1 | // =============================================================== // |
---|
2 | // // |
---|
3 | // File : aditem.cxx // |
---|
4 | // Purpose : item functions // |
---|
5 | // items are e.g. species, SAIs, genes, etc // |
---|
6 | // // |
---|
7 | // Institute of Microbiology (Technical University Munich) // |
---|
8 | // http://www.arb-home.de/ // |
---|
9 | // // |
---|
10 | // =============================================================== // |
---|
11 | |
---|
12 | #include "gb_local.h" |
---|
13 | |
---|
14 | #include <arbdbt.h> |
---|
15 | #include <arb_strbuf.h> |
---|
16 | |
---|
17 | |
---|
18 | GBDATA *GBT_find_or_create_item_rel_item_data(GBDATA *gb_item_data, const char *itemname, const char *id_field, const char *id, bool markCreated) { |
---|
19 | /* Search for a item with field 'id_field' set to given 'id' (id compare is case-insensitive) |
---|
20 | * If item does not exist, create it. |
---|
21 | * Newly created items are automatically marked, if 'markCreated' is true |
---|
22 | * items may be: species, genes, SAIs, ... |
---|
23 | */ |
---|
24 | |
---|
25 | GBDATA *gb_item = 0; |
---|
26 | GB_ERROR error = 0; |
---|
27 | |
---|
28 | if (!gb_item_data) error = "No container"; |
---|
29 | else { |
---|
30 | gb_item = GBT_find_item_rel_item_data(gb_item_data, id_field, id); |
---|
31 | if (!gb_item) { |
---|
32 | error = GB_push_transaction(gb_item_data); |
---|
33 | if (!error) { |
---|
34 | gb_item = GB_create_container(gb_item_data, itemname); // create a new item |
---|
35 | if (!gb_item) error = GB_await_error(); |
---|
36 | else { |
---|
37 | error = GBT_write_string(gb_item, id_field, id); // write item identifier |
---|
38 | if (!error && markCreated) GB_write_flag(gb_item, 1); // mark generated item |
---|
39 | } |
---|
40 | } |
---|
41 | error = GB_end_transaction(gb_item_data, error); |
---|
42 | } |
---|
43 | } |
---|
44 | |
---|
45 | if (!gb_item && !error) error = GB_await_error(); |
---|
46 | if (error) { |
---|
47 | gb_item = 0; |
---|
48 | GB_export_errorf("Can't create %s '%s': %s", itemname, id, error); |
---|
49 | } |
---|
50 | |
---|
51 | return gb_item; |
---|
52 | } |
---|
53 | |
---|
54 | GBDATA *GBT_find_or_create_species_rel_species_data(GBDATA *gb_species_data, const char *name) { |
---|
55 | return GBT_find_or_create_item_rel_item_data(gb_species_data, "species", "name", name, true); |
---|
56 | } |
---|
57 | |
---|
58 | GBDATA *GBT_find_or_create_species(GBDATA *gb_main, const char *name) { |
---|
59 | return GBT_find_or_create_species_rel_species_data(GBT_get_species_data(gb_main), name); |
---|
60 | } |
---|
61 | |
---|
62 | GBDATA *GBT_find_or_create_SAI(GBDATA *gb_main, const char *name) { |
---|
63 | //! Search for an SAI, when SAI does not exist, create it |
---|
64 | return GBT_find_or_create_item_rel_item_data(GBT_get_SAI_data(gb_main), "extended", "name", name, true); |
---|
65 | } |
---|
66 | |
---|
67 | |
---|
68 | // ------------------------------------ |
---|
69 | // some simple find procedures |
---|
70 | |
---|
71 | GBDATA *GBT_find_item_rel_item_data(GBDATA *gb_item_data, const char *id_field, const char *id_value) { |
---|
72 | /*! search for items starting at item container |
---|
73 | * |
---|
74 | * @param 'gb_item_data' is a container containing items |
---|
75 | * @param 'id_field' is a field containing a unique identifier for each item (e.g. 'name' for species) |
---|
76 | * |
---|
77 | * @return a pointer to an item with 'id_field' containing 'id_value' |
---|
78 | * or NULL (in this case an error MAY be exported) |
---|
79 | * |
---|
80 | * Note: If you expect the item to exist, use GBT_expect_item_rel_item_data() |
---|
81 | */ |
---|
82 | |
---|
83 | GBDATA *gb_item_id = GB_find_string(gb_item_data, id_field, id_value, GB_IGNORE_CASE, SEARCH_GRANDCHILD); |
---|
84 | return gb_item_id ? GB_get_father(gb_item_id) : 0; |
---|
85 | } |
---|
86 | |
---|
87 | static GBDATA *GBT_expect_item_rel_item_data(GBDATA *gb_item_data, const char *id_field, const char *id_value) { |
---|
88 | //! like GBT_find_item_rel_item_data(), but also exports an error if the item is not present |
---|
89 | |
---|
90 | GBDATA *gb_found = GBT_find_item_rel_item_data(gb_item_data, id_field, id_value); |
---|
91 | if (!gb_found && !GB_have_error()) { // item simply not exists |
---|
92 | GBDATA *gb_any = GB_find(gb_item_data, id_field, SEARCH_GRANDCHILD); |
---|
93 | const char *itemname = gb_any ? GB_read_key_pntr(GB_get_father(gb_any)) : "<item>"; |
---|
94 | GB_export_errorf("Could not find %s with %s '%s'", itemname, id_field, id_value); |
---|
95 | } |
---|
96 | return gb_found; |
---|
97 | } |
---|
98 | |
---|
99 | // -------------------------------------------------------------------------------- |
---|
100 | |
---|
101 | GBDATA *GBT_get_species_data(GBDATA *gb_main) { |
---|
102 | return GB_search(gb_main, "species_data", GB_CREATE_CONTAINER); |
---|
103 | } |
---|
104 | |
---|
105 | GBDATA *GBT_first_marked_species_rel_species_data(GBDATA *gb_species_data) { |
---|
106 | return GB_first_marked(gb_species_data, "species"); |
---|
107 | } |
---|
108 | |
---|
109 | GBDATA *GBT_first_marked_species(GBDATA *gb_main) { |
---|
110 | return GB_first_marked(GBT_get_species_data(gb_main), "species"); |
---|
111 | } |
---|
112 | GBDATA *GBT_next_marked_species(GBDATA *gb_species) { |
---|
113 | gb_assert(GB_has_key(gb_species, "species")); |
---|
114 | return GB_next_marked(gb_species, "species"); |
---|
115 | } |
---|
116 | |
---|
117 | GBDATA *GBT_first_species_rel_species_data(GBDATA *gb_species_data) { |
---|
118 | return GB_entry(gb_species_data, "species"); |
---|
119 | } |
---|
120 | GBDATA *GBT_first_species(GBDATA *gb_main) { |
---|
121 | return GB_entry(GBT_get_species_data(gb_main), "species"); |
---|
122 | } |
---|
123 | |
---|
124 | GBDATA *GBT_next_species(GBDATA *gb_species) { |
---|
125 | gb_assert(GB_has_key(gb_species, "species")); |
---|
126 | return GB_nextEntry(gb_species); |
---|
127 | } |
---|
128 | |
---|
129 | GBDATA *GBT_find_species_rel_species_data(GBDATA *gb_species_data, const char *name) { |
---|
130 | return GBT_find_item_rel_item_data(gb_species_data, "name", name); |
---|
131 | } |
---|
132 | GBDATA *GBT_find_species(GBDATA *gb_main, const char *name) { |
---|
133 | // Search for a species. |
---|
134 | // Return found species or NULL (in this case an error MAY be exported). |
---|
135 | // |
---|
136 | // Note: If you expect the species to exists, use GBT_expect_species! |
---|
137 | return GBT_find_item_rel_item_data(GBT_get_species_data(gb_main), "name", name); |
---|
138 | } |
---|
139 | |
---|
140 | GBDATA *GBT_expect_species(GBDATA *gb_main, const char *name) { |
---|
141 | // Returns an existing species or |
---|
142 | // NULL (in that case an error is exported) |
---|
143 | return GBT_expect_item_rel_item_data(GBT_get_species_data(gb_main), "name", name); |
---|
144 | } |
---|
145 | |
---|
146 | // -------------------------------------------------------------------------------- |
---|
147 | |
---|
148 | GBDATA *GBT_get_SAI_data(GBDATA *gb_main) { |
---|
149 | return GB_search(gb_main, "extended_data", GB_CREATE_CONTAINER); |
---|
150 | } |
---|
151 | |
---|
152 | // Search SAIs |
---|
153 | GBDATA *GBT_first_SAI_rel_SAI_data(GBDATA *gb_sai_data) { |
---|
154 | return GB_entry(gb_sai_data, "extended"); |
---|
155 | } |
---|
156 | GBDATA *GBT_first_SAI(GBDATA *gb_main) { |
---|
157 | return GB_entry(GBT_get_SAI_data(gb_main), "extended"); |
---|
158 | } |
---|
159 | |
---|
160 | GBDATA *GBT_next_SAI(GBDATA *gb_sai) { |
---|
161 | gb_assert(GB_has_key(gb_sai, "extended")); |
---|
162 | return GB_nextEntry(gb_sai); |
---|
163 | } |
---|
164 | |
---|
165 | GBDATA *GBT_find_SAI_rel_SAI_data(GBDATA *gb_sai_data, const char *name) { |
---|
166 | return GBT_find_item_rel_item_data(gb_sai_data, "name", name); |
---|
167 | } |
---|
168 | GBDATA *GBT_find_SAI(GBDATA *gb_main, const char *name) { |
---|
169 | // Search for a SAI. |
---|
170 | // Return found SAI or NULL (in this case an error MAY be exported). |
---|
171 | // |
---|
172 | // Note: If you expect the SAI to exist, use GBT_expect_SAI! |
---|
173 | return GBT_find_item_rel_item_data(GBT_get_SAI_data(gb_main), "name", name); |
---|
174 | } |
---|
175 | |
---|
176 | GBDATA *GBT_expect_SAI(GBDATA *gb_main, const char *name) { |
---|
177 | // Returns an existing SAI or |
---|
178 | // NULL (in that case an error is exported) |
---|
179 | return GBT_expect_item_rel_item_data(GBT_get_SAI_data(gb_main), "name", name); |
---|
180 | } |
---|
181 | |
---|
182 | // --------------------- |
---|
183 | // count items |
---|
184 | |
---|
185 | static long GBT_get_item_count(GBDATA *gb_parent_of_container, const char *item_container_name) { |
---|
186 | // returns elements stored in a container |
---|
187 | |
---|
188 | GBDATA *gb_item_data; |
---|
189 | long count = 0; |
---|
190 | |
---|
191 | GB_push_transaction(gb_parent_of_container); |
---|
192 | gb_item_data = GB_entry(gb_parent_of_container, item_container_name); |
---|
193 | if (gb_item_data) count = GB_number_of_subentries(gb_item_data); |
---|
194 | GB_pop_transaction(gb_parent_of_container); |
---|
195 | |
---|
196 | return count; |
---|
197 | } |
---|
198 | |
---|
199 | long GBT_get_species_count(GBDATA *gb_main) { |
---|
200 | return GBT_get_item_count(gb_main, "species_data"); |
---|
201 | } |
---|
202 | |
---|
203 | long GBT_get_SAI_count(GBDATA *gb_main) { |
---|
204 | return GBT_get_item_count(gb_main, "extended_data"); |
---|
205 | } |
---|
206 | |
---|
207 | // -------------------------------------------------------------------------------- |
---|
208 | |
---|
209 | static char *GBT_create_unique_item_identifier(GBDATA *gb_item_container, const char *id_field, const char *default_id) { |
---|
210 | // returns an identifier not used by items in 'gb_item_container' |
---|
211 | // 'id_field' is the entry that is used as identifier (e.g. 'name' for species) |
---|
212 | // 'default_id' will be suffixed with a number to generate a unique id |
---|
213 | // |
---|
214 | // Note: |
---|
215 | // * The resulting id may be longer than 8 characters |
---|
216 | // * This function is slow, so just use in extra-ordinary situations |
---|
217 | |
---|
218 | GBDATA *gb_item = GBT_find_item_rel_item_data(gb_item_container, id_field, default_id); |
---|
219 | char *unique_id; |
---|
220 | |
---|
221 | if (!gb_item) { |
---|
222 | unique_id = strdup(default_id); // default_id is unused |
---|
223 | } |
---|
224 | else { |
---|
225 | char *generated_id = (char*)malloc(strlen(default_id)+20); |
---|
226 | size_t min_num = 1; |
---|
227 | |
---|
228 | #define GENERATE_ID(num) sprintf(generated_id, "%s%zi", default_id, num); |
---|
229 | |
---|
230 | GENERATE_ID(min_num); |
---|
231 | gb_item = GBT_find_item_rel_item_data(gb_item_container, id_field, generated_id); |
---|
232 | |
---|
233 | if (gb_item) { |
---|
234 | size_t num_items = GB_number_of_subentries(gb_item_container); |
---|
235 | size_t max_num = 0; |
---|
236 | |
---|
237 | gb_assert(num_items); // otherwise deadlock below |
---|
238 | |
---|
239 | do { |
---|
240 | max_num += num_items; |
---|
241 | GENERATE_ID(max_num); |
---|
242 | gb_item = GBT_find_item_rel_item_data(gb_item_container, id_field, generated_id); |
---|
243 | } while (gb_item && max_num >= num_items); |
---|
244 | |
---|
245 | if (max_num<num_items) { // overflow |
---|
246 | char *uid; |
---|
247 | generated_id[0] = 'a'+GB_random(26); |
---|
248 | generated_id[1] = 'a'+GB_random(26); |
---|
249 | generated_id[2] = 0; |
---|
250 | |
---|
251 | uid = GBT_create_unique_item_identifier(gb_item_container, id_field, generated_id); |
---|
252 | strcpy(generated_id, uid); |
---|
253 | free(uid); |
---|
254 | } |
---|
255 | else { |
---|
256 | // max_num is unused |
---|
257 | while ((max_num-min_num)>1) { |
---|
258 | size_t mid = (min_num+max_num)/2; |
---|
259 | gb_assert(mid != min_num && mid != max_num); |
---|
260 | |
---|
261 | GENERATE_ID(mid); |
---|
262 | gb_item = GBT_find_item_rel_item_data(gb_item_container, id_field, generated_id); |
---|
263 | |
---|
264 | if (gb_item) min_num = mid; |
---|
265 | else max_num = mid; |
---|
266 | } |
---|
267 | GENERATE_ID(max_num); |
---|
268 | gb_assert(GBT_find_item_rel_item_data(gb_item_container, id_field, generated_id) == NULL); |
---|
269 | } |
---|
270 | } |
---|
271 | unique_id = generated_id; |
---|
272 | |
---|
273 | #undef GENERATE_ID |
---|
274 | } |
---|
275 | |
---|
276 | return unique_id; |
---|
277 | } |
---|
278 | |
---|
279 | char *GBT_create_unique_species_name(GBDATA *gb_main, const char *default_name) { |
---|
280 | return GBT_create_unique_item_identifier(GBT_get_species_data(gb_main), "name", default_name); |
---|
281 | } |
---|
282 | |
---|
283 | |
---|
284 | // -------------------------------- |
---|
285 | // mark and unmark species |
---|
286 | |
---|
287 | void GBT_mark_all(GBDATA *gb_main, int flag) { |
---|
288 | // flag == 0 -> unmark |
---|
289 | // flag == 1 -> mark |
---|
290 | // flag == 2 -> invert |
---|
291 | |
---|
292 | GBDATA *gb_species; |
---|
293 | GB_push_transaction(gb_main); |
---|
294 | |
---|
295 | if (flag == 2) { |
---|
296 | for (gb_species = GBT_first_species(gb_main); |
---|
297 | gb_species; |
---|
298 | gb_species = GBT_next_species(gb_species)) |
---|
299 | { |
---|
300 | GB_write_flag(gb_species, !GB_read_flag(gb_species)); |
---|
301 | } |
---|
302 | } |
---|
303 | else { |
---|
304 | gb_assert(flag == 0 || flag == 1); |
---|
305 | |
---|
306 | for (gb_species = GBT_first_species(gb_main); |
---|
307 | gb_species; |
---|
308 | gb_species = GBT_next_species(gb_species)) |
---|
309 | { |
---|
310 | GB_write_flag(gb_species, flag); |
---|
311 | } |
---|
312 | } |
---|
313 | GB_pop_transaction(gb_main); |
---|
314 | } |
---|
315 | void GBT_mark_all_that(GBDATA *gb_main, int flag, int (*condition)(GBDATA*, void*), void *cd) |
---|
316 | { |
---|
317 | GBDATA *gb_species; |
---|
318 | GB_push_transaction(gb_main); |
---|
319 | |
---|
320 | if (flag == 2) { |
---|
321 | for (gb_species = GBT_first_species(gb_main); |
---|
322 | gb_species; |
---|
323 | gb_species = GBT_next_species(gb_species)) |
---|
324 | { |
---|
325 | if (condition(gb_species, cd)) { |
---|
326 | GB_write_flag(gb_species, !GB_read_flag(gb_species)); |
---|
327 | } |
---|
328 | } |
---|
329 | } |
---|
330 | else { |
---|
331 | gb_assert(flag == 0 || flag == 1); |
---|
332 | |
---|
333 | for (gb_species = GBT_first_species(gb_main); |
---|
334 | gb_species; |
---|
335 | gb_species = GBT_next_species(gb_species)) |
---|
336 | { |
---|
337 | int curr_flag = GB_read_flag(gb_species); |
---|
338 | if (curr_flag != flag && condition(gb_species, cd)) { |
---|
339 | GB_write_flag(gb_species, flag); |
---|
340 | } |
---|
341 | } |
---|
342 | } |
---|
343 | GB_pop_transaction(gb_main); |
---|
344 | } |
---|
345 | |
---|
346 | long GBT_count_marked_species(GBDATA *gb_main) |
---|
347 | { |
---|
348 | long cnt = 0; |
---|
349 | GBDATA *gb_species_data; |
---|
350 | |
---|
351 | GB_push_transaction(gb_main); |
---|
352 | gb_species_data = GB_search(gb_main, "species_data", GB_CREATE_CONTAINER); |
---|
353 | GB_pop_transaction(gb_main); |
---|
354 | |
---|
355 | cnt = GB_number_of_marked_subentries(gb_species_data); |
---|
356 | return cnt; |
---|
357 | } |
---|
358 | |
---|
359 | char *GBT_store_marked_species(GBDATA *gb_main, int unmark_all) |
---|
360 | { |
---|
361 | /* stores the currently marked species in a string |
---|
362 | if (unmark_all != 0) then unmark them too |
---|
363 | */ |
---|
364 | |
---|
365 | GBS_strstruct *out = GBS_stropen(10000); |
---|
366 | GBDATA *gb_species; |
---|
367 | |
---|
368 | for (gb_species = GBT_first_marked_species(gb_main); |
---|
369 | gb_species; |
---|
370 | gb_species = GBT_next_marked_species(gb_species)) |
---|
371 | { |
---|
372 | GBS_strcat(out, GBT_read_name(gb_species)); |
---|
373 | GBS_chrcat(out, ';'); |
---|
374 | if (unmark_all) GB_write_flag(gb_species, 0); |
---|
375 | } |
---|
376 | |
---|
377 | GBS_str_cut_tail(out, 1); // remove trailing ';' |
---|
378 | return GBS_strclose(out); |
---|
379 | } |
---|
380 | |
---|
381 | NOT4PERL GB_ERROR GBT_with_stored_species(GBDATA *gb_main, const char *stored, species_callback doit, int *clientdata) { |
---|
382 | // call function 'doit' with all species stored in 'stored' |
---|
383 | |
---|
384 | #define MAX_NAME_LEN 20 |
---|
385 | char name[MAX_NAME_LEN+1]; |
---|
386 | GB_ERROR error = 0; |
---|
387 | |
---|
388 | while (!error) { |
---|
389 | const char *p = strchr(stored, ';'); |
---|
390 | int len = p ? (p-stored) : (int)strlen(stored); |
---|
391 | GBDATA *gb_species; |
---|
392 | |
---|
393 | gb_assert(len <= MAX_NAME_LEN); |
---|
394 | memcpy(name, stored, len); |
---|
395 | name[len] = 0; |
---|
396 | |
---|
397 | gb_species = GBT_find_species(gb_main, name); |
---|
398 | if (gb_species) { |
---|
399 | error = doit(gb_species, clientdata); |
---|
400 | } |
---|
401 | else { |
---|
402 | error = "Some stored species where not found."; |
---|
403 | } |
---|
404 | |
---|
405 | if (!p) break; |
---|
406 | stored = p+1; |
---|
407 | } |
---|
408 | #undef MAX_NAME_LEN |
---|
409 | return error; |
---|
410 | } |
---|
411 | |
---|
412 | static GB_ERROR restore_mark(GBDATA *gb_species, int *) { |
---|
413 | GB_write_flag(gb_species, 1); |
---|
414 | return 0; |
---|
415 | } |
---|
416 | |
---|
417 | GB_ERROR GBT_restore_marked_species(GBDATA *gb_main, const char *stored_marked) { |
---|
418 | /* restores the species-marks to a state currently saved |
---|
419 | * into 'stored_marked' by GBT_store_marked_species |
---|
420 | */ |
---|
421 | |
---|
422 | GBT_mark_all(gb_main, 0); // unmark all species |
---|
423 | return GBT_with_stored_species(gb_main, stored_marked, restore_mark, 0); |
---|
424 | } |
---|
425 | |
---|
426 | // --------------------------------- |
---|
427 | // read species information |
---|
428 | |
---|
429 | #if defined(WARN_TODO) |
---|
430 | #warning rename the following functions to make the difference more obvious?? |
---|
431 | #endif |
---|
432 | GB_CSTR GBT_read_name(GBDATA *gb_item) { |
---|
433 | GB_CSTR result = GBT_read_char_pntr(gb_item, "name"); |
---|
434 | if (!result) result = GBS_global_string("<unnamed_%s>", GB_read_key_pntr(gb_item)); |
---|
435 | return result; |
---|
436 | } |
---|
437 | |
---|
438 | const char *GBT_get_name(GBDATA *gb_item) { |
---|
439 | GBDATA *gb_name = GB_entry(gb_item, "name"); |
---|
440 | if (!gb_name) return 0; |
---|
441 | return GB_read_char_pntr(gb_name); |
---|
442 | } |
---|
443 | |
---|
444 | GBDATA **GBT_gen_species_array(GBDATA *gb_main, long *pspeccnt) |
---|
445 | { |
---|
446 | GBDATA *gb_species; |
---|
447 | GBDATA *gb_species_data = GBT_find_or_create(gb_main, "species_data", 7); |
---|
448 | GBDATA **result; |
---|
449 | *pspeccnt = 0; |
---|
450 | for (gb_species = GBT_first_species_rel_species_data(gb_species_data); |
---|
451 | gb_species; |
---|
452 | gb_species = GBT_next_species(gb_species)) { |
---|
453 | (*pspeccnt) ++; |
---|
454 | } |
---|
455 | result = (GBDATA **)malloc((size_t)(sizeof(GBDATA *)* (*pspeccnt))); |
---|
456 | *pspeccnt = 0; |
---|
457 | for (gb_species = GBT_first_species_rel_species_data(gb_species_data); |
---|
458 | gb_species; |
---|
459 | gb_species = GBT_next_species(gb_species)) { |
---|
460 | result[(*pspeccnt)++]=gb_species; |
---|
461 | } |
---|
462 | return result; |
---|
463 | } |
---|
464 | |
---|