aboutsummaryrefslogtreecommitdiff
path: root/lib
diff options
context:
space:
mode:
authorThomas Voss <mail@thomasvoss.com> 2024-05-04 04:01:45 +0200
committerThomas Voss <mail@thomasvoss.com> 2024-05-04 04:01:45 +0200
commitac1b4bcbaeaee7d2ef9132dcdc254f2d08691650 (patch)
tree90250966629653f0462cf17bc0b6f2476fb6d1fc /lib
parent8b923ba5e5bb37ea26350b4c1c688b8697706609 (diff)
Go all in on string views, and fix manuals
Diffstat (limited to 'lib')
-rw-r--r--lib/mbstring/u8chk.c14
-rw-r--r--lib/mbstring/u8chr.c32
-rw-r--r--lib/mbstring/u8cmp.c4
-rw-r--r--lib/mbstring/u8cspn.c8
-rw-r--r--lib/mbstring/u8haspfx.c4
-rw-r--r--lib/mbstring/u8hassfx.c4
-rw-r--r--lib/mbstring/u8len.c4
-rw-r--r--lib/mbstring/u8next.c15
-rw-r--r--lib/mbstring/u8rchr.c40
-rw-r--r--lib/mbstring/u8split.c16
-rw-r--r--lib/mbstring/u8spn.c10
-rw-r--r--lib/optparse/optparse.c9
-rw-r--r--lib/unicode/string/u8casefold.c4
-rw-r--r--lib/unicode/string/u8gcnt.c4
-rw-r--r--lib/unicode/string/u8gnext.c13
-rw-r--r--lib/unicode/string/u8lower.c16
-rw-r--r--lib/unicode/string/u8title.c22
-rw-r--r--lib/unicode/string/u8upper.c4
-rw-r--r--lib/unicode/string/u8wcnt.c4
-rw-r--r--lib/unicode/string/u8wcnt_human.c4
-rw-r--r--lib/unicode/string/u8wnext.c28
-rw-r--r--lib/unicode/string/u8wnext_human.c11
22 files changed, 131 insertions, 139 deletions
diff --git a/lib/mbstring/u8chk.c b/lib/mbstring/u8chk.c
index 2566bac..20c4f3f 100644
--- a/lib/mbstring/u8chk.c
+++ b/lib/mbstring/u8chk.c
@@ -1,17 +1,15 @@
#include "rune.h"
#include "mbstring.h"
-char8_t *
-(u8chk)(const char8_t *s, size_t n)
+const char8_t *
+u8chk(struct u8view sv)
{
- while (n) {
- rune ch;
- int m = u8tor(&ch, s);
+ int w;
+ rune ch;
+ while (w = u8next(&ch, &sv)) {
if (ch == RUNE_ERROR)
- return (char8_t *)s;
- n -= m;
- s += m;
+ return sv.p - w;
}
return nullptr;
diff --git a/lib/mbstring/u8chr.c b/lib/mbstring/u8chr.c
index 395a328..4831695 100644
--- a/lib/mbstring/u8chr.c
+++ b/lib/mbstring/u8chr.c
@@ -31,7 +31,7 @@
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */
-static char8_t *
+static const char8_t *
memmem2(const char8_t *h, size_t k, const char8_t *n)
{
uint16_t hw, nw;
@@ -40,12 +40,12 @@ memmem2(const char8_t *h, size_t k, const char8_t *n)
for (h += 2, k -= 2; k; k--, hw = hw << 8 | *h++) {
if (hw == nw)
- return (char8_t *)h - 2;
+ return h - 2;
}
- return hw == nw ? (char8_t *)h - 2 : nullptr;
+ return hw == nw ? h - 2 : nullptr;
}
-static char8_t *
+static const char8_t *
memmem3(const char8_t *h, size_t k, const char8_t *n)
{
uint32_t hw, nw;
@@ -54,12 +54,12 @@ memmem3(const char8_t *h, size_t k, const char8_t *n)
for (h += 3, k -= 3; k; k--, hw = (hw | *h++) << 8) {
if (hw == nw)
- return (char8_t *)h - 3;
+ return h - 3;
}
- return hw == nw ? (char8_t *)h - 3 : nullptr;
+ return hw == nw ? h - 3 : nullptr;
}
-static char8_t *
+static const char8_t *
memmem4(const char8_t *h, size_t k, const char8_t *n)
{
uint32_t hw, nw;
@@ -68,28 +68,28 @@ memmem4(const char8_t *h, size_t k, const char8_t *n)
for (h += 4, k -= 4; k; k--, hw = hw << 8 | *h++) {
if (hw == nw)
- return (char8_t *)h - 4;
+ return h - 4;
}
- return hw == nw ? (char8_t *)h - 4 : nullptr;
+ return hw == nw ? h - 4 : nullptr;
}
-char8_t *
-(u8chr)(const char8_t *s, size_t n, rune ch)
+const char8_t *
+u8chr(struct u8view sv, rune ch)
{
char8_t buf[U8_LEN_MAX];
int m = rtou8(buf, sizeof(buf), ch);
- if (n < (size_t)m)
+ if (sv.len < (size_t)m)
return nullptr;
switch (m) {
case 1:
- return memchr(s, ch, n);
+ return memchr(sv.p, ch, sv.len);
case 2:
- return memmem2(s, n, buf);
+ return memmem2(sv.p, sv.len, buf);
case 3:
- return memmem3(s, n, buf);
+ return memmem3(sv.p, sv.len, buf);
case 4:
- return memmem4(s, n, buf);
+ return memmem4(sv.p, sv.len, buf);
}
unreachable();
diff --git a/lib/mbstring/u8cmp.c b/lib/mbstring/u8cmp.c
index 8bd2400..0059020 100644
--- a/lib/mbstring/u8cmp.c
+++ b/lib/mbstring/u8cmp.c
@@ -3,7 +3,7 @@
#include "mbstring.h"
int
-u8cmp(const char8_t *x, size_t n, const char8_t *y, size_t m)
+u8cmp(struct u8view x, struct u8view y)
{
- return n != m ? (n > m ? +1 : -1) : memcmp(x, y, n);
+ return x.len != y.len ? (x.len > y.len ? +1 : -1) : memcmp(x.p, y.p, x.len);
}
diff --git a/lib/mbstring/u8cspn.c b/lib/mbstring/u8cspn.c
index 4892de4..827373f 100644
--- a/lib/mbstring/u8cspn.c
+++ b/lib/mbstring/u8cspn.c
@@ -1,13 +1,13 @@
#include "mbstring.h"
size_t
-u8cspn(const char8_t *s, size_t n, const rune *p, size_t m)
+u8cspn(struct u8view sv, const rune *p, size_t n)
{
rune ch;
- size_t k, l;
+ size_t k, w;
- for (k = 0; (l = u8next(&ch, &s, &n)); k += l) {
- for (size_t i = 0; i < m; i++) {
+ for (k = 0; w = u8next(&ch, &sv); k += w) {
+ for (size_t i = 0; i < n; i++) {
if (p[i] == ch)
goto found;
}
diff --git a/lib/mbstring/u8haspfx.c b/lib/mbstring/u8haspfx.c
index b6cea50..c61efbb 100644
--- a/lib/mbstring/u8haspfx.c
+++ b/lib/mbstring/u8haspfx.c
@@ -4,7 +4,7 @@
#include "mbstring.h"
bool
-u8haspfx(const char8_t *s, size_t n, const char8_t *pfx, size_t m)
+u8haspfx(struct u8view sv, struct u8view pfx)
{
- return n >= m && memeq(s, pfx, m);
+ return sv.len >= pfx.len && memeq(sv.p, pfx.p, pfx.len);
}
diff --git a/lib/mbstring/u8hassfx.c b/lib/mbstring/u8hassfx.c
index e31bb4b..8ea4456 100644
--- a/lib/mbstring/u8hassfx.c
+++ b/lib/mbstring/u8hassfx.c
@@ -4,7 +4,7 @@
#include "mbstring.h"
bool
-u8hassfx(const char8_t *s, size_t n, const char8_t *sfx, size_t m)
+u8hassfx(struct u8view sv, struct u8view sfx)
{
- return n >= m && memeq(s + n - m, sfx, m);
+ return sv.len >= sfx.len && memeq(sv.p + sv.len - sfx.len, sfx.p, sfx.len);
}
diff --git a/lib/mbstring/u8len.c b/lib/mbstring/u8len.c
index 217ab66..23c55c5 100644
--- a/lib/mbstring/u8len.c
+++ b/lib/mbstring/u8len.c
@@ -1,10 +1,10 @@
#include "mbstring.h"
size_t
-u8len(const char8_t *s, size_t n)
+u8len(struct u8view sv)
{
size_t m = 0;
- while (u8next(nullptr, &s, &n))
+ while (u8next(nullptr, &sv))
m++;
return m;
}
diff --git a/lib/mbstring/u8next.c b/lib/mbstring/u8next.c
index 82d2ad7..518de49 100644
--- a/lib/mbstring/u8next.c
+++ b/lib/mbstring/u8next.c
@@ -1,16 +1,15 @@
#include "mbstring.h"
int
-u8next(rune *ch, const char8_t **s, size_t *n)
+u8next(rune *ch, struct u8view *sv)
{
- rune _;
- int m = 0;
+ int n = 0;
- if (*n) {
- m = u8tor(ch ? ch : &_, *s);
- *n -= m;
- *s += m;
+ if (sv->len) {
+ rune _;
+ n = u8tor(ch ? ch : &_, sv->p);
+ VSHFT(sv, n);
}
- return m;
+ return n;
}
diff --git a/lib/mbstring/u8rchr.c b/lib/mbstring/u8rchr.c
index 09aa111..825f8fd 100644
--- a/lib/mbstring/u8rchr.c
+++ b/lib/mbstring/u8rchr.c
@@ -3,17 +3,17 @@
#include "mbstring.h"
-static char8_t *
+static const char8_t *
memrchr1(const char8_t *s, size_t k, const char8_t *n)
{
for (const char8_t *p = s + k - 1; k-- > 0; p--) {
if (*p == *n)
- return (char8_t *)p;
+ return p;
}
return nullptr;
}
-static char8_t *
+static const char8_t *
memrchr2(const char8_t *h, size_t k, const char8_t *n)
{
uint16_t hw, nw;
@@ -23,13 +23,13 @@ memrchr2(const char8_t *h, size_t k, const char8_t *n)
for (H -= 2, k -= 2; k; k--, hw = hw >> 8 | (*H-- << 8)) {
if (hw == nw)
- return (char8_t *)H + 1;
+ return H + 1;
}
- return hw == nw ? (char8_t *)H + 1 : nullptr;
+ return hw == nw ? H + 1 : nullptr;
}
-static char8_t *
+static const char8_t *
memrchr3(const char8_t *h, size_t k, const char8_t *n)
{
uint32_t hw, nw;
@@ -41,13 +41,13 @@ memrchr3(const char8_t *h, size_t k, const char8_t *n)
k--, hw = (hw >> 8 | (*H-- << 24)) & UINT32_C(0xFFFFFF00))
{
if (hw == nw)
- return (char8_t *)H + 1;
+ return H + 1;
}
- return hw == nw ? (char8_t *)H + 1 : nullptr;
+ return hw == nw ? H + 1 : nullptr;
}
-static char8_t *
+static const char8_t *
memrchr4(const char8_t *h, size_t k, const char8_t *n)
{
uint32_t hw, nw;
@@ -57,29 +57,29 @@ memrchr4(const char8_t *h, size_t k, const char8_t *n)
for (H -= 4, k -= 4; k; k--, hw = hw >> 8 | (*H-- << 24)) {
if (hw == nw)
- return (char8_t *)H + 1;
+ return H + 1;
}
- return hw == nw ? (char8_t *)H + 1 : nullptr;
+ return hw == nw ? H + 1 : nullptr;
}
-char8_t *
-(u8rchr)(const char8_t *s, size_t n, rune ch)
+const char8_t *
+u8rchr(struct u8view sv, rune ch)
{
char8_t buf[U8_LEN_MAX];
- int m = rtou8(buf, ch, sizeof(buf));
+ int n = rtou8(buf, ch, sizeof(buf));
- if (n < (size_t)m)
+ if (sv.len < (size_t)n)
return nullptr;
- switch (m) {
+ switch (n) {
case 1:
- return (char8_t *)memrchr1(s, n, buf);
+ return memrchr1(sv.p, sv.len, buf);
case 2:
- return (char8_t *)memrchr2(s, n, buf);
+ return memrchr2(sv.p, sv.len, buf);
case 3:
- return (char8_t *)memrchr3(s, n, buf);
+ return memrchr3(sv.p, sv.len, buf);
case 4:
- return (char8_t *)memrchr4(s, n, buf);
+ return memrchr4(sv.p, sv.len, buf);
}
unreachable();
diff --git a/lib/mbstring/u8split.c b/lib/mbstring/u8split.c
index 5ee3bc0..c26f48b 100644
--- a/lib/mbstring/u8split.c
+++ b/lib/mbstring/u8split.c
@@ -1,16 +1,16 @@
#include "mbstring.h"
struct u8view
-u8split(const char8_t **p, size_t *n, rune ch)
+u8split(struct u8view *rhs, rune ch)
{
- struct u8view lhs = {.p = *p};
- if ((*p = u8chr(*p, *n, ch)) == nullptr) {
- lhs.len = *n;
- *n = 0;
+ struct u8view lhs = {.p = rhs->p};
+ if ((rhs->p = u8chr(*rhs, ch)) == nullptr) {
+ lhs.len = rhs->len;
+ rhs->len = 0;
} else {
- lhs.len = *p - lhs.p;
- *n -= lhs.len;
- u8next(nullptr, p, n);
+ lhs.len = rhs->p - lhs.p;
+ rhs->len -= lhs.len;
+ u8next(nullptr, rhs);
}
return lhs;
}
diff --git a/lib/mbstring/u8spn.c b/lib/mbstring/u8spn.c
index 1cf45f2..d41fcbc 100644
--- a/lib/mbstring/u8spn.c
+++ b/lib/mbstring/u8spn.c
@@ -1,15 +1,15 @@
#include "mbstring.h"
size_t
-u8spn(const char8_t *s, size_t n, const rune *p, size_t m)
+u8spn(struct u8view sv, const rune *p, size_t n)
{
rune ch;
- size_t k = 0, l;
+ size_t k = 0, w;
- while ((l = u8next(&ch, &s, &n))) {
- for (size_t i = 0; i < m; i++) {
+ while (w = u8next(&ch, &sv)) {
+ for (size_t i = 0; i < n; i++) {
if (p[i] == ch) {
- k += l;
+ k += w;
goto found;
}
}
diff --git a/lib/optparse/optparse.c b/lib/optparse/optparse.c
index 407fa62..757dd47 100644
--- a/lib/optparse/optparse.c
+++ b/lib/optparse/optparse.c
@@ -44,11 +44,10 @@ optparse(struct optparse *st, const struct op_option *opts, size_t nopts)
st->optind++;
/* Skip ‘--’ */
- opt.p += 2;
- opt.len -= 2;
+ VSHFT(&opt, 2);
const struct op_option *o = nullptr;
- const char8_t *eq_p = u8chr(opt.p, '=', opt.len);
+ const char8_t *eq_p = u8chr(opt, '=');
struct u8view opt_no_eq = {
.p = opt.p,
.len = eq_p == nullptr ? opt.len : (size_t)(eq_p - opt.p),
@@ -56,7 +55,7 @@ optparse(struct optparse *st, const struct op_option *opts, size_t nopts)
for (size_t i = 0; i < nopts; i++) {
struct u8view lo = opts[i].longopt;
- if (lo.p == nullptr || !u8haspfx(U8_ARGS(lo), U8_ARGS(opt_no_eq)))
+ if (lo.p == nullptr || !u8haspfx(lo, opt_no_eq))
continue;
if (o != nullptr)
return error(st, OPT_MSG_INVALID, opt_no_eq);
@@ -146,7 +145,7 @@ rune
error_s(struct optparse *st, const char *msg, struct u8view s)
{
snprintf(st->errmsg, sizeof(st->errmsg), u8"%s — ‘%.*s’", msg,
- U8_PRI_ARGS(s));
+ SV_PRI_ARGS(s));
return -1;
}
diff --git a/lib/unicode/string/u8casefold.c b/lib/unicode/string/u8casefold.c
index 6c0b61d..2ab7c7c 100644
--- a/lib/unicode/string/u8casefold.c
+++ b/lib/unicode/string/u8casefold.c
@@ -3,13 +3,13 @@
#include "unicode/string.h"
size_t
-u8casefold(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
+u8casefold(char8_t *restrict dst, size_t dstn, struct u8view sv,
enum caseflags flags)
{
rune ch;
size_t n = 0;
- while (u8next(&ch, &src, &srcn)) {
+ while (u8next(&ch, &sv)) {
struct rview rv = uprop_get_cf(ch, flags & CF_LANG_AZ);
for (size_t i = 0; i < rv.len; i++) {
if (n >= dstn) {
diff --git a/lib/unicode/string/u8gcnt.c b/lib/unicode/string/u8gcnt.c
index 81a0f97..6dfc519 100644
--- a/lib/unicode/string/u8gcnt.c
+++ b/lib/unicode/string/u8gcnt.c
@@ -1,10 +1,10 @@
#include "unicode/string.h"
size_t
-u8gcnt(const char8_t *s, size_t n)
+u8gcnt(struct u8view sv)
{
size_t m = 0;
- while (u8gnext(nullptr, &s, &n))
+ while (u8gnext(nullptr, &sv))
m++;
return m;
}
diff --git a/lib/unicode/string/u8gnext.c b/lib/unicode/string/u8gnext.c
index a050bd5..3b0b410 100644
--- a/lib/unicode/string/u8gnext.c
+++ b/lib/unicode/string/u8gnext.c
@@ -20,17 +20,17 @@ static bool u8isgbrk(rune, rune, struct gbrk_state *);
_MLIB_DEFINE_BSEARCH(gbrk_prop, gbrk_prop_tbl, GBP_OTHER)
size_t
-u8gnext(struct u8view *g, const char8_t **s, size_t *n)
+u8gnext(struct u8view *g, struct u8view *sv)
{
int m;
rune ch1;
const char8_t *p;
struct gbrk_state gs = {0};
- if (*n == 0)
+ if (sv->len == 0)
return 0;
- p = *s;
+ p = sv->p;
if (g)
g->p = p;
p += u8tor(&ch1, p);
@@ -38,14 +38,13 @@ u8gnext(struct u8view *g, const char8_t **s, size_t *n)
for (;;) {
rune ch2;
- if ((size_t)(p - *s) >= *n)
+ if ((size_t)(p - sv->p) >= sv->len)
ch2 = 0;
else
m = u8tor(&ch2, p);
if (u8isgbrk(ch1, ch2, &gs)) {
- ptrdiff_t d = p - *s;
- *n -= d;
- *s = p;
+ ptrdiff_t d = p - sv->p;
+ VSHFT(sv, d);
if (g)
g->len = d;
return d;
diff --git a/lib/unicode/string/u8lower.c b/lib/unicode/string/u8lower.c
index 63fdae4..907077b 100644
--- a/lib/unicode/string/u8lower.c
+++ b/lib/unicode/string/u8lower.c
@@ -13,7 +13,7 @@ uprop_ccc_0_or_230(rune ch)
}
size_t
-u8lower(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
+u8lower(char8_t *restrict dst, size_t dstn, struct u8view sv,
enum caseflags flags)
{
struct lcctx ctx = {
@@ -32,21 +32,21 @@ u8lower(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
n = before_dot_cnt = more_above_cnt = 0;
- while (u8next(&ch, &src, &srcn)) {
+ while (u8next(&ch, &sv)) {
rune next = 0;
- if (srcn > 0)
- u8tor(&next, src);
+ if (sv.len > 0)
+ u8tor(&next, sv.p);
if (ctx.az_or_tr || ctx.lt) {
if (before_dot_cnt == 0 || more_above_cnt == 0) {
rune ch = 0;
before_dot_cnt = more_above_cnt = 0;
- struct u8view cpy = {src, srcn};
+ struct u8view cpy = sv;
do {
before_dot_cnt++;
more_above_cnt++;
- } while (u8next(&ch, U8_ARGSP(cpy)) && !uprop_ccc_0_or_230(ch));
+ } while (u8next(&ch, &cpy) && !uprop_ccc_0_or_230(ch));
if (ch != COMB_DOT_ABOVE)
before_dot_cnt = 0;
@@ -60,11 +60,11 @@ u8lower(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
if (final_sigma.after == 0) {
rune ch = 0;
- struct u8view cpy = {src, srcn};
+ struct u8view cpy = sv;
do
final_sigma.after++;
- while (u8next(&ch, U8_ARGSP(cpy)) && uprop_is_ci(ch));
+ while (u8next(&ch, &cpy) && uprop_is_ci(ch));
if (!uprop_is_cased(ch))
final_sigma.after = 0;
diff --git a/lib/unicode/string/u8title.c b/lib/unicode/string/u8title.c
index 01e9d2e..f4d9b7e 100644
--- a/lib/unicode/string/u8title.c
+++ b/lib/unicode/string/u8title.c
@@ -14,7 +14,7 @@ uprop_ccc_0_or_230(rune ch)
}
size_t
-u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
+u8title(char8_t *restrict dst, size_t dstn, struct u8view sv,
enum caseflags flags)
{
struct tcctx ctx_t;
@@ -26,7 +26,7 @@ u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
rune ch;
bool nl_IJ = false;
size_t n, before_dot_cnt, more_above_cnt;
- struct u8view word = {}, wcpy = {src, srcn};
+ struct u8view word = {}, wcpy = sv;
struct {
bool before;
size_t after;
@@ -39,9 +39,9 @@ u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
n = before_dot_cnt = more_above_cnt = 0;
- while (u8next(&ch, &src, &srcn)) {
- if (src > word.p + word.len) {
- u8wnext(&word, U8_ARGSP(wcpy));
+ while (u8next(&ch, &sv)) {
+ if (sv.p > word.p + word.len) {
+ u8wnext(&word, &wcpy);
ctx_t.after_soft_dotted = false;
state = TITLE;
}
@@ -50,12 +50,12 @@ u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
if (before_dot_cnt == 0 || more_above_cnt == 0) {
rune ch = 0;
before_dot_cnt = more_above_cnt = 0;
- struct u8view cpy = {src, srcn};
+ struct u8view cpy = sv;
do {
before_dot_cnt++;
more_above_cnt++;
- } while (u8next(&ch, U8_ARGSP(cpy)) && !uprop_ccc_0_or_230(ch));
+ } while (u8next(&ch, &cpy) && !uprop_ccc_0_or_230(ch));
if (ch != COMB_DOT_ABOVE)
before_dot_cnt = 0;
@@ -69,11 +69,11 @@ u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
if (final_sigma.after == 0) {
rune ch = 0;
- struct u8view cpy = {src, srcn};
+ struct u8view cpy = sv;
do
final_sigma.after++;
- while (u8next(&ch, U8_ARGSP(cpy)) && uprop_is_ci(ch));
+ while (u8next(&ch, &cpy) && uprop_is_ci(ch));
if (!uprop_is_cased(ch))
final_sigma.after = 0;
@@ -95,8 +95,8 @@ u8title(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
if (flags & CF_LANG_NL) {
rune next = 0;
- if (srcn > 0)
- u8tor(&next, src);
+ if (sv.len > 0)
+ u8tor(&next, sv.p);
nl_IJ =
(ch == 'i' || ch == 'I') && (next == 'j' || next == 'J');
}
diff --git a/lib/unicode/string/u8upper.c b/lib/unicode/string/u8upper.c
index 086a160..6d4026d 100644
--- a/lib/unicode/string/u8upper.c
+++ b/lib/unicode/string/u8upper.c
@@ -3,7 +3,7 @@
#include "unicode/string.h"
size_t
-u8upper(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
+u8upper(char8_t *restrict dst, size_t dstn, struct u8view sv,
enum caseflags flags)
{
struct ucctx ctx = {
@@ -15,7 +15,7 @@ u8upper(char8_t *restrict dst, size_t dstn, const char8_t *src, size_t srcn,
rune ch;
size_t n = 0;
- while (u8next(&ch, &src, &srcn)) {
+ while (u8next(&ch, &sv)) {
struct rview rv = uprop_get_uc(ch, ctx);
for (size_t i = 0; i < rv.len; i++) {
if (n >= dstn) {
diff --git a/lib/unicode/string/u8wcnt.c b/lib/unicode/string/u8wcnt.c
index f1b1742..f71faf5 100644
--- a/lib/unicode/string/u8wcnt.c
+++ b/lib/unicode/string/u8wcnt.c
@@ -1,10 +1,10 @@
#include "unicode/string.h"
size_t
-u8wcnt(const char8_t *s, size_t n)
+u8wcnt(struct u8view sv)
{
size_t m = 0;
- while (u8wnext(nullptr, &s, &n))
+ while (u8wnext(nullptr, &sv))
m++;
return m;
}
diff --git a/lib/unicode/string/u8wcnt_human.c b/lib/unicode/string/u8wcnt_human.c
index 6e70398..60e7f95 100644
--- a/lib/unicode/string/u8wcnt_human.c
+++ b/lib/unicode/string/u8wcnt_human.c
@@ -1,10 +1,10 @@
#include "unicode/string.h"
size_t
-u8wcnt_human(const char8_t *s, size_t n)
+u8wcnt_human(struct u8view sv)
{
size_t m = 0;
- while (u8wnext_human(nullptr, &s, &n))
+ while (u8wnext_human(nullptr, &sv))
m++;
return m;
}
diff --git a/lib/unicode/string/u8wnext.c b/lib/unicode/string/u8wnext.c
index 5e893c6..6655c5d 100644
--- a/lib/unicode/string/u8wnext.c
+++ b/lib/unicode/string/u8wnext.c
@@ -29,22 +29,20 @@ static size_t findwbrk(struct u8view);
static struct wbrk_state mkwbrkstate(struct u8view);
size_t
-u8wnext(struct u8view *w, const char8_t **s, size_t *n)
+u8wnext(struct u8view *w, struct u8view *sv)
{
- ASSUME(n != nullptr);
- ASSUME(s != nullptr);
- ASSUME(*s != nullptr);
+ ASSUME(sv != nullptr);
+ ASSUME(sv->p != nullptr);
- if (*n == 0)
+ if (sv->len == 0)
return 0;
- size_t off = findwbrk((struct u8view){*s, *n});
+ size_t off = findwbrk(*sv);
if (w != nullptr)
- *w = (struct u8view){*s, off};
+ *w = (struct u8view){sv->p, off};
- ASSUME(*n >= off);
- *s += off;
- *n -= off;
+ ASSUME(sv->len >= off);
+ VSHFT(sv, off);
return off;
}
@@ -196,13 +194,13 @@ mkwbrkstate(struct u8view sv)
rune ch;
for (size_t i = 0;
- i < lengthof(ws.raw.next) && u8next(&ch, U8_ARGSP(ws.raw_v)) != 0; i++)
+ i < lengthof(ws.raw.next) && u8next(&ch, &ws.raw_v) != 0; i++)
{
ws.raw.next[i] = mlib_lookup(ch);
}
for (size_t i = 0;
- i < lengthof(ws.raw.next) && u8next(&ch, U8_ARGSP(ws.skip_v)) != 0;)
+ i < lengthof(ws.raw.next) && u8next(&ch, &ws.skip_v) != 0;)
{
ws.skip.next[i] = mlib_lookup(ch);
if (!IS_IGNORE(ws.skip.next[i]))
@@ -224,10 +222,10 @@ advance(struct wbrk_state *ws)
ws->raw.prev[0] = ws->raw.next[0];
ws->raw.next[0] = ws->raw.next[1];
ws->raw.next[1] =
- u8next(&ch, U8_ARGSP(ws->raw_v)) != 0 ? mlib_lookup(ch) : WBRK_EOT;
+ u8next(&ch, &ws->raw_v) != 0 ? mlib_lookup(ch) : WBRK_EOT;
/* Increment the midpoint */
- u8next(nullptr, U8_ARGSP(ws->mid_v));
+ u8next(nullptr, &ws->mid_v);
/* Ignore ignorable properties */
if (!IS_IGNORE(ws->raw.prev[0])) {
@@ -237,7 +235,7 @@ advance(struct wbrk_state *ws)
ws->ri_parity = ws->ri_parity == 0 && ws->skip.prev[0] == WBRK_RI;
do {
- if (u8next(&ch, U8_ARGSP(ws->skip_v)) == 0) {
+ if (u8next(&ch, &ws->skip_v) == 0) {
ws->skip.next[1] = WBRK_EOT;
break;
}
diff --git a/lib/unicode/string/u8wnext_human.c b/lib/unicode/string/u8wnext_human.c
index d85abf1..953d942 100644
--- a/lib/unicode/string/u8wnext_human.c
+++ b/lib/unicode/string/u8wnext_human.c
@@ -4,17 +4,16 @@
#include "unicode/string.h"
size_t
-u8wnext_human(struct u8view *dst, const char8_t **s, size_t *n)
+u8wnext_human(struct u8view *dst, struct u8view *sv)
{
- ASSUME(n != nullptr);
- ASSUME(s != nullptr);
- ASSUME(*s != nullptr);
+ ASSUME(sv != nullptr);
+ ASSUME(sv->p != nullptr);
struct u8view w;
- while (u8wnext(&w, s, n)) {
+ while (u8wnext(&w, sv)) {
rune ch;
struct u8view cpy = w;
- while (u8next(&ch, U8_ARGSP(cpy))) {
+ while (u8next(&ch, &cpy)) {
if (uprop_get_gc(ch) & (GC_L | GC_N)) {
if (dst != nullptr)
*dst = w;