[disorder] / lib / test.c

/*
 * This file is part of DisOrder.
 * Copyright (C) 2005, 2007 Richard Kettlewell
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
 * USA
 */
/** @file lib/test.c @brief Library tests */

#include <config.h>
#include "types.h"

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <errno.h>
#include <ctype.h>
#include <assert.h>
#include <sys/types.h>
#include <sys/stat.h>

#include "utf8.h"
#include "mem.h"
#include "log.h"
#include "vector.h"
#include "charset.h"
#include "mime.h"
#include "hex.h"
#include "words.h"
#include "heap.h"
#include "unicode.h"
#include "inputline.h"

static int tests, errors;

/** @brief Checks that @p expr is nonzero */
#define insist(expr) do {				\
  if(!(expr)) {						\
    ++errors;						\
    fprintf(stderr, "%s:%d: error checking %s\n",	\
            __FILE__, __LINE__, #expr);			\
  }							\
  ++tests;						\
} while(0)

static const char *format(const char *s) {
  struct dynstr d;
  int c;
  char buf[10];
  
  dynstr_init(&d);
  while((c = (unsigned char)*s++)) {
    if(c >= ' ' && c <= '~')
      dynstr_append(&d, c);
    else {
      sprintf(buf, "\\x%02X", (unsigned)c);
      dynstr_append_string(&d, buf);
    }
  }
  dynstr_terminate(&d);
  return d.vec;
}

static const char *format_utf32(const uint32_t *s) {
  struct dynstr d;
  uint32_t c;
  char buf[64];
  
  dynstr_init(&d);
  while((c = *s++)) {
    if(c >= 32 && c <= 127)
      dynstr_append(&d, c);
    else {
      sprintf(buf, "\\x%04lX", (unsigned long)c);
      dynstr_append_string(&d, buf);
    }
  }
  dynstr_terminate(&d);
  return d.vec;
}

#define check_string(GOT, WANT) do {				\
  const char *g = GOT;						\
  const char *w = WANT;						\
								\
  if(w == 0) {							\
    fprintf(stderr, "%s:%d: %s returned 0\n",			\
            __FILE__, __LINE__, #GOT);				\
    ++errors;							\
  } else if(strcmp(w, g)) {					\
    fprintf(stderr, "%s:%d: %s returned:\n%s\nexpected:\n%s\n",	\
	    __FILE__, __LINE__, #GOT, format(g), format(w));	\
    ++errors;							\
  }								\
  ++tests;							\
 } while(0)

static uint32_t *ucs4parse(const char *s) {
  struct dynstr_ucs4 d;
  char *e;

  dynstr_ucs4_init(&d);
  while(*s) {
    errno = 0;
    dynstr_ucs4_append(&d, strtoul(s, &e, 0));
    if(errno) fatal(errno, "strtoul (%s)", s);
    s = e;
  }
  dynstr_ucs4_terminate(&d);
  return d.vec;
}

static void test_utf8(void) {
  /* Test validutf8, convert to UCS-4, check the answer is right,
   * convert back to UTF-8, check we got to where we started */
#define U8(CHARS, WORDS) do {			\
  uint32_t *w = ucs4parse(WORDS);		\
  uint32_t *ucs;				\
  char *u8;					\
						\
  insist(validutf8(CHARS));			\
  ucs = utf82ucs4(CHARS);			\
  insist(ucs != 0);				\
  insist(!ucs4cmp(w, ucs));			\
  u8 = ucs42utf8(ucs);				\
  insist(u8 != 0);				\
  insist(!strcmp(u8, CHARS));			\
} while(0)

  fprintf(stderr, "test_utf8\n");

  /* empty string */

  U8("", "");
  
  /* ASCII characters */

  U8(" !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~",
     "0x20 0x21 0x22 0x23 0x24 0x25 0x26 0x27 0x28 0x29 0x2a 0x2b 0x2c 0x2d "
     "0x2e 0x2f 0x30 0x31 0x32 0x33 0x34 0x35 0x36 0x37 0x38 0x39 0x3a "
     "0x3b 0x3c 0x3d 0x3e 0x3f 0x40 0x41 0x42 0x43 0x44 0x45 0x46 0x47 "
     "0x48 0x49 0x4a 0x4b 0x4c 0x4d 0x4e 0x4f 0x50 0x51 0x52 0x53 0x54 "
     "0x55 0x56 0x57 0x58 0x59 0x5a 0x5b 0x5c 0x5d 0x5e 0x5f 0x60 0x61 "
     "0x62 0x63 0x64 0x65 0x66 0x67 0x68 0x69 0x6a 0x6b 0x6c 0x6d 0x6e "
     "0x6f 0x70 0x71 0x72 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7a 0x7b "
     "0x7c 0x7d 0x7e");
  U8("\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\177",
     "0x1 0x2 0x3 0x4 0x5 0x6 0x7 0x8 0x9 0xa 0xb 0xc 0xd 0xe 0xf 0x10 "
     "0x11 0x12 0x13 0x14 0x15 0x16 0x17 0x18 0x19 0x1a 0x1b 0x1c 0x1d "
     "0x1e 0x1f 0x7f");

  /* from RFC3629 */

  /* UTF8-2      = %xC2-DF UTF8-tail */
  insist(!validutf8("\xC0\x80"));
  insist(!validutf8("\xC1\x80"));
  insist(!validutf8("\xC2\x7F"));
  U8("\xC2\x80", "0x80");
  U8("\xDF\xBF", "0x7FF");
  insist(!validutf8("\xDF\xC0"));

  /*  UTF8-3      = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
   *                %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
   */
  insist(!validutf8("\xE0\x9F\x80"));
  U8("\xE0\xA0\x80", "0x800");
  U8("\xE0\xBF\xBF", "0xFFF");
  insist(!validutf8("\xE0\xC0\xBF"));

  insist(!validutf8("\xE1\x80\x7F"));
  U8("\xE1\x80\x80", "0x1000");
  U8("\xEC\xBF\xBF", "0xCFFF");
  insist(!validutf8("\xEC\xC0\xBF"));
  
  U8("\xED\x80\x80", "0xD000");
  U8("\xED\x9F\xBF", "0xD7FF");
  insist(!validutf8("\xED\xA0\xBF"));

  insist(!validutf8("\xEE\x7f\x80"));
  U8("\xEE\x80\x80", "0xE000");
  U8("\xEF\xBF\xBF", "0xFFFF");
  insist(!validutf8("\xEF\xC0\xBF"));

  /*  UTF8-4      = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
   *                %xF4 %x80-8F 2( UTF8-tail )
   */
  insist(!validutf8("\xF0\x8F\x80\x80"));
  U8("\xF0\x90\x80\x80", "0x10000");
  U8("\xF0\xBF\xBF\xBF", "0x3FFFF");
  insist(!validutf8("\xF0\xC0\x80\x80"));

  insist(!validutf8("\xF1\x80\x80\x7F"));
  U8("\xF1\x80\x80\x80", "0x40000");
  U8("\xF3\xBF\xBF\xBF", "0xFFFFF");
  insist(!validutf8("\xF3\xC0\x80\x80"));

  insist(!validutf8("\xF4\x80\x80\x7F"));
  U8("\xF4\x80\x80\x80", "0x100000");
  U8("\xF4\x8F\xBF\xBF", "0x10FFFF");
  insist(!validutf8("\xF4\x90\x80\x80"));

  /* miscellaneous non-UTF-8 rubbish */
  insist(!validutf8("\x80"));
  insist(!validutf8("\xBF"));
  insist(!validutf8("\xC0"));
  insist(!validutf8("\xC0\x7F"));
  insist(!validutf8("\xC0\xC0"));
  insist(!validutf8("\xE0"));
  insist(!validutf8("\xE0\x7F"));
  insist(!validutf8("\xE0\xC0"));
  insist(!validutf8("\xE0\x80"));
  insist(!validutf8("\xE0\x80\x7f"));
  insist(!validutf8("\xE0\x80\xC0"));
  insist(!validutf8("\xF0"));
  insist(!validutf8("\xF0\x7F"));
  insist(!validutf8("\xF0\xC0"));
  insist(!validutf8("\xF0\x80"));
  insist(!validutf8("\xF0\x80\x7f"));
  insist(!validutf8("\xF0\x80\xC0"));
  insist(!validutf8("\xF0\x80\x80\x7f"));
  insist(!validutf8("\xF0\x80\x80\xC0"));
  insist(!validutf8("\xF5\x80\x80\x80"));
  insist(!validutf8("\xF8"));
}

static void test_mime(void) {
  char *t, *n, *v;

  fprintf(stderr, "test_mime\n");

  t = n = v = 0;
  insist(!mime_content_type("text/plain", &t, &n, &v));
  insist(!strcmp(t, "text/plain"));
  insist(n == 0);
  insist(v == 0);

  t = n = v = 0;
  insist(!mime_content_type("TEXT ((nested) comment) /plain", &t, &n, &v));
  insist(!strcmp(t, "text/plain"));
  insist(n == 0);
  insist(v == 0);

  t = n = v = 0;
  insist(!mime_content_type(" text/plain ; Charset=utf-8", &t, &n, &v));
  insist(!strcmp(t, "text/plain"));
  insist(!strcmp(n, "charset"));
  insist(!strcmp(v, "utf-8"));

  t = n = v = 0;
  insist(!mime_content_type("text/plain;charset = ISO-8859-1 ", &t, &n, &v));
  insist(!strcmp(t, "text/plain"));
  insist(!strcmp(n, "charset"));
  insist(!strcmp(v, "ISO-8859-1"));

  /* XXX mime_parse */
  /* XXX mime_multipart */
  /* XXX mime_rfc2388_content_disposition */

  check_string(mime_qp(""), "");
  check_string(mime_qp("foobar"), "foobar");
  check_string(mime_qp("foo=20bar"), "foo bar");
  check_string(mime_qp("x \r\ny"), "x\r\ny");
  check_string(mime_qp("x=\r\ny"), "xy");
  check_string(mime_qp("x= \r\ny"), "xy");
  check_string(mime_qp("x =\r\ny"), "x y");
  check_string(mime_qp("x = \r\ny"), "x y");

  /* from RFC2045 */
  check_string(mime_qp("Now's the time =\r\n"
"for all folk to come=\r\n"
" to the aid of their country."),
	       "Now's the time for all folk to come to the aid of their country.");

  check_string(mime_base64(""),  "");
  check_string(mime_base64("BBBB"), "\x04\x10\x41");
  check_string(mime_base64("////"), "\xFF\xFF\xFF");
  check_string(mime_base64("//BB"), "\xFF\xF0\x41");
  check_string(mime_base64("BBBB//BB////"),
	       "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
  check_string(mime_base64("B B B B  / / B B / / / /"),
	       "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
  check_string(mime_base64("B\r\nBBB.// B-B//~//"),
	       "\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
  check_string(mime_base64("BBBB="),
	       "\x04\x10\x41");
  check_string(mime_base64("BBBBx="),	/* not actually valid base64 */
	       "\x04\x10\x41");
  check_string(mime_base64("BBBB BB=="),
	       "\x04\x10\x41" "\x04");
  check_string(mime_base64("BBBB BBB="),
	       "\x04\x10\x41" "\x04\x10");
}

static void test_hex(void) {
  unsigned n;
  static const unsigned char h[] = { 0x00, 0xFF, 0x80, 0x7F };
  uint8_t *u;
  size_t ul;

  fprintf(stderr, "test_hex\n");

  for(n = 0; n <= UCHAR_MAX; ++n) {
    if(!isxdigit(n))
      insist(unhexdigitq(n) == -1);
  }
  insist(unhexdigitq('0') == 0);
  insist(unhexdigitq('1') == 1);
  insist(unhexdigitq('2') == 2);
  insist(unhexdigitq('3') == 3);
  insist(unhexdigitq('4') == 4);
  insist(unhexdigitq('5') == 5);
  insist(unhexdigitq('6') == 6);
  insist(unhexdigitq('7') == 7);
  insist(unhexdigitq('8') == 8);
  insist(unhexdigitq('9') == 9);
  insist(unhexdigitq('a') == 10);
  insist(unhexdigitq('b') == 11);
  insist(unhexdigitq('c') == 12);
  insist(unhexdigitq('d') == 13);
  insist(unhexdigitq('e') == 14);
  insist(unhexdigitq('f') == 15);
  insist(unhexdigitq('A') == 10);
  insist(unhexdigitq('B') == 11);
  insist(unhexdigitq('C') == 12);
  insist(unhexdigitq('D') == 13);
  insist(unhexdigitq('E') == 14);
  insist(unhexdigitq('F') == 15);
  check_string(hex(h, sizeof h), "00ff807f");
  check_string(hex(0, 0), "");
  u = unhex("00ff807f", &ul);
  insist(ul == 4);
  insist(memcmp(u, h, 4) == 0);
  u = unhex("00FF807F", &ul);
  insist(ul == 4);
  insist(memcmp(u, h, 4) == 0);
  u = unhex("", &ul);
  insist(ul == 0);
  fprintf(stderr, "2 ERROR reports expected {\n");
  insist(unhex("F", 0) == 0);
  insist(unhex("az", 0) == 0);
  fprintf(stderr, "}\n");
}

static void test_casefold(void) {
  uint32_t c, l;
  const char *input, *folded, *expected;

  fprintf(stderr, "test_casefold\n");
  
  for(c = 1; c < 256; ++c) {
    input = utf32_to_utf8(&c, 1, 0);
    folded = utf8_casefold_canon(input, strlen(input), 0);
    switch(c) {
    default:
      if((c >= 'A' && c <= 'Z')
	 || (c >= 0xC0 && c <= 0xDE && c != 0xD7))
	l = c ^ 0x20;
      else
	l = c;
      break;
    case 0xB5:				/* MICRO SIGN */
      l = 0x3BC;			/* GREEK SMALL LETTER MU */
      break;
    case 0xDF:				/* LATIN SMALL LETTER SHARP S */
      insist(!strcmp(folded, "ss"));
      l = 0;
      break;
    }
    if(l) {
      /* Case-folded data is now normalized */
      expected = ucs42utf8(utf32_decompose_canon(&l, 1, 0));
      if(strcmp(folded, expected)) {
	fprintf(stderr, "%s:%d: casefolding %#lx got '%s', expected '%s'\n",
		__FILE__, __LINE__, (unsigned long)c,
		format(folded), format(expected));
	++errors;
      }
      ++tests;
    }
  }
  check_string(casefold(""), "");
}

/** @brief Less-than comparison function for integer heap */
static inline int int_lt(int a, int b) { return a < b; }

/** @struct iheap
 * @brief A heap with @c int elements */
HEAP_TYPE(iheap, int, int_lt);
HEAP_DEFINE(iheap, int, int_lt);

/** @brief Tests for @ref heap.h */
static void test_heap(void) {
  struct iheap h[1];
  int n;
  int last = -1;

  fprintf(stderr, "test_heap\n");

  iheap_init(h);
  for(n = 0; n < 1000; ++n)
    iheap_insert(h, random() % 100);
  for(n = 0; n < 1000; ++n) {
    const int latest = iheap_remove(h);
    if(last > latest)
      fprintf(stderr, "should have %d <= %d\n", last, latest);
    insist(last <= latest);
    last = latest;
  }
  putchar('\n');
}

/** @brief Tests for @ref lib/unicode.h */
static void test_unicode(void) {
  FILE *fp;
  int lineno = 0;
  char *l, *lp;
  uint32_t buffer[1024];
  uint32_t *c[6], *NFD_c[6],  *NFKD_c[6]; /* 1-indexed */
  int cn, bn;

  fprintf(stderr, "test_unicode\n");
  if(!(fp = fopen("NormalizationTest.txt", "r"))) {
    system("wget http://www.unicode.org/Public/5.0.0/ucd/NormalizationTest.txt");
    chmod("NormalizationTest.txt", 0444);
    if(!(fp = fopen("NormalizationTest.txt", "r"))) {
      perror("NormalizationTest.txt");
      ++tests;				/* don't know how many... */
      ++errors;
      return;
    }
  }
  while(!inputline("NormalizationTest.txt", fp, &l, '\n')) {
    ++lineno;
    if(*l == '#' || *l == '@')
      continue;
    bn = 0;
    cn = 1;
    lp = l;
    c[cn++] = &buffer[bn];
    while(*lp && *lp != '#') {
      if(*lp == ' ') {
	++lp;
	continue;
      }
      if(*lp == ';') {
	buffer[bn++] = 0;
	if(cn == 6)
	  break;
	c[cn++] = &buffer[bn];
	++lp;
	continue;
      }
      buffer[bn++] = strtoul(lp, &lp, 16);
    }
    buffer[bn] = 0;
    assert(cn == 6);
    for(cn = 1; cn <= 5; ++cn) {
      NFD_c[cn] = utf32_decompose_canon(c[cn], utf32_len(c[cn]), 0);
      NFKD_c[cn] = utf32_decompose_compat(c[cn], utf32_len(c[cn]), 0);
    }
#define unt_check(T, A, B) do {					\
    ++tests;							\
    if(utf32_cmp(c[A], T##_c[B])) {				\
      fprintf(stderr, "L%d: c%d != "#T"(c%d)\n", lineno, A, B);	\
      fprintf(stderr, "    c%d:      %s\n",			\
              A, format_utf32(c[A]));				\
      fprintf(stderr, "%4s(c%d): %s\n",				\
              #T, B, format_utf32(T##_c[B]));			\
      ++errors;							\
    }								\
  } while(0)
    unt_check(NFD, 3, 1);
    unt_check(NFD, 3, 2);
    unt_check(NFD, 3, 3);
    unt_check(NFD, 5, 4);
    unt_check(NFD, 5, 5);
    unt_check(NFKD, 5, 1);
    unt_check(NFKD, 5, 2);
    unt_check(NFKD, 5, 3);
    unt_check(NFKD, 5, 4);
    unt_check(NFKD, 5, 5);
    for(cn = 1; cn <= 5; ++cn) {
      xfree(NFD_c[cn]);
      xfree(NFKD_c[cn]);
    }
    xfree(l);
  }
}

int main(void) {
  insist('\n' == 0x0A);
  insist('\r' == 0x0D);
  insist(' ' == 0x20);
  insist('0' == 0x30);
  insist('9' == 0x39);
  insist('A' == 0x41);
  insist('Z' == 0x5A);
  insist('a' == 0x61);
  insist('z' == 0x7A);
  /* addr.c */
  /* asprintf.c */
  /* authhash.c */
  /* basen.c */
  /* charset.c */
  /* client.c */
  /* configuration.c */
  /* event.c */
  /* fprintf.c */
  /* heap.c */
  test_heap();
  /* hex.c */
  test_hex();
  /* inputline.c */
  /* kvp.c */
  /* log.c */
  /* mem.c */
  /* mime.c */
  test_mime();
  /* mixer.c */
  /* plugin.c */
  /* printf.c */
  /* queue.c */
  /* sink.c */
  /* snprintf.c */
  /* split.c */
  /* syscalls.c */
  /* table.c */
  /* unicode.c */
  test_unicode();
  /* utf8.c */
  test_utf8();
  /* vector.c */
  /* words.c */
  test_casefold();
  /* XXX words() */
  /* wstat.c */
  fprintf(stderr,  "%d errors out of %d tests\n", errors, tests);
  return !!errors;
}
  
/*
Local Variables:
c-basic-offset:2
comment-column:40
End:
*/
Commit	Line	Data
460b9539	1	/*
460b9539	2	* This file is part of DisOrder.
033fd4e3	3	* Copyright (C) 2005, 2007 Richard Kettlewell
460b9539	4	*
	5	* This program is free software; you can redistribute it and/or modify
	6	* it under the terms of the GNU General Public License as published by
	7	* the Free Software Foundation; either version 2 of the License, or
	8	* (at your option) any later version.
	9	*
	10	* This program is distributed in the hope that it will be useful, but
	11	* WITHOUT ANY WARRANTY; without even the implied warranty of
	12	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
	13	* General Public License for more details.
	14	*
	15	* You should have received a copy of the GNU General Public License
	16	* along with this program; if not, write to the Free Software
	17	* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
	18	* USA
	19	*/
033fd4e3	20	/** @file lib/test.c @brief Library tests */
460b9539	21
	22	#include <config.h>
	23	#include "types.h"
	24
	25	#include <stdio.h>
	26	#include <string.h>
	27	#include <stdlib.h>
	28	#include <errno.h>
	29	#include <ctype.h>
033fd4e3	30	#include <assert.h>
e5a5a138 RK	31	#include <sys/types.h>
e5a5a138 RK	32	#include <sys/stat.h>
460b9539	33
	34	#include "utf8.h"
	35	#include "mem.h"
	36	#include "log.h"
	37	#include "vector.h"
	38	#include "charset.h"
	39	#include "mime.h"
	40	#include "hex.h"
	41	#include "words.h"
033fd4e3	42	#include "heap.h"
e5a5a138 RK	43	#include "unicode.h"
e5a5a138 RK	44	#include "inputline.h"
460b9539	45
	46	static int tests, errors;
	47
033fd4e3	48	/** @brief Checks that @p expr is nonzero */
460b9539	49	#define insist(expr) do { \
033fd4e3	50	if(!(expr)) { \
460b9539	51	++errors; \
	52	fprintf(stderr, "%s:%d: error checking %s\n", \
	53	__FILE__, __LINE__, #expr); \
	54	} \
	55	++tests; \
	56	} while(0)
	57
	58	static const char format(const char s) {
	59	struct dynstr d;
	60	int c;
	61	char buf[10];
	62
	63	dynstr_init(&d);
	64	while((c = (unsigned char)*s++)) {
	65	if(c >= ' ' && c <= '~')
	66	dynstr_append(&d, c);
	67	else {
	68	sprintf(buf, "\\x%02X", (unsigned)c);
	69	dynstr_append_string(&d, buf);
	70	}
	71	}
	72	dynstr_terminate(&d);
	73	return d.vec;
	74	}
	75
e5a5a138 RK	76	static const char format_utf32(const uint32_t s) {
	77	struct dynstr d;
	78	uint32_t c;
	79	char buf[64];
	80
	81	dynstr_init(&d);
	82	while((c = *s++)) {
	83	if(c >= 32 && c <= 127)
	84	dynstr_append(&d, c);
	85	else {
	86	sprintf(buf, "\\x%04lX", (unsigned long)c);
	87	dynstr_append_string(&d, buf);
	88	}
	89	}
	90	dynstr_terminate(&d);
	91	return d.vec;
	92	}
	93
460b9539	94	#define check_string(GOT, WANT) do { \
	95	const char *g = GOT; \
	96	const char *w = WANT; \
	97	\
	98	if(w == 0) { \
	99	fprintf(stderr, "%s:%d: %s returned 0\n", \
	100	__FILE__, __LINE__, #GOT); \
	101	++errors; \
	102	} else if(strcmp(w, g)) { \
	103	fprintf(stderr, "%s:%d: %s returned:\n%s\nexpected:\n%s\n", \
	104	__FILE__, __LINE__, #GOT, format(g), format(w)); \
	105	++errors; \
	106	} \
	107	++tests; \
	108	} while(0)
	109
	110	static uint32_t ucs4parse(const char s) {
	111	struct dynstr_ucs4 d;
	112	char *e;
	113
	114	dynstr_ucs4_init(&d);
	115	while(*s) {
	116	errno = 0;
	117	dynstr_ucs4_append(&d, strtoul(s, &e, 0));
	118	if(errno) fatal(errno, "strtoul (%s)", s);
	119	s = e;
	120	}
	121	dynstr_ucs4_terminate(&d);
	122	return d.vec;
	123	}
	124
	125	static void test_utf8(void) {
	126	/* Test validutf8, convert to UCS-4, check the answer is right,
	127	* convert back to UTF-8, check we got to where we started */
	128	#define U8(CHARS, WORDS) do { \
	129	uint32_t *w = ucs4parse(WORDS); \
	130	uint32_t *ucs; \
	131	char *u8; \
	132	\
	133	insist(validutf8(CHARS)); \
	134	ucs = utf82ucs4(CHARS); \
	135	insist(ucs != 0); \
	136	insist(!ucs4cmp(w, ucs)); \
	137	u8 = ucs42utf8(ucs); \
	138	insist(u8 != 0); \
	139	insist(!strcmp(u8, CHARS)); \
	140	} while(0)
	141
033fd4e3 RK	142	fprintf(stderr, "test_utf8\n");
033fd4e3 RK	143
460b9539	144	/* empty string */
	145
	146	U8("", "");
	147
	148	/* ASCII characters */
	149
	150	U8(" !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{\|}~",
	151	"0x20 0x21 0x22 0x23 0x24 0x25 0x26 0x27 0x28 0x29 0x2a 0x2b 0x2c 0x2d "
	152	"0x2e 0x2f 0x30 0x31 0x32 0x33 0x34 0x35 0x36 0x37 0x38 0x39 0x3a "
	153	"0x3b 0x3c 0x3d 0x3e 0x3f 0x40 0x41 0x42 0x43 0x44 0x45 0x46 0x47 "
	154	"0x48 0x49 0x4a 0x4b 0x4c 0x4d 0x4e 0x4f 0x50 0x51 0x52 0x53 0x54 "
	155	"0x55 0x56 0x57 0x58 0x59 0x5a 0x5b 0x5c 0x5d 0x5e 0x5f 0x60 0x61 "
	156	"0x62 0x63 0x64 0x65 0x66 0x67 0x68 0x69 0x6a 0x6b 0x6c 0x6d 0x6e "
	157	"0x6f 0x70 0x71 0x72 0x73 0x74 0x75 0x76 0x77 0x78 0x79 0x7a 0x7b "
	158	"0x7c 0x7d 0x7e");
	159	U8("\001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037\177",
	160	"0x1 0x2 0x3 0x4 0x5 0x6 0x7 0x8 0x9 0xa 0xb 0xc 0xd 0xe 0xf 0x10 "
	161	"0x11 0x12 0x13 0x14 0x15 0x16 0x17 0x18 0x19 0x1a 0x1b 0x1c 0x1d "
	162	"0x1e 0x1f 0x7f");
	163
	164	/* from RFC3629 */
	165
	166	/* UTF8-2 = %xC2-DF UTF8-tail */
	167	insist(!validutf8("\xC0\x80"));
	168	insist(!validutf8("\xC1\x80"));
	169	insist(!validutf8("\xC2\x7F"));
	170	U8("\xC2\x80", "0x80");
	171	U8("\xDF\xBF", "0x7FF");
	172	insist(!validutf8("\xDF\xC0"));
	173
	174	/* UTF8-3 = %xE0 %xA0-BF UTF8-tail / %xE1-EC 2( UTF8-tail ) /
	175	* %xED %x80-9F UTF8-tail / %xEE-EF 2( UTF8-tail )
	176	*/
	177	insist(!validutf8("\xE0\x9F\x80"));
	178	U8("\xE0\xA0\x80", "0x800");
	179	U8("\xE0\xBF\xBF", "0xFFF");
	180	insist(!validutf8("\xE0\xC0\xBF"));
	181
	182	insist(!validutf8("\xE1\x80\x7F"));
	183	U8("\xE1\x80\x80", "0x1000");
	184	U8("\xEC\xBF\xBF", "0xCFFF");
	185	insist(!validutf8("\xEC\xC0\xBF"));
	186
	187	U8("\xED\x80\x80", "0xD000");
	188	U8("\xED\x9F\xBF", "0xD7FF");
	189	insist(!validutf8("\xED\xA0\xBF"));
	190
	191	insist(!validutf8("\xEE\x7f\x80"));
	192	U8("\xEE\x80\x80", "0xE000");
	193	U8("\xEF\xBF\xBF", "0xFFFF");
	194	insist(!validutf8("\xEF\xC0\xBF"));
	195
	196	/* UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
	197	* %xF4 %x80-8F 2( UTF8-tail )
	198	*/
	199	insist(!validutf8("\xF0\x8F\x80\x80"));
	200	U8("\xF0\x90\x80\x80", "0x10000");
	201	U8("\xF0\xBF\xBF\xBF", "0x3FFFF");
	202	insist(!validutf8("\xF0\xC0\x80\x80"));
	203
	204	insist(!validutf8("\xF1\x80\x80\x7F"));
	205	U8("\xF1\x80\x80\x80", "0x40000");
	206	U8("\xF3\xBF\xBF\xBF", "0xFFFFF");
	207	insist(!validutf8("\xF3\xC0\x80\x80"));
208
209	insist(!validutf8("\xF4\x80\x80\x7F"));
210	U8("\xF4\x80\x80\x80", "0x100000");
211	U8("\xF4\x8F\xBF\xBF", "0x10FFFF");
212	insist(!validutf8("\xF4\x90\x80\x80"));
213
214	/* miscellaneous non-UTF-8 rubbish */
215	insist(!validutf8("\x80"));
216	insist(!validutf8("\xBF"));
217	insist(!validutf8("\xC0"));
218	insist(!validutf8("\xC0\x7F"));
219	insist(!validutf8("\xC0\xC0"));
220	insist(!validutf8("\xE0"));
221	insist(!validutf8("\xE0\x7F"));
222	insist(!validutf8("\xE0\xC0"));
223	insist(!validutf8("\xE0\x80"));
224	insist(!validutf8("\xE0\x80\x7f"));
225	insist(!validutf8("\xE0\x80\xC0"));
226	insist(!validutf8("\xF0"));
227	insist(!validutf8("\xF0\x7F"));
228	insist(!validutf8("\xF0\xC0"));
229	insist(!validutf8("\xF0\x80"));
230	insist(!validutf8("\xF0\x80\x7f"));
231	insist(!validutf8("\xF0\x80\xC0"));
232	insist(!validutf8("\xF0\x80\x80\x7f"));
233	insist(!validutf8("\xF0\x80\x80\xC0"));
234	insist(!validutf8("\xF5\x80\x80\x80"));
235	insist(!validutf8("\xF8"));
236	}
237
238	static void test_mime(void) {
239	char t, n, *v;
240
033fd4e3 RK	241	fprintf(stderr, "test_mime\n");
033fd4e3 RK	242
460b9539	243	t = n = v = 0;
	244	insist(!mime_content_type("text/plain", &t, &n, &v));
	245	insist(!strcmp(t, "text/plain"));
	246	insist(n == 0);
	247	insist(v == 0);
	248
	249	t = n = v = 0;
	250	insist(!mime_content_type("TEXT ((nested) comment) /plain", &t, &n, &v));
	251	insist(!strcmp(t, "text/plain"));
	252	insist(n == 0);
	253	insist(v == 0);
	254
	255	t = n = v = 0;
	256	insist(!mime_content_type(" text/plain ; Charset=utf-8", &t, &n, &v));
	257	insist(!strcmp(t, "text/plain"));
	258	insist(!strcmp(n, "charset"));
	259	insist(!strcmp(v, "utf-8"));
	260
	261	t = n = v = 0;
	262	insist(!mime_content_type("text/plain;charset = ISO-8859-1 ", &t, &n, &v));
	263	insist(!strcmp(t, "text/plain"));
	264	insist(!strcmp(n, "charset"));
	265	insist(!strcmp(v, "ISO-8859-1"));
	266
	267	/* XXX mime_parse */
	268	/* XXX mime_multipart */
	269	/* XXX mime_rfc2388_content_disposition */
	270
	271	check_string(mime_qp(""), "");
	272	check_string(mime_qp("foobar"), "foobar");
	273	check_string(mime_qp("foo=20bar"), "foo bar");
	274	check_string(mime_qp("x \r\ny"), "x\r\ny");
	275	check_string(mime_qp("x=\r\ny"), "xy");
	276	check_string(mime_qp("x= \r\ny"), "xy");
	277	check_string(mime_qp("x =\r\ny"), "x y");
	278	check_string(mime_qp("x = \r\ny"), "x y");
	279
	280	/* from RFC2045 */
	281	check_string(mime_qp("Now's the time =\r\n"
	282	"for all folk to come=\r\n"
	283	" to the aid of their country."),
	284	"Now's the time for all folk to come to the aid of their country.");
	285
	286	check_string(mime_base64(""), "");
	287	check_string(mime_base64("BBBB"), "\x04\x10\x41");
	288	check_string(mime_base64("////"), "\xFF\xFF\xFF");
	289	check_string(mime_base64("//BB"), "\xFF\xF0\x41");
	290	check_string(mime_base64("BBBB//BB////"),
	291	"\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
	292	check_string(mime_base64("B B B B / / B B / / / /"),
	293	"\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
	294	check_string(mime_base64("B\r\nBBB.// B-B//~//"),
	295	"\x04\x10\x41" "\xFF\xF0\x41" "\xFF\xFF\xFF");
	296	check_string(mime_base64("BBBB="),
	297	"\x04\x10\x41");
	298	check_string(mime_base64("BBBBx="), /* not actually valid base64 */
	299	"\x04\x10\x41");
	300	check_string(mime_base64("BBBB BB=="),
	301	"\x04\x10\x41" "\x04");
	302	check_string(mime_base64("BBBB BBB="),
	303	"\x04\x10\x41" "\x04\x10");
	304	}
	305
	306	static void test_hex(void) {
307	unsigned n;
308	static const unsigned char h[] = { 0x00, 0xFF, 0x80, 0x7F };
309	uint8_t *u;
310	size_t ul;
311
033fd4e3 RK	312	fprintf(stderr, "test_hex\n");
033fd4e3 RK	313
460b9539	314	for(n = 0; n <= UCHAR_MAX; ++n) {
	315	if(!isxdigit(n))
	316	insist(unhexdigitq(n) == -1);
	317	}
	318	insist(unhexdigitq('0') == 0);
	319	insist(unhexdigitq('1') == 1);
	320	insist(unhexdigitq('2') == 2);
	321	insist(unhexdigitq('3') == 3);
	322	insist(unhexdigitq('4') == 4);
	323	insist(unhexdigitq('5') == 5);
	324	insist(unhexdigitq('6') == 6);
	325	insist(unhexdigitq('7') == 7);
	326	insist(unhexdigitq('8') == 8);
	327	insist(unhexdigitq('9') == 9);
	328	insist(unhexdigitq('a') == 10);
	329	insist(unhexdigitq('b') == 11);
	330	insist(unhexdigitq('c') == 12);
	331	insist(unhexdigitq('d') == 13);
	332	insist(unhexdigitq('e') == 14);
	333	insist(unhexdigitq('f') == 15);
	334	insist(unhexdigitq('A') == 10);
	335	insist(unhexdigitq('B') == 11);
	336	insist(unhexdigitq('C') == 12);
	337	insist(unhexdigitq('D') == 13);
	338	insist(unhexdigitq('E') == 14);
	339	insist(unhexdigitq('F') == 15);
	340	check_string(hex(h, sizeof h), "00ff807f");
	341	check_string(hex(0, 0), "");
	342	u = unhex("00ff807f", &ul);
	343	insist(ul == 4);
	344	insist(memcmp(u, h, 4) == 0);
	345	u = unhex("00FF807F", &ul);
	346	insist(ul == 4);
	347	insist(memcmp(u, h, 4) == 0);
	348	u = unhex("", &ul);
	349	insist(ul == 0);
033fd4e3	350	fprintf(stderr, "2 ERROR reports expected {\n");
460b9539	351	insist(unhex("F", 0) == 0);
460b9539	352	insist(unhex("az", 0) == 0);
033fd4e3	353	fprintf(stderr, "}\n");
460b9539	354	}
	355
	356	static void test_casefold(void) {
e5a5a138 RK	357	uint32_t c, l;
e5a5a138 RK	358	const char input, folded, *expected;
460b9539	359
033fd4e3	360	fprintf(stderr, "test_casefold\n");
e5a5a138	361
460b9539	362	for(c = 1; c < 256; ++c) {
e5a5a138 RK	363	input = utf32_to_utf8(&c, 1, 0);
e5a5a138 RK	364	folded = utf8_casefold_canon(input, strlen(input), 0);
460b9539	365	switch(c) {
	366	default:
	367	if((c >= 'A' && c <= 'Z')
	368	\|\| (c >= 0xC0 && c <= 0xDE && c != 0xD7))
	369	l = c ^ 0x20;
	370	else
	371	l = c;
	372	break;
	373	case 0xB5: /* MICRO SIGN */
e5a5a138	374	l = 0x3BC; /* GREEK SMALL LETTER MU */
460b9539	375	break;
460b9539	376	case 0xDF: /* LATIN SMALL LETTER SHARP S */
e5a5a138	377	insist(!strcmp(folded, "ss"));
460b9539	378	l = 0;
	379	break;
	380	}
	381	if(l) {
e5a5a138 RK	382	/* Case-folded data is now normalized */
	383	expected = ucs42utf8(utf32_decompose_canon(&l, 1, 0));
	384	if(strcmp(folded, expected)) {
460b9539	385	fprintf(stderr, "%s:%d: casefolding %#lx got '%s', expected '%s'\n",
460b9539	386	__FILE__, __LINE__, (unsigned long)c,
e5a5a138	387	format(folded), format(expected));
460b9539	388	++errors;
	389	}
	390	++tests;
	391	}
	392	}
	393	check_string(casefold(""), "");
	394	}
	395
033fd4e3 RK	396	/** @brief Less-than comparison function for integer heap */
	397	static inline int int_lt(int a, int b) { return a < b; }
	398
dab22732 RK	399	/** @struct iheap
dab22732 RK	400	* @brief A heap with @c int elements */
033fd4e3	401	HEAP_TYPE(iheap, int, int_lt);
8e3fe3d8	402	HEAP_DEFINE(iheap, int, int_lt);
033fd4e3 RK	403
	404	/** @brief Tests for @ref heap.h */
	405	static void test_heap(void) {
	406	struct iheap h[1];
	407	int n;
	408	int last = -1;
	409
	410	fprintf(stderr, "test_heap\n");
	411
	412	iheap_init(h);
	413	for(n = 0; n < 1000; ++n)
	414	iheap_insert(h, random() % 100);
	415	for(n = 0; n < 1000; ++n) {
	416	const int latest = iheap_remove(h);
	417	if(last > latest)
	418	fprintf(stderr, "should have %d <= %d\n", last, latest);
	419	insist(last <= latest);
	420	last = latest;
	421	}
	422	putchar('\n');
	423	}
	424
e5a5a138 RK	425	/** @brief Tests for @ref lib/unicode.h */
	426	static void test_unicode(void) {
	427	FILE *fp;
	428	int lineno = 0;
	429	char l, lp;
	430	uint32_t buffer[1024];
	431	uint32_t c[6], NFD_c[6], NFKD_c[6]; / 1-indexed */
	432	int cn, bn;
	433
	434	fprintf(stderr, "test_unicode\n");
	435	if(!(fp = fopen("NormalizationTest.txt", "r"))) {
	436	system("wget http://www.unicode.org/Public/5.0.0/ucd/NormalizationTest.txt");
	437	chmod("NormalizationTest.txt", 0444);
	438	if(!(fp = fopen("NormalizationTest.txt", "r"))) {
	439	perror("NormalizationTest.txt");
	440	++tests; /* don't know how many... */
	441	++errors;
	442	return;
	443	}
	444	}
	445	while(!inputline("NormalizationTest.txt", fp, &l, '\n')) {
	446	++lineno;
	447	if(l == '#' \|\| l == '@')
	448	continue;
	449	bn = 0;
	450	cn = 1;
	451	lp = l;
	452	c[cn++] = &buffer[bn];
	453	while(lp && lp != '#') {
	454	if(*lp == ' ') {
	455	++lp;
	456	continue;
	457	}
	458	if(*lp == ';') {
	459	buffer[bn++] = 0;
	460	if(cn == 6)
	461	break;
	462	c[cn++] = &buffer[bn];
	463	++lp;
	464	continue;
	465	}
	466	buffer[bn++] = strtoul(lp, &lp, 16);
	467	}
	468	buffer[bn] = 0;
	469	assert(cn == 6);
	470	for(cn = 1; cn <= 5; ++cn) {
	471	NFD_c[cn] = utf32_decompose_canon(c[cn], utf32_len(c[cn]), 0);
	472	NFKD_c[cn] = utf32_decompose_compat(c[cn], utf32_len(c[cn]), 0);
	473	}
	474	#define unt_check(T, A, B) do { \
	475	++tests; \
	476	if(utf32_cmp(c[A], T##_c[B])) { \
	477	fprintf(stderr, "L%d: c%d != "#T"(c%d)\n", lineno, A, B); \
	478	fprintf(stderr, " c%d: %s\n", \
	479	A, format_utf32(c[A])); \
	480	fprintf(stderr, "%4s(c%d): %s\n", \
	481	#T, B, format_utf32(T##_c[B])); \
	482	++errors; \
	483	} \
	484	} while(0)
	485	unt_check(NFD, 3, 1);
	486	unt_check(NFD, 3, 2);
	487	unt_check(NFD, 3, 3);
	488	unt_check(NFD, 5, 4);
489	unt_check(NFD, 5, 5);
490	unt_check(NFKD, 5, 1);
491	unt_check(NFKD, 5, 2);
492	unt_check(NFKD, 5, 3);
493	unt_check(NFKD, 5, 4);
494	unt_check(NFKD, 5, 5);
495	for(cn = 1; cn <= 5; ++cn) {
496	xfree(NFD_c[cn]);
497	xfree(NFKD_c[cn]);
498	}
499	xfree(l);
500	}
501	}
502
460b9539	503	int main(void) {
	504	insist('\n' == 0x0A);
	505	insist('\r' == 0x0D);
	506	insist(' ' == 0x20);
	507	insist('0' == 0x30);
	508	insist('9' == 0x39);
	509	insist('A' == 0x41);
	510	insist('Z' == 0x5A);
	511	insist('a' == 0x61);
	512	insist('z' == 0x7A);
	513	/* addr.c */
	514	/* asprintf.c */
	515	/* authhash.c */
	516	/* basen.c */
	517	/* charset.c */
	518	/* client.c */
	519	/* configuration.c */
	520	/* event.c */
	521	/* fprintf.c */
033fd4e3 RK	522	/* heap.c */
033fd4e3 RK	523	test_heap();
460b9539	524	/* hex.c */
	525	test_hex();
	526	/* inputline.c */
	527	/* kvp.c */
	528	/* log.c */
	529	/* mem.c */
	530	/* mime.c */
	531	test_mime();
	532	/* mixer.c */
	533	/* plugin.c */
	534	/* printf.c */
	535	/* queue.c */
	536	/* sink.c */
	537	/* snprintf.c */
	538	/* split.c */
	539	/* syscalls.c */
	540	/* table.c */
e5a5a138 RK	541	/* unicode.c */
e5a5a138 RK	542	test_unicode();
460b9539	543	/* utf8.c */
	544	test_utf8();
	545	/* vector.c */
	546	/* words.c */
	547	test_casefold();
	548	/* XXX words() */
	549	/* wstat.c */
	550	fprintf(stderr, "%d errors out of %d tests\n", errors, tests);
	551	return !!errors;
	552	}
	553
	554	/*
	555	Local Variables:
	556	c-basic-offset:2
	557	comment-column:40
	558	End:
	559	*/
	560