[systemd-commits] 3 commits - .gitignore Makefile.am src/libudev src/shared src/test TODO
Dave Reisner
dreisner at kemper.freedesktop.org
Thu Sep 19 08:59:29 PDT 2013
.gitignore | 1
Makefile.am | 14 +
TODO | 1
src/libudev/libudev-util.c | 5
src/shared/device-nodes.c | 74 +++++++++
src/shared/device-nodes.h | 23 +++
src/shared/utf8.c | 325 +++++++++----------------------------------
src/shared/utf8.h | 5
src/shared/util.c | 4
src/test/test-device-nodes.c | 55 +++++++
src/test/test-utf8.c | 63 +++++---
11 files changed, 283 insertions(+), 287 deletions(-)
New commits:
commit e7363c59d69b71a4327429719e24cab9020e2796
Author: Dave Reisner <dreisner at archlinux.org>
Date: Wed Sep 18 12:32:23 2013 -0400
test-utf8: add more tests for public functions
diff --git a/src/test/test-utf8.c b/src/test/test-utf8.c
index b5a833e..7bd0db1 100644
--- a/src/test/test-utf8.c
+++ b/src/test/test-utf8.c
@@ -34,9 +34,43 @@ static void test_utf8_is_valid(void) {
assert_se(!utf8_is_valid("\341\204"));
}
+static void test_ascii_is_valid(void) {
+ assert_se(ascii_is_valid("alsdjf\t\vbarr\nba z"));
+ assert_se(!ascii_is_valid("\342\204\242"));
+ assert_se(!ascii_is_valid("\341\204"));
+}
+
+static void test_ascii_filter(void) {
+ char *f;
+
+ f = ascii_filter("alsdjf\t\vbarr\nba z");
+ assert_se(streq(f, "alsdjf\t\vbarr\nba z"));
+ free(f);
+
+ f = ascii_filter("\342\204\242");
+ assert_se(streq(f, ""));
+ free(f);
+
+ f = ascii_filter("foo\341\204bar");
+ assert_se(streq(f, "foobar"));
+ free(f);
+}
+
+static void test_utf8_encoded_valid_unichar(void) {
+ assert_se(utf8_encoded_valid_unichar("\342\204\242") == 3);
+ assert_se(utf8_encoded_valid_unichar("\302\256") == 2);
+ assert_se(utf8_encoded_valid_unichar("a") == 1);
+ assert_se(utf8_encoded_valid_unichar("\341\204") < 0);
+ assert_se(utf8_encoded_valid_unichar("\341\204\341\204") < 0);
+
+}
+
int main(int argc, char *argv[]) {
test_utf8_is_valid();
test_utf8_is_printable();
+ test_ascii_is_valid();
+ test_ascii_filter();
+ test_utf8_encoded_valid_unichar();
return 0;
}
commit 8f6ce71fe79d897b67157d92869db87ee2042af6
Author: Dave Reisner <dreisner at archlinux.org>
Date: Wed Sep 18 12:12:04 2013 -0400
device-nodes: move device node specific code to own file
In the process, rename udev_encode_string which is poorly named for what
it does. It deals specifically with encoding names that udev creates and
has its own rules: utf8 is valid but some ascii is not (e.g. path
separators), and everything else is simply escaped. Rename it to
encode_devnode_name.
diff --git a/.gitignore b/.gitignore
index deeee53..8115d4d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -101,6 +101,7 @@
/test-cgroup-util
/test-daemon
/test-date
+/test-device-nodes
/test-efivars
/test-engine
/test-env-replace
diff --git a/Makefile.am b/Makefile.am
index 8d70ad3..89a5c86 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -642,6 +642,8 @@ libsystemd_shared_la_SOURCES = \
src/shared/list.h \
src/shared/macro.h \
src/shared/def.h \
+ src/shared/device-nodes.c \
+ src/shared/device-nodes.h \
src/shared/sparse-endian.h \
src/shared/util.c \
src/shared/util.h \
@@ -1137,7 +1139,8 @@ tests += \
test-time \
test-hashmap \
test-list \
- test-tables
+ test-tables \
+ test-device-nodes
EXTRA_DIST += \
test/sched_idle_bad.service \
@@ -1149,6 +1152,15 @@ EXTRA_DIST += \
EXTRA_DIST += \
src/test/test-helper.h
+test_device_nodes_SOURCES = \
+ src/test/test-device-nodes.c
+
+test_device_nodes_CFLAGS = \
+ $(AM_CFLAGS)
+
+test_device_nodes_LDADD = \
+ libsystemd-shared.la
+
test_engine_SOURCES = \
src/test/test-engine.c
diff --git a/src/libudev/libudev-util.c b/src/libudev/libudev-util.c
index d54430c..b5b9db6 100644
--- a/src/libudev/libudev-util.c
+++ b/src/libudev/libudev-util.c
@@ -32,6 +32,7 @@
#include <sys/stat.h>
#include <sys/param.h>
+#include "device-nodes.h"
#include "libudev.h"
#include "libudev-private.h"
#include "utf8.h"
@@ -344,7 +345,7 @@ int util_replace_chars(char *str, const char *white)
while (str[i] != '\0') {
int len;
- if (is_utf8_encoding_whitelisted(str[i], white)) {
+ if (whitelisted_char_for_devnode(str[i], white)) {
i++;
continue;
}
@@ -392,7 +393,7 @@ int util_replace_chars(char *str, const char *white)
**/
_public_ int udev_util_encode_string(const char *str, char *str_enc, size_t len)
{
- return udev_encode_string(str, str_enc, len);
+ return encode_devnode_name(str, str_enc, len);
}
/*
diff --git a/src/shared/device-nodes.c b/src/shared/device-nodes.c
new file mode 100644
index 0000000..986553e
--- /dev/null
+++ b/src/shared/device-nodes.c
@@ -0,0 +1,74 @@
+/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
+
+/***
+ This file is part of systemd.
+
+ Copyright 2012 Lennart Poettering
+
+ systemd is free software; you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1 of the License, or
+ (at your option) any later version.
+
+ systemd 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
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with systemd; If not, see <http://www.gnu.org/licenses/>.
+***/
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <sys/types.h>
+
+#include "device-nodes.h"
+#include "utf8.h"
+
+int whitelisted_char_for_devnode(char c, const char *white) {
+ if ((c >= '0' && c <= '9') ||
+ (c >= 'A' && c <= 'Z') ||
+ (c >= 'a' && c <= 'z') ||
+ strchr("#+-.:=@_", c) != NULL ||
+ (white != NULL && strchr(white, c) != NULL))
+ return 1;
+ return 0;
+}
+
+int encode_devnode_name(const char *str, char *str_enc, size_t len) {
+ size_t i, j;
+
+ if (str == NULL || str_enc == NULL)
+ return -1;
+
+ for (i = 0, j = 0; str[i] != '\0'; i++) {
+ int seqlen;
+
+ seqlen = utf8_encoded_valid_unichar(&str[i]);
+ if (seqlen > 1) {
+ if (len-j < (size_t)seqlen)
+ goto err;
+ memcpy(&str_enc[j], &str[i], seqlen);
+ j += seqlen;
+ i += (seqlen-1);
+ } else if (str[i] == '\\' || !whitelisted_char_for_devnode(str[i], NULL)) {
+ if (len-j < 4)
+ goto err;
+ sprintf(&str_enc[j], "\\x%02x", (unsigned char) str[i]);
+ j += 4;
+ } else {
+ if (len-j < 1)
+ goto err;
+ str_enc[j] = str[i];
+ j++;
+ }
+ }
+ if (len-j < 1)
+ goto err;
+ str_enc[j] = '\0';
+ return 0;
+err:
+ return -1;
+}
diff --git a/src/shared/device-nodes.h b/src/shared/device-nodes.h
new file mode 100644
index 0000000..a98195a
--- /dev/null
+++ b/src/shared/device-nodes.h
@@ -0,0 +1,23 @@
+/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
+
+/***
+ This file is part of systemd.
+
+ Copyright 2012 Lennart Poettering
+
+ systemd is free software; you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1 of the License, or
+ (at your option) any later version.
+
+ systemd 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
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with systemd; If not, see <http://www.gnu.org/licenses/>.
+***/
+
+int encode_devnode_name(const char *str, char *str_enc, size_t len);
+int whitelisted_char_for_devnode(char c, const char *additional);
diff --git a/src/shared/utf8.c b/src/shared/utf8.c
index 732f0f0..c3d97cc 100644
--- a/src/shared/utf8.c
+++ b/src/shared/utf8.c
@@ -285,49 +285,3 @@ int utf8_encoded_valid_unichar(const char *str) {
return len;
}
-
-int is_utf8_encoding_whitelisted(char c, const char *white) {
- if ((c >= '0' && c <= '9') ||
- (c >= 'A' && c <= 'Z') ||
- (c >= 'a' && c <= 'z') ||
- strchr("#+-.:=@_", c) != NULL ||
- (white != NULL && strchr(white, c) != NULL))
- return 1;
- return 0;
-}
-
-int udev_encode_string(const char *str, char *str_enc, size_t len) {
- size_t i, j;
-
- if (str == NULL || str_enc == NULL)
- return -1;
-
- for (i = 0, j = 0; str[i] != '\0'; i++) {
- int seqlen;
-
- seqlen = utf8_encoded_valid_unichar(&str[i]);
- if (seqlen > 1) {
- if (len-j < (size_t)seqlen)
- goto err;
- memcpy(&str_enc[j], &str[i], seqlen);
- j += seqlen;
- i += (seqlen-1);
- } else if (str[i] == '\\' || !is_utf8_encoding_whitelisted(str[i], NULL)) {
- if (len-j < 4)
- goto err;
- sprintf(&str_enc[j], "\\x%02x", (unsigned char) str[i]);
- j += 4;
- } else {
- if (len-j < 1)
- goto err;
- str_enc[j] = str[i];
- j++;
- }
- }
- if (len-j < 1)
- goto err;
- str_enc[j] = '\0';
- return 0;
-err:
- return -1;
-}
diff --git a/src/shared/utf8.h b/src/shared/utf8.h
index 22e1346..96a03ea 100644
--- a/src/shared/utf8.h
+++ b/src/shared/utf8.h
@@ -35,5 +35,3 @@ char *ascii_filter(const char *s);
char *utf16_to_utf8(const void *s, size_t length);
int utf8_encoded_valid_unichar(const char *str);
-int is_utf8_encoding_whitelisted(char c, const char *white);
-int udev_encode_string(const char *str, char *str_enc, size_t len);
diff --git a/src/shared/util.c b/src/shared/util.c
index 2b76a5c..2009553 100644
--- a/src/shared/util.c
+++ b/src/shared/util.c
@@ -73,7 +73,7 @@
#include "hashmap.h"
#include "env-util.h"
#include "fileio.h"
-#include "utf8.h"
+#include "device-nodes.h"
int saved_argc = 0;
char **saved_argv = NULL;
@@ -3509,7 +3509,7 @@ static char *tag_to_udev_node(const char *tagvalue, const char *by) {
if (t == NULL)
return NULL;
- if (udev_encode_string(u, t, enc_len) < 0)
+ if (encode_devnode_name(u, t, enc_len) < 0)
return NULL;
if (asprintf(&dn, "/dev/disk/by-%s/%s", by, t) < 0)
diff --git a/src/test/test-device-nodes.c b/src/test/test-device-nodes.c
new file mode 100644
index 0000000..2f3dedb
--- /dev/null
+++ b/src/test/test-device-nodes.c
@@ -0,0 +1,55 @@
+/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
+
+/***
+ This file is part of systemd.
+
+ Copyright 2013 Dave Reisner
+
+ systemd is free software; you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1 of the License, or
+ (at your option) any later version.
+
+ systemd 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
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with systemd; If not, see <http://www.gnu.org/licenses/>.
+***/
+
+#include <sys/types.h>
+
+#include "device-nodes.h"
+#include "util.h"
+
+/* helpers for test_encode_devnode_name */
+static char *do_encode_string(const char *in) {
+ size_t out_len = strlen(in) * 4;
+ char *out = malloc(out_len);
+
+ assert_se(out);
+ assert_se(encode_devnode_name(in, out, out_len) >= 0);
+ puts(out);
+
+ return out;
+}
+
+static bool expect_encoded_as(const char *in, const char *expected) {
+ _cleanup_free_ char *encoded = do_encode_string(in);
+ return streq(encoded, expected);
+}
+
+static void test_encode_devnode_name(void) {
+ assert_se(expect_encoded_as("systemd sucks", "systemd\\x20sucks"));
+ assert_se(expect_encoded_as("pinkiepie", "pinkiepie"));
+ assert_se(expect_encoded_as("valÃd\\ųtf8", "valÃd\\x5cųtf8"));
+ assert_se(expect_encoded_as("s/ash/ng", "s\\x2fash\\x2fng"));
+}
+
+int main(int argc, char *argv[]) {
+ test_encode_devnode_name();
+
+ return 0;
+}
diff --git a/src/test/test-utf8.c b/src/test/test-utf8.c
index 26cc37b..b5a833e 100644
--- a/src/test/test-utf8.c
+++ b/src/test/test-utf8.c
@@ -19,34 +19,9 @@
along with systemd; If not, see <http://www.gnu.org/licenses/>.
***/
-
#include "utf8.h"
#include "util.h"
-/* helpers for test_udev_encode_string */
-static char *do_encode_string(const char *in) {
- size_t out_len = strlen(in) * 4;
- char *out = malloc(out_len);
-
- assert_se(out);
- assert_se(udev_encode_string(in, out, out_len) >= 0);
- puts(out);
-
- return out;
-}
-
-static bool expect_encoded_as(const char *in, const char *expected) {
- _cleanup_free_ char *encoded = do_encode_string(in);
- return streq(encoded, expected);
-}
-
-static void test_udev_encode_string(void) {
- assert_se(expect_encoded_as("systemd sucks", "systemd\\x20sucks"));
- assert_se(expect_encoded_as("pinkiepie", "pinkiepie"));
- assert_se(expect_encoded_as("valÃd\\ųtf8", "valÃd\\x5cųtf8"));
- assert_se(expect_encoded_as("s/ash/ng", "s\\x2fash\\x2fng"));
-}
-
static void test_utf8_is_printable(void) {
assert_se(utf8_is_printable("ascii is valid\tunicode", 22));
assert_se(utf8_is_printable("\342\204\242", 3));
@@ -55,14 +30,13 @@ static void test_utf8_is_printable(void) {
static void test_utf8_is_valid(void) {
assert_se(utf8_is_valid("ascii is valid unicode"));
- assert_se(utf8_is_valid("\341\204\242"));
+ assert_se(utf8_is_valid("\342\204\242"));
assert_se(!utf8_is_valid("\341\204"));
}
int main(int argc, char *argv[]) {
test_utf8_is_valid();
test_utf8_is_printable();
- test_udev_encode_string();
return 0;
}
commit 7991ac34ab08421415b907e42775c5539a4a5bbb
Author: Dave Reisner <dreisner at archlinux.org>
Date: Wed Sep 18 11:52:14 2013 -0400
shared/utf8: merge implementations, remove cruft
This unifies the utf8 handling code which was previously duplicated in
udev and systemd.
diff --git a/TODO b/TODO
index c0f51de..01bc993 100644
--- a/TODO
+++ b/TODO
@@ -602,7 +602,6 @@ Features:
* udev:
- remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n)
- move to LGPL
- - unify utf8 validator code with shared/
- kill scsi_id
- add trigger --subsystem-match=usb/usb_device device
diff --git a/src/shared/utf8.c b/src/shared/utf8.c
index 1a68394..732f0f0 100644
--- a/src/shared/utf8.c
+++ b/src/shared/utf8.c
@@ -51,8 +51,6 @@
#include "utf8.h"
#include "util.h"
-#define FILTER_CHAR '_'
-
static inline bool is_unicode_valid(uint32_t ch) {
if (ch >= 0x110000) /* End of unicode space */
@@ -67,17 +65,6 @@ static inline bool is_unicode_valid(uint32_t ch) {
return true;
}
-static inline bool is_continuation_char(uint8_t ch) {
- if ((ch & 0xc0) != 0x80) /* 10xxxxxx */
- return false;
- return true;
-}
-
-static inline void merge_continuation_char(uint32_t *u_ch, uint8_t ch) {
- *u_ch <<= 6;
- *u_ch |= ch & 0x3f;
-}
-
static bool is_unicode_control(uint32_t ch) {
/*
@@ -90,163 +77,97 @@ static bool is_unicode_control(uint32_t ch) {
(0x7F <= ch && ch <= 0x9F);
}
-bool utf8_is_printable(const char* str, size_t length) {
- uint32_t val = 0;
- uint32_t min = 0;
- const uint8_t *p;
+/* count of characters used to encode one unicode char */
+static int utf8_encoded_expected_len(const char *str) {
+ unsigned char c = (unsigned char)str[0];
- assert(str);
+ if (c < 0x80)
+ return 1;
+ if ((c & 0xe0) == 0xc0)
+ return 2;
+ if ((c & 0xf0) == 0xe0)
+ return 3;
+ if ((c & 0xf8) == 0xf0)
+ return 4;
+ if ((c & 0xfc) == 0xf8)
+ return 5;
+ if ((c & 0xfe) == 0xfc)
+ return 6;
+ return 0;
+}
- for (p = (const uint8_t*) str; length; p++, length--) {
- if (*p < 128) {
- val = *p;
- } else {
- if ((*p & 0xe0) == 0xc0) { /* 110xxxxx two-char seq. */
- min = 128;
- val = (uint32_t) (*p & 0x1e);
- goto ONE_REMAINING;
- } else if ((*p & 0xf0) == 0xe0) { /* 1110xxxx three-char seq.*/
- min = (1 << 11);
- val = (uint32_t) (*p & 0x0f);
- goto TWO_REMAINING;
- } else if ((*p & 0xf8) == 0xf0) { /* 11110xxx four-char seq */
- min = (1 << 16);
- val = (uint32_t) (*p & 0x07);
- } else
- return false;
-
- p++;
- length--;
- if (!length || !is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- TWO_REMAINING:
- p++;
- length--;
- if (!is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- ONE_REMAINING:
- p++;
- length--;
- if (!is_continuation_char(*p))
- return false;
- merge_continuation_char(&val, *p);
-
- if (val < min)
- return false;
- }
+/* decode one unicode char */
+static int utf8_encoded_to_unichar(const char *str) {
+ int unichar;
+ int len;
+ int i;
- if (is_unicode_control(val))
- return false;
+ len = utf8_encoded_expected_len(str);
+ switch (len) {
+ case 1:
+ return (int)str[0];
+ case 2:
+ unichar = str[0] & 0x1f;
+ break;
+ case 3:
+ unichar = (int)str[0] & 0x0f;
+ break;
+ case 4:
+ unichar = (int)str[0] & 0x07;
+ break;
+ case 5:
+ unichar = (int)str[0] & 0x03;
+ break;
+ case 6:
+ unichar = (int)str[0] & 0x01;
+ break;
+ default:
+ return -1;
}
- return true;
+ for (i = 1; i < len; i++) {
+ if (((int)str[i] & 0xc0) != 0x80)
+ return -1;
+ unichar <<= 6;
+ unichar |= (int)str[i] & 0x3f;
+ }
+
+ return unichar;
}
-static char* utf8_validate(const char *str, char *output) {
- uint32_t val = 0;
- uint32_t min = 0;
- const uint8_t *p, *last;
- int size;
- uint8_t *o;
+bool utf8_is_printable(const char* str, size_t length) {
+ const uint8_t *p;
assert(str);
- o = (uint8_t*) output;
- for (p = (const uint8_t*) str; *p; p++) {
- if (*p < 128) {
- if (o)
- *o = *p;
- } else {
- last = p;
-
- if ((*p & 0xe0) == 0xc0) { /* 110xxxxx two-char seq. */
- size = 2;
- min = 128;
- val = (uint32_t) (*p & 0x1e);
- goto ONE_REMAINING;
- } else if ((*p & 0xf0) == 0xe0) { /* 1110xxxx three-char seq.*/
- size = 3;
- min = (1 << 11);
- val = (uint32_t) (*p & 0x0f);
- goto TWO_REMAINING;
- } else if ((*p & 0xf8) == 0xf0) { /* 11110xxx four-char seq */
- size = 4;
- min = (1 << 16);
- val = (uint32_t) (*p & 0x07);
- } else
- goto error;
-
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- TWO_REMAINING:
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- ONE_REMAINING:
- p++;
- if (!is_continuation_char(*p))
- goto error;
- merge_continuation_char(&val, *p);
-
- if (val < min)
- goto error;
-
- if (!is_unicode_valid(val))
- goto error;
-
- if (o) {
- memcpy(o, last, (size_t) size);
- o += size;
- }
-
- continue;
-
- error:
- if (o) {
- *o = FILTER_CHAR;
- p = last; /* We retry at the next character */
- } else
- goto failure;
- }
+ for (p = (const uint8_t*) str; length; p++) {
+ int encoded_len = utf8_encoded_valid_unichar((const char *)p);
+ int32_t val = utf8_encoded_to_unichar((const char*)p);
- if (o)
- o++;
- }
+ if (encoded_len < 0 || val < 0 || is_unicode_control(val))
+ return false;
- if (o) {
- *o = '\0';
- return output;
+ length -= encoded_len;
}
- return (char*) str;
-
-failure:
- return NULL;
-}
-
-char* utf8_is_valid (const char *str) {
- return utf8_validate(str, NULL);
+ return true;
}
-char* utf8_filter (const char *str) {
- char *new_str;
+const char *utf8_is_valid(const char *str) {
+ const uint8_t *p;
assert(str);
- new_str = malloc(strlen(str) + 1);
- if (!new_str)
- return NULL;
+ for (p = (const uint8_t*) str; *p; ) {
+ int len = utf8_encoded_valid_unichar((const char *)p);
+
+ if (len < 0)
+ return NULL;
+
+ p += len;
+ }
- return utf8_validate(str, new_str);
+ return str;
}
char *ascii_is_valid(const char *str) {
@@ -318,64 +239,6 @@ char *utf16_to_utf8(const void *s, size_t length) {
return r;
}
-/* count of characters used to encode one unicode char */
-static int utf8_encoded_expected_len(const char *str) {
- unsigned char c = (unsigned char)str[0];
-
- if (c < 0x80)
- return 1;
- if ((c & 0xe0) == 0xc0)
- return 2;
- if ((c & 0xf0) == 0xe0)
- return 3;
- if ((c & 0xf8) == 0xf0)
- return 4;
- if ((c & 0xfc) == 0xf8)
- return 5;
- if ((c & 0xfe) == 0xfc)
- return 6;
- return 0;
-}
-
-/* decode one unicode char */
-static int utf8_encoded_to_unichar(const char *str) {
- int unichar;
- int len;
- int i;
-
- len = utf8_encoded_expected_len(str);
- switch (len) {
- case 1:
- return (int)str[0];
- case 2:
- unichar = str[0] & 0x1f;
- break;
- case 3:
- unichar = (int)str[0] & 0x0f;
- break;
- case 4:
- unichar = (int)str[0] & 0x07;
- break;
- case 5:
- unichar = (int)str[0] & 0x03;
- break;
- case 6:
- unichar = (int)str[0] & 0x01;
- break;
- default:
- return -1;
- }
-
- for (i = 1; i < len; i++) {
- if (((int)str[i] & 0xc0) != 0x80)
- return -1;
- unichar <<= 6;
- unichar |= (int)str[i] & 0x3f;
- }
-
- return unichar;
-}
-
/* expected size used to encode one unicode char */
static int utf8_unichar_to_encoded_len(int unichar) {
if (unichar < 0x80)
diff --git a/src/shared/utf8.h b/src/shared/utf8.h
index 7a5608c..22e1346 100644
--- a/src/shared/utf8.h
+++ b/src/shared/utf8.h
@@ -25,12 +25,11 @@
#include "macro.h"
-char *utf8_is_valid(const char *s) _pure_;
+const char *utf8_is_valid(const char *s) _pure_;
char *ascii_is_valid(const char *s) _pure_;
bool utf8_is_printable(const char* str, size_t length) _pure_;
-char *utf8_filter(const char *s);
char *ascii_filter(const char *s);
char *utf16_to_utf8(const void *s, size_t length);
diff --git a/src/test/test-utf8.c b/src/test/test-utf8.c
index d2b9771..26cc37b 100644
--- a/src/test/test-utf8.c
+++ b/src/test/test-utf8.c
@@ -47,6 +47,12 @@ static void test_udev_encode_string(void) {
assert_se(expect_encoded_as("s/ash/ng", "s\\x2fash\\x2fng"));
}
+static void test_utf8_is_printable(void) {
+ assert_se(utf8_is_printable("ascii is valid\tunicode", 22));
+ assert_se(utf8_is_printable("\342\204\242", 3));
+ assert_se(!utf8_is_printable("\341\204", 2));
+}
+
static void test_utf8_is_valid(void) {
assert_se(utf8_is_valid("ascii is valid unicode"));
assert_se(utf8_is_valid("\341\204\242"));
@@ -55,5 +61,8 @@ static void test_utf8_is_valid(void) {
int main(int argc, char *argv[]) {
test_utf8_is_valid();
+ test_utf8_is_printable();
test_udev_encode_string();
+
+ return 0;
}
More information about the systemd-commits
mailing list