utf8proc

commit c6858e955c6a6f0c991d48cc8818062863bceb13
parent 243875b456f23717dfc4b0de4d275b23145c91d1
Author: Steven G. Johnson <stevenj@mit.edu>
Date:   Sun, 29 Mar 2020 10:44:42 -0400

use unsigned char more consistently, silence -Wextra compiler warnings (#188)


Diffstat:
M
Makefile
 | 
2
+-
M
test/graphemetest.c
 | 
8
++++----
M
test/iterate.c
 | 
4
+++-
M
test/normtest.c
 | 
8
++++----
M
test/tests.c
 | 
10
+++++-----
M
test/tests.h
 | 
6
+++---

6 files changed, 20 insertions(+), 18 deletions(-)
diff --git a/Makefile b/Makefile
@@ -11,7 +11,7 @@ PERL=perl
 CFLAGS ?= -O2
 PICFLAG = -fPIC
 C99FLAG = -std=c99
-WCFLAGS = -Wall -pedantic
+WCFLAGS = -Wall -Wextra -pedantic
 UCFLAGS = $(CPPFLAGS) $(CFLAGS) $(PICFLAG) $(C99FLAG) $(WCFLAGS) -DUTF8PROC_EXPORTS $(UTF8PROC_DEFINES)
 LDFLAG_SHARED = -shared
 SOFLAG = -Wl,-soname
diff --git a/test/graphemetest.c b/test/graphemetest.c
@@ -2,7 +2,7 @@
 
 int main(int argc, char **argv)
 {
-    char buf[8192];
+    unsigned char buf[8192];
     FILE *f = argc > 1 ? fopen(argv[1], "r") : NULL;
     utf8proc_uint8_t src[1024];
 
@@ -18,18 +18,18 @@ int main(int argc, char **argv)
 
         while (buf[bi]) {
             bi = skipspaces(buf, bi);
-            if ((uint8_t)buf[bi] == 0xc3 && (uint8_t)buf[bi+1] == 0xb7) { /* U+00f7 = grapheme break */
+            if (buf[bi] == 0xc3 && buf[bi+1] == 0xb7) { /* U+00f7 = grapheme break */
                 src[si++] = '/';
                 bi += 2;
             }
-            else if ((uint8_t)buf[bi] == 0xc3 && (uint8_t)buf[bi+1] == 0x97) { /* U+00d7 = no break */
+            else if (buf[bi] == 0xc3 && buf[bi+1] == 0x97) { /* U+00d7 = no break */
                 bi += 2;
             }
             else if (buf[bi] == '#') { /* start of comments */
                 break;
             }
 	    else { /* hex-encoded codepoint */
-                size_t len = encode((char*) (src + si), buf + bi) - 1;
+                size_t len = encode((unsigned char*) (src + si), buf + bi) - 1;
                 while (src[si]) ++si; /* advance to NUL termination */
                 bi += len;
             }
diff --git a/test/iterate.c b/test/iterate.c
@@ -35,6 +35,8 @@ int main(int argc, char **argv)
     uint32_t byt;
     unsigned char buf[16];
 
+    (void) argc; (void) argv; /* unused */
+
     tests = error = 0;
 
     // Check valid sequences that were considered valid erroneously before
@@ -54,7 +56,7 @@ int main(int argc, char **argv)
         CHECKVALID(3, 0xbe, 4);
         CHECKVALID(3, 0xbf, 4);
     }
-    
+
     // Continuation byte not after lead
     for (byt = 0x80; byt < 0xc0; byt++) {
         CHECKINVALID(0, byt, 1);
diff --git a/test/normtest.c b/test/normtest.c
@@ -1,17 +1,17 @@
 #include "tests.h"
 
 #define CHECK_NORM(NRM, norm, src) {                                 \
-    char *src_norm = (char*) utf8proc_ ## NRM((utf8proc_uint8_t*) src);      \
-    check(!strcmp(norm, src_norm),                                  \
+    unsigned char *src_norm = (unsigned char*) utf8proc_ ## NRM((utf8proc_uint8_t*) src);      \
+    check(!strcmp((char *) norm, (char *) src_norm),                                  \
           "normalization failed for %s -> %s", src, norm);          \
     free(src_norm);                                                 \
 }
 
 int main(int argc, char **argv)
 {
-     char buf[8192];
+     unsigned char buf[8192];
      FILE *f = argc > 1 ? fopen(argv[1], "r") : NULL;
-     char source[1024], NFC[1024], NFD[1024], NFKC[1024], NFKD[1024];
+     unsigned char source[1024], NFC[1024], NFD[1024], NFKC[1024], NFKD[1024];
 
      check(f != NULL, "error opening NormalizationTest.txt");
      while (simple_getline(buf, f) > 0) {
diff --git a/test/tests.c b/test/tests.c
@@ -17,7 +17,7 @@ void check(int cond, const char *format, ...)
      }
 }
 
-size_t skipspaces(const char *buf, size_t i)
+size_t skipspaces(const unsigned char *buf, size_t i)
 {
     while (isspace(buf[i])) ++i;
     return i;
@@ -27,7 +27,7 @@ size_t skipspaces(const char *buf, size_t i)
    separated by whitespace, and terminated by any character not in
    [0-9a-fA-F] or whitespace, then stores the corresponding utf8 string
    in dest, returning the number of bytes read from buf */
-size_t encode(char *dest, const char *buf)
+size_t encode(unsigned char *dest, const unsigned char *buf)
 {
      size_t i = 0, j, d = 0;
      for (;;) {
@@ -39,19 +39,19 @@ size_t encode(char *dest, const char *buf)
                dest[d] = 0; /* NUL-terminate destination string */
                return i + 1;
           }
-          check(sscanf(buf + i, "%x", (unsigned int *)&c) == 1, "invalid hex input %s", buf+i);
+          check(sscanf((char *) (buf + i), "%x", (unsigned int *)&c) == 1, "invalid hex input %s", buf+i);
           i = j; /* skip to char after hex input */
           d += utf8proc_encode_char(c, (utf8proc_uint8_t *) (dest + d));
      }
 }
 
 /* simplistic, portable replacement for getline, sufficient for our tests */
-size_t simple_getline(char buf[8192], FILE *f) {
+size_t simple_getline(unsigned char buf[8192], FILE *f) {
     size_t i = 0;
     while (i < 8191) {
         int c = getc(f);
         if (c == EOF || c == '\n') break;
-        buf[i++] = (char) ((uint8_t) c);
+        buf[i++] = (unsigned char) c;
     }
     buf[i] = 0;
     return i;
diff --git a/test/tests.h b/test/tests.h
@@ -22,6 +22,6 @@
 extern size_t lineno;
 
 void check(int cond, const char *format, ...);
-size_t skipspaces(const char *buf, size_t i);
-size_t encode(char *dest, const char *buf);
-size_t simple_getline(char buf[8192], FILE *f);
+size_t skipspaces(const unsigned char *buf, size_t i);
+size_t encode(unsigned char *dest, const unsigned char *buf);
+size_t simple_getline(unsigned char buf[8192], FILE *f);