')",
298 symbol->name().c_str())));
300 /* For variable symbols, return the type of the variable value */
301 if (symtype == gmx::SelectionParserSymbol::VariableSymbol)
303 const gmx::SelectionTreeElementPointer &var = symbol->variableValue();
304 /* Return simple tokens for constant variables */
305 if (var->type == SEL_CONST)
310 yylval->i = var->v.u.i[0];
313 yylval->r = var->v.u.r[0];
318 GMX_THROW(gmx::InternalError("Unsupported variable type"));
321 yylval->sel = new gmx::SelectionTreeElementPointer(var);
324 case INT_VALUE: return VARIABLE_NUMERIC;
325 case REAL_VALUE: return VARIABLE_NUMERIC;
326 case POS_VALUE: return VARIABLE_POS;
327 case GROUP_VALUE: return VARIABLE_GROUP;
330 GMX_THROW(gmx::InternalError("Unsupported variable type"));
332 /* This position should not be reached. */
334 /* For position symbols, we need to return KEYWORD_POS, but we also need
335 * some additional handling. */
336 if (symtype == gmx::SelectionParserSymbol::PositionSymbol)
338 state->bMatchOf = true;
339 yylval->str = gmx_strdup(symbol->name().c_str());
340 state->prev_pos_kw = 2;
343 /* Should not be reached */
348 _gmx_sel_lexer_add_token(YYLTYPE *yylloc, const char *str, int len,
349 gmx_sel_lexer_t *state)
351 yylloc->startIndex = yylloc->endIndex = state->pselstr.size();
352 /* Do nothing if the string is empty, or if it is a space and there is
353 * no other text yet, or if there already is a space. */
354 if (!str || len == 0 || strlen(str) == 0
355 || (str[0] == ' ' && str[1] == 0
356 && (state->pselstr.empty() || state->pselstr.back() == ' ')))
364 /* Append the token to the stored string */
365 state->pselstr.append(str, len);
366 yylloc->endIndex = state->pselstr.size();
370 _gmx_sel_init_lexer(yyscan_t *scannerp, struct gmx_ana_selcollection_t *sc,
371 gmx::TextWriter *statusWriter, int maxnr,
372 bool bGroups, struct gmx_ana_indexgrps_t *grps)
374 int rc = _gmx_sel_yylex_init(scannerp);
377 // TODO: Throw a more representative exception.
378 GMX_THROW(gmx::InternalError("Lexer initialization failed"));
381 gmx_sel_lexer_t *state = new gmx_sel_lexer_t;
384 state->bGroups = bGroups;
386 state->nexpsel = (maxnr > 0 ? static_cast<int>(sc->sel.size()) + maxnr : -1);
388 state->statusWriter = statusWriter;
390 state->currentLocation.startIndex = 0;
391 state->currentLocation.endIndex = 0;
393 snew(state->mstack, 20);
394 state->mstack_alloc = 20;
397 state->nextparam = nullptr;
398 state->nextMethodSymbol = nullptr;
399 state->prev_pos_kw = 0;
400 state->bBoolNo = false;
401 state->bMatchOf = false;
402 state->bMatchBool = false;
403 state->bCmdStart = true;
404 state->bBuffer = false;
406 _gmx_sel_yyset_extra(state, *scannerp);
410 _gmx_sel_free_lexer(yyscan_t scanner)
412 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
414 sfree(state->mstack);
417 _gmx_sel_yy_delete_buffer(state->buffer, scanner);
420 _gmx_sel_yylex_destroy(scanner);
424 _gmx_sel_lexer_set_exception(yyscan_t scanner,
425 const std::exception_ptr &ex)
427 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
428 state->exception = ex;
432 _gmx_sel_lexer_rethrow_exception_if_occurred(yyscan_t scanner)
434 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
435 if (state->exception)
437 std::exception_ptr ex = state->exception;
438 state->exception = std::exception_ptr();
439 std::rethrow_exception(ex);
444 _gmx_sel_lexer_get_status_writer(yyscan_t scanner)
446 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
447 return state->statusWriter;
450 struct gmx_ana_selcollection_t *
451 _gmx_sel_lexer_selcollection(yyscan_t scanner)
453 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
458 _gmx_sel_lexer_has_groups_set(yyscan_t scanner)
460 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
461 return state->bGroups;
464 struct gmx_ana_indexgrps_t *
465 _gmx_sel_lexer_indexgrps(yyscan_t scanner)
467 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
472 _gmx_sel_lexer_exp_selcount(yyscan_t scanner)
474 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
475 return state->nexpsel;
479 _gmx_sel_lexer_pselstr(yyscan_t scanner)
481 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
482 return state->pselstr.c_str();
486 _gmx_sel_lexer_set_current_location(yyscan_t scanner,
487 const gmx::SelectionLocation &location)
489 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
490 state->currentLocation = location;
493 const gmx::SelectionLocation &
494 _gmx_sel_lexer_get_current_location(yyscan_t scanner)
496 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
497 return state->currentLocation;
501 _gmx_sel_lexer_get_current_text(yyscan_t scanner)
503 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
504 return _gmx_sel_lexer_get_text(scanner, state->currentLocation);
508 _gmx_sel_lexer_get_text(yyscan_t scanner,
509 const gmx::SelectionLocation &location)
511 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
512 const int startIndex = location.startIndex;
513 const int endIndex = location.endIndex;
514 if (startIndex >= endIndex)
516 return std::string();
518 return state->pselstr.substr(startIndex, endIndex - startIndex);
522 _gmx_sel_lexer_clear_pselstr(yyscan_t scanner)
524 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
525 state->pselstr.clear();
529 _gmx_sel_lexer_clear_method_stack(yyscan_t scanner)
531 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
537 _gmx_sel_finish_method(yyscan_t scanner)
539 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
548 _gmx_sel_set_lex_input_file(yyscan_t scanner, FILE *fp)
550 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
552 state->bBuffer = true;
553 state->buffer = _gmx_sel_yy_create_buffer(fp, YY_BUF_SIZE, scanner);
554 _gmx_sel_yy_switch_to_buffer(state->buffer, scanner);
558 _gmx_sel_set_lex_input_str(yyscan_t scanner, const char *str)
560 gmx_sel_lexer_t *state = _gmx_sel_yyget_extra(scanner);
564 _gmx_sel_yy_delete_buffer(state->buffer, scanner);
566 state->bBuffer = true;
567 state->buffer = _gmx_sel_yy_scan_string(str, scanner);