void RegEx::ClearMatchList(void)
{
if (matchlist)
{
pcre_free_substring_list(matchlist);
matchlist = NULL;
}
}
/* Public Methods */
gboolean
mud_regex_check(MudRegex *regex,
const gchar *data,
guint length,
const gchar *rx,
gint ovector[1020])
{
pcre *re = NULL;
const gchar *error = NULL;
gint errorcode;
gint erroroffset;
gint rc;
if(!MUD_IS_REGEX(regex))
return FALSE;
re = pcre_compile2(rx, 0, &errorcode, &error, &erroroffset, NULL);
if(!re)
{
gint i;
/*
This should never be called since we check the regex validity
at entry time. But You Never Know(tm) so its here to catch
any runtime errors that cosmic rays, evil magic, errant gconf
editing, and Monday mornings might produce.
*/
g_warning("Error in Regex! - ErrCode: %d - %s", errorcode, error);
printf("--> %s\n ", rx);
for(i = 0; i < erroroffset - 1; i++)
printf(" ");
printf("^\n");
return FALSE;
}
rc = pcre_exec(re, NULL, data, length, 0, 0, ovector, 1020);
if(rc < 0)
return FALSE;
if(regex->priv->substring_list)
pcre_free_substring_list(regex->priv->substring_list);
pcre_get_substring_list(data, ovector, rc, ®ex->priv->substring_list);
regex->priv->substring_count = rc;
return TRUE;
}
/* The expected input file is the result of anathons analysis */
int get_data(char *inFileName, Dat *dat){
FILE *inFile = 0; /* input file */
char line[80]; /* line */
unsigned int allocated = 64; /* allocation counter */
pcre *re; /* regular expression */
const char *error; /* error message string */
int erroffset; /* error offset */
int ovector[OVECCOUNT]; /* match vector */
int rc; /* match return value */
char **substring_list; /* substring list */
/* initialise/allocate memory for set of (64) frag entries */
dat->nData = 0;
dat->data = safe_malloc(allocated * sizeof(Ang));
/* read data */
inFile = safe_open(inFileName, "r");
/* compile regexp */
re = pcre_compile("^.{7}\t.{4}\t.{6}\t(........)\t(........)\t(........)", 0, &error, &erroffset, NULL);
/* count the number of models */
while(fgets(line, 80, inFile) != NULL )
{
rc = pcre_exec(re, NULL, line, strlen(line), 0, 0, ovector, OVECCOUNT);
if (rc == 4){
pcre_get_substring_list(line, ovector, rc, (const char ***)&substring_list);
dat->data[dat->nData].phi1 = atof(substring_list[1]);
dat->data[dat->nData].phi2 = atof(substring_list[2]);
dat->data[dat->nData].theta = atof(substring_list[3]);
++dat->nData;
/* free substring_list memory */
pcre_free_substring_list((const char **)substring_list);
}
/* allocate more memory if needed */
if (dat->nData == allocated) {
allocated += 64;
dat->data = safe_realloc(dat->data, allocated * sizeof(Ang));
}
}
/* free regexp */
pcre_free(re);
assert(dat->nData > 1);
/* close file handle */
fclose(inFile);
return 0;
}
static void
mud_regex_finalize (GObject *object)
{
MudRegex *regex;
GObjectClass *parent_class;
regex = MUD_REGEX(object);
if(regex->priv->substring_list)
pcre_free_substring_list(regex->priv->substring_list);
parent_class = g_type_class_peek_parent(G_OBJECT_GET_CLASS(object));
parent_class->finalize(object);
}
std::vector<std::string> Pattern::groups (void)
{
const char ** stringList;
pcre_get_substring_list (_subject.c_str(),
_ovector,
_count,
&stringList);
std::vector<std::string> matchedSubstrings;
for (int i = 1; stringList[i] != NULL; i++)
matchedSubstrings.push_back (stringList[i]);
pcre_free_substring_list (stringList);
return matchedSubstrings;
}
static int
zpcre_get_substrings(char *arg, int *ovec, int ret, char *receptacle)
{
char **captures, **matches;
if(!pcre_get_substring_list(arg, ovec, ret, (const char ***)&captures)) {
matches = zarrdup(&captures[1]); /* first one would be entire string */
if (receptacle == NULL)
setaparam("match", matches);
else
setaparam(receptacle, matches);
pcre_free_substring_list((const char **)captures);
}
return 0;
}
static int
zpcre_get_substrings(char *arg, int *ovec, int ret, char *matchvar,
char *substravar, int want_offset_pair, int matchedinarr,
int want_begin_end)
{
char **captures, *match_all, **matches;
char offset_all[50];
int capture_start = 1;
if (matchedinarr)
capture_start = 0;
if (matchvar == NULL)
matchvar = "MATCH";
if (substravar == NULL)
substravar = "match";
/* captures[0] will be entire matched string, [1] first substring */
if (!pcre_get_substring_list(arg, ovec, ret, (const char ***)&captures)) {
int nelem = arrlen(captures)-1;
/* Set to the offsets of the complete match */
if (want_offset_pair) {
sprintf(offset_all, "%d %d", ovec[0], ovec[1]);
setsparam("ZPCRE_OP", ztrdup(offset_all));
}
match_all = metafy(captures[0], -1, META_DUP);
setsparam(matchvar, match_all);
/*
* If we're setting match, mbegin, mend we only do
* so if there were parenthesised matches, for consistency
* (c.f. regex.c).
*/
if (!want_begin_end || nelem) {
char **x, **y;
y = &captures[capture_start];
matches = x = (char **) zalloc(sizeof(char *) * (arrlen(y) + 1));
do {
if (*y)
*x++ = metafy(*y, -1, META_DUP);
else
*x++ = NULL;
} while (*y++);
setaparam(substravar, matches);
}
if (want_begin_end) {
char *ptr = arg;
zlong offs = 0;
/* Count the characters before the match */
MB_METACHARINIT();
while (ptr < arg + ovec[0]) {
offs++;
ptr += MB_METACHARLEN(ptr);
}
setiparam("MBEGIN", offs + !isset(KSHARRAYS));
/* Add on the characters in the match */
while (ptr < arg + ovec[1]) {
offs++;
ptr += MB_METACHARLEN(ptr);
}
setiparam("MEND", offs + !isset(KSHARRAYS) - 1);
if (nelem) {
char **mbegin, **mend, **bptr, **eptr;
int i, *ipair;
bptr = mbegin = zalloc(sizeof(char*)*(nelem+1));
eptr = mend = zalloc(sizeof(char*)*(nelem+1));
for (ipair = ovec + 2, i = 0;
i < nelem;
ipair += 2, i++, bptr++, eptr++)
{
char buf[DIGBUFSIZE];
ptr = arg;
offs = 0;
/* Find the start offset */
MB_METACHARINIT();
while (ptr < arg + ipair[0]) {
offs++;
ptr += MB_METACHARLEN(ptr);
}
convbase(buf, offs + !isset(KSHARRAYS), 10);
*bptr = ztrdup(buf);
/* Continue to the end offset */
while (ptr < arg + ipair[1]) {
offs++;
ptr += MB_METACHARLEN(ptr);
}
convbase(buf, offs + !isset(KSHARRAYS) - 1, 10);
*eptr = ztrdup(buf);
}
*bptr = *eptr = NULL;
setaparam("mbegin", mbegin);
setaparam("mend", mend);
}
}
pcre_free_substring_list((const char **)captures);
}
return 0;
}
int main(int argc, char **argv)
{
FILE *infile = stdin;
int options = 0;
int study_options = 0;
int op = 1;
int timeit = 0;
int showinfo = 0;
int showstore = 0;
int size_offsets = 45;
int size_offsets_max;
int *offsets;
#if !defined NOPOSIX
int posix = 0;
#endif
int debug = 0;
int done = 0;
unsigned char buffer[30000];
unsigned char dbuffer[1024];
/* Static so that new_malloc can use it. */
outfile = stdout;
/* Scan options */
while (argc > 1 && argv[op][0] == '-')
{
char *endptr;
if (strcmp(argv[op], "-s") == 0 || strcmp(argv[op], "-m") == 0)
showstore = 1;
else if (strcmp(argv[op], "-t") == 0) timeit = 1;
else if (strcmp(argv[op], "-i") == 0) showinfo = 1;
else if (strcmp(argv[op], "-d") == 0) showinfo = debug = 1;
else if (strcmp(argv[op], "-o") == 0 && argc > 2 &&
((size_offsets = (int)strtoul(argv[op+1], &endptr, 10)), *endptr == 0))
{
op++;
argc--;
}
#if !defined NOPOSIX
else if (strcmp(argv[op], "-p") == 0) posix = 1;
#endif
else
{
printf("** Unknown or malformed option %s\n", argv[op]);
printf("Usage: pcretest [-d] [-i] [-o <n>] [-p] [-s] [-t] [<input> [<output>]]\n");
printf(" -d debug: show compiled code; implies -i\n"
" -i show information about compiled pattern\n"
" -o <n> set size of offsets vector to <n>\n");
#if !defined NOPOSIX
printf(" -p use POSIX interface\n");
#endif
printf(" -s output store information\n"
" -t time compilation and execution\n");
return 1;
}
op++;
argc--;
}
/* Get the store for the offsets vector, and remember what it was */
size_offsets_max = size_offsets;
offsets = malloc(size_offsets_max * sizeof(int));
if (offsets == NULL)
{
printf("** Failed to get %d bytes of memory for offsets vector\n",
size_offsets_max * sizeof(int));
return 1;
}
/* Sort out the input and output files */
if (argc > 1)
{
infile = fopen(argv[op], "r");
if (infile == NULL)
{
printf("** Failed to open %s\n", argv[op]);
return 1;
}
}
if (argc > 2)
{
outfile = fopen(argv[op+1], "w");
if (outfile == NULL)
{
printf("** Failed to open %s\n", argv[op+1]);
return 1;
}
}
/* Set alternative malloc function */
pcre_malloc = new_malloc;
/* Heading line, then prompt for first regex if stdin */
fprintf(outfile, "PCRE version %s\n\n", pcre_version());
/* Main loop */
while (!done)
{
pcre *re = NULL;
pcre_extra *extra = NULL;
#if !defined NOPOSIX /* There are still compilers that require no indent */
regex_t preg;
int do_posix = 0;
#endif
const char *error;
unsigned char *p, *pp, *ppp;
const unsigned char *tables = NULL;
int do_study = 0;
int do_debug = debug;
int do_G = 0;
int do_g = 0;
int do_showinfo = showinfo;
int do_showrest = 0;
int utf8 = 0;
int erroroffset, len, delimiter;
if (infile == stdin) printf(" re> ");
if (fgets((char *)buffer, sizeof(buffer), infile) == NULL) break;
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
p = buffer;
while (isspace(*p)) p++;
if (*p == 0) continue;
/* Get the delimiter and seek the end of the pattern; if is isn't
complete, read more. */
delimiter = *p++;
if (isalnum(delimiter) || delimiter == '\\')
{
fprintf(outfile, "** Delimiter must not be alphameric or \\\n");
goto SKIP_DATA;
}
pp = p;
for(;;)
{
while (*pp != 0)
{
if (*pp == '\\' && pp[1] != 0) pp++;
else if (*pp == delimiter) break;
pp++;
}
if (*pp != 0) break;
len = sizeof(buffer) - (pp - buffer);
if (len < 256)
{
fprintf(outfile, "** Expression too long - missing delimiter?\n");
goto SKIP_DATA;
}
if (infile == stdin) printf(" > ");
if (fgets((char *)pp, len, infile) == NULL)
{
fprintf(outfile, "** Unexpected EOF\n");
done = 1;
goto CONTINUE;
}
if (infile != stdin) fprintf(outfile, "%s", (char *)pp);
}
/* If the first character after the delimiter is backslash, make
the pattern end with backslash. This is purely to provide a way
of testing for the error message when a pattern ends with backslash. */
if (pp[1] == '\\') *pp++ = '\\';
/* Terminate the pattern at the delimiter */
*pp++ = 0;
/* Look for options after final delimiter */
options = 0;
study_options = 0;
log_store = showstore; /* default from command line */
while (*pp != 0)
{
switch (*pp++)
{
case 'g': do_g = 1; break;
case 'i': options |= PCRE_CASELESS; break;
case 'm': options |= PCRE_MULTILINE; break;
case 's': options |= PCRE_DOTALL; break;
case 'x': options |= PCRE_EXTENDED; break;
case '+': do_showrest = 1; break;
case 'A': options |= PCRE_ANCHORED; break;
case 'D': do_debug = do_showinfo = 1; break;
case 'E': options |= PCRE_DOLLAR_ENDONLY; break;
case 'G': do_G = 1; break;
case 'I': do_showinfo = 1; break;
case 'M': log_store = 1; break;
#if !defined NOPOSIX
case 'P': do_posix = 1; break;
#endif
case 'S': do_study = 1; break;
case 'U': options |= PCRE_UNGREEDY; break;
case 'X': options |= PCRE_EXTRA; break;
case '8': options |= PCRE_UTF8; utf8 = 1; break;
case 'L':
ppp = pp;
while (*ppp != '\n' && *ppp != ' ') ppp++;
*ppp = 0;
if (setlocale(LC_CTYPE, (const char *)pp) == NULL)
{
fprintf(outfile, "** Failed to set locale \"%s\"\n", pp);
goto SKIP_DATA;
}
tables = pcre_maketables();
pp = ppp;
break;
case '\n': case ' ': break;
default:
fprintf(outfile, "** Unknown option '%c'\n", pp[-1]);
goto SKIP_DATA;
}
}
/* Handle compiling via the POSIX interface, which doesn't support the
timing, showing, or debugging options, nor the ability to pass over
local character tables. */
#if !defined NOPOSIX
if (posix || do_posix)
{
int rc;
int cflags = 0;
if ((options & PCRE_CASELESS) != 0) cflags |= REG_ICASE;
if ((options & PCRE_MULTILINE) != 0) cflags |= REG_NEWLINE;
rc = regcomp(&preg, (char *)p, cflags);
/* Compilation failed; go back for another re, skipping to blank line
if non-interactive. */
if (rc != 0)
{
(void)regerror(rc, &preg, (char *)buffer, sizeof(buffer));
fprintf(outfile, "Failed: POSIX code %d: %s\n", rc, buffer);
goto SKIP_DATA;
}
}
/* Handle compiling via the native interface */
else
#endif /* !defined NOPOSIX */
{
if (timeit)
{
register int i;
clock_t time_taken;
clock_t start_time = clock();
for (i = 0; i < LOOPREPEAT; i++)
{
re = pcre_compile((char *)p, options, &error, &erroroffset, tables);
if (re != NULL) free(re);
}
time_taken = clock() - start_time;
fprintf(outfile, "Compile time %.3f milliseconds\n",
((double)time_taken * 1000.0) /
((double)LOOPREPEAT * (double)CLOCKS_PER_SEC));
}
re = pcre_compile((char *)p, options, &error, &erroroffset, tables);
/* Compilation failed; go back for another re, skipping to blank line
if non-interactive. */
if (re == NULL)
{
fprintf(outfile, "Failed: %s at offset %d\n", error, erroroffset);
SKIP_DATA:
if (infile != stdin)
{
for (;;)
{
if (fgets((char *)buffer, sizeof(buffer), infile) == NULL)
{
done = 1;
goto CONTINUE;
}
len = (int)strlen((char *)buffer);
while (len > 0 && isspace(buffer[len-1])) len--;
if (len == 0) break;
}
fprintf(outfile, "\n");
}
goto CONTINUE;
}
/* Compilation succeeded; print data if required. There are now two
info-returning functions. The old one has a limited interface and
returns only limited data. Check that it agrees with the newer one. */
if (do_showinfo)
{
unsigned long int get_options;
int old_first_char, old_options, old_count;
int count, backrefmax, first_char, need_char;
size_t size;
if (do_debug) print_internals(re);
new_info(re, NULL, PCRE_INFO_OPTIONS, &get_options);
new_info(re, NULL, PCRE_INFO_SIZE, &size);
new_info(re, NULL, PCRE_INFO_CAPTURECOUNT, &count);
new_info(re, NULL, PCRE_INFO_BACKREFMAX, &backrefmax);
new_info(re, NULL, PCRE_INFO_FIRSTCHAR, &first_char);
new_info(re, NULL, PCRE_INFO_LASTLITERAL, &need_char);
old_count = pcre_info(re, &old_options, &old_first_char);
if (count < 0) fprintf(outfile,
"Error %d from pcre_info()\n", count);
else
{
if (old_count != count) fprintf(outfile,
"Count disagreement: pcre_fullinfo=%d pcre_info=%d\n", count,
old_count);
if (old_first_char != first_char) fprintf(outfile,
"First char disagreement: pcre_fullinfo=%d pcre_info=%d\n",
first_char, old_first_char);
if (old_options != (int)get_options) fprintf(outfile,
"Options disagreement: pcre_fullinfo=%ld pcre_info=%d\n",
get_options, old_options);
}
if (size != gotten_store) fprintf(outfile,
"Size disagreement: pcre_fullinfo=%d call to malloc for %d\n",
size, gotten_store);
fprintf(outfile, "Capturing subpattern count = %d\n", count);
if (backrefmax > 0)
fprintf(outfile, "Max back reference = %d\n", backrefmax);
if (get_options == 0) fprintf(outfile, "No options\n");
else fprintf(outfile, "Options:%s%s%s%s%s%s%s%s%s\n",
((get_options & PCRE_ANCHORED) != 0)? " anchored" : "",
((get_options & PCRE_CASELESS) != 0)? " caseless" : "",
((get_options & PCRE_EXTENDED) != 0)? " extended" : "",
((get_options & PCRE_MULTILINE) != 0)? " multiline" : "",
((get_options & PCRE_DOTALL) != 0)? " dotall" : "",
((get_options & PCRE_DOLLAR_ENDONLY) != 0)? " dollar_endonly" : "",
((get_options & PCRE_EXTRA) != 0)? " extra" : "",
((get_options & PCRE_UNGREEDY) != 0)? " ungreedy" : "",
((get_options & PCRE_UTF8) != 0)? " utf8" : "");
if (((((real_pcre *)re)->options) & PCRE_ICHANGED) != 0)
fprintf(outfile, "Case state changes\n");
if (first_char == -1)
{
fprintf(outfile, "First char at start or follows \\n\n");
}
else if (first_char < 0)
{
fprintf(outfile, "No first char\n");
}
else
{
if (isprint(first_char))
fprintf(outfile, "First char = \'%c\'\n", first_char);
else
fprintf(outfile, "First char = %d\n", first_char);
}
if (need_char < 0)
{
fprintf(outfile, "No need char\n");
}
else
{
if (isprint(need_char))
fprintf(outfile, "Need char = \'%c\'\n", need_char);
else
fprintf(outfile, "Need char = %d\n", need_char);
}
}
/* If /S was present, study the regexp to generate additional info to
help with the matching. */
if (do_study)
{
if (timeit)
{
register int i;
clock_t time_taken;
clock_t start_time = clock();
for (i = 0; i < LOOPREPEAT; i++)
extra = pcre_study(re, study_options, &error);
time_taken = clock() - start_time;
if (extra != NULL) free(extra);
fprintf(outfile, " Study time %.3f milliseconds\n",
((double)time_taken * 1000.0)/
((double)LOOPREPEAT * (double)CLOCKS_PER_SEC));
}
extra = pcre_study(re, study_options, &error);
if (error != NULL)
fprintf(outfile, "Failed to study: %s\n", error);
else if (extra == NULL)
fprintf(outfile, "Study returned NULL\n");
else if (do_showinfo)
{
uschar *start_bits = NULL;
new_info(re, extra, PCRE_INFO_FIRSTTABLE, &start_bits);
if (start_bits == NULL)
fprintf(outfile, "No starting character set\n");
else
{
int i;
int c = 24;
fprintf(outfile, "Starting character set: ");
for (i = 0; i < 256; i++)
{
if ((start_bits[i/8] & (1<<(i%8))) != 0)
{
if (c > 75)
{
fprintf(outfile, "\n ");
c = 2;
}
if (isprint(i) && i != ' ')
{
fprintf(outfile, "%c ", i);
c += 2;
}
else
{
fprintf(outfile, "\\x%02x ", i);
c += 5;
}
}
}
fprintf(outfile, "\n");
}
}
}
}
/* Read data lines and test them */
for (;;)
{
unsigned char *q;
unsigned char *bptr = dbuffer;
int *use_offsets = offsets;
int use_size_offsets = size_offsets;
int count, c;
int copystrings = 0;
int getstrings = 0;
int getlist = 0;
int gmatched = 0;
int start_offset = 0;
int g_notempty = 0;
options = 0;
if (infile == stdin) printf("data> ");
if (fgets((char *)buffer, sizeof(buffer), infile) == NULL)
{
done = 1;
goto CONTINUE;
}
if (infile != stdin) fprintf(outfile, "%s", (char *)buffer);
len = (int)strlen((char *)buffer);
while (len > 0 && isspace(buffer[len-1])) len--;
buffer[len] = 0;
if (len == 0) break;
p = buffer;
while (isspace(*p)) p++;
q = dbuffer;
while ((c = *p++) != 0)
{
int i = 0;
int n = 0;
if (c == '\\') switch ((c = *p++))
{
case 'a': c = 7; break;
case 'b': c = '\b'; break;
case 'e': c = 27; break;
case 'f': c = '\f'; break;
case 'n': c = '\n'; break;
case 'r': c = '\r'; break;
case 't': c = '\t'; break;
case 'v': c = '\v'; break;
case '0': case '1': case '2': case '3':
case '4': case '5': case '6': case '7':
c -= '0';
while (i++ < 2 && isdigit(*p) && *p != '8' && *p != '9')
c = c * 8 + *p++ - '0';
break;
case 'x':
/* Handle \x{..} specially - new Perl thing for utf8 */
if (*p == '{')
{
unsigned char *pt = p;
c = 0;
while (isxdigit(*(++pt)))
c = c * 16 + tolower(*pt) - ((isdigit(*pt))? '0' : 'W');
if (*pt == '}')
{
unsigned char buffer[8];
int ii, utn;
utn = ord2utf8(c, buffer);
for (ii = 0; ii < utn - 1; ii++) *q++ = buffer[ii];
c = buffer[ii]; /* Last byte */
p = pt + 1;
break;
}
/* Not correct form; fall through */
}
/* Ordinary \x */
c = 0;
while (i++ < 2 && isxdigit(*p))
{
c = c * 16 + tolower(*p) - ((isdigit(*p))? '0' : 'W');
p++;
}
break;
case 0: /* Allows for an empty line */
p--;
continue;
case 'A': /* Option setting */
options |= PCRE_ANCHORED;
continue;
case 'B':
options |= PCRE_NOTBOL;
continue;
case 'C':
while(isdigit(*p)) n = n * 10 + *p++ - '0';
copystrings |= 1 << n;
continue;
case 'G':
while(isdigit(*p)) n = n * 10 + *p++ - '0';
getstrings |= 1 << n;
continue;
case 'L':
getlist = 1;
continue;
case 'N':
options |= PCRE_NOTEMPTY;
continue;
case 'O':
while(isdigit(*p)) n = n * 10 + *p++ - '0';
if (n > size_offsets_max)
{
size_offsets_max = n;
free(offsets);
use_offsets = offsets = malloc(size_offsets_max * sizeof(int));
if (offsets == NULL)
{
printf("** Failed to get %d bytes of memory for offsets vector\n",
size_offsets_max * sizeof(int));
return 1;
}
}
use_size_offsets = n;
if (n == 0) use_offsets = NULL;
continue;
case 'Z':
options |= PCRE_NOTEOL;
continue;
}
*q++ = c;
}
*q = 0;
len = q - dbuffer;
/* Handle matching via the POSIX interface, which does not
support timing. */
#if !defined NOPOSIX
if (posix || do_posix)
{
int rc;
int eflags = 0;
regmatch_t *pmatch = malloc(sizeof(regmatch_t) * use_size_offsets);
if ((options & PCRE_NOTBOL) != 0) eflags |= REG_NOTBOL;
if ((options & PCRE_NOTEOL) != 0) eflags |= REG_NOTEOL;
rc = regexec(&preg, (const char *)bptr, use_size_offsets, pmatch, eflags);
if (rc != 0)
{
(void)regerror(rc, &preg, (char *)buffer, sizeof(buffer));
fprintf(outfile, "No match: POSIX code %d: %s\n", rc, buffer);
}
else
{
size_t i;
for (i = 0; i < use_size_offsets; i++)
{
if (pmatch[i].rm_so >= 0)
{
fprintf(outfile, "%2d: ", (int)i);
pchars(dbuffer + pmatch[i].rm_so,
pmatch[i].rm_eo - pmatch[i].rm_so, utf8);
fprintf(outfile, "\n");
if (i == 0 && do_showrest)
{
fprintf(outfile, " 0+ ");
pchars(dbuffer + pmatch[i].rm_eo, len - pmatch[i].rm_eo, utf8);
fprintf(outfile, "\n");
}
}
}
}
free(pmatch);
}
/* Handle matching via the native interface - repeats for /g and /G */
else
#endif /* !defined NOPOSIX */
for (;; gmatched++) /* Loop for /g or /G */
{
if (timeit)
{
register int i;
clock_t time_taken;
clock_t start_time = clock();
for (i = 0; i < LOOPREPEAT; i++)
count = pcre_exec(re, extra, (char *)bptr, len,
start_offset, options | g_notempty, use_offsets, use_size_offsets);
time_taken = clock() - start_time;
fprintf(outfile, "Execute time %.3f milliseconds\n",
((double)time_taken * 1000.0)/
((double)LOOPREPEAT * (double)CLOCKS_PER_SEC));
}
count = pcre_exec(re, extra, (char *)bptr, len,
start_offset, options | g_notempty, use_offsets, use_size_offsets);
if (count == 0)
{
fprintf(outfile, "Matched, but too many substrings\n");
count = use_size_offsets/3;
}
/* Matched */
if (count >= 0)
{
int i;
for (i = 0; i < count * 2; i += 2)
{
if (use_offsets[i] < 0)
fprintf(outfile, "%2d: <unset>\n", i/2);
else
{
fprintf(outfile, "%2d: ", i/2);
pchars(bptr + use_offsets[i], use_offsets[i+1] - use_offsets[i], utf8);
fprintf(outfile, "\n");
if (i == 0)
{
if (do_showrest)
{
fprintf(outfile, " 0+ ");
pchars(bptr + use_offsets[i+1], len - use_offsets[i+1], utf8);
fprintf(outfile, "\n");
}
}
}
}
for (i = 0; i < 32; i++)
{
if ((copystrings & (1 << i)) != 0)
{
char copybuffer[16];
int rc = pcre_copy_substring((char *)bptr, use_offsets, count,
i, copybuffer, sizeof(copybuffer));
if (rc < 0)
fprintf(outfile, "copy substring %d failed %d\n", i, rc);
else
fprintf(outfile, "%2dC %s (%d)\n", i, copybuffer, rc);
}
}
for (i = 0; i < 32; i++)
{
if ((getstrings & (1 << i)) != 0)
{
const char *substring;
int rc = pcre_get_substring((char *)bptr, use_offsets, count,
i, &substring);
if (rc < 0)
fprintf(outfile, "get substring %d failed %d\n", i, rc);
else
{
fprintf(outfile, "%2dG %s (%d)\n", i, substring, rc);
/* free((void *)substring); */
pcre_free_substring(substring);
}
}
}
if (getlist)
{
const char **stringlist;
int rc = pcre_get_substring_list((char *)bptr, use_offsets, count,
&stringlist);
if (rc < 0)
fprintf(outfile, "get substring list failed %d\n", rc);
else
{
for (i = 0; i < count; i++)
fprintf(outfile, "%2dL %s\n", i, stringlist[i]);
if (stringlist[i] != NULL)
fprintf(outfile, "string list not terminated by NULL\n");
/* free((void *)stringlist); */
pcre_free_substring_list(stringlist);
}
}
}
/* Failed to match. If this is a /g or /G loop and we previously set
g_notempty after a null match, this is not necessarily the end.
We want to advance the start offset, and continue. Fudge the offset
values to achieve this. We won't be at the end of the string - that
was checked before setting g_notempty. */
else
{
if (g_notempty != 0)
{
use_offsets[0] = start_offset;
use_offsets[1] = start_offset + 1;
}
else
{
if (gmatched == 0) /* Error if no previous matches */
{
if (count == -1) fprintf(outfile, "No match\n");
else fprintf(outfile, "Error %d\n", count);
}
break; /* Out of the /g loop */
}
}
/* If not /g or /G we are done */
if (!do_g && !do_G) break;
/* If we have matched an empty string, first check to see if we are at
the end of the subject. If so, the /g loop is over. Otherwise, mimic
what Perl's /g options does. This turns out to be rather cunning. First
we set PCRE_NOTEMPTY and PCRE_ANCHORED and try the match again at the
same point. If this fails (picked up above) we advance to the next
character. */
g_notempty = 0;
if (use_offsets[0] == use_offsets[1])
{
if (use_offsets[0] == len) break;
g_notempty = PCRE_NOTEMPTY | PCRE_ANCHORED;
}
/* For /g, update the start offset, leaving the rest alone */
if (do_g) start_offset = use_offsets[1];
/* For /G, update the pointer and length */
else
{
bptr += use_offsets[1];
len -= use_offsets[1];
}
} /* End of loop for /g and /G */
} /* End of loop for data lines */
CONTINUE:
#if !defined NOPOSIX
if (posix || do_posix) regfree(&preg);
#endif
if (re != NULL) free(re);
if (extra != NULL) free(extra);
if (tables != NULL)
{
free((void *)tables);
setlocale(LC_CTYPE, "C");
}
}
fprintf(outfile, "\n");
return 0;
}
static int mod_ssi_handle_request(server *srv, connection *con, plugin_data *p) {
stream s;
#ifdef HAVE_PCRE_H
int i, n;
#define N 10
int ovec[N * 3];
#endif
stat_cache_entry *sce = NULL;
/* get a stream to the file */
array_reset(p->ssi_vars);
array_reset(p->ssi_cgi_env);
buffer_copy_string_len(p->timefmt, CONST_STR_LEN("%a, %d %b %Y %H:%M:%S %Z"));
p->sizefmt = 0;
build_ssi_cgi_vars(srv, con, p);
p->if_is_false = 0;
/* Reset the modified time of included files */
include_file_last_mtime = 0;
if (HANDLER_ERROR == stat_cache_get_entry(srv, con, con->physical.path, &sce)) {
log_error_write(srv, __FILE__, __LINE__, "SB", "stat_cache_get_entry failed: ", con->physical.path);
return -1;
}
if (-1 == stream_open(&s, con->physical.path)) {
log_error_write(srv, __FILE__, __LINE__, "sb",
"stream-open: ", con->physical.path);
return -1;
}
/**
* <!--#element attribute=value attribute=value ... -->
*
* config DONE
* errmsg -- missing
* sizefmt DONE
* timefmt DONE
* echo DONE
* var DONE
* encoding -- missing
* exec DONE
* cgi -- never
* cmd DONE
* fsize DONE
* file DONE
* virtual DONE
* flastmod DONE
* file DONE
* virtual DONE
* include DONE
* file DONE
* virtual DONE
* printenv DONE
* set DONE
* var DONE
* value DONE
*
* if DONE
* elif DONE
* else DONE
* endif DONE
*
*
* expressions
* AND, OR DONE
* comp DONE
* ${...} -- missing
* $... DONE
* '...' DONE
* ( ... ) DONE
*
*
*
* ** all DONE **
* DATE_GMT
* The current date in Greenwich Mean Time.
* DATE_LOCAL
* The current date in the local time zone.
* DOCUMENT_NAME
* The filename (excluding directories) of the document requested by the user.
* DOCUMENT_URI
* The (%-decoded) URL path of the document requested by the user. Note that in the case of nested include files, this is not then URL for the current document.
* LAST_MODIFIED
* The last modification date of the document requested by the user.
* USER_NAME
* Contains the owner of the file which included it.
*
*/
#ifdef HAVE_PCRE_H
for (i = 0; (n = pcre_exec(p->ssi_regex, NULL, s.start, s.size, i, 0, ovec, N * 3)) > 0; i = ovec[1]) {
const char **l;
/* take everything from last offset to current match pos */
if (!p->if_is_false) chunkqueue_append_file(con->write_queue, con->physical.path, i, ovec[0] - i);
pcre_get_substring_list(s.start, ovec, n, &l);
process_ssi_stmt(srv, con, p, l, n, sce);
pcre_free_substring_list(l);
}
switch(n) {
case PCRE_ERROR_NOMATCH:
/* copy everything/the rest */
chunkqueue_append_file(con->write_queue, con->physical.path, i, s.size - i);
break;
default:
log_error_write(srv, __FILE__, __LINE__, "sd",
"execution error while matching: ", n);
break;
}
#endif
stream_close(&s);
con->file_started = 1;
con->file_finished = 1;
con->mode = p->id;
if (p->conf.content_type->used <= 1) {
response_header_overwrite(srv, con, CONST_STR_LEN("Content-Type"), CONST_STR_LEN("text/html"));
} else {
response_header_overwrite(srv, con, CONST_STR_LEN("Content-Type"), CONST_BUF_LEN(p->conf.content_type));
}
{
/* Generate "ETag" & "Last-Modified" headers */
time_t lm_time = 0;
buffer *mtime = NULL;
etag_mutate(con->physical.etag, sce->etag);
response_header_overwrite(srv, con, CONST_STR_LEN("ETag"), CONST_BUF_LEN(con->physical.etag));
if (sce->st.st_mtime > include_file_last_mtime)
lm_time = sce->st.st_mtime;
else
lm_time = include_file_last_mtime;
mtime = strftime_cache_get(srv, lm_time);
response_header_overwrite(srv, con, CONST_STR_LEN("Last-Modified"), CONST_BUF_LEN(mtime));
}
/* Reset the modified time of included files */
include_file_last_mtime = 0;
/* reset physical.path */
buffer_reset(con->physical.path);
return 0;
}
bool eas::Decode(const char *s, Message &message)
{
const char *errptr;
int erroffset;
pcre *re = pcre_compile(
"^ZCZC-(\\w+)-(\\w+)((?:-[^+-]+){1,31})\\+(\\d{2})(\\d{2})-(\\d{3})(\\d{2})(\\d{2})-([^-]+)-",
// 1 2 3 4 5 6 7 8 9
0,
&errptr,
&erroffset,
NULL);
if (re == NULL) {
return false;
}
int ovector[3*10];
int r = pcre_exec(
re,
NULL,
s,
strlen(s),
0,
0,
ovector,
sizeof(ovector)/sizeof(ovector[0]));
pcre_free(re);
if (r < 0) {
return false;
}
const char **matches;
pcre_get_substring_list(s, ovector, r, &matches);
int yday = atoi(matches[6]);
time_t now = time(0);
struct tm *tt;
for (;;) {
tt = gmtime(&now);
if (1+tt->tm_yday == yday) {
break;
} else if (1+tt->tm_yday < yday) {
now += 86400;
} else if (1+tt->tm_yday > yday) {
now -= 86400;
}
}
tt->tm_hour = atoi(matches[7]);
tt->tm_min = atoi(matches[8]);
tt->tm_sec = 0;
message.raw = s;
message.originator = matches[1];
message.originator_desc = getOriginatorDesc(message.originator);
message.event = matches[2];
message.event_desc = getEventDesc(message.event);
message.areas.clear();
string a;
for (const char *p = matches[3]; ; p++) {
if (*p == '-' || *p == 0) {
if (*p == '-') {
p++;
}
if (!a.empty()) {
Message::Area area;
area.code = a;
if (a.length() == 6 && strspn(a.c_str(), "0123456789") == 6) {
area.part = a[0] - '0';
area.state = 10*(a[1] - '0') + (a[2] - '0');
area.county = 10*(10*(a[3] - '0') + (a[4] - '0')) + (a[5] - '0');
}
area.desc = getAreaDesc(area);
message.areas.push_back(area);
}
a.erase();
if (*p == 0) {
break;
}
}
a += *p;
}
message.issued = mkgmtime(tt);
message.received = time(0);
message.purge = message.issued + 60*(60*atoi(matches[4]) + atoi(matches[5]));
message.sender = matches[9];
message.sender_desc = getSenderDesc(message.sender);
pcre_free_substring_list(matches);
return true;
}
void
mud_regex_substring_clear(const gchar **substring_list)
{
pcre_free_substring_list(substring_list);
}