static int find_best_del_candidate(const int q1, const int q2,
const uint32_t* cigarstring1,
const int numcigarops1,
const int q3, const int q4,
const uint32_t* cigarstring2,
const int numcigarops2,
const int readlength)
{
forceassert(q1 == 0);
forceassert(q3 <= q2);
int i, matches, bestmatches = 0, mm, bestmm = INT_MAX, index = -1;
for(i = q3; i <= q2; i++){
matches = count_matches(cigarstring1, numcigarops1, q1, i,
cigarstring2, numcigarops2, i, q4, &mm);
// printf("Matches: %d, Mismatches: %d\n", matches, mm);
forceassert(matches <= readlength);
if(matches > bestmatches){
bestmatches = matches;
bestmm = mm;
index = i;
}else if((matches == bestmatches) && (mm < bestmm)){
bestmatches = matches;
bestmm = mm;
index = i;
}
if((matches == readlength) && (mm == 0)) break;
}
forceassert(index != -1);
// printf("%d : %d : %d\n", index, bestmatches, bestmm);
return index;
}
void expect_matches(int n,
const std::string& stan_code,
const std::string& target,
bool print_model = false) {
std::string model_cpp = model_to_cpp(stan_code);
if (print_model)
std::cout << model_cpp << std::endl;
EXPECT_EQ(n, count_matches(target,model_cpp));
}
TEST(LangCompiler, emptyProgram) {
std::stringstream msgs, stan_lang_in, cpp_out;
std::string model_name = "m";
stan_lang_in << "";
stan::lang::compile(&msgs, stan_lang_in, cpp_out, model_name);
EXPECT_EQ(1,count_matches("WARNING: empty program",msgs.str()));
EXPECT_EQ(std::string::npos, cpp_out.str().find("int main("));
// can't test equivalence of "" and "model { }" because of the
// recording of the positions in the file
}
/* Attempt to parse specific OS.
*
* On success, a malloc'd string containing the OS is returned. */
static char *
parse_os (const char *str, char *tkn, char *os_type, int idx)
{
char *b;
int spaces = 0;
xstrncpy (os_type, os[idx][1], OPESYS_TYPE_LEN);
/* Windows */
if ((strstr (str, "Windows")) != NULL)
return conf.real_os && (b = get_real_win (tkn)) ? b : xstrdup (os[idx][0]);
/* Android */
if ((strstr (tkn, "Android")) != NULL) {
tkn = parse_android (tkn);
return conf.real_os ? get_real_android (tkn) : xstrdup (tkn);
}
/* iOS */
if (strstr (tkn, "iPad") || strstr (tkn, "iPod"))
return xstrdup (parse_ios (tkn, 4));
if (strstr (tkn, "iPhone"))
return xstrdup (parse_ios (tkn, 6));
/* Mac OS X */
if ((strstr (tkn, "OS X")) != NULL) {
tkn = parse_osx (tkn);
return conf.real_os ? get_real_mac_osx (tkn) : xstrdup (tkn);
}
/* Darwin - capture the first part of agents such as:
* Slack/248000 CFNetwork/808.0.2 Darwin/16.0.0 */
if ((strstr (tkn, "Darwin")) != NULL) {
if ((b = strchr (str, ' ')))
*b = 0;
return xstrdup (str);
}
/* all others */
spaces = count_matches (os[idx][0], ' ');
return alloc_string (parse_others (tkn, spaces));
}
/* Given a user agent, determine the operating system used.
*
* ###NOTE: The size of the list is proportional to the run time,
* which makes this pretty slow
*
* On error, NULL is returned.
* On success, a malloc'd string containing the OS is returned. */
char *
verify_os (const char *str, char *os_type)
{
char *a, *b;
int spaces = 0;
size_t i;
if (str == NULL || *str == '\0')
return NULL;
for (i = 0; i < ARRAY_SIZE (os); i++) {
if ((a = strstr (str, os[i][0])) == NULL)
continue;
xstrncpy (os_type, os[i][1], OPESYS_TYPE_LEN);
/* Windows */
if ((strstr (str, "Windows")) != NULL) {
return conf.real_os && (b = get_real_win (a)) ? b : xstrdup (os[i][0]);
}
/* Android */
if ((strstr (a, "Android")) != NULL) {
a = parse_android (a);
return conf.real_os ? get_real_android (a) : xstrdup (a);
}
/* Mac OS X */
if ((strstr (a, "OS X")) != NULL) {
a = parse_osx (a);
return conf.real_os ? get_real_mac_osx (a) : xstrdup (a);
}
/* all others */
spaces = count_matches (os[i][0], ' ');
return alloc_string (parse_others (a, spaces));
}
xstrncpy (os_type, "Unknown", OPESYS_TYPE_LEN);
return alloc_string ("Unknown");
}
/* render a list of agents if available */
void
load_agent_list (WINDOW * main_win, char *addr)
{
char buf[256];
char *ptr_value;
GAgents *agents = NULL;
GMenu *menu;
int c, quit = 1, delims = 0;
int i, n = 0, alloc = 0;
int y, x, list_h, list_w, menu_w, menu_h;
void *value_ptr = NULL;
WINDOW *win;
if (!conf.list_agents)
return;
getmaxyx (stdscr, y, x);
list_h = y / 2; /* list window - height */
list_w = x - 4; /* list window - width */
menu_h = list_h - AGENTS_MENU_Y - 1; /* menu window - height */
menu_w = list_w - AGENTS_MENU_X - AGENTS_MENU_X; /* menu window - width */
#ifdef HAVE_LIBTOKYOCABINET
value_ptr = tc_db_get_str (ht_hosts_agents, addr);
#else
value_ptr = g_hash_table_lookup (ht_hosts_agents, addr);
#endif
if (value_ptr != NULL) {
ptr_value = (char *) value_ptr;
delims = count_matches (ptr_value, '|');
n = ((strlen (ptr_value) + menu_w - 1) / menu_w) + delims + 1;
agents = new_gagents (n);
alloc = split_agent_str (ptr_value, agents, menu_w);
#ifdef HAVE_LIBTOKYOCABINET
free (value_ptr);
#endif
}
win = newwin (list_h, list_w, (y - list_h) / 2, (x - list_w) / 2);
keypad (win, TRUE);
wborder (win, '|', '|', '-', '-', '+', '+', '+', '+');
/* create a new instance of GMenu and make it selectable */
menu = new_gmenu (win, menu_h, menu_w, AGENTS_MENU_Y, AGENTS_MENU_X);
/* add items to GMenu */
menu->items = (GItem *) xcalloc (alloc, sizeof (GItem));
for (i = 0; i < alloc; ++i) {
menu->items[i].name = alloc_string (agents[i].agents);
menu->items[i].checked = 0;
menu->size++;
}
post_gmenu (menu);
snprintf (buf, sizeof buf, "User Agents for %s", addr);
draw_header (win, buf, " %s", 1, 1, list_w - 2, 1, 0);
mvwprintw (win, 2, 2, "[UP/DOWN] to scroll - [q] to close window");
wrefresh (win);
while (quit) {
c = wgetch (stdscr);
switch (c) {
case KEY_DOWN:
gmenu_driver (menu, REQ_DOWN);
draw_header (win, "", "%s", 3, 2, CONF_MENU_W, 0, 0);
break;
case KEY_UP:
gmenu_driver (menu, REQ_UP);
draw_header (win, "", "%s", 3, 2, CONF_MENU_W, 0, 0);
break;
case KEY_RESIZE:
case 'q':
quit = 0;
break;
}
wrefresh (win);
}
/* clean stuff up */
for (i = 0; i < alloc; ++i)
free (menu->items[i].name);
free (menu->items);
free (menu);
for (i = 0; i < alloc; ++i)
free (agents[i].agents);
if (agents)
free (agents);
touchwin (main_win);
close_win (win);
wrefresh (main_win);
}
void expect_matches(int n,
const std::string& stan_code,
const std::string& target) {
std::string model_cpp = model_to_cpp(stan_code);
EXPECT_EQ(n, count_matches(target,model_cpp));
}
int main(int argc, char **argv) {
char **fasta_files;
int seed_len, i, n_fasta_files;
ChrTable *chr_tab;
SeedTable *seed_tab;
Seq *seq;
gzFile gzf, out_gzf;
char *out_filename;
if(argc < 4) {
fprintf(stderr, "usage: %s <seed_len> <chromInfo.txt> "
"<output_seed_index.gz> [chr1.fa.gz [chr2.fa.gz [...]]]\n",
argv[0]);
exit(2);
}
seed_len = util_parse_long(argv[1]);
fasta_files = &argv[4];
n_fasta_files = argc - 4;
out_filename = argv[3];
/* read chromosomes and make table containing offsets for both
* forward and reverse strands (so we can represent genomic
* coordinates with a single long integer).
*/
chr_tab = chr_table_read(argv[2]);
fprintf(stderr, "there are %d chromosomes, total length: %u\n",
chr_tab->n_chr, chr_tab->total_chr_len);
/* create a table to hold seed matches */
fprintf(stderr, "initializing seed table\n");
seed_tab = seed_table_new(seed_len);
/* open an output file to write seed table to */
if(util_file_exists(out_filename)) {
my_err("output file %s already exists\n", out_filename);
exit(2);
}
out_gzf = util_must_gzopen(out_filename, "wb");
/*
* first pass: count number of matches to each seed
*/
seq = seq_new();
for(i = 0; i < n_fasta_files; i++) {
fprintf(stderr, "reading sequence from file %s\n", fasta_files[i]);
gzf = util_must_gzopen(fasta_files[i], "rb");
while(seq_read_fasta_record(seq, gzf)) {
fprintf(stderr, "%s %ld\n", seq->name, seq->len);
fprintf(stderr, "counting seed matches\n");
count_matches(chr_tab, seed_tab, seq);
}
gzclose(gzf);
}
/*
* second pass: store location of each match
*/
for(i = 0; i < n_fasta_files; i++) {
fprintf(stderr, "reading sequence from file %s\n", fasta_files[i]);
gzf = util_must_gzopen(fasta_files[i], "rb");
while(seq_read_fasta_record(seq, gzf)) {
fprintf(stderr, "%s %ld\n", seq->name, seq->len);
fprintf(stderr, "recording seed match positions\n");
add_matches(chr_tab, seed_tab, seq);
}
gzclose(gzf);
}
seq_free(seq);
/* write table to file in binary format */
fprintf(stderr, "writing seed table to file %s\n", out_filename);
seed_table_write(seed_tab, out_gzf);
gzclose(out_gzf);
chr_table_free(chr_tab);
seed_table_free(seed_tab);
return 0;
}
static int
parse_format (GLogItem * glog, const char *fmt, const char *date_format,
char *str)
{
const char *p;
double serve_secs;
int special = 0;
struct tm tm;
unsigned long long bandw, serve_time;
if (str == NULL || *str == '\0')
return 1;
memset (&tm, 0, sizeof (tm));
/* iterate over the log format */
for (p = fmt; *p; p++) {
if (*p == '%') {
special++;
continue;
}
if (special && *p != '\0') {
char *pch, *sEnd, *bEnd, *tkn = NULL, *end = NULL;
errno = 0;
bandw = 0;
serve_time = 0;
serve_secs = 0;
switch (*p) {
/* date */
case 'd':
if (glog->date)
return 1;
/* parse date format including dates containing spaces,
* i.e., syslog date format (Jul 15 20:10:56) */
tkn = parse_string (&str, p[1], count_matches (date_format, ' ') + 1);
if (tkn == NULL)
return 1;
end = strptime (tkn, date_format, &tm);
if (end == NULL || *end != '\0') {
free (tkn);
return 1;
}
glog->date = tkn;
break;
/* remote hostname (IP only) */
case 'h':
if (glog->host)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
if (invalid_ipaddr (tkn)) {
free (tkn);
return 1;
}
glog->host = tkn;
break;
/* request method */
case 'm':
if (glog->method)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
if (!extract_method (tkn)) {
free (tkn);
return 1;
}
glog->method = tkn;
break;
/* request not including method or protocol */
case 'U':
if (glog->req)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL || *tkn == '\0')
return 1;
if ((glog->req = decode_url (tkn)) == NULL)
return 1;
free (tkn);
break;
/* request protocol */
case 'H':
if (glog->protocol)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
if (invalid_protocol (tkn)) {
free (tkn);
return 1;
}
glog->protocol = tkn;
break;
/* request, including method + protocol */
case 'r':
if (glog->req)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
glog->req = parse_req (tkn, &glog->method, &glog->protocol);
free (tkn);
break;
/* Status Code */
case 's':
if (glog->status)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
strtol (tkn, &sEnd, 10);
if (tkn == sEnd || *sEnd != '\0' || errno == ERANGE) {
free (tkn);
return 1;
}
glog->status = tkn;
break;
/* size of response in bytes - excluding HTTP headers */
case 'b':
if (glog->resp_size)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
bandw = strtol (tkn, &bEnd, 10);
if (tkn == bEnd || *bEnd != '\0' || errno == ERANGE)
bandw = 0;
glog->resp_size = bandw;
conf.bandwidth = 1;
free (tkn);
break;
/* referrer */
case 'R':
if (glog->ref)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
tkn = alloc_string ("-");
if (tkn != NULL && *tkn == '\0') {
free (tkn);
tkn = alloc_string ("-");
}
if (strcmp (tkn, "-") != 0)
extract_referer_site (tkn, glog->site);
glog->ref = tkn;
break;
/* user agent */
case 'u':
if (glog->agent)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn != NULL && *tkn != '\0') {
/* Make sure the user agent is decoded (i.e.: CloudFront)
* and replace all '+' with ' ' (i.e.: w3c) */
glog->agent = char_replace (decode_url (tkn), '+', ' ');
free (tkn);
break;
} else if (tkn != NULL && *tkn == '\0') {
free (tkn);
tkn = alloc_string ("-");
}
/* must be null */
else {
tkn = alloc_string ("-");
}
glog->agent = tkn;
break;
/* time taken to serve the request, in seconds */
case 'T':
if (glog->serve_time)
return 1;
/* ignore seconds if we have microseconds */
if (strstr (fmt, "%D") != NULL)
break;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
if (strchr (tkn, '.') != NULL)
serve_secs = strtod (tkn, &bEnd);
else
serve_secs = strtoull (tkn, &bEnd, 10);
if (tkn == bEnd || *bEnd != '\0' || errno == ERANGE)
serve_secs = 0;
/* convert it to microseconds */
if (serve_secs > 0)
glog->serve_time = serve_secs * SECS;
else
glog->serve_time = 0;
conf.serve_usecs = 1;
free (tkn);
break;
/* time taken to serve the request, in microseconds */
case 'D':
if (glog->serve_time)
return 1;
tkn = parse_string (&str, p[1], 1);
if (tkn == NULL)
return 1;
serve_time = strtoull (tkn, &bEnd, 10);
if (tkn == bEnd || *bEnd != '\0' || errno == ERANGE)
serve_time = 0;
glog->serve_time = serve_time;
conf.serve_usecs = 1;
free (tkn);
break;
/* everything else skip it */
default:
if ((pch = strchr (str, p[1])) != NULL)
str += pch - str;
}
if ((str == NULL) || (*str == '\0'))
return 0;
special = 0;
} else if (special && isspace (p[0])) {
return 1;
} else
str++;
}
return 0;
}
