From dd20a05a04a0fc35074cdb46e3fac0a010d217f6 Mon Sep 17 00:00:00 2001 From: Karel Zak Date: Wed, 14 Jan 2009 01:52:11 +0100 Subject: [PATCH] blkid: add udev string encoding routines Signed-off-by: Karel Zak --- libs/blkid/src/Makefile.am | 2 +- libs/blkid/src/blkid.h | 4 +- libs/blkid/src/blkid.sym | 2 + libs/blkid/src/encode.c | 288 +++++++++++++++++++++++++++++++++ libs/blkid/src/probers/btrfs.c | 2 +- 5 files changed, 295 insertions(+), 3 deletions(-) create mode 100644 libs/blkid/src/encode.c diff --git a/libs/blkid/src/Makefile.am b/libs/blkid/src/Makefile.am index cf43a734..e21fbc20 100644 --- a/libs/blkid/src/Makefile.am +++ b/libs/blkid/src/Makefile.am @@ -8,7 +8,7 @@ AM_CPPFLAGS += -I$(top_builddir)/libs/blkid/src lib_LIBRARIES = libblkid.a libblkid_a_SOURCES = cache.c dev.c devname.c devno.c getsize.c llseek.c \ probe.c read.c resolve.c save.c tag.c version.c verify.c \ - blkid.h blkidP.h list.h probers/probers.h \ + encode.c blkid.h blkidP.h list.h probers/probers.h \ $(top_srcdir)/lib/blkdev.c \ $(top_srcdir)/lib/linux_version.c \ $(top_srcdir)/lib/md5.c diff --git a/libs/blkid/src/blkid.h b/libs/blkid/src/blkid.h index be6f7e06..26f04b89 100644 --- a/libs/blkid/src/blkid.h +++ b/libs/blkid/src/blkid.h @@ -111,7 +111,9 @@ extern int blkid_parse_version_string(const char *ver_string); extern int blkid_get_library_version(const char **ver_string, const char **date_string); - +/* encode.c */ +extern int blkid_encode_string(const char *str, char *str_enc, size_t len); +extern int blkid_safe_string(const char *str, char *str_safe, size_t len); /* probe.c */ extern int blkid_known_fstype(const char *fstype); diff --git a/libs/blkid/src/blkid.sym b/libs/blkid/src/blkid.sym index 51bdde97..105f3ff9 100644 --- a/libs/blkid/src/blkid.sym +++ b/libs/blkid/src/blkid.sym @@ -7,6 +7,7 @@ blkid_devno_to_devname; blkid_dev_set_search; blkid_do_probe; + blkid_encode_string; blkid_find_dev_with_tag; blkid_free_probe; blkid_gc_cache; @@ -34,6 +35,7 @@ blkid_probe_set_request; blkid_put_cache; blkid_reset_probe; + blkid_safe_string; blkid_tag_iterate_begin; blkid_tag_iterate_end; blkid_tag_next; diff --git a/libs/blkid/src/encode.c b/libs/blkid/src/encode.c new file mode 100644 index 00000000..0c00ec94 --- /dev/null +++ b/libs/blkid/src/encode.c @@ -0,0 +1,288 @@ + +/* + * encode.c - string convertion routines (mostly for compatibility with + * udev/volume_id) + * + * Copyright (C) 2008 Kay Sievers + * Copyright (C) 2009 Karel Zak + * + * This file may be redistributed under the terms of the + * GNU Lesser General Public License. + */ +#include +#include +#include +#include +#include +#include +#include + +#include "blkdev.h" +#include "blkidP.h" + +#define UDEV_ALLOWED_CHARS_INPUT "/ $%?," + +/* 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) + return 1; + if (unichar < 0x800) + return 2; + if (unichar < 0x10000) + return 3; + if (unichar < 0x200000) + return 4; + if (unichar < 0x4000000) + return 5; + return 6; +} + +/* check if unicode char has a valid numeric range */ +static int utf8_unichar_valid_range(int unichar) +{ + if (unichar > 0x10ffff) + return 0; + if ((unichar & 0xfffff800) == 0xd800) + return 0; + if ((unichar > 0xfdcf) && (unichar < 0xfdf0)) + return 0; + if ((unichar & 0xffff) == 0xffff) + return 0; + return 1; +} + +/* validate one encoded unicode char and return its length */ +static int utf8_encoded_valid_unichar(const char *str) +{ + int len; + int unichar; + int i; + + len = utf8_encoded_expected_len(str); + if (len == 0) + return -1; + + /* ascii is valid */ + if (len == 1) + return 1; + + /* check if expected encoded chars are available */ + for (i = 0; i < len; i++) + if ((str[i] & 0x80) != 0x80) + return -1; + + unichar = utf8_encoded_to_unichar(str); + + /* check if encoded length matches encoded value */ + if (utf8_unichar_to_encoded_len(unichar) != len) + return -1; + + /* check if value has valid range */ + if (!utf8_unichar_valid_range(unichar)) + return -1; + + return len; +} + +static int replace_whitespace(const char *str, char *to, size_t len) +{ + size_t i, j; + + /* strip trailing whitespace */ + len = strnlen(str, len); + while (len && isspace(str[len-1])) + len--; + + /* strip leading whitespace */ + i = 0; + while (isspace(str[i]) && (i < len)) + i++; + + j = 0; + while (i < len) { + /* substitute multiple whitespace with a single '_' */ + if (isspace(str[i])) { + while (isspace(str[i])) + i++; + to[j++] = '_'; + } + to[j++] = str[i++]; + } + to[j] = '\0'; + return 0; +} + +static int is_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; +} + +/* allow chars in whitelist, plain ascii, hex-escaping and valid utf8 */ +static int replace_chars(char *str, const char *white) +{ + size_t i = 0; + int replaced = 0; + + while (str[i] != '\0') { + int len; + + if (is_whitelisted(str[i], white)) { + i++; + continue; + } + + /* accept hex encoding */ + if (str[i] == '\\' && str[i+1] == 'x') { + i += 2; + continue; + } + + /* accept valid utf8 */ + len = utf8_encoded_valid_unichar(&str[i]); + if (len > 1) { + i += len; + continue; + } + + /* if space is allowed, replace whitespace with ordinary space */ + if (isspace(str[i]) && white != NULL && strchr(white, ' ') != NULL) { + str[i] = ' '; + i++; + replaced++; + continue; + } + + /* everything else is replaced with '_' */ + str[i] = '_'; + i++; + replaced++; + } + return replaced; +} + +/** + * blkid_encode_string: + * @str: input string to be encoded + * @str_enc: output string to store the encoded input string + * @len: maximum size of the output string, which may be + * four times as long as the input string + * + * Encode all potentially unsafe characters of a string to the + * corresponding hex value prefixed by '\x'. + * + * Returns: 0 if the entire string was copied, non-zero otherwise. + **/ +int blkid_encode_string(const char *str, char *str_enc, size_t len) +{ + size_t i, j; + + if (str == NULL || str_enc == NULL || len == 0) + return -1; + + str_enc[0] = '\0'; + for (i = 0, j = 0; str[i] != '\0'; i++) { + int seqlen; + + seqlen = utf8_encoded_valid_unichar(&str[i]); + if (seqlen > 1) { + memcpy(&str_enc[j], &str[i], seqlen); + j += seqlen; + i += (seqlen-1); + } else if (str[i] == '\\' || !is_whitelisted(str[i], NULL)) { + sprintf(&str_enc[j], "\\x%02x", (unsigned char) str[i]); + j += 4; + } else { + str_enc[j] = str[i]; + j++; + } + if (j+3 >= len) + goto err; + } + str_enc[j] = '\0'; + return 0; +err: + return -1; +} + +/** + * blkid_safe_string: + * @str: input string + * @str_safe: output string + * @len: size of output string + * + * Allows plain ascii, hex-escaping and valid utf8. Replaces all whitespaces + * with '_'. + */ +int blkid_safe_string(const char *str, char *str_safe, size_t len) +{ + replace_whitespace(str, str_safe, sizeof(len)); + replace_chars(str_safe, UDEV_ALLOWED_CHARS_INPUT); + return 0; +} diff --git a/libs/blkid/src/probers/btrfs.c b/libs/blkid/src/probers/btrfs.c index b7590f78..3205e3d9 100644 --- a/libs/blkid/src/probers/btrfs.c +++ b/libs/blkid/src/probers/btrfs.c @@ -73,7 +73,7 @@ static int probe_btrfs(blkid_probe pr, const struct blkid_idmag *mag) sizeof(bfs->label)); blkid_probe_set_uuid(pr, bfs->fsid); - blkid_probe_set_uuid_as(pr, bfs->dev_item.uuid, "SUB_UUID"); + blkid_probe_set_uuid_as(pr, bfs->dev_item.uuid, "UUID_SUB"); return 0; } -- 2.39.5