2 * A basic toolset containing miscellaneous functions for string manipluation,
3 * encoding/decoding, and a bunch of other stuff.
11 #include <sys/types.h>
18 #if TIME_WITH_SYS_TIME
19 # include <sys/time.h>
23 # include <sys/time.h>
29 #include "libcitadel.h"
35 typedef unsigned char byte; /* Byte type */
37 /* Base64 encoding table */
38 const byte etable[256] = {
39 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81,
40 82, 83, 84, 85, 86, 87, 88, 89, 90, 97, 98, 99, 100, 101, 102, 103,
41 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117,
42 118, 119, 120, 121, 122, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 43,
43 47, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
44 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
45 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
46 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
47 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
48 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
49 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
50 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
51 0, 0, 0, 0, 0, 0, 0, 0, 0
54 /* Base64 decoding table */
55 const byte dtable[256] = {
56 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
57 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
58 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
59 128, 62, 128, 128, 128, 63, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,
60 128, 128, 128, 0, 128, 128, 128, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
61 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 128, 128, 128,
62 128, 128, 128, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39,
63 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 128, 128, 128, 128,
64 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
65 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
66 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
67 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
68 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
69 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
70 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
71 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
72 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
77 char *safestrncpy(char *dest, const char *src, size_t n)
81 if (dest == NULL || src == NULL) {
82 fprintf(stderr, "safestrncpy: NULL argument\n");
88 if (dest[i] == 0) return(dest);
98 * num_tokens() - discover number of parameters/tokens in a string
100 int num_tokens(const char *source, char tok)
103 const char *ptr = source;
105 if (source == NULL) {
109 while (*ptr != '\0') {
118 //extern void cit_backtrace(void);
122 * extract_token() - a string tokenizer
123 * returns -1 if not found, or length of token.
125 long extract_token(char *dest, const char *source, int parmnum, char separator, int maxlen)
127 const char *s; //* source * /
128 int len = 0; //* running total length of extracted string * /
129 int current_token = 0; //* token currently being processed * /
138 //lprintf (CTDL_DEBUG, "test >: n: %d sep: %c source: %s \n willi \n", parmnum, separator, source);
148 if (*s == separator) {
151 if ( (current_token == parmnum) &&
157 else if ((current_token > parmnum) || (len >= maxlen)) {
164 if (current_token < parmnum) {
165 //lprintf (CTDL_DEBUG,"test <!: %s\n", dest);
168 //lprintf (CTDL_DEBUG,"test <: %d; %s\n", len, dest);
175 * extract_token() - a string tokenizer
177 long extract_token(char *dest, const char *source, int parmnum, char separator, int maxlen)
180 const char *s; // source
186 //lprintf (CTDL_DEBUG, "test >: n: %d sep: %c source: %s \n willi \n", parmnum, separator, source);
189 // Locate desired parameter
191 while (count < parmnum) {
192 // End of string, bail!
197 if (*s == separator) {
203 //lprintf (CTDL_DEBUG,"test <!: %s\n", dest);
204 return -1; // Parameter not found
207 for (d = dest; *s && *s != separator && ++len<maxlen; s++, d++) {
211 //lprintf (CTDL_DEBUG,"test <: %d; %s\n", len, dest);
218 * remove_token() - a tokenizer that kills, maims, and destroys
220 void remove_token(char *source, int parmnum, char separator)
222 char *d, *s; /* dest, source */
225 /* Find desired parameter */
227 while (count < parmnum) {
228 /* End of string, bail! */
233 if (*d == separator) {
238 if (!d) return; /* Parameter not found */
240 /* Find next parameter */
242 while (*s && *s != separator) {
249 else if (d == source)
263 * extract_int() - extract an int parm w/o supplying a buffer
265 int extract_int(const char *source, int parmnum)
269 if (extract_token(buf, source, parmnum, '|', sizeof buf) > 0)
276 * extract_long() - extract an long parm w/o supplying a buffer
278 long extract_long(const char *source, int parmnum)
282 if (extract_token(buf, source, parmnum, '|', sizeof buf) > 0)
290 * extract_unsigned_long() - extract an unsigned long parm
292 unsigned long extract_unsigned_long(const char *source, int parmnum)
296 if (extract_token(buf, source, parmnum, '|', sizeof buf) > 0)
297 return strtoul(buf, NULL, 10);
304 * CtdlDecodeBase64() and CtdlEncodeBase64() are adaptations of code by John Walker.
307 size_t CtdlEncodeBase64(char *dest, const char *source, size_t sourcelen, int linebreaks)
309 int i, hiteof = FALSE;
315 byte igroup[3], ogroup[4];
318 igroup[0] = igroup[1] = igroup[2] = 0;
319 for (n = 0; n < 3; n++) {
320 if (spos >= sourcelen) {
325 igroup[n] = (byte) c;
328 ogroup[0] = etable[igroup[0] >> 2];
330 etable[((igroup[0] & 3) << 4) |
333 etable[((igroup[1] & 0xF) << 2) |
335 ogroup[3] = etable[igroup[2] & 0x3F];
338 * Replace characters in output stream with "=" pad
339 * characters if fewer than three characters were
340 * read from the end of the input stream.
349 for (i = 0; i < 4; i++) {
350 dest[dpos++] = ogroup[i];
354 if ( (linebreaks) && (thisline > 70) ) {
362 if ( (linebreaks) && (thisline > 70) ) {
375 * Convert base64-encoded to binary. Returns the length of the decoded data.
376 * It will stop after reading 'length' bytes.
378 int CtdlDecodeBase64(char *dest, const char *source, size_t length)
385 byte a[4], b[4], o[3];
387 for (i = 0; i < 4; i++) {
388 if (spos >= length) {
399 if (dtable[c] & 0x80) {
400 /* Ignoring errors: discard invalid character. */
405 b[i] = (byte) dtable[c];
407 o[0] = (b[0] << 2) | (b[1] >> 4);
408 o[1] = (b[1] << 4) | (b[2] >> 2);
409 o[2] = (b[2] << 6) | b[3];
410 i = a[2] == '=' ? 1 : (a[3] == '=' ? 2 : 3);
411 if (i>=1) dest[dpos++] = o[0];
412 if (i>=2) dest[dpos++] = o[1];
413 if (i>=3) dest[dpos++] = o[2];
423 * if we send out non ascii subjects, we encode it this way.
425 char *rfc2047encode(char *line, long length)
427 char *AlreadyEncoded;
430 #define UTF8_HEADER "=?UTF-8?B?"
432 /* check if we're already done */
433 AlreadyEncoded = strstr(line, "=?");
434 if ((AlreadyEncoded != NULL) &&
435 ((strstr(AlreadyEncoded, "?B?") != NULL)||
436 (strstr(AlreadyEncoded, "?Q?") != NULL)))
441 result = (char*) malloc(strlen(UTF8_HEADER) + 4 + length * 2);
442 strncpy (result, UTF8_HEADER, strlen (UTF8_HEADER));
443 CtdlEncodeBase64(result + strlen(UTF8_HEADER), line, length, 0);
444 end = strlen (result);
453 * Strip leading and trailing spaces from a string
455 void striplt(char *buf)
460 if (buf==NULL) return;
461 if (IsEmptyStr(buf)) return;
463 while ((!IsEmptyStr(buf)) && (isspace(buf[len - 1])))
465 if (IsEmptyStr(buf)) return;
467 while ((!IsEmptyStr(buf)) && (isspace(buf[a])))
470 memmove(buf, &buf[a], len - a + 1);
478 * \brief check for the presence of a character within a string (returns count)
479 * \param st the string to examine
480 * \param ch the char to search
481 * \return the position inside of st
483 int haschar(const char *st,int ch)
489 while (!IsEmptyStr(ptr))
503 * Format a date/time stamp for output
504 * seconds is whether to print the seconds
506 void fmt_date(char *buf, size_t n, time_t thetime, int seconds) {
510 /* Month strings for date conversions ... this needs to be localized eventually */
511 char *fmt_date_months[12] = {
512 "Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"
516 localtime_r(&thetime, &tm);
519 if (hour == 0) hour = 12;
520 else if (hour > 12) hour = hour - 12;
523 snprintf(buf, n, "%s %d %4d %d:%02d:%02d%s",
524 fmt_date_months[tm.tm_mon],
530 ( (tm.tm_hour >= 12) ? "pm" : "am" )
533 snprintf(buf, n, "%s %d %4d %d:%02d%s",
534 fmt_date_months[tm.tm_mon],
539 ( (tm.tm_hour >= 12) ? "pm" : "am" )
547 * Determine whether the specified message number is contained within the
548 * specified sequence set.
550 int is_msg_in_sequence_set(char *mset, long msgnum) {
553 char setstr[128], lostr[128], histr[128];
556 num_sets = num_tokens(mset, ',');
557 for (s=0; s<num_sets; ++s) {
558 extract_token(setstr, mset, s, ',', sizeof setstr);
560 extract_token(lostr, setstr, 0, ':', sizeof lostr);
561 if (num_tokens(setstr, ':') >= 2) {
562 extract_token(histr, setstr, 1, ':', sizeof histr);
563 if (!strcmp(histr, "*")) {
564 snprintf(histr, sizeof histr, "%ld", LONG_MAX);
568 strcpy(histr, lostr);
573 if ((msgnum >= lo) && (msgnum <= hi)) return(1);
580 * \brief Utility function to "readline" from memory
581 * \param start Location in memory from which we are reading.
582 * \param buf the buffer to place the string in.
583 * \param maxlen Size of string buffer
584 * \return Pointer to the source memory right after we stopped reading.
586 char *memreadline(char *start, char *buf, int maxlen)
590 int len = 0; /**< tally our own length to avoid strlen() delays */
596 if ((len + 1 < (maxlen)) && (ch != 13) && (ch != 10)) {
599 if ((ch == 10) || (ch == 0)) {
608 * \brief Utility function to "readline" from memory
609 * \param start Location in memory from which we are reading.
610 * \param buf the buffer to place the string in.
611 * \param maxlen Size of string buffer
612 * \param retlen the length of the returned string
613 * \return Pointer to the source memory right after we stopped reading.
615 char *memreadlinelen(char *start, char *buf, int maxlen, int *retlen)
619 int len = 0; /**< tally our own length to avoid strlen() delays */
625 if ((len + 1 < (maxlen)) && (ch != 13) && (ch != 10)) {
628 if ((ch == 10) || (ch == 0)) {
640 * Strip a boundarized substring out of a string (for example, remove
641 * parentheses and anything inside them).
643 void stripout(char *str, char leftboundary, char rightboundary) {
648 for (a = 0; a < strlen(str); ++a) {
649 if (str[a] == leftboundary) lb = a;
650 if (str[a] == rightboundary) rb = a;
653 if ( (lb > 0) && (rb > lb) ) {
654 strcpy(&str[lb - 1], &str[rb + 1]);
657 else if ( (lb == 0) && (rb > lb) ) {
658 strcpy(str, &str[rb + 1]);
665 * Reduce a string down to a boundarized substring (for example, remove
666 * parentheses and anything outside them).
668 void stripallbut(char *str, char leftboundary, char rightboundary) {
671 for (a = 0; a < strlen(str); ++ a) {
672 if (str[a] == leftboundary) strcpy(str, &str[a+1]);
675 for (a = 0; a < strlen(str); ++ a) {
676 if (str[a] == rightboundary) str[a] = 0;
681 char *myfgets(char *s, int size, FILE *stream) {
682 char *ret = fgets(s, size, stream);
686 nl = strchr(s, '\n');
696 * \brief Escape a string for feeding out as a URL.
697 * \param outbuf the output buffer
698 * \param oblen the size of outbuf to sanitize
699 * \param strbuf the input buffer
701 void urlesc(char *outbuf, size_t oblen, char *strbuf)
703 int a, b, c, len, eclen, olen;
704 char *ec = " +#&;`'|*?-~<>^()[]{}/$\"\\";
707 len = strlen(strbuf);
710 for (a = 0; a < len; ++a) {
712 for (b = 0; b < eclen; ++b) {
713 if (strbuf[a] == ec[b])
717 snprintf(&outbuf[olen], oblen - olen, "%%%02x", strbuf[a]);
721 outbuf[olen ++] = strbuf[a];
729 * In our world, we want strcpy() to be able to work with overlapping strings.
734 char *strcpy(char *dest, const char *src) {
735 memmove(dest, src, (strlen(src) + 1) );
741 * Generate a new, globally unique UID parameter for a calendar etc. object
743 void generate_uuid(char *buf) {
746 sprintf(buf, "%lx-%lx-%x",
754 * bmstrcasestr() -- case-insensitive substring search
756 * This uses the Boyer-Moore search algorithm and is therefore quite fast.
757 * The code is roughly based on the strstr() replacement from 'tin' written
760 char *bmstrcasestr(char *text, char *pattern) {
762 register unsigned char *p, *t;
763 register int i, j, *delta;
769 textlen = strlen (text);
770 patlen = strlen (pattern);
772 /* algorithm fails if pattern is empty */
773 if ((p1 = patlen) == 0)
776 /* code below fails (whenever i is unsigned) if pattern too long */
782 for (i = 0; i <= 255; i++)
784 for (p = (unsigned char *) pattern, i = p1; --i > 0;)
785 delta[tolower(*p++)] = i;
788 * From now on, we want patlen - 1.
789 * In the loop below, p points to the end of the pattern,
790 * t points to the end of the text to be tested against the
791 * pattern, and i counts the amount of text remaining, not
792 * including the part to be tested.
795 p = (unsigned char *) pattern + p1;
796 t = (unsigned char *) text + p1;
797 i = textlen - patlen;
799 if (tolower(p[0]) == tolower(t[0])) {
800 if (strncasecmp ((const char *)(p - p1), (const char *)(t - p1), p1) == 0) {
801 return ((char *)t - p1);
804 j = delta[tolower(t[0])];
816 * Local replacement for controversial C library function that generates
817 * names for temporary files. Included to shut up compiler warnings.
819 void CtdlMakeTempFileName(char *name, int len) {
822 while (i++, i < 100) {
823 snprintf(name, len, "/tmp/ctdl.%04lx.%04x",
827 if (!access(name, F_OK)) {
836 * Determine whether the specified message number is contained within the specified set.
837 * Returns nonzero if the specified message number is in the specified message set string.
839 int is_msg_in_mset(char *mset, long msgnum) {
842 char setstr[SIZ], lostr[SIZ], histr[SIZ]; /* was 1024 */
846 * Now set it for all specified messages.
848 num_sets = num_tokens(mset, ',');
849 for (s=0; s<num_sets; ++s) {
850 extract_token(setstr, mset, s, ',', sizeof setstr);
852 extract_token(lostr, setstr, 0, ':', sizeof lostr);
853 if (num_tokens(setstr, ':') >= 2) {
854 extract_token(histr, setstr, 1, ':', sizeof histr);
855 if (!strcmp(histr, "*")) {
856 snprintf(histr, sizeof histr, "%ld", LONG_MAX);
860 strcpy(histr, lostr);
865 if ((msgnum >= lo) && (msgnum <= hi)) return(1);
873 * \brief searches for a paternn within asearch string
874 * \param search the string to search
875 * \param patn the pattern to find in string
876 * \returns position in string
878 int pattern2(char *search, char *patn)
882 len = strlen (search);
883 plen = strlen (patn);
884 for (a = 0; a < len; ++a) {
885 if (!strncasecmp(&search[a], patn, plen))
893 * \brief Strip leading and trailing spaces from a string; with premeasured and adjusted length.
894 * \param buf the string to modify
895 * \param len length of the string.
897 void stripltlen(char *buf, int *len)
900 if (*len == 0) return;
901 while ((*len > delta) && (isspace(buf[delta]))){
904 memmove (buf, &buf[delta], *len - delta + 1);
907 if (*len == 0) return;
908 while (isspace(buf[(*len) - 1])){
909 buf[--(*len)] = '\0';