2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2009,2010,2011,2012,2013 by the GROMACS development team.
5 * Copyright (c) 2014,2015,2016,2017,2018 by the GROMACS development team.
6 * Copyright (c) 2019,2020,2021, by the GROMACS development team, led by
7 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
8 * and including many others, as listed in the AUTHORS file in the
9 * top-level source directory and at http://www.gromacs.org.
11 * GROMACS is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU Lesser General Public License
13 * as published by the Free Software Foundation; either version 2.1
14 * of the License, or (at your option) any later version.
16 * GROMACS is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 * Lesser General Public License for more details.
21 * You should have received a copy of the GNU Lesser General Public
22 * License along with GROMACS; if not, see
23 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
24 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
26 * If you want to redistribute modifications to GROMACS, please
27 * consider that scientific software is very special. Version
28 * control is crucial - bugs must be traceable. We will be happy to
29 * consider code for inclusion in the official distribution, but
30 * derived work must not be called official GROMACS. Details are found
31 * in the README & COPYING files - if they are missing, get the
32 * official version at http://www.gromacs.org.
34 * To help us fund GROMACS development, we humbly ask that you cite
35 * the research papers on the package. Check out http://www.gromacs.org.
39 * Implements the \p same selection method.
41 * \author Teemu Murtola <teemu.murtola@gmail.com>
42 * \ingroup module_selection
49 #include "gromacs/utility/arraysize.h"
50 #include "gromacs/utility/exceptions.h"
51 #include "gromacs/utility/smalloc.h"
54 #include "parsetree.h"
56 #include "selmethod.h"
57 #include "selmethod_impl.h"
61 * Data structure for the \p same selection method.
63 * To avoid duplicate initialization code, the same data structure is used
64 * for matching both integer and string keywords; hence the unions.
66 * \ingroup module_selection
70 /** Value for each atom to match. */
78 * Number of values in the \p as array.
80 * For string values, this is actually the number of values in the
81 * \p as_s_sorted array.
84 /** Values to match against. */
92 * Separate array for sorted \p as.s array.
94 * The array of strings returned as the output value of a parameter should
95 * not be messed with to avoid memory corruption (the pointers in the array
96 * may be reused for several evaluations), so we keep our own copy for
100 /** Whether simple matching can be used. */
105 * Allocates data for the \p same selection method.
107 * \param[in] npar Not used (should be 2).
108 * \param[in,out] param Method parameters (should point to a copy of
109 * ::smparams_same_int or ::smparams_same_str).
110 * \returns Pointer to the allocated data (\ref t_methoddata_same).
112 static void* init_data_same(int npar, gmx_ana_selparam_t* param);
114 * Initializes the \p same selection method.
116 * \param top Not used.
117 * \param npar Not used (should be 2).
118 * \param param Initialized method parameters (should point to a copy of
119 * ::smparams_same_int or ::smparams_same_str).
120 * \param data Pointer to \ref t_methoddata_same to initialize.
121 * \returns 0 on success, -1 on failure.
123 static void init_same(const gmx_mtop_t* top, int npar, gmx_ana_selparam_t* param, void* data);
124 /** Frees the data allocated for the \p same selection method. */
125 static void free_data_same(void* data);
127 * Initializes the evaluation of the \p same selection method for a frame.
129 * \param[in] context Not used.
130 * \param data Should point to a \ref t_methoddata_same.
132 * Sorts the \c data->as.i array and removes identical values for faster and
135 static void init_frame_same_int(const gmx::SelMethodEvalContext& context, void* data);
136 /** Evaluates the \p same selection method. */
137 static void evaluate_same_int(const gmx::SelMethodEvalContext& context,
139 gmx_ana_selvalue_t* out,
142 * Initializes the evaluation of the \p same selection method for a frame.
144 * \param[in] context Not used.
145 * \param data Should point to a \ref t_methoddata_same.
147 * Sorts the \c data->as.s array and removes identical values for faster and
150 static void init_frame_same_str(const gmx::SelMethodEvalContext& context, void* data);
151 /** Evaluates the \p same selection method. */
152 static void evaluate_same_str(const gmx::SelMethodEvalContext& context,
154 gmx_ana_selvalue_t* out,
157 /** Parameters for the \p same selection method. */
158 static gmx_ana_selparam_t smparams_same_int[] = {
159 { nullptr, { INT_VALUE, -1, { nullptr } }, nullptr, SPAR_DYNAMIC | SPAR_ATOMVAL },
160 { "as", { INT_VALUE, -1, { nullptr } }, nullptr, SPAR_DYNAMIC | SPAR_VARNUM },
163 /** Parameters for the \p same selection method. */
164 static gmx_ana_selparam_t smparams_same_str[] = {
165 { nullptr, { STR_VALUE, -1, { nullptr } }, nullptr, SPAR_DYNAMIC | SPAR_ATOMVAL },
166 { "as", { STR_VALUE, -1, { nullptr } }, nullptr, SPAR_DYNAMIC | SPAR_VARNUM },
169 /** Help text for the \p same selection method. */
170 static const char* const help_same[] = {
173 " same KEYWORD as ATOM_EXPR",
176 "The keyword [TT]same[tt] can be used to select all atoms for which",
177 "the given [TT]KEYWORD[tt] matches any of the atoms in [TT]ATOM_EXPR[tt].",
178 "Keywords that evaluate to integer or string values are supported.",
181 /*! \internal \brief Selection method data for the \p same method. */
182 gmx_ana_selmethod_t sm_same = {
186 asize(smparams_same_int),
193 &init_frame_same_int,
196 { "same KEYWORD as ATOM_EXPR", "Extending selections", asize(help_same), help_same },
200 * Selection method data for the \p same method.
202 * This selection method is used for matching string keywords. The parser
203 * never sees this method; _gmx_selelem_custom_init_same() replaces sm_same
204 * with this method in cases where it is required.
206 static gmx_ana_selmethod_t sm_same_str = {
210 asize(smparams_same_str),
217 &init_frame_same_str,
220 { nullptr, nullptr, 0, nullptr },
223 static void* init_data_same(int /* npar */, gmx_ana_selparam_t* param)
225 t_methoddata_same* data;
228 data->as_s_sorted = nullptr;
229 param[1].nvalptr = &data->nas;
234 * \param[in,out] method The method to initialize.
235 * \param[in,out] params Pointer to the first parameter.
236 * \param[in] scanner Scanner data structure.
238 * If \p *method is not a \c same method, this function returns
241 void _gmx_selelem_custom_init_same(gmx_ana_selmethod_t** method,
242 const gmx::SelectionParserParameterListPointer& params,
246 /* Do nothing if this is not a same method. */
247 if (!*method || (*method)->name != sm_same.name || params->empty())
252 const gmx::SelectionParserValueList& kwvalues = params->front().values();
253 if (kwvalues.size() != 1 || !kwvalues.front().hasExpressionValue()
254 || kwvalues.front().expr->type != SEL_EXPRESSION)
256 GMX_THROW(gmx::InvalidInputError("'same' should be followed by a single keyword"));
258 gmx_ana_selmethod_t* kwmethod = kwvalues.front().expr->u.expr.method;
259 if (kwmethod->type == STR_VALUE)
261 *method = &sm_same_str;
264 /* We do custom processing for the "as" parameter. */
265 gmx::SelectionParserParameterList::iterator asparam = ++params->begin();
266 if (asparam != params->end() && asparam->name() == sm_same.param[1].name)
268 const gmx::SelectionParserValueList& asvalues = asparam->values();
269 if (asvalues.size() != 1 || !asvalues.front().hasExpressionValue())
271 // TODO: Think about providing more informative context.
272 GMX_THROW(gmx::InvalidInputError(
273 "'same ... as' should be followed by a single expression"));
275 const gmx::SelectionTreeElementPointer& child = asvalues.front().expr;
276 /* Create a second keyword evaluation element for the keyword given as
277 * the first parameter, evaluating the keyword in the group given by the
278 * second parameter. */
279 gmx::SelectionTreeElementPointer kwelem =
280 _gmx_sel_init_keyword_evaluator(kwmethod, child, scanner);
281 /* Replace the second parameter with one with a value from \p kwelem. */
282 std::string pname = asparam->name();
283 *asparam = gmx::SelectionParserParameter::createFromExpression(pname, kwelem);
287 static void init_same(const gmx_mtop_t* /* top */, int /* npar */, gmx_ana_selparam_t* param, void* data)
289 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
291 d->val.ptr = param[0].val.u.ptr;
292 d->as.ptr = param[1].val.u.ptr;
293 if (param[1].val.type == STR_VALUE)
295 snew(d->as_s_sorted, d->nas);
297 if (!(param[0].flags & SPAR_ATOMVAL))
300 gmx::InvalidInputError("The 'same' selection keyword combined with a "
301 "non-keyword does not make sense"));
306 * \param data Data to free (should point to a \ref t_methoddata_same).
308 static void free_data_same(void* data)
310 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
312 sfree(d->as_s_sorted);
317 * Helper function for comparison of two integers.
319 static int cmp_int(const void* a, const void* b)
321 if (*reinterpret_cast<const int*>(a) < *reinterpret_cast<const int*>(b))
325 if (*reinterpret_cast<const int*>(a) > *reinterpret_cast<const int*>(b))
332 static void init_frame_same_int(const gmx::SelMethodEvalContext& /*context*/, void* data)
334 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
337 /* Collapse adjacent values, and check whether the array is sorted. */
343 for (i = 1, j = 0; i < d->nas; ++i)
345 if (d->as.i[i] != d->as.i[j])
347 if (d->as.i[i] < d->as.i[j])
352 d->as.i[j] = d->as.i[i];
359 qsort(d->as.i, d->nas, sizeof(d->as.i[0]), &cmp_int);
360 /* More identical values may become adjacent after sorting. */
361 for (i = 1, j = 0; i < d->nas; ++i)
363 if (d->as.i[i] != d->as.i[j])
366 d->as.i[j] = d->as.i[i];
374 * See sel_updatefunc() for description of the parameters.
375 * \p data should point to a \c t_methoddata_same.
377 * Calculates which values in \c data->val.i can be found in \c data->as.i
378 * (assumed sorted), and writes the corresponding atoms to output.
379 * If \c data->val is sorted, uses a linear scan of both arrays, otherwise a
380 * binary search of \c data->as is performed for each block of values in
383 static void evaluate_same_int(const gmx::SelMethodEvalContext& /*context*/,
385 gmx_ana_selvalue_t* out,
388 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
397 /* If we are sorted, we can do a simple linear scan. */
398 while (i < d->nas && d->as.i[i] < d->val.i[j])
405 /* If not, we must do a binary search of all the values. */
412 int itry = (i1 + i2) / 2;
413 if (d->as.i[itry] <= d->val.i[j])
422 i = (d->as.i[i1] == d->val.i[j] ? i1 : d->nas);
424 /* Check whether the value was found in the as list. */
425 if (i == d->nas || d->as.i[i] != d->val.i[j])
427 /* If not, skip all atoms with the same value. */
428 int tmpval = d->val.i[j];
430 while (j < g->isize && d->val.i[j] == tmpval)
437 /* Copy all the atoms with this value to the output. */
438 while (j < g->isize && d->val.i[j] == d->as.i[i])
440 out->u.g->index[out->u.g->isize++] = g->index[j];
444 if (j < g->isize && d->val.i[j] < d->val.i[j - 1])
452 * Helper function for comparison of two strings.
454 static int cmp_str(const void* a, const void* b)
456 return strcmp(*static_cast<char* const*>(a), *static_cast<char* const*>(b));
459 static void init_frame_same_str(const gmx::SelMethodEvalContext& /*context*/, void* data)
461 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
464 /* Collapse adjacent values.
465 * For strings, it's unlikely that the values would be sorted originally,
466 * so set bSorted always to false. */
472 d->as_s_sorted[0] = d->as.s[0];
473 for (i = 1, j = 0; i < d->nas; ++i)
475 if (strcmp(d->as.s[i], d->as_s_sorted[j]) != 0)
478 d->as_s_sorted[j] = d->as.s[i];
483 qsort(d->as_s_sorted, d->nas, sizeof(d->as_s_sorted[0]), &cmp_str);
484 /* More identical values may become adjacent after sorting. */
485 for (i = 1, j = 0; i < d->nas; ++i)
487 if (strcmp(d->as_s_sorted[i], d->as_s_sorted[j]) != 0)
490 d->as_s_sorted[j] = d->as_s_sorted[i];
497 * See sel_updatefunc() for description of the parameters.
498 * \p data should point to a \c t_methoddata_same.
500 * Calculates which strings in \c data->val.s can be found in \c data->as.s
501 * (assumed sorted), and writes the corresponding atoms to output.
502 * A binary search of \c data->as is performed for each block of values in
505 static void evaluate_same_str(const gmx::SelMethodEvalContext& /*context*/,
507 gmx_ana_selvalue_t* out,
510 t_methoddata_same* d = static_cast<t_methoddata_same*>(data);
517 /* Do a binary search of the strings. */
521 ptr = bsearch(&d->val.s[j], d->as_s_sorted, d->nas, sizeof(d->as_s_sorted[0]), &cmp_str);
523 /* Check whether the value was found in the as list. */
526 /* If not, skip all atoms with the same value. */
527 const char* tmpval = d->val.s[j];
529 while (j < g->isize && strcmp(d->val.s[j], tmpval) == 0)
536 const char* tmpval = d->val.s[j];
537 /* Copy all the atoms with this value to the output. */
538 while (j < g->isize && strcmp(d->val.s[j], tmpval) == 0)
540 out->u.g->index[out->u.g->isize++] = g->index[j];