bst_t bst_remove(bst_t bst, index_t index) {
if (bst_type(bst) == isNotEmpty) {
switch (index_compare(index, bst)) {
case EQ:
switch (has_child(bst)) {
case none:
bst = bst_destroy(bst);
break;
case both:
case justRigth:
pair_destroy(bst->pair);
bst->pair = pair_copy((bst->der)->pair); /*copia el par q le sigue (der) */
bst->der =
bst_remove(bst->der, pair_fst((bst->der)->pair));
break;
case justLeft:
pair_destroy(bst->pair);
bst->pair = pair_copy((bst->izq)->pair); /*copia el par q le sigue (der) */
bst->izq =
bst_remove(bst->izq, pair_fst((bst->izq)->pair));
break;
}
break;
case LT:
bst->izq = bst_remove(bst->izq, index);
break;
case GT:
bst->der = bst_remove(bst->der, index);
break;
}
}
return (bst);
}
map_iterator_t map_insert(map_t* pt_map, const pair_t* cpt_pair)
{
pair_t t_elempair;
map_iterator_t t_result;
#ifdef CSTL_MAP_AVL_TREE
avl_tree_result_pair_t t_avlresult;
#else
rb_tree_result_pair_t t_rbresult;
#endif
assert(pt_map != NULL && cpt_pair != NULL);
assert(_same_map_pair_type(&pt_map->_t_pair, cpt_pair));
/* initialize the new pair */
t_elempair = _create_pair(
pt_map->_t_pair._t_firsttypesize, pt_map->_t_pair._sz_firsttypename,
pt_map->_t_pair._t_secondtypesize, pt_map->_t_pair._sz_secondtypename);
pair_init(&t_elempair);
memcpy(t_elempair.first, cpt_pair->first, t_elempair._t_firsttypesize);
memcpy(t_elempair.second, cpt_pair->second, t_elempair._t_secondtypesize);
/* insert pair into tree */
#ifdef CSTL_MAP_AVL_TREE
t_avlresult = _avl_tree_insert_unique(_GET_MAP_AVL_TREE(pt_map), &t_elempair);
if(t_avlresult._t_second._t_bool)
{
t_result = t_avlresult._t_first;
}
else
{
t_result = map_end(pt_map);
pair_destroy(&t_elempair);
}
#else
t_rbresult = _rb_tree_insert_unique(_GET_MAP_RB_TREE(pt_map), &t_elempair);
if(t_rbresult._t_second._t_bool)
{
t_result = t_rbresult._t_first;
}
else
{
t_result = map_end(pt_map);
pair_destroy(&t_elempair);
}
#endif
_GET_CONTAINER(&t_result) = pt_map;
_GET_MAP_CONTAINER_TYPE(&t_result) = _MAP_CONTAINER;
_GET_MAP_ITERATOR_TYPE(&t_result) = _BIDIRECTIONAL_ITERATOR;
return t_result;
}
list_t list_remove(list_t list, index_t index) {
list_t current = list,last=NULL;
while(current != NULL && !index_is_equal(pair_fst(current->elem),index)) {
last = current;
current = current->next;
}
if(current!=NULL) {
if(last==NULL) {
list = current->next;
} else {
last->next = current->next;
current->next = NULL;
}
current->elem = pair_destroy(current->elem);
free(current);
current = NULL;
last = NULL;
}
return list;
}
bst_t bst_destroy(bst_t bst) {
switch (bst_type(bst)) {
case isNull:
break;
case isEmpty:
case isNotEmpty:
if (bst->izq != NULL) {
bst_destroy(bst->izq); /*Destruye subarbol izquierdo */
}
if (bst->der != NULL) {
bst_destroy(bst->der); /*Destruye subarbol derecho */
}
bst->pair = pair_destroy(bst->pair);
free(bst);
bst = NULL;
break;
}
return (bst);
}
void multimap_destroy(multimap_t* pt_multimap)
{
assert(pt_multimap != NULL);
/* destroy pair */
pair_destroy(&pt_multimap->_t_pair);
/* destroy tree */
#ifdef CSTL_MULTIMAP_AVL_TREE
_avl_tree_destroy(_GET_MULTIMAP_AVL_TREE(pt_multimap));
#else
_rb_tree_destroy(_GET_MULTIMAP_RB_TREE(pt_multimap));
#endif
}
list_t list_destroy(list_t list) {
list_t current;
while(list!= NULL)
{
current = list;
list = list->next;
current->elem = pair_destroy(current->elem);
free(current);
}
current = NULL;
return list;
}
bst_t bst_remove(bst_t bst, index_t index) {
bst_t current;
if (bst != NULL) {
if (index_is_less_than(index, pair_fst(bst->pair))) {
bst->left = bst_remove(bst->left, index);
} else if (index_is_equal(pair_fst(bst->pair), index) &&
bst->left == NULL) {
current = bst->right;
bst->pair = pair_destroy(bst->pair);
free(bst);
bst = current;
} else if (index_is_equal(pair_fst(bst->pair), index) &&
bst->left != NULL) {
bst->pair = pair_destroy(bst->pair);
bst->pair = bst_max(bst->left);
bst->left = delete_max(bst->left);
} else {
bst->right = bst_remove(bst->right, index);
}
}
return (bst);
}
bst_t delete_max(bst_t bst) {
bst_t current;
if (bst != NULL) {
if (bst->right == NULL) {
current = bst->left;
/* Free the allocated resources */
bst->pair = pair_destroy(bst->pair);
free(bst);
bst = current;
} else {
bst->right = delete_max(bst->right);
}
}
return (bst);
}
void* _map_at_varg(map_t* pt_map, va_list val_elemlist)
{
pair_t t_elempair;
bool_t t_result = false;
void* pv_value = NULL;
#ifdef CSTL_MAP_AVL_TREE
avl_tree_result_pair_t t_avlresult;
#else
rb_tree_result_pair_t t_rbresult;
#endif
assert(pt_map != NULL);
t_elempair = _create_pair(
pt_map->_t_pair._t_firsttypesize, pt_map->_t_pair._sz_firsttypename,
pt_map->_t_pair._t_secondtypesize, pt_map->_t_pair._sz_secondtypename);
pair_init(&t_elempair);
_get_varg_value(
t_elempair.first,
val_elemlist,
t_elempair._t_firsttypesize,
t_elempair._sz_firsttypename);
memset(t_elempair.second, 0x00, t_elempair._t_secondtypesize);
#ifdef CSTL_MAP_AVL_TREE
t_avlresult =
_avl_tree_insert_unique(_GET_MAP_AVL_TREE(pt_map), &t_elempair);
pv_value = ((pair_t*)((avlnode_t*)
_GET_AVL_TREE_COREPOS(&t_avlresult._t_first))->_pc_data)->second;
t_result = t_avlresult._t_second._t_bool;
#else
t_rbresult =
_rb_tree_insert_unique(_GET_MAP_RB_TREE(pt_map), &t_elempair);
pv_value = ((pair_t*)((rbnode_t*)
_GET_RB_TREE_COREPOS(&t_rbresult._t_first))->_pc_data)->second;
t_result = t_rbresult._t_second._t_bool;
#endif
assert(pv_value != NULL);
/* destroy the pair when inserting failed */
if(!t_result)
{
pair_destroy(&t_elempair);
}
return pv_value;
}
bst_t bst_destroy(bst_t bst) {
/* To destroy the current node, first we need to destroy its pair,
* and its children if it have */
if (bst != NULL) {
bst->pair = pair_destroy(bst->pair);
if (bst->left != NULL) {
bst->left = bst_destroy(bst->left);
}
if (bst->right != NULL) {
bst->right = bst_destroy(bst->right);
}
/* Free the allocated resources */
free(bst);
}
bst = NULL;
return (bst);
}
static void _multimap_destroy_elem(void* pv_value)
{
assert(pv_value != NULL);
pair_destroy((pair_t*)pv_value);
}
/**
* @brief main function for the copyright agent
*
* The copyright agent is used to automatically locate copyright statements
* found in code.
*
* There are 3 ways to use the copyright agent:
* 1. Command Line Analysis :: test a file from the command line
* 2. Agent Based Analysis :: waits for commands from stdin
* 3. Accuracy Test :: tests the accuracy of the copyright agent
*
* +-----------------------+
* | Command Line Analysis |
* +-----------------------+
*
* To analyze a file from the command line:
* -C <filename> :: run copyright agent from command line
* -d :: turn on debugging information
* -T <Copyright Statements | URLs| Emails> :: Copyright Statements | URLs |Emails
*
* example:
* $ ./copyright -C myfiletoscan
*
* +----------------------+
* | Agent Based Analysis |
* +----------------------+
*
* To run the copyright agent as an agent simply run with no command line args
* -i :: initialize a connection to the database
* -d :: turn on debugging information
*
* example:
* $ upload_pk | ./copyright
*
* +---------------+
* | Accuracy Test |
* +---------------+
*
* To test the accuracy of the copyright agent run with a -t. Make sure to run the
* accuracy tests in the source directory with the testdata directory:
* -t :: run the accuracy analysis
*
* example:
* $ ./copyright -t
*
* Running the tests will create 3 files:
* 1. Matches: contains all of the matches found by the copyright agent, information
* includes what file the match was found in, the dictionary element
* that it matched, the name that it matched and the text that was found
* 2. False_Positives: contains all of the false positives found by the agent,
* information in the file includes the file the false positive was
* in, the dictionary match, the name match, and the text
* 3. Flase_Negatives: contains all of the false negatives found by the agent,
* information in the file includes the file the false negative was
* in, and the text of the false negative
*
* NOTE: -d will produces the exact same style of Matches file that the accuracy
* testing does. Currently this is the only thing that -d will produce
*
* @param argc the number of command line arguments
* @param argv the command line arguments
* @return 0 on a successful program execution
*/
int main(int argc, char** argv)
{
/* primitives */
char sql[512]; // buffer for database access
int c, i = -1; // temporary int containers
int num_files = 0; // the number of rows in a job
int ars_pk = 0; // the args primary key
int user_pk = 0;
long upload_pk = 0; // the upload primary key
long agent_pk = 0; // the agents primary key
char *SVN_REV = NULL;
char *VERSION = NULL;
char agent_rev[myBUFSIZ];
char copy_buf[FILENAME_MAX];
char name_buf[FILENAME_MAX];
int report_type = 7; // defaul as all. binary xxx 1st number as email, 2nd number as url, 3rd number as statement
int cli_run = 0; // when run from command line, that mean -C option is set; 1: yes, 0: no
/* Database structs */
PGconn* pgConn = NULL; // the connection to Database
PGresult* pgResult = NULL; // result of a database access
/* copyright structs */
copyright copy; // the work horse of the copyright agent
pair curr; // pair to push into the file list
/* verbose data */
FILE* mout = NULL;
/* set the output streams */
cout = stdout;
cerr = stdout;
cin = stdin;
/* connect to the scheduler */
fo_scheduler_connect(&argc, argv, &pgConn);
/* initialize complex data strcutres */
memset(copy_buf, '\0', sizeof(copy_buf));
memset(name_buf, '\0', sizeof(copy_buf));
snprintf(copy_buf, sizeof(copy_buf),
"%s/mods-enabled/copyright/agent/copyright.dic",
sysconfigdir);
snprintf(name_buf, sizeof(name_buf),
"%s/mods-enabled/copyright/agent/names.dic",
sysconfigdir);
if(!copyright_init(©, copy_buf, name_buf))
{
fprintf(cerr, "FATAL %s.%d: copyright initialization failed\n", __FILE__, __LINE__);
fprintf(cerr, "FATAL %s\n", strerror(errno));
fflush(cerr);
return 1;
}
/* parse the command line options */
while((c = getopt(argc, argv, "T:dc:C:tiVvh")) != -1)
{
switch(c)
{
case 'v': /* debugging */
mout = fopen("Matches", "w");
if(!mout)
{
fprintf(cerr, "ERROR could not open Matches for logging\n");
fflush(cerr);
}
else
{
verbose = 1;
}
break;
case 'C': /* run from command line */
cli_run = 1;
pair_init(&curr, string_function_registry(), int_function_registry());
pair_set_first(curr, optarg);
pair_set_second(curr, &i);
num_files++;
break;
case 'T': /* report type, Copyright Statements | URLs| Emails */
report_type = atoi(optarg);
printf("report_type is:%d\n", report_type);
break;
case 't': /* run accuracy testing */
run_test_files(copy);
copyright_destroy(copy);
return 0;
case 'i': /* initialize database connections */
copyright_destroy(copy);
PQfinish(pgConn);
return 0;
case 'V':
printf("%s", BuildVersion);
copyright_destroy(copy);
PQfinish(pgConn);
return 0;
default: /* error, print usage */
copyright_usage(argv[0]);
return 3;
}
}
/** run from command line */
if (1 == cli_run) {
perform_analysis(pgConn, copy, curr, agent_pk, mout, report_type);
pair_destroy(curr);
}
/* if there are no files in the file list then the agent is begin run from */
/* the scheduler, open the database and grab the files to be analyzed */
if(num_files == 0)
{
/* create the sql copy structure */
sqlcpy = fo_sqlCopyCreate(pgConn, "copyright", 32768, 7,
"agent_fk", "pfile_fk", "copy_startbyte", "copy_endbyte", "content", "hash", "type");
/* book keeping */
pair_init(&curr, string_function_registry(), int_function_registry());
db_connected = 1;
SVN_REV = fo_sysconfig("copyright", "SVN_REV");
VERSION = fo_sysconfig("copyright", "VERSION");
sprintf(agent_rev, "%s.%s", VERSION, SVN_REV);
agent_pk = fo_GetAgentKey(pgConn, AGENT_NAME, 0, agent_rev, AGENT_DESC);
/* make sure that we are connected to the database */
if(!check_copyright_table(pgConn))
{
return 5;
}
user_pk = fo_scheduler_userID(); /* get user_pk for user who queued the agent */
/* enter the main agent loop */
while(fo_scheduler_next())
{
upload_pk = atol(fo_scheduler_current());
/* Check Permissions */
if (GetUploadPerm(pgConn, upload_pk, user_pk) < PERM_WRITE)
{
LOG_ERROR("You have no update permissions on upload %ld", upload_pk);
continue;
}
ars_pk = fo_WriteARS(pgConn, 0, upload_pk, agent_pk, AGENT_ARS, NULL, 0);
sprintf(sql, fetch_pfile, upload_pk, agent_pk, agent_pk);
pgResult = PQexec(pgConn, sql);
num_files = PQntuples(pgResult);
for(i = 0; i < num_files; i++)
{
c = atoi(PQgetvalue(pgResult, i, PQfnumber(pgResult, "pfile_pk")));
pair_set_first(curr, PQgetvalue(pgResult, i, PQfnumber(pgResult, "pfilename")));
pair_set_second(curr, &c);
perform_analysis(pgConn, copy, curr, agent_pk, mout, REPORTALL);
}
fo_WriteARS(pgConn, ars_pk, upload_pk, agent_pk, AGENT_ARS, NULL, 1);
PQclear(pgResult);
}
pair_destroy(curr);
}
if(db_connected)
{
fo_sqlCopyDestroy(sqlcpy, 1);
PQfinish(pgConn);
}
if(verbose)
{
fclose(mout);
}
copyright_destroy(copy);
fo_scheduler_disconnect(0);
return 0;
}
static void pair_free_object(zend_object *object)
{
Pair *pair = (Pair*) object;
zend_object_std_dtor(&pair->std);
pair_destroy(pair);
}
