void test_cvector_remove()
{
cvector vec;
int tester = 1;
/* start the test */
/* run the functions */
cvector_init(&vec, int_function_registry());
cvector_push_back(vec, &tester);
cvector_remove(vec, vec->data);
/* test the results */
FO_ASSERT_EQUAL(vec->size, 0);
FO_ASSERT_PTR_EQUAL(vec->data[0], NULL);
cvector_destroy(vec);
}
void test_cvector_remove()
{
cvector vec;
int tester = 1;
/* start the test */
printf("Test cvector_remove:");
/* run the functions */
cvector_init(&vec, int_function_registry());
cvector_push_back(vec, &tester);
cvector_remove(vec, vec->data);
/* test the results */
CU_ASSERT_EQUAL(vec->size, 0);
CU_ASSERT_EQUAL(vec->data[0], NULL);
cvector_destroy(vec);
test_failure();
printf("\n");
}
/**
* @brief runs the labeled test files to determine accuracy
*
* This function will open each pair of files in the testdata directory to
* analyze how accurate the copyright agent is. This function will respond with
* the number of false negatives, false positives, and correct answers for each
* file and total tally of these numbers. This will also produce 3 files, one
* containing all matches that the copyright agent found, all the things that it
* didn't find, and all of the false positives.
*/
void run_test_files(copyright copy)
{
/* locals */
cvector compare;
copyright_iterator iter;
cvector_iterator curr;
FILE* istr, * m_out, * n_out, * p_out;
char buffer[READMAX + 1];
char file_name[FILENAME_MAX];
char copy_buf[FILENAME_MAX];
char name_buf[FILENAME_MAX];
char* first, * last, * loc, tmp;
int i, matches, correct = 0, falsep = 0, falsen = 0;
/* grab the copyright files */
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);
/* create data structures */
copyright_init(©, copy_buf, name_buf);
cvector_init(&compare, string_function_registry());
/* open the logging files */
m_out = fopen("Matches", "w");
n_out = fopen("False_Negatives", "w");
p_out = fopen("False_Positives", "w");
/* big problem if any of the log files didn't open correctly */
if(!m_out || !n_out || !p_out)
{
fprintf(cerr, "ERROR did not successfully open one of the log files\n");
fprintf(cerr, "ERROR the files that needed to be opened were:\n");
fprintf(cerr, "ERROR Matches, False_Positives, False_Negatives\n");
exit(-1);
}
/* loop over every file in the test directory */
for(i = 0; i < TESTFILE_NUMBER; i++)
{
sprintf(file_name, "%s%d_raw", test_dir, i);
/* attempt to open the labeled test file */
istr = fopen(file_name, "r");
if(!istr)
{
fprintf(cerr, "ERROR Must run testing from correct directory. The\n");
fprintf(cerr, "ERROR correct directory is installation dependent but\n");
fprintf(cerr, "ERROR the working directory should include the folder:\n");
fprintf(cerr, "ERROR %s\n", test_dir);
exit(-1);
}
/* initialize the buffer and read in any information */
memset(buffer, '\0', sizeof(buffer));
buffer[fread(buffer, sizeof(char), READMAX, istr)] = '\0';
matches = 0;
/* set everything in the buffer to lower case */
for(first = buffer; *first; first++)
{
*first = tolower(*first);
}
/* loop through and find all <s>...</s> tags */
loc = buffer;
while((first = strstr(loc, "<s>")) != NULL)
{
last = strstr(loc, "</s>");
if(last == NULL)
{
fprintf(cerr, "ERROR unmatched \"<s>\"\n");
fprintf(cerr, "ERROR in file: \"%s\"\n", file_name);
exit(-1);
}
if(last <= first)
{
fprintf(cerr, "ERROR unmatched \"</s>\"\n");
fprintf(cerr, "ERROR in file: \"%s\"\n", file_name);
exit(-1);
}
tmp = *last;
*last = 0;
cvector_push_back(compare, first + 3);
*last = tmp;
loc = last + 4;
}
/* close the previous file and open the corresponding raw data */
fclose(istr);
file_name[strlen(file_name) - 4] = '\0';
istr = fopen(file_name, "r");
if(!istr)
{
fprintf(cerr, "ERROR Unmatched file in the test directory");
fprintf(cerr, "ERROR File with no match: \"%s\"_raw\n", file_name);
fprintf(cerr, "ERROR File that caused error: \"%s\"\n", file_name);
}
/* perform the analysis on the current file */
copyright_analyze(copy, istr, REPORTALL);
fclose(istr);
/* loop over every match that the copyright object found */
for(iter = copyright_begin(copy); iter != copyright_end(copy); iter++)
{
cvector_iterator best = cvector_begin(compare);
char score[2048];
char dst[2048];
memset(dst, '\0', sizeof(dst));
memset(score, '\0', sizeof(score));
/* log the coyright entry */
fprintf(m_out, "====%s================================\n", file_name);
fprintf(m_out, "DICT: %s\tNAME: %s\n",copy_entry_dict(*iter), copy_entry_name(*iter));
fprintf(m_out, "TEXT[%s]\n",copy_entry_text(*iter));
/* loop over the vector looking for matches */
for(curr = cvector_begin(compare); curr != cvector_end(compare); curr++)
{
if(longest_common(dst, copy_entry_text(*iter), (char*)*curr) > strlen(score))
{
strcpy(score, dst);
best = curr;
}
}
/* log the entry as found if it matched something in compare */
if(cvector_size(compare) != 0 &&
(strcmp(copy_entry_dict(*iter), "by") || strlen(score) > THRESHOLD))
{
cvector_remove(compare, best);
matches++;
}
else if(!strcmp(copy_entry_dict(*iter), "email") || !strcmp(copy_entry_dict(*iter), "url"))
{
matches++;
}
else
{
fprintf(p_out, "====%s================================\n", file_name);
fprintf(p_out, "DICT: %s\tNAME: %s\n",copy_entry_dict(*iter), copy_entry_name(*iter));
fprintf(p_out, "TEXT[%s]\n",copy_entry_text(*iter));
}
}
/* log all the false negatives */
for(curr = cvector_begin(compare); curr != cvector_end(compare); curr++)
{
fprintf(n_out, "====%s================================\n", file_name);
fprintf(n_out, "%s\n", (char*)*curr);
}
fprintf(cout, "====%s================================\n", file_name);
fprintf(cout, "Correct: %d\n", matches);
fprintf(cout, "False Positives: %d\n", copyright_size(copy) - matches);
fprintf(cout, "False Negatives: %d\n", cvector_size(compare));
/* clean up for the next file */
correct += matches;
falsep += copyright_size(copy) - matches;
falsen += cvector_size(compare);
cvector_clear(compare);
}
fprintf(cout, "==== Totals ================================\n");
fprintf(cout, "Total Found: %d\n", correct + falsep);
fprintf(cout, "Correct: %d\n", correct);
fprintf(cout, "False Positives: %d\n", falsep);
fprintf(cout, "False Negatives: %d\n", falsen);
fclose(m_out);
fclose(n_out);
fclose(p_out);
copyright_destroy(copy);
cvector_destroy(compare);
}