Context Navigation

source: tags/ms_r18q1/ARBDB/adlang1.cxx

Visit:

Last change on this file was 17110, checked in by westram, 6 years ago
reintegrates '' into 'trunk' implements #735 adds: log:branches/addtest@17040:17044 log:branches/tree@16921:17109
Property svn:eol-style set to `native` Property svn:keywords set to `Author Date Id Revision`
File size: 90.6 KB

Line
1	// =============================================================== //
2	// //
3	// File : adlang1.cxx //
4	// Purpose : //
5	// //
6	// Institute of Microbiology (Technical University Munich) //
7	// http://www.arb-home.de/ //
8	// //
9	// =============================================================== //
10
11	#include "gb_aci_impl.h"
12	#include "gb_key.h"
13
14	#include "TreeNode.h"
15
16	#include <aw_awar_defs.hxx>
17
18	#include <adGene.h>
19	#include <ad_cb.h>
20
21	#include <arb_defs.h>
22	#include <arb_strbuf.h>
23	#include <arb_file.h>
24	#include <arb_strarray.h>
25	#include <arb_sort.h>
26
27	#include <cctype>
28	#include <cmath>
29	#include <algorithm>
30
31	// hook for 'export_sequence'
32
33	static gb_export_sequence_cb get_export_sequence = NULp;
34
35	NOT4PERL void GB_set_export_sequence_hook(gb_export_sequence_cb escb) {
36	gb_assert(!get_export_sequence \|\| !escb); // avoid unwanted overwrite
37	get_export_sequence = escb;
38	}
39
40	using namespace GBL_IMPL;
41
42	namespace GBL_IMPL {
43	// global ACI/SRT debug switch
44	int traceACI = 0;
45	int traceIndent = 0;
46
47	void print_trace(const char *text) {
48	FILE *DUMPOUT = stdout;
49
50	gb_assert(traceACI>0);
51	if (traceIndent>0) {
52	ConstStrArray line;
53	GBT_split_string(line, text, "\n", true);
54	for (unsigned L = 0; L<line.size(); ++L) {
55	for (int i = 0; i<traceIndent; ++i) {
56	fputc(' ', DUMPOUT);
57	}
58	fputs(line[L], DUMPOUT);
59	fputc('\n', DUMPOUT);
60	}
61	}
62	else {
63	fputs(text, DUMPOUT);
64	}
65	fflush(DUMPOUT);
66	}
67
68	GB_ERROR trace_params(const GBL_streams& param, gbl_param ppara, const char com) {
69	GB_ERROR error = NULp;
70	int i;
71
72	int argc = param.size();
73	for (i=0; i<argc; i++) {
74	gbl_param *para;
75	const char *argument = param.get(i);
76
77	for (para = ppara; para && !error; para = para->next) {
78	if (para->param_name) { // NULp means param is inactive (see PARAM_IF)
79	int len = strlen(para->param_name);
80
81	if (strncmp(argument, para->param_name, len) == 0) {
82	const char *value = argument+len; // set to start of value
83
84	if (para->type == GB_BIT) {
85	// GB_BIT is special cause param_name does NOT contain trailing '='
86
87	if (!value[0]) { // only 'name' -> handle like 'name=1'
88	;
89	}
90	else if (value[0] == '=') {
91	value++;
92	}
93	}
94
95	switch (para->type) {
96	case GB_STRING:
97	(const char *)para->varaddr = value;
98	break;
99
100	case GB_INT:
101	STATIC_ASSERT(sizeof(int) == sizeof(nat)); // assumed by GBL_PARAM_UINT
102	(int )para->varaddr = atoi(value);
103	break;
104
105	case GB_BIT:
106	// 'param=' is same as 'param' or 'param=1' (historical reason, don't change)
107	(int )para->varaddr = (value[0] ? atoi(value) : 1);
108	break;
109
110	case GB_BYTE:
111	(char )para->varaddr = *value; // this may use the terminal zero-byte (e.g. for p1 in 'p0=0,p1=,p2=2' )
112	if (value[0] && value[1]) { // found at least 2 chars
113	GB_warningf("Only one character expected in value '%s' of param '%s' - rest is ignored", value, para->param_name);
114	}
115	break;
116
117	default:
118	gb_assert(0);
119	error = GBS_global_string("Parameter '%s': Unknown type %i (internal error)", para->param_name, para->type);
120	break;
121	}
122	break; // accept parameter
123	}
124	}
125	}
126
127	if (!error && !para) { // no parameter found for argument
128	int pcount = 0;
129	gbl_param **params;
130	int k;
131	char *res;
132	GBS_strstruct *str = GBS_stropen(1000);
133	GB_ERROR err;
134
135
136	for (para = ppara; para; para = para->next) pcount++;
137	ARB_calloc(params, pcount);
138	for (k = 0, para = ppara; para; para = para->next) params[k++] = para;
139
140
141	for (pcount--; pcount>=0; pcount--) {
142	para = params[pcount];
143	if (para->param_name) {
144	GBS_strcat(str, " ");
145	GBS_strcat(str, para->param_name);
146	switch (para->type) {
147	case GB_STRING: GBS_strcat(str, "STRING"); break;
148	case GB_INT: GBS_strcat(str, "INT"); break;
149	case GB_FLOAT: GBS_strcat(str, "FLOAT"); break;
150	case GB_BYTE: GBS_strcat(str, "CHAR"); break;
151	case GB_BIT: GBS_strcat(str, " "); break;
152	default: gb_assert (0); GBS_strcat(str, "????"); break;
153	}
154	GBS_strcat(str, "\t\t;");
155	GBS_strcat(str, para->help_text);
156	GBS_strcat(str, "\n");
157	}
158	}
159	freenull(params);
160	res = GBS_strclose(str);
161	err = GB_export_errorf("Unknown Parameter '%s' in command '%s'\n PARAMETERS:\n%s", argument, com, res);
162	free(res);
163	return err;
164	}
165	}
166
167	return error;
168	}
169	};
170
171
172
173	// -------------------------
174	// String functions
175
176	static int gbl_stricmp(const char s1, const char s2) {
177	// case insensitive strcmp
178	int i;
179	for (i = 0; ; ++i) {
180	char c1 = tolower(s1[i]);
181	char c2 = tolower(s2[i]);
182
183	if (c1 == c2) {
184	if (!c1) break; // equal strings
185	}
186	else {
187	if (c1<c2) return -1;
188	return 1;
189	}
190	}
191	return 0;
192	}
193	static int gbl_strincmp(const char s1, const char s2, int size2) {
194	// case insensitive strcmp
195	int i;
196	for (i = 0; i<size2; ++i) {
197	char c1 = tolower(s1[i]);
198	char c2 = tolower(s2[i]);
199
200	if (c1 == c2) {
201	if (!c1) break; // equal strings
202	}
203	else {
204	if (c1<c2) return -1;
205	return 1;
206	}
207	}
208	return 0;
209	}
210	static const char gbl_stristr(const char haystack, const char *needle) {
211	// case insensitive strstr
212	const char *hp = haystack;
213	char c1 = toupper(needle[0]);
214	char c2 = tolower(c1);
215	int needle_size = strlen(needle);
216
217	if (c1 == c2) {
218	hp = strchr(hp, c1);
219	while (hp) {
220	if (gbl_strincmp(hp, needle, needle_size) == 0) return hp;
221	hp = strchr(hp+1, c1);
222	}
223	}
224	else {
225	while (hp) {
226	const char *h1 = strchr(hp, c1);
227	const char *h2 = strchr(hp, c2);
228
229	if (h1 && h2) {
230	if (h1<h2) {
231	if (gbl_strincmp(h1, needle, needle_size) == 0) return h1;
232	hp = h1+1;
233	}
234	else {
235	gb_assert(h1>h2);
236	if (gbl_strincmp(h2, needle, needle_size) == 0) return h2;
237	hp = h2+1;
238	}
239	}
240	else {
241	if (h1) { hp = h1; }
242	else if (h2) { hp = h2; c1 = c2; }
243	else { hp = NULp; }
244
245	while (hp) {
246	if (gbl_strincmp(hp, needle, needle_size) == 0) return hp;
247	hp = strchr(hp+1, c1);
248	}
249	}
250	}
251	}
252	return NULp;
253	}
254
255	inline int approve_pos(int pos, int len) { return pos<0 ? (-pos<len ? len+pos : 0) : pos; }
256
257	static GB_ERROR gbl_mid_streams(const GBL_streams& arg_input, GBL_streams& arg_out, int start, int end) {
258	// used as well to copy all streams (e.g. by 'dd')
259	for (int i=0; i<arg_input.size(); i++) {
260	const char *p = arg_input.get(i);
261	int len = strlen(p);
262
263	int s = approve_pos(start, len);
264	int e = approve_pos(end, len);
265
266	char *res;
267	if (s >= len \|\| e<s) {
268	res = ARB_strdup("");
269	}
270	else {
271	gb_assert(s >= 0);
272	res = ARB_strpartdup(p+s, p+e);
273	}
274	arg_out.insert(res);
275	}
276	return NULp;
277	}
278
279	static GB_ERROR gbl_trace(GBL_command_arguments *args) {
280	int tmp_trace;
281
282	EXPECT_PARAMS(args, 1, "0\|1");
283
284	tmp_trace = atoi(args->get_param(0));
285	if (tmp_trace<0 \|\| tmp_trace>1) return GBS_global_string("Illegal value %i to trace", tmp_trace);
286
287	if (tmp_trace != traceACI) {
288	traceACI = 1;
289	print_trace(GBS_global_string("%sctivated ACI trace\n", tmp_trace ? "A" : "De-a"));
290	traceACI = tmp_trace;
291	}
292
293	return gbl_mid_streams(args->input, args->output, 0, -1); // copy all streams
294	}
295
296	/* ---------------------------------------------------------------------------------------
297	* Binary operators work on pairs of values.
298	* Three different operational modes are implemented for all binary operators:
299	*
300	* 1. inputstreams\|operator
301	*
302	* The number of inputstreams has to be even and the operator will be
303	* applied to pairs of them.
304	*
305	* Example : a;b;c;d;e;f \| plus
306	* Result : a+b;c+d;e+f
307	*
308	* 2. inputstreams\|operator(x)
309	*
310	* The number of inputstreams has to be at least 1.
311	* The operator is applied to each inputstream.
312	*
313	* Example : a;b;c \| plus(d)
314	* Result : a+d;b+d;c+d
315	*
316	* 3. operator(x, y)
317	*
318	* @@@ this decription does not match behavior!
319	* @@@ check description in helpfile as well
320	*
321	* Inputstreams will be ignored and the operator is applied
322	* to the arguments
323	*
324	* Example : a;b \| plus(c,d)
325	* Result : c+d
326	*/
327
328	template <typename T>
329	GB_ERROR gbl_apply_binary_operator(GBL_command_arguments args, char (op)(const char , const char *, T), T client_data) {
330	GB_ERROR error = NULp;
331	switch (args->param_count()) {
332	case 0:
333	gb_assert(args->set_params_checked());
334	if (args->input.size() == 0) error = "Expect at least two input streams if called with 0 parameters";
335	else if (args->input.size()%2) error = "Expect an even number of input streams if called with 0 parameters";
336	else {
337	int inputpairs = args->input.size()/2;
338	int i;
339	for (i = 0; i<inputpairs; ++i) {
340	PASS_2_OUT(args, op(args->input.get(i2), args->input.get(i2+1), client_data));
341	}
342	}
343	break;
344
345	case 1:
346	gb_assert(args->set_params_checked());
347	if (args->input.size() == 0) error = "Expect at least one input stream if called with 1 parameter";
348	else {
349	int i;
350	const char *argument = args->get_param(0);
351	for (i = 0; i<args->input.size(); ++i) {
352	PASS_2_OUT(args, op(args->input.get(i), argument, client_data));
353	}
354	}
355	break;
356
357	case 2:
358	gb_assert(args->set_params_checked());
359	for (int i = 0; i<args->input.size(); ++i) {
360	char *result1 = args->get_callEnv().interpret_subcommand(args->input.get(i), args->get_param(0)); // @@@ EVALUATED_PARAM (#768)
361	if (!result1) error = GB_await_error();
362	else {
363	char *result2 = args->get_callEnv().interpret_subcommand(args->input.get(i), args->get_param(1)); // @@@ EVALUATED_PARAM (#768)
364	if (!result2) error = GB_await_error();
365	else {
366	PASS_2_OUT(args, op(result1, result2, client_data));
367	free(result2);
368	}
369	free(result1);
370	}
371	}
372	break;
373
374	default:
375	error = check_optional_parameters(args, 0, NULp, 2, "Expr1[,Expr2]", true, false);
376	break;
377	}
378
379	return error;
380	}
381
382	// --------------------------------
383	// escape/unescape strings
384
385	static char unEscapeString(const char escapedString) {
386	// replaces all \x by x
387	char *result = nulldup(escapedString);
388	char *to = result;
389	char *from = result;
390
391	while (1) {
392	char c = *from++;
393	if (!c) break;
394
395	if (c=='\\') {
396	to++ = from++;
397	}
398	else {
399	*to++ = c;
400	}
401	}
402	*to = 0;
403	return result;
404	}
405	static char escapeString(const char unescapedString) {
406	// replaces all '\' and '"' by '\\' and '\"'
407	int len = strlen(unescapedString);
408	char result = ARB_alloc<char>(2len+1);
409	char *to = result;
410	const char *from = unescapedString;
411
412	while (1) {
413	char c = *from++;
414	if (!c) break;
415
416	if (c=='\\' \|\| c == '\"') {
417	*to++ = '\\';
418	*to++ = c;
419	}
420	else {
421	*to++ = c;
422	}
423	}
424	*to = 0;
425	return result;
426	}
427
428	// ---------------------------------
429	// the commands themselves:
430
431	static GB_ERROR gbl_quote(GBL_command_arguments *args) {
432	EXPECT_NO_PARAM(args);
433
434	for (int i=0; i<args->input.size(); i++) {
435	FORMAT_2_OUT(args, "\"%s\"", args->input.get(i));
436	}
437	return NULp;
438	}
439	static GB_ERROR gbl_unquote(GBL_command_arguments *args) {
440	EXPECT_NO_PARAM(args);
441
442	for (int i=0; i<args->input.size(); i++) {
443	const char *str = args->input.get(i);
444	int len = strlen(str);
445
446	if (str[0] == '\"' && str[len-1] == '\"') {
447	PASS_2_OUT(args, ARB_strpartdup(str+1, str+len-2));
448	}
449	else {
450	IN_2_OUT(args, i);
451	}
452	}
453	return NULp;
454	}
455
456	static GB_ERROR gbl_escape(GBL_command_arguments *args) {
457	EXPECT_NO_PARAM(args);
458
459	for (int i=0; i<args->input.size(); i++) {
460	char *escaped = escapeString(args->input.get(i));
461	PASS_2_OUT(args, escaped);
462	}
463	return NULp;
464	}
465	static GB_ERROR gbl_unescape(GBL_command_arguments *args) {
466	EXPECT_NO_PARAM(args);
467
468	for (int i=0; i<args->input.size(); i++) {
469	char *unescaped = unEscapeString(args->input.get(i));
470	PASS_2_OUT(args, unescaped);
471	}
472	return NULp;
473	}
474
475	static GB_ERROR gbl_command(GBL_command_arguments *args) {
476	EXPECT_PARAMS(args, 1, "\"ACI command\"");
477
478	GB_ERROR error = NULp;
479	char *command = unEscapeString(args->get_param(0));
480
481	if (traceACI) {
482	print_trace(GBS_global_string("executing command '%s'\n", command));
483	}
484
485	for (int i=0; i<args->input.size() && !error; i++) {
486	char *result = args->get_callEnv().interpret_subcommand(args->input.get(i), command);
487	if (!result) error = GB_await_error();
488	else PASS_2_OUT(args, result);
489	}
490	free(command);
491	return error;
492	}
493
494	static GB_ERROR gbl_eval(GBL_command_arguments *args) {
495	EXPECT_PARAMS(args, 1, "\"expression evaluating to ACI command\"");
496
497	GB_ERROR error = NULp;
498	char *to_eval = unEscapeString(args->get_param(0));
499	TRACE_ACI(GBS_global_string("evaluating '%s'\n", to_eval));
500
501	char *command = args->get_callEnv().interpret_subcommand("", to_eval); // evaluate independent
502	if (!command) error = GB_await_error();
503	else {
504	TRACE_ACI(GBS_global_string("executing '%s'\n", command));
505
506	for (int i=0; i<args->input.size() && !error; i++) {
507	char *result = args->get_callEnv().interpret_subcommand(args->input.get(i), command);
508	if (!result) error = GB_await_error();
509	else PASS_2_OUT(args, result);
510	}
511	free(command);
512	}
513	free(to_eval);
514	return error;
515	}
516
517	class DefinedCommands : virtual Noncopyable {
518	GB_HASH *cmds;
519	public:
520	DefinedCommands() { cmds = GBS_create_dynaval_hash(100, GB_MIND_CASE, GBS_dynaval_free); }
521	~DefinedCommands() { GBS_free_hash(cmds); }
522
523	void set(const char name, char cmd) { GBS_dynaval_free(GBS_write_hash(cmds, name, (long)cmd)); } // takes ownership of 'cmd'!
524	const char get(const char name) const { return (const char *)GBS_read_hash(cmds, name); }
525	};
526
527	static DefinedCommands defined_commands;
528
529	static GB_ERROR gbl_define(GBL_command_arguments *args) {
530	COMMAND_DROPS_INPUT_STREAMS(args);
531	EXPECT_PARAMS(args, 2, "name, \"ACI command\"");
532
533	const char *name = args->get_param(0);
534	char *cmd = unEscapeString(args->get_param(1));
535
536	defined_commands.set(name, cmd);
537	TRACE_ACI(GBS_global_string("defining command '%s'='%s'\n", name, cmd));
538	return NULp;
539	}
540
541	static GB_ERROR gbl_do(GBL_command_arguments *args) {
542	EXPECT_PARAMS(args, 1, "definedCommandName");
543
544	GB_ERROR error = NULp;
545	const char *name = args->get_param(0);
546	const char *cmd = defined_commands.get(name);
547
548	if (!cmd) {
549	error = GBS_global_string("Can't do undefined command '%s' - use define(%s, ...) first", name, name);
550	}
551	else {
552	TRACE_ACI(GBS_global_string("executing defined command '%s'='%s' on %i streams\n", name, cmd, args->input.size()));
553
554	for (int i=0; i<args->input.size() && !error; i++) {
555	char *result = args->get_callEnv().interpret_subcommand(args->input.get(i), cmd);
556	if (!result) error = GB_await_error();
557	else PASS_2_OUT(args, result);
558	}
559	}
560	return error;
561	}
562
563	static GB_ERROR gbl_streams(GBL_command_arguments *args) {
564	EXPECT_NO_PARAM(args);
565
566	FORMAT_2_OUT(args, "%i", args->input.size());
567	return NULp;
568	}
569
570	static GB_ERROR apply_to_origin(GBL_command_arguments *args, bool organism) {
571	EXPECT_PARAMS(args, 1, "\"ACI command\"");
572
573	GB_ERROR error = NULp;
574	if (!GEN_is_pseudo_gene_species(args->get_ref())) {
575	error = GBS_global_string("'%s' applies to gene-species only", args->get_cmdName());
576	}
577	else {
578	GBDATA *gb_origin = NULp;
579	if (organism) {
580	gb_origin = GEN_find_origin_organism(args->get_ref(), NULp);
581	}
582	else {
583	gb_origin = GEN_find_origin_gene(args->get_ref(), NULp);
584	}
585
586	if (!error && !gb_origin) error = GB_await_error();
587
588	if (!error) {
589	char *command = unEscapeString(args->get_param(0));
590	GBL_call_env callEnv(gb_origin, args->get_env()); // refer to gb_origin for subcommands
591
592	for (int i=0; i<args->input.size() && !error; i++) {
593	char *result = callEnv.interpret_subcommand(args->input.get(i), command);
594	if (!result) error = GB_await_error();
595	else PASS_2_OUT(args, result);
596	}
597
598	free(command);
599	}
600	}
601
602	return error;
603	}
604
605	static GB_ERROR gbl_origin_gene(GBL_command_arguments *args) { return apply_to_origin(args, false); }
606	static GB_ERROR gbl_origin_organism(GBL_command_arguments *args) { return apply_to_origin(args, true); }
607
608	class Tab {
609	bool tab[256];
610	public:
611	Tab(bool take, const char *invert) {
612	bool init = !take;
613	for (int i = 0; i<256; ++i) tab[i] = init;
614	for (int i = 0; invert[i]; ++i) tab[safeCharIndex(invert[i])] = take;
615	}
616	bool operator[](int i) const { return tab[i]; }
617	};
618
619	inline GB_ERROR count_by_tab(GBL_command_arguments *args, const Tab& tab) {
620	for (int i=0; i<args->input.size(); ++i) {
621	long sum = 0; // count frequencies
622	const char *p = args->input.get(i);
623
624	while (p) sum += tab[safeCharIndex((p++))];
625	FORMAT_2_OUT(args, "%li", sum);
626	}
627	return NULp;
628	}
629	inline GB_ERROR remove_by_tab(GBL_command_arguments *args, const Tab& tab) {
630	for (int i=0; i<args->input.size(); ++i) {
631	GBS_strstruct *strstruct = GBS_stropen(1000);
632	for (const char p = args->input.get(i); p; p++) {
633	if (!tab[(unsigned int)*p]) {
634	GBS_chrcat(strstruct, *p);
635	}
636	}
637	PASS_2_OUT(args, GBS_strclose(strstruct));
638	}
639	return NULp;
640	}
641
642	static GB_ERROR gbl_count(GBL_command_arguments *args) {
643	EXPECT_PARAMS(args, 1, "\"characters to count\"");
644	return count_by_tab(args, Tab(true, args->get_param(0)));
645	}
646	static GB_ERROR gbl_len(GBL_command_arguments *args) {
647	EXPECT_OPTIONAL_PARAMS(args, 0, NULp, 1, "\"characters not to count\"");
648	const char *exclude = args->get_optional_param(0, "");
649	return count_by_tab(args, Tab(false, exclude));
650	}
651	static GB_ERROR gbl_remove(GBL_command_arguments *args) {
652	EXPECT_PARAMS(args, 1, "\"characters to remove\"");
653	return remove_by_tab(args, Tab(true, args->get_param(0)));
654	}
655	static GB_ERROR gbl_keep(GBL_command_arguments *args) {
656	EXPECT_PARAMS(args, 1, "\"characters to keep\"");
657	return remove_by_tab(args, Tab(false, args->get_param(0)));
658	}
659
660
661	static char binop_compare(const char arg1, const char *arg2, bool case_sensitive) {
662	int result;
663
664	if (case_sensitive) result = strcmp(arg1, arg2);
665	else result = gbl_stricmp(arg1, arg2);
666
667	return GBS_global_string_copy("%i", result<0 ? -1 : (result>0 ? 1 : 0));
668	}
669	static char binop_equals(const char arg1, const char *arg2, bool case_sensitive) {
670	int result;
671
672	if (case_sensitive) result = strcmp(arg1, arg2);
673	else result = gbl_stricmp(arg1, arg2);
674
675	return GBS_global_string_copy("%i", result == 0 ? 1 : 0);
676	}
677	static char binop_contains(const char arg1, const char *arg2, bool case_sensitive) {
678	const char *found = NULp;
679
680	if (!arg2[0]) return strdup("0"); // do not report matches of empty string
681
682	if (case_sensitive) found = strstr(arg1, arg2);
683	else found = gbl_stristr(arg1, arg2);
684
685	return GBS_global_string_copy("%ti", found ? (found-arg1)+1 : 0);
686	}
687	static char binop_partof(const char arg1, const char *arg2, bool case_sensitive) {
688	return binop_contains(arg2, arg1, case_sensitive);
689	}
690
691	static GB_ERROR gbl_compare (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_compare, true); }
692	static GB_ERROR gbl_icompare (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_compare, false); }
693	static GB_ERROR gbl_equals (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_equals, true); }
694	static GB_ERROR gbl_iequals (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_equals, false); }
695	static GB_ERROR gbl_contains (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_contains, true); }
696	static GB_ERROR gbl_icontains(GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_contains, false); }
697	static GB_ERROR gbl_partof (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_partof, true); }
698	static GB_ERROR gbl_ipartof (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, binop_partof, false); }
699
700	static GB_ERROR gbl_isEmpty(GBL_command_arguments *args) {
701	EXPECT_NO_PARAM(args);
702	for (int i=0; i<args->input.size(); i++) {
703	const char *str = args->input.get(i);
704	FORMAT_2_OUT(args, "%i", str[0] == 0);
705	}
706	return NULp;
707	}
708	static GB_ERROR gbl_inRange(GBL_command_arguments *args) {
709	EXPECT_PARAMS(args, 2, "low,high");
710
711	double low = strtod(args->get_param(0), NULp);
712	double high = strtod(args->get_param(1), NULp);
713
714	for (int i=0; i<args->input.size(); i++) {
715	double val = strtod(args->input.get(i), NULp);
716	bool inRange = low<=val && val<=high;
717	FORMAT_2_OUT(args, "%i", inRange);
718	}
719	return NULp;
720	}
721
722
723	static GB_ERROR gbl_translate(GBL_command_arguments *args) {
724	EXPECT_OPTIONAL_PARAMS(args, 2, "old,new", 1, "other");
725
726	const char *other = args->get_optional_param(2, NULp);
727	if (other && (other[0] == 0 \|\| other[1] != 0)) {
728	return "third parameter of translate has to be one character (i.e. \"-\")";
729	}
730	const char replace_other = other ? other[0] : 0;
731
732	// build translation table :
733	unsigned char tab[256];
734	{
735	const unsigned char o = (const unsigned char )args->get_param(0);
736	const unsigned char n = (const unsigned char )args->get_param(1);
737	char used[256];
738
739	if (strlen((const char )o) != strlen((const char )n)) {
740	return "arguments 1 and 2 of translate should be strings with identical length";
741	}
742
743	for (int i = 0; i<256; ++i) { // IRRELEVANT_LOOP
744	tab[i] = replace_other ? replace_other : i; // replace unused or identity translation
745	used[i] = 0;
746	}
747
748	for (int i = 0; o[i]; ++i) {
749	if (used[o[i]]) return GBS_global_string("character '%c' used twice in argument 1 of translate", o[i]);
750	used[o[i]] = 1;
751	tab[o[i]] = n[i]; // real translation
752	}
753	}
754
755	for (int i=0; i<args->input.size(); i++) {
756	GBS_strstruct *strstruct = GBS_stropen(1000);
757	for (const char p = args->input.get(i); p; p++) {
758	GBS_chrcat(strstruct, tab[(unsigned char)*p]);
759	}
760	PASS_2_OUT(args, GBS_strclose(strstruct));
761	}
762	return NULp;
763	}
764
765
766	static GB_ERROR gbl_echo(GBL_command_arguments *args) {
767	ACCEPT_ANY_PARAMS(args);
768	COMMAND_DROPS_INPUT_STREAMS(args);
769	for (int i=0; i<args->param_count(); i++) PARAM_2_OUT(args, i);
770	return NULp;
771	}
772
773	static GB_ERROR gbl_dd(GBL_command_arguments *args) {
774	EXPECT_NO_PARAM(args);
775	return gbl_mid_streams(args->input, args->output, 0, -1); // copy all streams
776	}
777
778	enum Case { UPPER, LOWER, CAPS };
779
780	static GB_ERROR convert_case(GBL_command_arguments *args, Case convTo) {
781	EXPECT_NO_PARAM(args);
782
783	for (int i=0; i<args->input.size(); i++) {
784	char *p = ARB_strdup(args->input.get(i));
785	bool last_was_alnum = false;
786
787	for (char *pp = p; pp[0]; ++pp) {
788	switch (convTo) {
789	case LOWER: pp[0] = tolower(pp[0]); break;
790	case UPPER: pp[0] = toupper(pp[0]); break;
791	case CAPS: {
792	bool alnum = isalnum(pp[0]);
793	if (alnum) pp[0] = (last_was_alnum ? tolower : toupper)(pp[0]);
794	last_was_alnum = alnum;
795	break;
796	}
797	default: gb_assert(0); break;
798	}
799	}
800
801	PASS_2_OUT(args, p);
802	}
803
804	return NULp;
805	}
806
807	static GB_ERROR gbl_caps (GBL_command_arguments *args) { return convert_case(args, CAPS); }
808	static GB_ERROR gbl_upper(GBL_command_arguments *args) { return convert_case(args, UPPER); }
809	static GB_ERROR gbl_lower(GBL_command_arguments *args) { return convert_case(args, LOWER); }
810
811	static GB_ERROR gbl_head(GBL_command_arguments *args) {
812	EXPECT_PARAMS(args, 1, "length_of_head");
813	int start = atoi(args->get_param(0));
814	if (start <= 0) return gbl_mid_streams(args->input, args->output, 1, 0); // empty all streams
815	return gbl_mid_streams(args->input, args->output, 0, start-1);
816	}
817	static GB_ERROR gbl_tail(GBL_command_arguments *args) {
818	EXPECT_PARAMS(args, 1, "length_of_tail");
819	int end = atoi(args->get_param(0));
820	if (end <= 0) return gbl_mid_streams(args->input, args->output, 1, 0); // empty all streams
821	return gbl_mid_streams(args->input, args->output, -end, -1);
822	}
823
824	inline GB_ERROR mid(GBL_command_arguments *args, int start_index) {
825	EXPECT_PARAMS(args, 2, "start,end");
826	return gbl_mid_streams(args->input, args->output, atoi(args->get_param(0))-start_index, atoi(args->get_param(1))-start_index);
827	}
828	static GB_ERROR gbl_mid0(GBL_command_arguments *args) { return mid(args, 0); }
829	static GB_ERROR gbl_mid (GBL_command_arguments *args) { return mid(args, 1); }
830
831	static GB_ERROR tab(GBL_command_arguments *args, bool pretab) {
832	EXPECT_PARAMS(args, 1, "tabstop");
833
834	int tab = atoi(args->get_param(0));
835	for (int i=0; i<args->input.size(); i++) {
836	int len = strlen(args->input.get(i));
837	if (len >= tab) IN_2_OUT(args, i);
838	else {
839	char *p = ARB_alloc<char>(tab+1);
840	if (pretab) {
841	int spaces = tab-len;
842	for (int j = 0; j<spaces; ++j) p[j] = ' ';
843	strcpy(p+spaces, args->input.get(i));
844	}
845	else {
846	strcpy(p, args->input.get(i));
847	for (int j=len; j<tab; j++) p[j] = ' ';
848	p[tab] = 0;
849	}
850	PASS_2_OUT(args, p);
851	}
852	}
853	return NULp;
854	}
855	static GB_ERROR gbl_tab (GBL_command_arguments *args) { return tab(args, false); }
856	static GB_ERROR gbl_pretab(GBL_command_arguments *args) { return tab(args, true); }
857
858	static GB_ERROR gbl_crop(GBL_command_arguments *args) {
859	EXPECT_PARAMS(args, 1, "\"chars_to_crop\"");
860
861	const char *chars_to_crop = args->get_param(0);
862	for (int i=0; i<args->input.size(); i++) {
863	const char *s = args->input.get(i);
864	while (s[0] && strchr(chars_to_crop, s[0])) s++; // crop at beg of line
865
866	int len = strlen(s);
867	char *p = ARB_alloc<char>(len+1);
868	strcpy(p, s);
869
870	{
871	char *pe = p+len-1;
872
873	while (pe >= p && strchr(chars_to_crop, pe[0])) { // crop at end of line
874	--pe;
875	}
876	gb_assert(pe >= (p-1));
877	pe[1] = 0;
878	}
879	PASS_2_OUT(args, p);
880	}
881	return NULp;
882	}
883
884
885
886	static GB_ERROR gbl_cut(GBL_command_arguments *args) {
887	EXPECT_PARAMS_PASSED(args, "streamnumber[,streamnumber]+");
888
889	for (int i=0; i<args->param_count(); i++) {
890	int stream = atoi(args->get_param(i));
891	EXPECT_LEGAL_STREAM_INDEX(args, stream);
892	IN_2_OUT(args, bio2info(stream));
893	}
894	return NULp;
895	}
896	static GB_ERROR gbl_drop(GBL_command_arguments *args) {
897	EXPECT_PARAMS_PASSED(args, "streamnumber[,streamnumber]+");
898
899	GB_ERROR error = NULp;
900	bool *dropped = ARB_alloc<bool>(args->input.size());
901
902	for (int i=0; i<args->input.size(); ++i) dropped[i] = false;
903
904	for (int i=0; i<args->param_count() && !error; ++i) {
905	int stream = atoi(args->get_param(i));
906	error = check_valid_stream_index(args, stream);
907	if (!error) dropped[bio2info(stream)] = true;
908	}
909
910	if (!error) {
911	for (int i=0; i<args->input.size(); ++i) {
912	if (!dropped[i]) IN_2_OUT(args, i);
913	}
914	}
915	free(dropped);
916
917	return error;
918	}
919
920	static GB_ERROR gbl_dropempty(GBL_command_arguments *args) {
921	EXPECT_NO_PARAM(args);
922
923	for (int i=0; i<args->input.size(); ++i) {
924	if (args->input.get(i)[0]) { // if non-empty
925	IN_2_OUT(args, i);
926	}
927	}
928	return NULp;
929	}
930
931	static GB_ERROR gbl_dropzero(GBL_command_arguments *args) {
932	EXPECT_NO_PARAM(args);
933
934	for (int i=0; i<args->input.size(); ++i) {
935	if (atoi(args->input.get(i))) { // if non-zero
936	IN_2_OUT(args, i);
937	}
938	}
939	return NULp;
940	}
941
942	static GB_ERROR gbl_swap(GBL_command_arguments *args) {
943	EXPECT_OPTIONAL_PARAMS(args, 0, NULp, 2, "streamnumber,streamnumber");
944
945	if (args->input.size()<2) return "need at least two input streams";
946
947	int swap1;
948	int swap2;
949	if (args->param_count() == 0) {
950	swap1 = args->input.size()-1;
951	swap2 = args->input.size()-2;
952	}
953	else {
954	gb_assert(args->param_count() == 2);
955
956	swap1 = atoi(args->get_param(0));
957	swap2 = atoi(args->get_param(1));
958
959	EXPECT_LEGAL_STREAM_INDEX(args, swap1);
960	EXPECT_LEGAL_STREAM_INDEX(args, swap2);
961
962	swap1 = bio2info(swap1);
963	swap2 = bio2info(swap2);
964	}
965
966	for (int i = 0; i<args->input.size(); ++i) {
967	int j = i == swap1 ? swap2 : (i == swap2 ? swap1 : i);
968	IN_2_OUT(args, j);
969	}
970
971	return NULp;
972	}
973
974	static GB_ERROR backfront_stream(GBL_command_arguments *args, int toback) {
975	EXPECT_PARAMS(args, 1, "streamnumber");
976	if (args->input.size()<1) return "need at least one input stream";
977
978	int stream_to_move = atoi(args->get_param(0));
979	EXPECT_LEGAL_STREAM_INDEX(args, stream_to_move);
980	stream_to_move = bio2info(stream_to_move);
981
982	if (!toback) IN_2_OUT(args, stream_to_move);
983	for (int i = 0; i<args->input.size(); ++i) {
984	if (i != stream_to_move) IN_2_OUT(args, i);
985	}
986	if (toback) IN_2_OUT(args, stream_to_move);
987
988	return NULp;
989	}
990	static GB_ERROR gbl_toback (GBL_command_arguments *args) { return backfront_stream(args, 1); }
991	static GB_ERROR gbl_tofront(GBL_command_arguments *args) { return backfront_stream(args, 0); }
992
993	static GB_ERROR gbl_merge(GBL_command_arguments *args) {
994	EXPECT_OPTIONAL_PARAMS(args, 0, NULp, 1, "\"separator\"");
995	const char *separator = args->get_optional_param(0, NULp);
996
997	if (args->input.size()) {
998	GBS_strstruct *str = GBS_stropen(1000);
999	GBS_strcat(str, args->input.get(0));
1000
1001	for (int i = 1; i<args->input.size(); ++i) {
1002	if (separator) GBS_strcat(str, separator);
1003	GBS_strcat(str, args->input.get(i));
1004	}
1005
1006	PASS_2_OUT(args, GBS_strclose(str));
1007	}
1008	return NULp;
1009	}
1010
1011	static GB_ERROR gbl_split(GBL_command_arguments *args) {
1012	EXPECT_OPTIONAL_PARAMS_CUSTOM(args, 0, NULp, 2, "\"separator\"[,mode]", true, false);
1013
1014	const char *separator = args->get_optional_param(0, "\n");
1015	int split_mode = atoi(args->get_optional_param(1, "0")); // 0: remove separator, 1: split before separator, 2: split behind separator
1016
1017	if (split_mode<0 \|\| split_mode>2) return GBS_global_string("Illegal split mode '%i' (valid: 0..2)", split_mode);
1018
1019	{
1020	size_t sepLen = strlen(separator);
1021
1022	for (int i = 0; i<args->input.size(); ++i) {
1023	const char *in = args->input.get(i);
1024	const char *from = in; // search from here
1025
1026	while (in) {
1027	const char *splitAt = strstr(from, separator);
1028	if (splitAt) {
1029	size_t len;
1030	char *copy;
1031
1032	if (split_mode == 2) splitAt += sepLen; // split behind separator
1033
1034	len = splitAt-in;
1035	copy = ARB_strndup(in, len);
1036
1037	PASS_2_OUT(args, copy);
1038
1039	in = splitAt + (split_mode == 0 ? sepLen : 0);
1040	from = in+(split_mode == 1 ? sepLen : 0);
1041	}
1042	else {
1043	COPY_2_OUT(args, in); // last part
1044	in = NULp;
1045	}
1046	}
1047	}
1048	}
1049
1050	return NULp;
1051	}
1052
1053	// ----------------------------------
1054	// Extended string functions
1055
1056	static char do_extract_words(const char source, const char *chars, float minlen, bool sort_output) {
1057	/* extract all words in a text that:
1058	* if minlen < 1.0 -> contain more than minlen*len_of_text characters that also exists in chars
1059	* if minlen > 1.0 -> contain more than minlen characters that also exists in chars
1060	*/
1061
1062	char *s = ARB_strdup(source);
1063	char *ps = ARB_calloc<char>((strlen(source)>>1) + 1);
1064	GBS_strstruct *strstruct = GBS_stropen(1000);
1065	char *f = s;
1066	int count = 0;
1067	char *p; // @@@ fix locals
1068	char *h;
1069	int cnt;
1070	int len;
1071	int iminlen = (int) (minlen+.5);
1072
1073	while ((p = strtok(f, " \t,;:\|"))) {
1074	f = NULp;
1075	cnt = 0;
1076	len = strlen(p);
1077	for (h=p; *h; h++) {
1078	if (strchr(chars, *h)) cnt++;
1079	}
1080
1081	if (minlen == 1.0) {
1082	if (cnt != len) continue;
1083	}
1084	else if (minlen > 1.0) {
1085	if (cnt < iminlen) continue;
1086	}
1087	else {
1088	if (len < 3 \|\| cnt < minlen*len) continue;
1089	}
1090	ps[count] = p;
1091	count ++;
1092	}
1093	if (sort_output) {
1094	GB_sort((void **)ps, 0, count, GB_string_comparator, NULp);
1095	}
1096	for (cnt = 0; cnt<count; cnt++) {
1097	if (cnt) {
1098	GBS_chrcat(strstruct, ' ');
1099	}
1100	GBS_strcat(strstruct, ps[cnt]);
1101	}
1102
1103	free(ps);
1104	free(s);
1105	return GBS_strclose(strstruct);
1106	}
1107
1108	static GB_ERROR gbl_extract_words(GBL_command_arguments *args) {
1109	EXPECT_PARAMS(args, 2, "\"chars\", minchars");
1110
1111	float len = atof(args->get_param(1));
1112	for (int i=0; i<args->input.size(); i++) {
1113	char *res = do_extract_words(args->input.get(i), args->get_param(0), len, 1);
1114	gb_assert(res);
1115	PASS_2_OUT(args, res);
1116	}
1117	return NULp;
1118	}
1119
1120	static GB_ERROR gbl_extract_sequence(GBL_command_arguments *args) {
1121	EXPECT_PARAMS(args, 2, "\"chars\",minFrequency");
1122
1123	const char *chars = args->get_param(0);
1124	float minFreq = atof(args->get_param(1));
1125
1126	if (minFreq <0.0 \|\| minFreq > 1.0) return GBS_global_string("Illegal minFrequency=%f (allowed: ]0.0 .. 1.0[)", minFreq);
1127
1128	for (int i=0; i<args->input.size(); i++) {
1129	char *res = do_extract_words(args->input.get(i), chars, minFreq, 0);
1130	gb_assert(res);
1131	PASS_2_OUT(args, res);
1132	}
1133	return NULp;
1134	}
1135
1136	static GB_ERROR gbl_checksum(GBL_command_arguments *args) {
1137	GBL_BEGIN_PARAMS;
1138	GBL_PARAM_STRING(exclude, "exclude=", "", "Remove given characters before calculating");
1139	GBL_PARAM_BIT (upper, "toupper", 0, "Convert all characters to uppercase before calculating");
1140	GBL_TRACE_PARAMS(args);
1141	GBL_END_PARAMS;
1142
1143	for (int i=0; i<args->input.size(); i++) {
1144	long id = GBS_checksum(args->input.get(i), upper, exclude);
1145	FORMAT_2_OUT(args, "%lX", id);
1146	}
1147	return NULp;
1148	}
1149
1150	static GB_ERROR gbl_gcgchecksum(GBL_command_arguments *args) {
1151	EXPECT_NO_PARAM(args);
1152
1153	for (int i=0; i<args->input.size(); i++) {
1154	long id = GBS_gcgchecksum(args->input.get(i));
1155	FORMAT_2_OUT(args, "%li", id);
1156	}
1157	return NULp;
1158	}
1159
1160	// ------------
1161	// SRT
1162
1163	static GB_ERROR gbl_srt(GBL_command_arguments *args) {
1164	EXPECT_PARAMS_PASSED(args, "expr[,expr]+");
1165
1166	GB_ERROR error = NULp;
1167	for (int i=0; i<args->input.size() && !error; i++) {
1168	char *modsource = NULp;
1169
1170	for (int j=0; j<args->param_count() && !error; j++) {
1171	char *hs = GBS_string_eval_in_env(modsource ? modsource : args->input.get(i), args->get_param(j), args->get_callEnv());
1172
1173	if (hs) freeset(modsource, hs);
1174	else {
1175	error = GB_await_error();
1176	free(modsource);
1177	}
1178	}
1179
1180	if (!error) {
1181	if (modsource) PASS_2_OUT(args, modsource);
1182	else IN_2_OUT(args, i);
1183	}
1184	}
1185	return error;
1186	}
1187
1188	// -----------------------------
1189	// Calculator Functions
1190
1191	struct binop_pair {
1192	int (*INT) (int, int);
1193	double (*DOUBLE)(double, double);
1194	binop_pair(int (INT_)(int, int), double (DOUBLE_)(double, double)) : INT(INT_), DOUBLE(DOUBLE_) {}
1195	};
1196
1197	static char apply_numeric_binop(const char arg1, const char arg2, int (num_bin_op)(int,int)) {
1198	int v1 = atoi(arg1);
1199	int v2 = atoi(arg2);
1200	int result = num_bin_op(v1, v2);
1201
1202	return GBS_global_string_copy("%i", result);
1203	}
1204
1205	static char apply_double_binop(const char arg1, const char arg2, double (num_bin_op)(double,double)) {
1206	double v1 = strtod(arg1, NULp);
1207	double v2 = strtod(arg2, NULp);
1208	double result = num_bin_op(v1, v2);
1209
1210	return GBS_global_string_copy("%g", result);
1211	}
1212
1213	static char apply_auto_numeric_binop(const char arg1, const char *arg2, binop_pair multiop) {
1214	// argument type detection (int vs double)
1215	int i1 = atoi(arg1);
1216	int i2 = atoi(arg2);
1217	double d1 = strtod(arg1, NULp);
1218	double d2 = strtod(arg2, NULp);
1219
1220	if (double(i1) == d1 \|\| double(i2) == d2) {
1221	int result = multiop.INT(i1, i2);
1222	return GBS_global_string_copy("%i", result);
1223	}
1224	else {
1225	double result = multiop.DOUBLE(d1, d2);
1226	return GBS_global_string_copy("%g", result);
1227	}
1228	}
1229
1230
1231
1232	template <typename T> static T binop_plus (T v1, T v2) { return v1+v2; }
1233	template <typename T> static T binop_minus (T v1, T v2) { return v1-v2; }
1234	template <typename T> static T binop_mult (T v1, T v2) { return v1*v2; }
1235	template <typename T> static T binop_div (T v1, T v2) { return v2 ? v1/v2 : 0; }
1236	template <typename T> static T binop_per_cent(T v1, T v2) { return v2 ? (v1*100)/v2 : 0; }
1237
1238	static int binop_rest(int i1, int i2) { return i2 ? i1%i2 : 0; }
1239
1240
1241	static GB_ERROR gbl_plus (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_plus<int>); }
1242	static GB_ERROR gbl_fplus (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_double_binop, binop_plus<double>); }
1243	static GB_ERROR gbl_minus (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_minus<int>); }
1244	static GB_ERROR gbl_fminus (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_double_binop, binop_minus<double>); }
1245	static GB_ERROR gbl_mult (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_mult<int>); }
1246	static GB_ERROR gbl_fmult (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_double_binop, binop_mult<double>); }
1247	static GB_ERROR gbl_div (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_div<int>); }
1248	static GB_ERROR gbl_fdiv (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_double_binop, binop_div<double>); }
1249	static GB_ERROR gbl_rest (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_rest); }
1250	static GB_ERROR gbl_per_cent (GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_numeric_binop, binop_per_cent<int>); }
1251	static GB_ERROR gbl_fper_cent(GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_double_binop, binop_per_cent<double>); }
1252
1253	template <typename T> static T binop_isAbove(T i1, T i2) { return i1>i2; }
1254	template <typename T> static T binop_isBelow(T i1, T i2) { return i1<i2; }
1255	template <typename T> static T binop_isEqual(T i1, T i2) { return i1 == i2; }
1256
1257	static GB_ERROR gbl_isAbove(GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_auto_numeric_binop, binop_pair(binop_isAbove<int>, binop_isAbove<double>)); }
1258	static GB_ERROR gbl_isBelow(GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_auto_numeric_binop, binop_pair(binop_isBelow<int>, binop_isBelow<double>)); }
1259	static GB_ERROR gbl_isEqual(GBL_command_arguments *args) { return gbl_apply_binary_operator(args, apply_auto_numeric_binop, binop_pair(binop_isEqual<int>, binop_isEqual<double>)); }
1260
1261	inline double float_shift_factor(int digits) {
1262	if (digits<0) {
1263	return 1.0/float_shift_factor(-digits);
1264	}
1265	int factor = 1;
1266	while (digits>0) { // LOOP_VECTORIZED=* // @@@ vectorized only 2x with 4.9.3; 3x with 5.x
1267	factor *= 10;
1268	--digits;
1269	}
1270	return factor;
1271	}
1272
1273	static GB_ERROR gbl_round(GBL_command_arguments *args) {
1274	EXPECT_PARAMS(args, 1, "digits");
1275	int digits = atoi(args->get_param(0));
1276
1277	double factor = float_shift_factor(digits);
1278	for (int i=0; i<args->input.size(); ++i) {
1279	double val = strtod(args->input.get(i), NULp);
1280	val = round(val*factor)/factor;
1281	FORMAT_2_OUT(args, "%g", val);
1282	}
1283	return NULp;
1284	}
1285
1286
1287
1288	// boolean operators
1289
1290	static GB_ERROR gbl_not(GBL_command_arguments *args) {
1291	EXPECT_NO_PARAM(args);
1292
1293	for (int i=0; i<args->input.size(); ++i) {
1294	const char *s = args->input.get(i);
1295	int val = atoi(s);
1296	FORMAT_2_OUT(args, "%i", !val);
1297	}
1298	return NULp;
1299	}
1300
1301	static GB_ERROR gbl_and(GBL_command_arguments *args) {
1302	EXPECT_NO_PARAM(args);
1303	bool conjunction = true;
1304	for (int i=0; conjunction && i<args->input.size(); ++i) {
1305	const char *s = args->input.get(i);
1306	conjunction = conjunction && atoi(s);
1307	}
1308	FORMAT_2_OUT(args, "%i", conjunction);
1309	return NULp;
1310	}
1311	static GB_ERROR gbl_or(GBL_command_arguments *args) {
1312	EXPECT_NO_PARAM(args);
1313	bool disjunction = false;
1314	for (int i=0; !disjunction && i<args->input.size(); ++i) {
1315	const char *s = args->input.get(i);
1316	disjunction = disjunction \|\| atoi(s);
1317	}
1318	FORMAT_2_OUT(args, "%i", disjunction);
1319	return NULp;
1320	}
1321
1322	static GB_ERROR gbl_select(GBL_command_arguments *args) {
1323	ACCEPT_ANY_PARAMS(args);
1324
1325	GB_ERROR error = NULp;
1326	for (int i=0; i<args->input.size() && !error; i++) {
1327	int paraidx = atoi(args->input.get(i));
1328	error = check_valid_param_index(args, paraidx);
1329	if (!error) {
1330	char *result = args->get_callEnv().interpret_subcommand("", args->get_param(paraidx)); // @@@ EVALUATED_PARAM (#768)
1331	if (!result) error = GB_await_error();
1332	else PASS_2_OUT(args, result);
1333	}
1334	}
1335	return error;
1336	}
1337
1338	static GB_ERROR gbl_readdb(GBL_command_arguments *args) {
1339	EXPECT_PARAMS_PASSED(args, "fieldname[,fieldname]+");
1340	COMMAND_DROPS_INPUT_STREAMS(args);
1341
1342	GBS_strstruct *strstr = GBS_stropen(1024);
1343	for (int i=0; i<args->param_count(); i++) {
1344	char *val = GBT_read_as_string(args->get_ref(), args->get_param(i));
1345	if (val) {
1346	GBS_strcat(strstr, val);
1347	free(val);
1348	}
1349	}
1350	PASS_2_OUT(args, GBS_strclose(strstr));
1351	return NULp;
1352	}
1353
1354
1355	enum GBT_ITEM_TYPE {
1356	GBT_ITEM_UNKNOWN,
1357	GBT_ITEM_SPECIES,
1358	GBT_ITEM_GENE
1359	};
1360
1361	static GBT_ITEM_TYPE identify_gb_item(GBDATA *gb_item) {
1362	/* returns: GBT_ITEM_UNKNOWN -> unknown database_item
1363	* GBT_ITEM_SPECIES -> /species_data/species
1364	* GBT_ITEM_GENE -> /species_data/species/gene_data/gene */
1365
1366	GBT_ITEM_TYPE res = GBT_ITEM_UNKNOWN;
1367	if (gb_item) {
1368	GBDATA *gb_father = GB_get_father(gb_item);
1369	if (gb_father) {
1370	const char *key = GB_KEY(gb_item);
1371
1372	if (strcmp(key, "species") == 0 &&
1373	strcmp(GB_KEY(gb_father), "species_data") == 0) {
1374	res = GBT_ITEM_SPECIES;
1375	}
1376	else if (strcmp(key, "gene") == 0 &&
1377	strcmp(GB_KEY(gb_father), "gene_data") == 0 &&
1378	identify_gb_item(GB_get_father(gb_father)) == GBT_ITEM_SPECIES) {
1379	res = GBT_ITEM_GENE;
1380	}
1381	}
1382	}
1383	return res;
1384	}
1385
1386	// --------------------------------------------------------------------------------
1387	// taxonomy caching
1388
1389	#if defined(DEBUG)
1390	// #define DUMP_TAXONOMY_CACHING
1391	#endif
1392
1393
1394	#define GROUP_COUNT_CHARS 6 // characters in taxonomy-key reserved for group-counter (hex number)
1395	#define BITS_PER_HEXCHAR 4
1396	#define MAX_GROUPS (1 << (GROUP_COUNT_CHARS*BITS_PER_HEXCHAR)) // resulting number of groups
1397
1398	struct cached_taxonomy {
1399	char *tree_name; // tree for which taxonomy is cached here
1400	int groups; // number of named groups in tree (at time of caching)
1401	GB_HASH taxonomy; / keys: "!species", ">XXXXgroup" and "<root>".
1402	* Species and groups contain their first parent (i.e. '>XXXXgroup' or '<root>').
1403	* Species not in hash are not members of tree.
1404	* The 'XXXX' in groupname is simply a counter to avoid multiple groups with same name.
1405	* The group-db-entries are stored in hash as pointers ('>>%p') and
1406	* point to their own group entry ('>XXXXgroup')
1407	*
1408	* Note: the number of 'X's in 'XXXX' above is defined by GROUP_COUNT_CHARS!
1409	*/
1410	};
1411
1412	static void free_cached_taxonomy(cached_taxonomy *ct) {
1413	free(ct->tree_name);
1414	GBS_free_hash(ct->taxonomy);
1415	free(ct);
1416	}
1417
1418	static void build_taxonomy_rek(TreeNode node, GB_HASH tax_hash, const char parent_group, int group_counter) {
1419	if (node->is_leaf()) {
1420	GBDATA *gb_species = node->gb_node;
1421	if (gb_species) { // not zombie
1422	GBS_write_hash(tax_hash, GBS_global_string("!%s", GBT_read_name(gb_species)), (long)ARB_strdup(parent_group));
1423	}
1424	}
1425	else {
1426	if (node->has_group_info()) { // node with name
1427	char *hash_entry;
1428	const char *hash_binary_entry;
1429	(*group_counter)++;
1430
1431	gb_assert((*group_counter)<MAX_GROUPS); // overflow - increase GROUP_COUNT_CHARS
1432
1433	TreeNode *keelTarget = node->keelTarget();
1434
1435	hash_entry = GBS_global_string_copy(">%0*x%s%s",
1436	GROUP_COUNT_CHARS, *group_counter,
1437	keelTarget ? "!" : "",
1438	node->name);
1439	GBS_write_hash(tax_hash, hash_entry, (long)ARB_strdup(parent_group));
1440
1441	hash_binary_entry = GBS_global_string(">>%p", node->gb_node);
1442	GBS_write_hash(tax_hash, hash_binary_entry, (long)ARB_strdup(hash_entry));
1443
1444	if (keelTarget) { // keeled group (projected to son)
1445	if (keelTarget->is_leftson()) {
1446	build_taxonomy_rek(node->get_leftson(), tax_hash, hash_entry, group_counter); // pass down hash_entry only to keelTarget
1447	build_taxonomy_rek(node->get_rightson(), tax_hash, parent_group, group_counter);
1448	}
1449	else {
1450	build_taxonomy_rek(node->get_leftson(), tax_hash, parent_group, group_counter);
1451	build_taxonomy_rek(node->get_rightson(), tax_hash, hash_entry, group_counter);
1452	}
1453	}
1454	else { // normal group
1455	build_taxonomy_rek(node->get_leftson(), tax_hash, hash_entry, group_counter); // pass down hash_entry to both sons
1456	build_taxonomy_rek(node->get_rightson(), tax_hash, hash_entry, group_counter);
1457	}
1458
1459	free(hash_entry);
1460	}
1461	else {
1462	build_taxonomy_rek(node->get_leftson(), tax_hash, parent_group, group_counter);
1463	build_taxonomy_rek(node->get_rightson(), tax_hash, parent_group, group_counter);
1464	}
1465	}
1466	}
1467
1468	static GB_HASH *cached_taxonomies = NULp;
1469
1470	static bool is_cached_taxonomy(const char /key/, long val, void cl_ct) {
1471	cached_taxonomy ct1 = (cached_taxonomy )val;
1472	cached_taxonomy ct2 = (cached_taxonomy )cl_ct;
1473
1474	return ct1 == ct2;
1475	}
1476
1477	static const char tree_of_cached_taxonomy(cached_taxonomy ct) {
1478	/* search the hash to find the correct cached taxonomy.
1479	* searching for tree name does not work, because the tree possibly already was deleted
1480	*/
1481	const char *tree = GBS_hash_next_element_that(cached_taxonomies, NULp, is_cached_taxonomy, ct);
1482	#ifdef DUMP_TAXONOMY_CACHING
1483	if (tree) printf("tree_of_cached_taxonomy: tree='%s' ct->tree_name='%s'\n", tree, ct->tree_name);
1484	#endif // DUMP_TAXONOMY_CACHING
1485	return tree;
1486	}
1487
1488	static void flush_taxonomy_cb(GBDATA gbd, cached_taxonomy ct) {
1489	/* this cb is bound all tree db members below "/tree_data/tree_xxx" which
1490	* may have an effect on the displayed taxonomy
1491	* it invalidates cached taxonomies for that tree (when changed or deleted)
1492	*/
1493
1494	GB_ERROR error = NULp;
1495	const char *found = tree_of_cached_taxonomy(ct);
1496
1497	if (found) {
1498	#ifdef DUMP_TAXONOMY_CACHING
1499	fprintf(stderr, "Deleting cached taxonomy ct=%p (tree='%s')\n", ct, found);
1500	#endif // DUMP_TAXONOMY_CACHING
1501	GBS_write_hash(cached_taxonomies, found, 0); // delete cached taxonomy from hash
1502	free_cached_taxonomy(ct);
1503	}
1504	#ifdef DUMP_TAXONOMY_CACHING
1505	else {
1506	fprintf(stderr, "No tree found for cached_taxonomies ct=%p (already deleted?)\n", ct);
1507	}
1508	#endif // DUMP_TAXONOMY_CACHING
1509
1510	if (!GB_inside_callback(gbd, GB_CB_DELETE)) {
1511	GB_remove_all_callbacks_to(gbd, GB_CB_CHANGED_OR_DELETED, CASTSIG(GB_CB, flush_taxonomy_cb));
1512	}
1513
1514	if (found && !error) {
1515	GBDATA *gb_main = GB_get_gb_main_during_cb();
1516	if (gb_main) {
1517	GBDATA *gb_tree_refresh = GB_search(gb_main, AWAR_TREE_REFRESH, GB_INT);
1518	if (!gb_tree_refresh) {
1519	error = GBS_global_string("%s (while trying to force refresh)", GB_await_error());
1520	}
1521	else {
1522	GB_touch(gb_tree_refresh); // Note : force tree update
1523	}
1524	}
1525	}
1526
1527	if (error) {
1528	fprintf(stderr, "Error in flush_taxonomy_cb: %s\n", error);
1529	}
1530	}
1531
1532	static void flush_taxonomy_if_new_group_cb(GBDATA gb_tree, cached_taxonomy ct) {
1533	// detects the creation of new groups and call flush_taxonomy_cb() manually
1534	#ifdef DUMP_TAXONOMY_CACHING
1535	fputs("flush_taxonomy_if_new_group_cb() has been called\n", stderr);
1536	#endif // DUMP_TAXONOMY_CACHING
1537
1538	const char *tree_name = tree_of_cached_taxonomy(ct);
1539	if (tree_name) {
1540	int groups = 0;
1541	GBDATA *gb_group_node;
1542
1543	for (gb_group_node = GB_entry(gb_tree, "node");
1544	gb_group_node;
1545	gb_group_node = GB_nextEntry(gb_group_node))
1546	{
1547	if (GB_entry(gb_group_node, "group_name")) {
1548	groups++; // count named groups only
1549	}
1550	}
1551
1552	#ifdef DUMP_TAXONOMY_CACHING
1553	fprintf(stderr, "cached_groups=%i counted_groups=%i\n", ct->groups, groups);
1554	#endif // DUMP_TAXONOMY_CACHING
1555	if (groups != ct->groups) {
1556	#ifdef DUMP_TAXONOMY_CACHING
1557	fprintf(stderr, "Number of groups changed -> invoking flush_taxonomy_cb() manually\n");
1558	#endif // DUMP_TAXONOMY_CACHING
1559	flush_taxonomy_cb(gb_tree, ct);
1560	}
1561	}
1562	#ifdef DUMP_TAXONOMY_CACHING
1563	else {
1564	fprintf(stderr, "cached taxonomy no longer valid.\n");
1565	}
1566	#endif // DUMP_TAXONOMY_CACHING
1567	}
1568
1569	static cached_taxonomy get_cached_taxonomy(GBDATA gb_main, const char tree_name, GB_ERROR error) {
1570	long cached;
1571	*error = NULp;
1572	if (!cached_taxonomies) {
1573	cached_taxonomies = GBS_create_hash(20, GB_IGNORE_CASE);
1574	}
1575	cached = GBS_read_hash(cached_taxonomies, tree_name);
1576	if (!cached) {
1577	TreeNode *tree = GBT_read_tree(gb_main, tree_name, new SimpleRoot);
1578	if (!tree) *error = GB_await_error();
1579	else *error = GBT_link_tree(tree, gb_main, false, NULp, NULp);
1580
1581	if (!*error) {
1582	GBDATA *gb_tree = GBT_find_tree(gb_main, tree_name);
1583	if (!gb_tree) {
1584	*error = GBS_global_string("Can't find tree '%s'", tree_name);
1585	}
1586	else {
1587	cached_taxonomy *ct = ARB_alloc<cached_taxonomy>(1);
1588	long nodes = GBT_count_leafs(tree);
1589	int group_counter = 0;
1590
1591	ct->tree_name = ARB_strdup(tree_name);
1592	ct->taxonomy = GBS_create_dynaval_hash(int(nodes), GB_IGNORE_CASE, GBS_dynaval_free);
1593	ct->groups = 0; // counted below
1594
1595	build_taxonomy_rek(tree, ct->taxonomy, "<root>", &group_counter);
1596	cached = (long)ct;
1597	GBS_write_hash(cached_taxonomies, tree_name, (long)ct);
1598
1599	GB_remove_all_callbacks_to(gb_tree, GB_CB_SON_CREATED, CASTSIG(GB_CB, flush_taxonomy_if_new_group_cb));
1600	GB_add_callback(gb_tree, GB_CB_SON_CREATED, makeDatabaseCallback(flush_taxonomy_if_new_group_cb, ct));
1601
1602	{
1603	GBDATA *gb_tree_entry = GB_entry(gb_tree, "tree");
1604	GBDATA *gb_group_node;
1605
1606	if (gb_tree_entry) {
1607	GB_remove_all_callbacks_to(gb_tree_entry, GB_CB_CHANGED_OR_DELETED, CASTSIG(GB_CB, flush_taxonomy_cb));
1608	GB_add_callback(gb_tree_entry, GB_CB_CHANGED_OR_DELETED, makeDatabaseCallback(flush_taxonomy_cb, ct));
1609	}
1610
1611	// add callbacks for all node/group_name subentries
1612	for (gb_group_node = GB_entry(gb_tree, "node");
1613	gb_group_node;
1614	gb_group_node = GB_nextEntry(gb_group_node))
1615	{
1616	GBDATA *gb_group_name = GB_entry(gb_group_node, "group_name");
1617	if (gb_group_name) { // group with id = 0 has no name
1618	GB_remove_all_callbacks_to(gb_group_name, GB_CB_CHANGED_OR_DELETED, CASTSIG(GB_CB, flush_taxonomy_cb));
1619	GB_add_callback(gb_group_name, GB_CB_CHANGED_OR_DELETED, makeDatabaseCallback(flush_taxonomy_cb, ct));
1620	ct->groups++;
1621	}
1622	}
1623	}
1624	#ifdef DUMP_TAXONOMY_CACHING
1625	fprintf(stderr, "Created taxonomy hash for '%s' (ct=%p)\n", tree_name, ct);
1626	#endif // DUMP_TAXONOMY_CACHING
1627	}
1628	}
1629
1630	destroy(tree);
1631	}
1632
1633	if (!*error) {
1634	cached_taxonomy ct = (cached_taxonomy)cached;
1635	gb_assert(ct);
1636	return ct;
1637	}
1638
1639	return NULp;
1640	}
1641
1642	static char get_taxonomy_string(GB_HASH tax_hash, const char group_key, int depth, GB_ERROR error) {
1643	long found;
1644	char *result = NULp;
1645
1646	gb_assert(depth>0);
1647	gb_assert(!(group_key[0] == '>' && group_key[1] == '>')); // internal group-pointers not allowed here!
1648
1649	found = GBS_read_hash(tax_hash, group_key);
1650	if (found) {
1651	const char parent_group_key = (const char )found;
1652	if (strcmp(parent_group_key, "<root>") == 0) { // root reached
1653	result = ARB_strdup(group_key+(GROUP_COUNT_CHARS+1)); // return own group name
1654	}
1655	else {
1656	if (depth>1) {
1657	char *parent_name = get_taxonomy_string(tax_hash, parent_group_key, depth-1, error);
1658	if (parent_name) {
1659	result = GBS_global_string_copy("%s/%s", parent_name, group_key+(GROUP_COUNT_CHARS+1));
1660	free(parent_name);
1661	}
1662	else {
1663	error = GBS_global_string("In get_taxonomy_string(%s): %s", group_key, error);
1664	result = NULp;
1665	}
1666	}
1667	else {
1668	result = ARB_strdup(group_key+(GROUP_COUNT_CHARS+1)); // return own group name
1669	}
1670	}
1671	}
1672	else {
1673	*error = GBS_global_string("Not in tax_hash: '%s'", group_key);
1674	}
1675	return result;
1676	}
1677
1678	static const char get_taxonomy(GBDATA gb_species_or_group, const char tree_name, bool is_current_tree, int depth, GB_ERROR error) {
1679	GBDATA *gb_main = GB_get_root(gb_species_or_group);
1680	cached_taxonomy *tax = get_cached_taxonomy(gb_main, tree_name, error);
1681	const char *result = NULp;
1682
1683	if (tax) {
1684	GBDATA *gb_name = GB_entry(gb_species_or_group, "name");
1685	GBDATA *gb_group_name = GB_entry(gb_species_or_group, "group_name");
1686
1687	if (gb_name && !gb_group_name) { // it's a species
1688	char *name = GB_read_string(gb_name);
1689	if (name) {
1690	GB_HASH *tax_hash = tax->taxonomy;
1691	long found = GBS_read_hash(tax_hash, GBS_global_string("!%s", name));
1692
1693	if (found) {
1694	const char parent_group = (const char )found;
1695
1696	if (strcmp(parent_group, "<root>") == 0) {
1697	result = ""; // not member of any group
1698	}
1699	else {
1700	static char *parent = NULp;
1701
1702	freeset(parent, get_taxonomy_string(tax_hash, parent_group, depth, error));
1703	result = parent;
1704	}
1705	}
1706	else {
1707	result = GBS_global_string("Species '%s' not in '%s'", name, tree_name);
1708	}
1709	free(name);
1710	}
1711	else {
1712	*error = GBS_global_string("Species without 'name' entry!");
1713	}
1714	}
1715	else if (gb_group_name && !gb_name) { // it's a group
1716	char *group_name = GB_read_string(gb_group_name);
1717	if (group_name) {
1718	if (is_current_tree) {
1719	GB_HASH *tax_hash = tax->taxonomy;
1720	long found = GBS_read_hash(tax_hash, GBS_global_string(">>%p", gb_species_or_group));
1721
1722	if (found) {
1723	static char *full_group = NULp;
1724	const char group_id = (const char )found;
1725
1726	freeset(full_group, get_taxonomy_string(tax_hash, group_id, depth, error));
1727	result = full_group;
1728	}
1729	else {
1730	result = GBS_global_string("Group '%s' not in '%s'", group_name, tree_name);
1731	}
1732	}
1733	else {
1734	*error = "It's not possible to specify the tree name in taxonomy() for groups";
1735	}
1736	free(group_name);
1737	}
1738	else {
1739	*error = "Group without 'group_name' entry";
1740	}
1741	}
1742	else if (gb_group_name) {
1743	*error = "Container has 'name' and 'group_name' entry - can't detect container type";
1744	}
1745	else {
1746	*error = "Container has neither 'name' nor 'group_name' entry - can't detect container type";
1747	}
1748	}
1749
1750	return result;
1751	}
1752
1753	static GB_ERROR gbl_taxonomy(GBL_command_arguments *args) {
1754	GB_ERROR error = check_optional_parameters(args, 1, "count", 1, "tree_name", false, true);
1755	if (!error) {
1756	COMMAND_DROPS_INPUT_STREAMS(args);
1757
1758	char *tree_name = NULp;
1759	bool is_current_tree = false;
1760	int depth = -1;
1761	char *result = NULp;
1762
1763	if (args->param_count() == 1) { // only 'depth'
1764	if (!args->get_treename()) {
1765	result = ARB_strdup("No default tree");
1766	}
1767	else {
1768	tree_name = ARB_strdup(args->get_treename());
1769	depth = atoi(args->get_param(0));
1770	is_current_tree = true;
1771	}
1772	}
1773	else { // 'tree_name', 'depth'
1774	tree_name = ARB_strdup(args->get_param(0));
1775	depth = atoi(args->get_param(1));
1776	}
1777
1778	if (!result) {
1779	if (depth<1) {
1780	error = GBS_global_string("Illegal depth '%i' (allowed 1..n)", depth);
1781	}
1782	if (!error) {
1783	const char *taxonomy_string = get_taxonomy(args->get_ref(), tree_name, is_current_tree, depth, &error);
1784	if (taxonomy_string) result = ARB_strdup(taxonomy_string);
1785	}
1786	}
1787
1788	gb_assert(contradicted(result, error));
1789	if (result) PASS_2_OUT(args, result);
1790	free(tree_name);
1791	}
1792	return error;
1793	}
1794
1795	static GB_ERROR gbl_sequence(GBL_command_arguments *args) {
1796	COMMAND_DROPS_INPUT_STREAMS(args);
1797
1798	GB_ERROR error = check_no_parameter(args);
1799	if (!error) {
1800	switch (identify_gb_item(args->get_ref())) {
1801	case GBT_ITEM_UNKNOWN: {
1802	error = "'sequence' used for unknown item";
1803	break;
1804	}
1805	case GBT_ITEM_SPECIES: {
1806	char *use = GBT_get_default_alignment(args->get_gb_main());
1807
1808	if (!use) {
1809	error = GB_have_error() ? GB_await_error() : "no default alignment defined";
1810	}
1811	else {
1812	GBDATA *gb_seq = GBT_find_sequence(args->get_ref(), use);
1813
1814	if (gb_seq) PASS_2_OUT(args, GB_read_string(gb_seq));
1815	else COPY_2_OUT(args, ""); // if current alignment does not exist -> return empty string
1816
1817	free(use);
1818	}
1819	break;
1820	}
1821	case GBT_ITEM_GENE: {
1822	char *seq = GBT_read_gene_sequence(args->get_ref(), true, 0);
1823
1824	if (!seq) error = GB_await_error();
1825	else PASS_2_OUT(args, seq);
1826
1827	break;
1828	}
1829	}
1830	}
1831	return error;
1832	}
1833
1834	static GB_ERROR gbl_export_sequence(GBL_command_arguments *args) {
1835	COMMAND_DROPS_INPUT_STREAMS(args);
1836
1837	GB_ERROR error = check_no_parameter(args);
1838	if (!error) {
1839	switch (identify_gb_item(args->get_ref())) {
1840	case GBT_ITEM_UNKNOWN: {
1841	error = "'export_sequence' used for unknown item";
1842	break;
1843	}
1844	case GBT_ITEM_SPECIES: {
1845	if (!get_export_sequence) {
1846	error = "No export-sequence-hook defined (can't use 'export_sequence' here)";
1847	}
1848	else {
1849	size_t len;
1850	const char *seq = get_export_sequence(args->get_ref(), &len, &error);
1851
1852	gb_assert(error \|\| seq);
1853
1854	if (seq) PASS_2_OUT(args, ARB_strduplen(seq, len));
1855	}
1856	break;
1857	}
1858	case GBT_ITEM_GENE: {
1859	error = "'export_sequence' cannot be used for gene";
1860	break;
1861	}
1862	}
1863	}
1864	return error;
1865	}
1866
1867	static GB_ERROR gbl_ali_name(GBL_command_arguments *args) {
1868	COMMAND_DROPS_INPUT_STREAMS(args);
1869
1870	GB_ERROR error = check_no_parameter(args);
1871	if (!error) {
1872	GBDATA *gb_main = args->get_gb_main();
1873	char *use = GBT_get_default_alignment(gb_main);
1874	PASS_2_OUT(args, use);
1875	}
1876	return error;
1877	}
1878
1879	static GB_ERROR gbl_sequence_type(GBL_command_arguments *args) {
1880	COMMAND_DROPS_INPUT_STREAMS(args);
1881
1882	GB_ERROR error = check_no_parameter(args);
1883	if (!error) {
1884	GBDATA *gb_main = args->get_gb_main();
1885	char *use = GBT_get_default_alignment(gb_main);
1886	PASS_2_OUT(args, GBT_get_alignment_type_string(gb_main, use));
1887	free(use);
1888	}
1889
1890	return error;
1891	}
1892
1893	static GB_ERROR format(GBL_command_arguments *args, bool simple_format) {
1894	// simple_format: true = "format", false="format_sequence"
1895
1896	GB_ERROR error = NULp;
1897	int ic;
1898
1899	GBL_BEGIN_PARAMS;
1900	GBL_PARAM_INT(firsttab, "firsttab=", 10, "Indent first line");
1901	GBL_PARAM_INT(tab, "tab=", 10, "Indent not first line");
1902	GBL_PARAM_INT(width, "width=", 50, "Sequence width (bases only)");
1903
1904	// "format_sequence"-only
1905	GBL_PARAM_BIT (numleft, PARAM_IF(!simple_format, "numleft"), 0, "Numbers left of sequence");
1906	GBL_PARAM_INT (numright, PARAM_IF(!simple_format, "numright="), 0, "Numbers right of sequence (specifies width; -1 -> auto-width)");
1907	GBL_PARAM_UINT(gap, PARAM_IF(!simple_format, "gap="), 10, "Insert ' ' every n sequence characters");
1908
1909	// "format"-only
1910	GBL_PARAM_STRING(nl, PARAM_IF(simple_format, "nl="), " ", "Break line at characters 'str' if wrapping needed");
1911	GBL_PARAM_STRING(forcenl, PARAM_IF(simple_format, "forcenl="), "\n", "Always break line at characters 'str'");
1912
1913	GBL_TRACE_PARAMS(args);
1914	GBL_END_PARAMS;
1915
1916	if (tab < 0) tab = 0;
1917	if (firsttab < 0) firsttab = 0;
1918
1919	if (width == 0) return "Illegal zero width";
1920	if (numleft && numright != 0) return "You may only specify 'numleft' OR 'numright', not both.";
1921
1922	if (gap<1) gap = UINT_MAX;
1923
1924	for (ic = 0; ic<args->input.size(); ++ic) {
1925	const char *src = args->input.get(ic);
1926	size_t data_size = strlen(src);
1927	size_t needed_size;
1928	size_t line_size;
1929	int numright_used = numright;
1930
1931	if (numright_used<0) {
1932	numright_used = calc_digits(data_size);
1933	}
1934
1935	{
1936	size_t lines;
1937
1938	if (simple_format) {
1939	lines = data_size/2 + 1; // worst case
1940	line_size = tab + (width>0 ? width : data_size) + 1;
1941	}
1942	else {
1943	size_t gapsPerLine = (width-1)/gap;
1944	lines = data_size/width+1;
1945	line_size = tab + width + gapsPerLine + 1;
1946
1947	if (numright_used) {
1948	// add space for numright
1949	line_size += numright_used+1; // plus space
1950	}
1951	}
1952
1953	needed_size = lines*line_size + firsttab + 1 + 10;
1954	}
1955
1956	char *result = ARB_alloc<char>(needed_size);
1957	if (!result) {
1958	error = GBS_global_string("Out of memory (tried to alloc %zu bytes)", needed_size);
1959	}
1960	else {
1961	char *dst = result;
1962	size_t rest_data = data_size;
1963
1964	if (simple_format) {
1965	/* format string w/o gaps or numleft
1966	* does word-wrapping at chars in nl
1967	*/
1968
1969	// build wrap table
1970	unsigned char isWrapChar[256];
1971	memset(isWrapChar, 0, sizeof(isWrapChar));
1972	for (int i = 0; nl[i]; ++i) isWrapChar[(unsigned char)nl[i]] = 1;
1973	for (int i = 0; forcenl[i]; ++i) isWrapChar[(unsigned char)forcenl[i]] = 2;
1974
1975	if (firsttab>0) {
1976	memset(dst, ' ', firsttab);
1977	dst += firsttab;
1978	}
1979
1980	while (width>0 && rest_data>unsigned(width)) {
1981	int take;
1982	int move;
1983	int took;
1984
1985	for (take = width; take > 0; --take) {
1986	if (isWrapChar[(unsigned char)src[take]]) break;
1987	}
1988	if (take <= 0) { // no wrap character found -> hard wrap at width
1989	take = move = width;
1990	}
1991	else { // soft wrap at last found wrap character
1992	move = take+1;
1993	}
1994
1995	for (took = 0; took<take; took++) {
1996	char c = src[took];
1997	if (isWrapChar[(unsigned char)c] == 2) { // forced newline
1998	take = took;
1999	move = take+1;
2000	break;
2001	}
2002	dst[took] = c;
2003	}
2004
2005	dst += take;
2006	src += move;
2007	rest_data -= move;
2008
2009	if (rest_data>0) {
2010	*dst++ = '\n';
2011	if (tab>0) {
2012	memset(dst, ' ', tab);
2013	dst += tab;
2014	}
2015	}
2016	}
2017
2018	if (rest_data>0) {
2019	size_t j, k;
2020	for (j = 0, k = 0; j<rest_data; ++j) {
2021	char c = src[j];
2022
2023	if (isWrapChar[(unsigned char)c] == 2) {
2024	dst[k++] = '\n';
2025	if (tab>0) {
2026	memset(dst+k, ' ', tab);
2027	k += tab;
2028	}
2029	}
2030	else {
2031	dst[k++] = c;
2032	}
2033	}
2034	src += j;
2035	dst += k;
2036	rest_data = 0;
2037	}
2038	}
2039	else {
2040	// "format_sequence" with gaps and numleft
2041	char *format = NULp;
2042	const char *src_start = src;
2043	const char *dst_linestart = dst;
2044
2045	if (numleft) {
2046	/* Warning: Be very careful, when you change format strings here!
2047	* currently all format strings result in '%u' or '%-##u' (where # are digits)
2048	*/
2049	if (firsttab>0) {
2050	char *firstFormat = GBS_global_string_copy("%%-%iu ", firsttab-1);
2051	dst += sprintf(dst, firstFormat, (unsigned)1);
2052	free(firstFormat);
2053	}
2054	else {
2055	dst += sprintf(dst, "%u ", (unsigned)1);
2056	}
2057	format = tab>0 ? GBS_global_string_copy("%%-%iu ", tab-1) : ARB_strdup("%u ");
2058	}
2059	else if (firsttab>0) {
2060	memset(dst, ' ', firsttab);
2061	dst += firsttab;
2062	}
2063
2064	while (rest_data>0) {
2065	size_t take = (width>0 && rest_data>unsigned(width)) ? width : rest_data;
2066
2067	rest_data -= take;
2068
2069	while (take>gap) {
2070	memcpy(dst, src, gap);
2071	dst += gap;
2072	src += gap;
2073	*dst++ = ' ';
2074	take -= gap;
2075	}
2076
2077	memcpy(dst, src, take);
2078	dst += take;
2079	src += take;
2080
2081	if (numright_used) {
2082	if (rest_data) *dst++ = ' ';
2083	else {
2084	// fill in missing spaces for proper alignment of numright
2085	size_t currSize = dst-dst_linestart;
2086	size_t wantSize = line_size-numright_used-1;
2087	if (currSize<wantSize) {
2088	size_t spaces = wantSize-currSize;
2089	memset(dst, ' ', spaces);
2090	dst += spaces;
2091	}
2092	}
2093	unsigned int num = (src-src_start);
2094	dst += sprintf(dst, "%*u", numright_used, num);
2095	}
2096
2097	if (rest_data>0) {
2098	*dst++ = '\n';
2099	dst_linestart = dst;
2100	if (numleft) {
2101	unsigned int num = (src-src_start)+1; // this goes to the '%u' (see comment above)
2102	dst += sprintf(dst, format, num);
2103	}
2104	else if (tab>0) {
2105	memset(dst, ' ', tab);
2106	dst += tab;
2107	}
2108	}
2109	}
2110
2111	free(format);
2112	}
2113
2114	*dst++ = 0; // close str
2115
2116	#if defined(DEBUG)
2117	{ // check for array overflow
2118	size_t used_size = dst-result;
2119	gb_assert(used_size <= needed_size);
2120	ARB_realloc(result, used_size);
2121	}
2122	#endif // DEBUG
2123	}
2124
2125	if (!error) PASS_2_OUT(args, result);
2126	else free(result);
2127	}
2128	return error;
2129	}
2130
2131	static GB_ERROR gbl_format (GBL_command_arguments *args) { return format(args, true); }
2132	static GB_ERROR gbl_format_sequence(GBL_command_arguments *args) { return format(args, false); }
2133
2134
2135	static char gbl_read_seq_sai_or_species(GBDATA gb_main, const char species, const char sai, const char ali, size_t seqLen) {
2136	/* Reads the alignment 'ali' of 'species' or 'sai'.
2137	* If 'ali' is NULp, use default alignment.
2138	* Returns NULp in case of error (which is exported then)
2139	*/
2140
2141	char *seq = NULp;
2142	GB_ERROR error = NULp;
2143
2144	int sources = !!species + !!sai;
2145	if (sources != 1) {
2146	error = "Either parameters 'species' or 'SAI' must be specified";
2147	}
2148	else {
2149	GBDATA *gb_item = NULp;
2150	const char *what = NULp;
2151	const char *name = NULp;
2152
2153	if (species) {
2154	gb_item = GBT_find_species(gb_main, species);
2155	what = "species";
2156	name = species;
2157	}
2158	else {
2159	gb_item = GBT_find_SAI(gb_main, sai);
2160	what = "SAI";
2161	name = sai;
2162	}
2163
2164	if (!gb_item) error = GBS_global_string("Can't find %s '%s'", what, name);
2165	else {
2166	char *freeMe = NULp;
2167
2168	if (!ali) {
2169	ali = freeMe = GBT_get_default_alignment(gb_main);
2170	if (!ali) error = "can't detect default alignment";
2171	}
2172
2173	if (ali) {
2174	GBDATA *gb_ali = GB_entry(gb_item, ali);
2175
2176	if (gb_ali) {
2177	GBDATA *gb_seq;
2178
2179	for (gb_seq = GB_child(gb_ali); gb_seq; gb_seq = GB_nextChild(gb_seq)) {
2180	long type = GB_read_type(gb_seq);
2181	if (type == GB_BITS) {
2182	seq = GB_read_bits(gb_seq, '-', '+');
2183	if (seqLen) *seqLen = GB_read_bits_count(gb_seq);
2184	break;
2185	}
2186	if (type == GB_STRING) {
2187	seq = GB_read_string(gb_seq);
2188	if (seqLen) *seqLen = GB_read_string_count(gb_seq);
2189	break;
2190	}
2191	}
2192	}
2193
2194	if (!seq) error = GBS_global_string("%s '%s' has no (usable) data in alignment '%s'", what, name, ali);
2195	}
2196	free(freeMe);
2197	}
2198	}
2199
2200	if (error) {
2201	gb_assert(!seq);
2202	GB_export_error(error);
2203	}
2204
2205	return seq;
2206	}
2207
2208	struct common_filter_params {
2209	const char *align;
2210	const char *sai;
2211	const char *species;
2212	int first;
2213	int pairwise;
2214	};
2215
2216	#define GBL_COMMON_FILTER_PARAMS \
2217	common_filter_params common_param; \
2218	GBL_STRUCT_PARAM_STRING(common_param, align, "align=", NULp, "alignment to use (defaults to default alignment)"); \
2219	GBL_STRUCT_PARAM_STRING(common_param, sai, "SAI=", NULp, "Use default sequence of given SAI as a filter"); \
2220	GBL_STRUCT_PARAM_STRING(common_param, species, "species=", NULp, "Use default sequence of given species as a filter"); \
2221	GBL_STRUCT_PARAM_BIT (common_param, first, "first=", 0, "Use 1st stream as filter for other streams"); \
2222	GBL_STRUCT_PARAM_BIT (common_param, pairwise, "pairwise=", 0, "Use 1st stream as filter for 2nd, 3rd for 4th, ...")
2223
2224	typedef char* (filter_fun)(const char seq, const char filter, size_t flen, void param);
2225	/* Note:
2226	* filter_fun has to return a heap copy of the filter-result.
2227	* if 'flen' != 0, it contains the length of 'filter'
2228	* 'param' may be any client data
2229	*/
2230
2231	static GB_ERROR apply_filters(GBL_command_arguments args, common_filter_params common, filter_fun filter_one, void *param) {
2232	GB_ERROR error = NULp;
2233
2234	if (args->input.size()==0) error = "No input stream";
2235	else {
2236	int methodCount = !!common->sai + !!common->species + !!common->pairwise + !!common->first;
2237
2238	if (methodCount != 1) error = "Need exactly one of the parameters 'SAI', 'species', 'pairwise' or 'first'";
2239	else {
2240	if (common->pairwise) {
2241	if (args->input.size() % 2) error = "Using 'pairwise' requires an even number of input streams";
2242	else {
2243	int i;
2244	for (i = 1; i<args->input.size(); i += 2) {
2245	PASS_2_OUT(args, filter_one(args->input.get(i), args->input.get(i-1), 0, param));
2246	}
2247	}
2248	}
2249	else {
2250	int i = 0;
2251	char *filter = NULp;
2252	size_t flen = 0;
2253
2254	if (common->first) {
2255	if (args->input.size()<2) error = "Using 'first' needs at least 2 input streams";
2256	else {
2257	const char *in = args->input.get(i++);
2258	gb_assert(in);
2259
2260	flen = strlen(in);
2261	filter = ARB_strduplen(in, flen);
2262	}
2263	}
2264	else {
2265	filter = gbl_read_seq_sai_or_species(args->get_gb_main(), common->species, common->sai, common->align, &flen);
2266	if (!filter) error = GB_await_error();
2267	}
2268
2269	gb_assert(filter \|\| error);
2270	if (filter) {
2271	for (; i<args->input.size(); ++i) {
2272	PASS_2_OUT(args, filter_one(args->input.get(i), filter, flen, param));
2273	}
2274	}
2275	free(filter);
2276	}
2277	}
2278	}
2279	return error;
2280	}
2281
2282	// -------------------------
2283	// calculate diff
2284
2285	struct diff_params {
2286	char equalC;
2287	char diffC;
2288	};
2289	static char calc_diff(const char seq, const char filter, size_t /flen/, void paramP) {
2290	// filters 'seq' through 'filter'
2291	// - replace all equal positions by 'equal_char' (if != 0)
2292	// - replace all differing positions by 'diff_char' (if != 0)
2293
2294	diff_params param = (diff_params)paramP;
2295	char equal_char = param->equalC;
2296	char diff_char = param->diffC;
2297
2298	char *result = ARB_strdup(seq);
2299	int p;
2300
2301	for (p = 0; result[p] && filter[p]; ++p) {
2302	if (result[p] == filter[p]) {
2303	if (equal_char) result[p] = equal_char;
2304	}
2305	else {
2306	if (diff_char) result[p] = diff_char;
2307	}
2308	}
2309
2310	// if 'seq' is longer than 'filter' and diff_char is given
2311	// -> fill rest of 'result' with 'diff_char'
2312	if (diff_char) {
2313	for (; result[p]; ++p) {
2314	result[p] = diff_char;
2315	}
2316	}
2317
2318	return result;
2319	}
2320	static GB_ERROR gbl_diff(GBL_command_arguments *args) {
2321	GBL_BEGIN_PARAMS;
2322	GBL_COMMON_FILTER_PARAMS;
2323
2324	diff_params param;
2325	GBL_STRUCT_PARAM_CHAR(param, equalC, "equal=", '.', "symbol for equal characters");
2326	GBL_STRUCT_PARAM_CHAR(param, diffC, "differ=", 0, "symbol for diff characters (default: use char from input stream)");
2327
2328	GBL_TRACE_PARAMS(args);
2329	GBL_END_PARAMS;
2330
2331	return apply_filters(args, &common_param, calc_diff, &param);
2332	}
2333
2334	// -------------------------
2335	// standard filter
2336
2337	enum filter_function { FP_FILTER, FP_MODIFY };
2338
2339	struct filter_params { // used by gbl_filter and gbl_change_gc
2340	filter_function function;
2341
2342	const char *include;
2343	const char *exclude;
2344
2345	// FP_MODIFY only:
2346	int change_pc;
2347	const char *change_to;
2348	};
2349
2350	static char filter_seq(const char seq, const char filter, size_t flen, void paramP) {
2351	filter_params param = (filter_params)paramP;
2352
2353	size_t slen = strlen(seq);
2354	if (!flen) flen = strlen(filter);
2355	size_t mlen = slen<flen ? slen : flen;
2356
2357	GBS_strstruct *out = GBS_stropen(mlen+1); // +1 to avoid invalid, zero-length buffer
2358
2359	const char *charset;
2360	int include;
2361
2362	if (param->include) {
2363	charset = param->include;
2364	include = 1;
2365	}
2366	else {
2367	gb_assert(param->exclude);
2368	charset = param->exclude;
2369	include = 0;
2370	}
2371
2372	size_t pos = 0;
2373	size_t rest = slen;
2374	size_t ctl = 0;
2375	if (param->function == FP_MODIFY) ctl = strlen(param->change_to);
2376
2377	int inset = 1; // 1 -> check chars in charset, 0 -> check chars NOT in charset
2378	while (rest) {
2379	size_t count;
2380	if (pos >= flen) { // behind filter
2381	// trigger last loop
2382	count = rest;
2383	inset = 0; // if 'include' -> 'applies' will get false, otherwise true
2384	// (meaning is: behind filter nothing can match 'include' or 'exclude')
2385	}
2386	else {
2387	count = (inset ? strspn : strcspn)(filter+pos, charset); // count how many chars are 'inset'
2388	}
2389	if (count) {
2390	int applies = !!include == !!inset; // true -> 'filter' matches 'include' or doesn't match 'exclude'
2391	if (count>rest) count = rest;
2392
2393	switch (param->function) {
2394	case FP_FILTER:
2395	if (applies) GBS_strncat(out, seq+pos, count);
2396	break;
2397
2398	case FP_MODIFY:
2399	if (applies) { // then modify
2400	size_t i;
2401	for (i = 0; i<count; i++) {
2402	char c = seq[pos+i];
2403	if (isalpha(c) && GB_random(100)<param->change_pc) c = param->change_to[GB_random(ctl)];
2404	GBS_chrcat(out, c);
2405	}
2406	}
2407	else { // otherwise simply copy
2408	GBS_strncat(out, seq+pos, count);
2409	}
2410	break;
2411	}
2412
2413	pos += count;
2414	rest -= count;
2415	}
2416	inset = 1-inset; // toggle
2417	}
2418	return GBS_strclose(out);
2419	}
2420
2421	static GB_ERROR gbl_filter(GBL_command_arguments *args) {
2422	GBL_BEGIN_PARAMS;
2423	GBL_COMMON_FILTER_PARAMS;
2424
2425	filter_params param;
2426	GBL_STRUCT_PARAM_STRING(param, exclude, "exclude=", NULp, "Exclude colums");
2427	GBL_STRUCT_PARAM_STRING(param, include, "include=", NULp, "Include colums");
2428	param.function = FP_FILTER;
2429
2430	GBL_TRACE_PARAMS(args);
2431	GBL_END_PARAMS;
2432
2433	GB_ERROR error = NULp;
2434	int inOrEx = !!param.include + !!param.exclude;
2435
2436	if (inOrEx != 1) error = "Need exactly one parameter of: 'include', 'exclude'";
2437	else error = apply_filters(args, &common_param, filter_seq, &param);
2438
2439	return error;
2440	}
2441
2442	static GB_ERROR gbl_change_gc(GBL_command_arguments *args) {
2443	GBL_BEGIN_PARAMS;
2444	GBL_COMMON_FILTER_PARAMS;
2445
2446	filter_params param;
2447	GBL_STRUCT_PARAM_STRING(param, exclude, "exclude=", NULp, "Exclude colums");
2448	GBL_STRUCT_PARAM_STRING(param, include, "include=", NULp, "Include colums");
2449	GBL_STRUCT_PARAM_INT (param, change_pc, "change=", 0, "percentage of changed columns (default: silently change nothing)");
2450	GBL_STRUCT_PARAM_STRING(param, change_to, "to=", "GC", "change to one of this");
2451	param.function = FP_MODIFY;
2452
2453	GBL_TRACE_PARAMS(args);
2454	GBL_END_PARAMS;
2455
2456	GB_ERROR error = NULp;
2457	int inOrEx = !!param.include + !!param.exclude;
2458
2459	if (inOrEx != 1) error = "Need exactly one parameter of: 'include', 'exclude'";
2460	else {
2461	error = apply_filters(args, &common_param, filter_seq, &param);
2462	}
2463
2464	return error;
2465	}
2466
2467	static GB_ERROR gbl_exec(GBL_command_arguments *args) {
2468	EXPECT_PARAMS_PASSED(args, "command[,arguments]+");
2469
2470	// write inputstreams to temp file:
2471	GB_ERROR error = NULp;
2472	char *inputname;
2473	int i;
2474	{
2475	char *filename = GB_unique_filename("arb_exec_input", "tmp");
2476	FILE *out = GB_fopen_tempfile(filename, "wt", &inputname);
2477
2478	if (!out) error = GB_await_error();
2479	else {
2480	for (i=0; i<args->input.size(); i++) {
2481	fprintf(out, "%s\n", args->input.get(i));
2482	}
2483	fclose(out);
2484	}
2485	free(filename);
2486	}
2487
2488	if (!error) {
2489	// build shell command to execute
2490	char *sys;
2491	{
2492	GBS_strstruct *str = GBS_stropen(1000);
2493
2494	GBS_strcat(str, args->get_param(0));
2495	for (i=1; i<args->param_count(); i++) {
2496	GBS_strcat(str, " \'");
2497	GBS_strcat(str, args->get_param(i)); // @@@ use GBK_singlequote here?
2498	GBS_chrcat(str, '\'');
2499	}
2500	GBS_strcat(str, " <");
2501	GBS_strcat(str, inputname);
2502
2503	sys = GBS_strclose(str);
2504	}
2505
2506	char *result = NULp;
2507	{
2508	FILE *in = popen(sys, "r");
2509	if (in) {
2510	GBS_strstruct *str = GBS_stropen(4096);
2511
2512	while ((i=getc(in)) != EOF) { GBS_chrcat(str, i); }
2513	result = GBS_strclose(str);
2514	pclose(in);
2515	}
2516	else {
2517	error = GBS_global_string("Cannot execute shell command '%s'", sys);
2518	}
2519	}
2520
2521	if (!error) {
2522	gb_assert(result);
2523	PASS_2_OUT(args, result);
2524	}
2525
2526	free(sys);
2527	}
2528
2529	gb_assert(GB_is_privatefile(inputname, false));
2530	GB_unlink_or_warn(inputname, &error);
2531	free(inputname);
2532
2533	return error;
2534	}
2535
2536
2537	static GBL_command_definition gbl_command_table[] = {
2538	{ "ali_name", gbl_ali_name },
2539	{ "caps", gbl_caps },
2540	{ "change", gbl_change_gc },
2541	{ "checksum", gbl_checksum },
2542	{ "command", gbl_command },
2543	{ "compare", gbl_compare },
2544	{ "icompare", gbl_icompare },
2545	{ "contains", gbl_contains },
2546	{ "icontains", gbl_icontains },
2547	{ "count", gbl_count },
2548	{ "crop", gbl_crop },
2549	{ "cut", gbl_cut },
2550	{ "dd", gbl_dd },
2551	{ "define", gbl_define },
2552	{ "diff", gbl_diff },
2553	{ "div", gbl_div },
2554	{ "fdiv", gbl_fdiv },
2555	{ "do", gbl_do },
2556	{ "drop", gbl_drop },
2557	{ "dropempty", gbl_dropempty },
2558	{ "dropzero", gbl_dropzero },
2559	{ "echo", gbl_echo },
2560	{ "equals", gbl_equals },
2561	{ "iequals", gbl_iequals },
2562	{ "escape", gbl_escape },
2563	{ "unescape", gbl_unescape },
2564	{ "eval", gbl_eval },
2565	{ "exec", gbl_exec },
2566	{ "export_sequence", gbl_export_sequence },
2567	{ "extract_sequence", gbl_extract_sequence },
2568	{ "extract_words", gbl_extract_words },
2569	{ "filter", gbl_filter },
2570	{ "format", gbl_format },
2571	{ "format_sequence", gbl_format_sequence },
2572	{ "gcgchecksum", gbl_gcgchecksum },
2573	{ "head", gbl_head },
2574	{ "inRange", gbl_inRange },
2575	{ "isAbove", gbl_isAbove },
2576	{ "isBelow", gbl_isBelow },
2577	{ "isEqual", gbl_isEqual },
2578	{ "isEmpty", gbl_isEmpty },
2579	{ "keep", gbl_keep },
2580	{ "left", gbl_head },
2581	{ "len", gbl_len },
2582	{ "lower", gbl_lower },
2583	{ "merge", gbl_merge },
2584	{ "mid", gbl_mid },
2585	{ "mid0", gbl_mid0 },
2586	{ "minus", gbl_minus },
2587	{ "fminus", gbl_fminus },
2588	{ "mult", gbl_mult },
2589	{ "fmult", gbl_fmult },
2590	{ "and", gbl_and },
2591	{ "or", gbl_or },
2592	{ "not", gbl_not },
2593	{ "origin_gene", gbl_origin_gene },
2594	{ "origin_organism", gbl_origin_organism },
2595	{ "partof", gbl_partof },
2596	{ "ipartof", gbl_ipartof },
2597	{ "per_cent", gbl_per_cent },
2598	{ "fper_cent", gbl_fper_cent },
2599	{ "plus", gbl_plus },
2600	{ "fplus", gbl_fplus },
2601	{ "pretab", gbl_pretab },
2602	{ "quote", gbl_quote },
2603	{ "unquote", gbl_unquote },
2604	{ "readdb", gbl_readdb },
2605	{ "remove", gbl_remove },
2606	{ "rest", gbl_rest },
2607	{ "right", gbl_tail },
2608	{ "round", gbl_round },
2609	{ "select", gbl_select },
2610	{ "sequence", gbl_sequence },
2611	{ "sequence_type", gbl_sequence_type },
2612	{ "split", gbl_split },
2613	{ "srt", gbl_srt },
2614	{ "streams", gbl_streams },
2615	{ "swap", gbl_swap },
2616	{ "tab", gbl_tab },
2617	{ "tail", gbl_tail },
2618	{ "taxonomy", gbl_taxonomy },
2619	{ "toback", gbl_toback },
2620	{ "tofront", gbl_tofront },
2621	{ "trace", gbl_trace },
2622	{ "translate", gbl_translate },
2623	{ "upper", gbl_upper },
2624
2625	{ NULp, NULp }
2626	};
2627
2628	const GBL_command_lookup_table& ACI_get_standard_commands() {
2629	static GBL_command_lookup_table clt(gbl_command_table, ARRAY_ELEMS(gbl_command_table)-1);
2630	return clt;
2631	}
2632

Note: See TracBrowser for help on using the repository browser.

Download in other formats: