/*
* ------------------------------------------------------------------------
*
* "rcqpCmd_attribute_size(SEXP inAttribute)" --
*
*
*
* ------------------------------------------------------------------------
*/
SEXP rcqpCmd_attribute_size(SEXP inAttribute)
{
SEXP result = R_NilValue;
char * a;
Attribute * attribute;
int size;
int found = 0;
if (!isString(inAttribute) || length(inAttribute) != 1) error("argument 'attribute' must be a string");
PROTECT(inAttribute);
a = (char*)CHAR(STRING_ELT(inAttribute,0));
/* Need to try all possible attribute types */
attribute = cqi_lookup_attribute(a, ATT_POS);
if (attribute != NULL) {
size = cl_max_cpos(attribute);
if (size < 0) {
UNPROTECT(1);
rcqp_send_error();
} else {
found = 1;
}
} else {
attribute = cqi_lookup_attribute(a, ATT_STRUC);
if (attribute != NULL) {
size = cl_max_struc(attribute);
if (size < 0) {
size = 0;
} else {
found = 1;
}
} else {
attribute = cqi_lookup_attribute(a, ATT_ALIGN);
if (attribute != NULL) {
size = cl_max_alg(attribute);
if (size < 0) {
UNPROTECT(1);
rcqp_send_error();
} else {
found = 1;
}
} else {
UNPROTECT(1);
rcqp_error_code(cqi_errno);
}
}
}
if (found) {
result = PROTECT(allocVector(INTSXP, 1));
INTEGER(result)[0] = size;
}
UNPROTECT(2);
return result;
}
void
do_cqi_cl_attribute_size(void)
{
char *a;
Attribute *attribute;
int size;
a = cqi_read_string(); /* need to try all possible attribute types */
if (server_debug)
Rprintf( "CQi: CQI_CL_ATTRIBUTE_SIZE('%s')\n", a);
attribute = cqi_lookup_attribute(a, ATT_POS);
if (attribute != NULL) {
size = cl_max_cpos(attribute);
if (size < 0) {
send_cl_error();
}
else {
cqi_data_int(size);
}
}
else {
attribute = cqi_lookup_attribute(a, ATT_STRUC);
if (attribute != NULL) {
size = cl_max_struc(attribute);
if (size < 0) {
/* send_cl_error(); */
/* current version of CL considers 0 regions a data access error condition, but we want to allow that */
cqi_data_int(0);
}
else {
cqi_data_int(size);
}
}
else {
attribute = cqi_lookup_attribute(a, ATT_ALIGN);
if (attribute != NULL) {
size = cl_max_alg(attribute);
if (size < 0) {
send_cl_error();
}
else {
cqi_data_int(size);
}
}
else {
cqi_command(cqi_errno); /* return errno from the last lookup */
}
}
}
free(a);
}
/**
* Prints statistical information about a corpus to STDOUT.
*
* Each corpus attribute gets info printed about it:
* tokens and types for a P-attribute, number of instances
* of regions for an S-attribute, number of alignment
* blocks for an A-attribute.
*
* @param corpus The corpus to analyse.
*/
void
describecorpus_show_statistics (Corpus *corpus)
{
Attribute *a;
int tokens, types, regions, blocks;
for (a = corpus->attributes; a; a = a->any.next) {
switch(a->any.type) {
case ATT_POS:
Rprintf("p-ATT %-16s ", a->any.name);
tokens = cl_max_cpos(a);
types = cl_max_id(a);
if ((tokens > 0) && (types > 0))
Rprintf("%10d tokens, %8d types", tokens, types);
else
Rprintf(" NO DATA");
break;
case ATT_STRUC:
Rprintf("s-ATT %-16s ", a->any.name);
regions = cl_max_struc(a);
if (regions >= 0) {
Rprintf("%10d regions", regions);
if (cl_struc_values(a))
Rprintf(" (with annotations)");
}
else
Rprintf(" NO DATA");
break;
case ATT_ALIGN:
Rprintf("a-ATT %-16s ", a->any.name);
blocks = cl_max_alg(a);
if (blocks >= 0) {
Rprintf("%10d alignment blocks", blocks);
if (cl_has_extended_alignment(a))
Rprintf(" (extended)");
}
else
Rprintf(" NO DATA");
break;
default:
Rprintf("??? %-16s (unknown attribute type)", a->any.name);
break;
}
Rprintf("\n");
}
Rprintf("\n");
}
/**
* Main function for cwb-s-encode.
*
* @param argc Number of command-line arguments.
* @param argv Command-line arguments.
*/
int
main(int argc, char **argv)
{
int input_line;
int start, end;
char *annot;
char buf[CL_MAX_LINE_LENGTH];
Attribute *att;
int V_switch, values, S_annotations_dropped;
int i, N;
progname = argv[0];
sencode_parse_options(argc, argv);
/* -a mode: read existing regions into memory */
if (add_to_existing) {
if (corpus == NULL) {
Rprintf( "Error: You have to specify source corpus (-C <corpus>) for -a switch.\n");
rcqp_receive_error(1);
}
att = cl_new_attribute(corpus, new_satt.name, ATT_STRUC);
if ((att != NULL) && (cl_max_struc(att) > 0)) {
V_switch = new_satt.store_values;
values = cl_struc_values(att);
if (V_switch && (!values)) {
Rprintf( "Error: Existing regions of -V attribute have no annotations.\n");
rcqp_receive_error(1);
}
else if ((!V_switch) && values) {
Rprintf( "Error: Existing regions of -S attributes have annotations.\n");
rcqp_receive_error(1);
}
if (!silent)
Rprintf("[Loading previous <%s> regions]\n", new_satt.name);
N = cl_max_struc(att);
for (i = 0; i < N; i++) {
cl_struc2cpos(att, i, &start, &end);
annot = cl_struc2str(att, i);
SL_insert(start, end, annot);
}
}
else {
if (!silent)
Rprintf("[No <%s> regions defined (skipped)]\n", new_satt.name);
}
}
/* loop reading input (stdin or -f <file>) */
if (in_memory && (!silent))
Rprintf("[Reading input data]\n");
input_line = 0;
S_annotations_dropped = 0;
while (fgets(buf, CL_MAX_LINE_LENGTH, text_fd)) {
input_line++;
/* check for buffer overflow */
if (strlen(buf) >= (CL_MAX_LINE_LENGTH - 1)) {
Rprintf( "BUFFER OVERFLOW, input line #%d is too long:\n>> %s", input_line, buf);
rcqp_receive_error(1);
}
if (! sencode_parse_line(buf, &start, &end, &annot)) {
Rprintf( "FORMAT ERROR on line #%d:\n>> %s", input_line, buf);
rcqp_receive_error(1);
}
if (new_satt.store_values && (annot == NULL)) {
Rprintf( "MISSING ANNOTATION on line #%d:\n>> %s", input_line, buf);
rcqp_receive_error(1);
}
if ((!new_satt.store_values) && (annot != NULL)) {
if (! S_annotations_dropped)
Rprintf( "WARNING: Annotation for -S attribute ignored on line #%d (warning issued only once):\n>> %s", input_line, buf);
S_annotations_dropped++;
}
if ((start <= new_satt.last_cpos) || (end < start)) {
Rprintf( "RANGE INCONSISTENCY on line #%d:\n>> %s(end of previous region was %d)\n", input_line, buf, new_satt.last_cpos);
rcqp_receive_error(1);
}
if (annot != NULL && set_att != set_none) {
/* convert set annotation into standard syntax */
annot = sencode_check_set(annot);
if (annot == NULL) {
Rprintf( "SET ANNOTATION SYNTAX ERROR on line #%d:\n>> %s", input_line, buf);
rcqp_receive_error(1);
}
}
/* debugging output */
if (debug) {
Rprintf( "[%d, %d]", start, end);
if (annot != NULL)
Rprintf( " <%s>", annot);
Rprintf( "\n");
}
/* in -M mode, store this region in memory; otherwise write it to the disk files */
if (in_memory)
SL_insert(start, end, annot);
else
sencode_write_region(start, end, annot);
cl_free(annot);
}
/* in -M mode, write data to disk now that we have finished looping across input data */
if (in_memory) {
SL item;
if (!silent)
Rprintf("[Creating encoded disk file(s)]\n");
SL_rewind();
while ((item = SL_next()) != NULL)
sencode_write_region(item->start, item->end, item->annot);
}
/* close files */
sencode_close_files();
if (S_annotations_dropped > 0)
Rprintf( "Warning: %d annotation values dropped for -S attribute '%s'.\n", S_annotations_dropped, new_satt.name);
rcqp_receive_error(0);
}
