Package: unzip Version: 5.52-11 Severity: important Tags: patch In current version, can not open shift-jis zip file on UTF-8 environment. These problems are in shift-jis, cp949(extended euc-kr) and other.
Details for this problem are in this URL [0], and patch to evade this problem is taken in Ubuntu. I confirmed the patch was extracted, applied, ,checked , and the problem was evaded. I attached this patch. Please check this problem and apply patch. Best regards, Nobuhiro [0]: https://bugs.launchpad.net/ubuntu/+source/unzip/+bug/203609 -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux 2.6.25-1-686-bigmem (SMP w/2 CPU cores) Locale: LANG=ja_JP.UTF-8, LC_CTYPE=ja_JP.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages unzip depends on: ii libc6 2.7-11 GNU C Library: Shared libraries unzip recommends no packages. -- no debconf information
diff -urN a/unix/unix.c b/unix/unix.c --- a/unix/unix.c 2008-04-23 22:52:13.000000000 +0900 +++ b/unix/unix.c 2008-04-23 21:47:24.000000000 +0900 @@ -30,6 +30,9 @@ #define UNZIP_INTERNAL #include "unzip.h" +#include <iconv.h> +#include <langinfo.h> + #ifdef SCO_XENIX # define SYSNDIR #else /* SCO Unix, AIX, DNIX, TI SysV, Coherent 4.x, ... */ @@ -1685,3 +1688,90 @@ } } #endif /* QLZIP */ + + +typedef struct { + char *local_charset; + char *archive_charset; +} CHARSET_MAP; + +/* A mapping of local <-> archive charsets used by default to convert filenames + * of DOS/Windows Zip archives. Currently very basic. */ +static CHARSET_MAP dos_charset_map[] = { + { "ANSI_X3.4-1968", "CP850" }, + { "ISO-8859-1", "CP850" }, + { "CP1252", "CP850" }, + { "UTF-8", "CP866" }, + { "KOI8-R", "CP866" }, + { "KOI8-U", "CP866" }, + { "ISO-8859-5", "CP866" } +}; + +char OEM_CP[MAX_CP_NAME] = ""; +char ISO_CP[MAX_CP_NAME] = ""; + +/* Try to guess the default value of OEM_CP based on the current locale. + * ISO_CP is left alone for now. */ +void init_conversion_charsets() +{ + const char *local_charset; + int i; + + /* Make a guess only if OEM_CP not already set. */ + if(*OEM_CP == '\0') { + local_charset = nl_langinfo(CODESET); + for(i = 0; i < sizeof(dos_charset_map)/sizeof(CHARSET_MAP); i++) + if(!strcasecmp(local_charset, dos_charset_map[i].local_charset)) { + strncpy(OEM_CP, dos_charset_map[i].archive_charset, + sizeof(OEM_CP)); + break; + } + } +} + +/* Convert a string from one encoding to the current locale using iconv(). + * Be as non-intrusive as possible. If error is encountered during covertion + * just leave the string intact. */ +static void charset_to_intern(char *string, char *from_charset) +{ + iconv_t cd; + char *s,*d, *buf; + size_t slen, dlen, buflen; + const char *local_charset; + + if(*from_charset == '\0') + return; + + buf = NULL; + local_charset = nl_langinfo(CODESET); + + if((cd = iconv_open(local_charset, from_charset)) == (iconv_t)-1) + return; + + slen = strlen(string); + s = string; + dlen = buflen = 2*slen; + d = buf = malloc(buflen + 1); + if(!d) + goto cleanup; + bzero(buf,buflen); + if(iconv(cd, &s, &slen, &d, &dlen) == (size_t)-1) + goto cleanup; + strncpy(string, buf, buflen); + + cleanup: + free(buf); + iconv_close(cd); +} + +/* Convert a string from OEM_CP to the current locale charset. */ +inline void oem_intern(char *string) +{ + charset_to_intern(string, OEM_CP); +} + +/* Convert a string from ISO_CP to the current locale charset. */ +inline void iso_intern(char *string) +{ + charset_to_intern(string, ISO_CP); +} diff -urN a/unix/unxcfg.h b/unix/unxcfg.h --- a/unix/unxcfg.h 2008-04-23 22:52:13.000000000 +0900 +++ b/unix/unxcfg.h 2008-04-23 21:47:24.000000000 +0900 @@ -142,4 +142,30 @@ /* wild_dir, dirname, wildname, matchname[], dirnamelen, have_dirname, */ /* and notfirstcall are used by do_wild(). */ + +#define MAX_CP_NAME 25 + +#ifdef SETLOCALE +# undef SETLOCALE +#endif +#define SETLOCALE(category, locale) setlocale(category, locale) +#include <locale.h> + +#ifdef _ISO_INTERN +# undef _ISO_INTERN +#endif +#define _ISO_INTERN(str1) iso_intern(str1) + +#ifdef _OEM_INTERN +# undef _OEM_INTERN +#endif +#ifndef IZ_OEM2ISO_ARRAY +# define IZ_OEM2ISO_ARRAY +#endif +#define _OEM_INTERN(str1) oem_intern(str1) + +void iso_intern(char *); +void oem_intern(char *); +void init_conversion_charsets(void); + #endif /* !__unxcfg_h */ diff -urN a/unzip.c b/unzip.c --- a/unzip.c 2008-04-23 22:52:13.000000000 +0900 +++ b/unzip.c 2008-04-23 21:47:24.000000000 +0900 @@ -309,11 +309,21 @@ -2 just filenames but allow -h/-t/-z -l long Unix \"ls -l\" format\n\ -v verbose, multi-page format\n"; +#ifndef UNIX static ZCONST char Far ZipInfoUsageLine3[] = "miscellaneous options:\n\ -h print header line -t print totals for listed files or for all\n\ -z print zipfile comment %c-T%c print file times in sortable decimal format\ \n %c-C%c be case-insensitive %s\ -x exclude filenames that follow from listing\n"; +#else /* UNIX */ +static ZCONST char Far ZipInfoUsageLine3[] = "miscellaneous options:\n\ + -h print header line -t print totals for listed files or for all\n\ + -z print zipfile comment %c-T%c print file times in sortable decimal format\ +\n %c-C%c be case-insensitive %s\ + -x exclude filenames that follow from listing\n\ + -O CHARSET specify a character encoding for DOS, Windows and OS/2 archives\n\ + -I CHARSET specify a character encoding for UNIX and other archives\n"; +#endif /* !UNIX */ #ifdef MORE #ifdef VMS static ZCONST char Far ZipInfoUsageLine4[] = @@ -518,7 +528,7 @@ #else /* !VMS */ # ifdef COPYRIGHT_CLEAN static ZCONST char Far UnzipUsageLine1[] = "\ -UnZip %d.%d%d%s of %s, by Debian. Original by Info-ZIP.\ +UnZip %d.%d%d%s of %s, by Ubuntu. Original by Info-ZIP.\ \n\n"; # else static ZCONST char Far UnzipUsageLine1[] = "\ @@ -598,6 +608,7 @@ #endif /* ?VM_CMS */ #endif /* ?MACOS */ +#ifndef UNIX static ZCONST char Far UnzipUsageLine4[] = "\ modifiers: -q quiet mode (-qq => quieter)\n\ -n never overwrite existing files -a auto-convert any text files\n\ @@ -605,6 +616,17 @@ -j junk paths (do not make directories) -v be verbose/print version info\n\ %c-C%c match filenames case-insensitively %c-L%c make (some) names \ lowercase\n %-42s %c-V%c retain VMS version numbers\n%s"; +#else /* UNIX */ +static ZCONST char Far UnzipUsageLine4[] = "\ +modifiers: -q quiet mode (-qq => quieter)\n\ + -n never overwrite existing files -a auto-convert any text files\n\ + -o overwrite files WITHOUT prompting -aa treat ALL files as text\n \ + -j junk paths (do not make directories) -v be verbose/print version info\n\ + %c-C%c match filenames case-insensitively %c-L%c make (some) names \ +lowercase\n %-42s %c-V%c retain VMS version numbers\n%s\n\ + -O CHARSET specify a character encoding for DOS, Windows and OS/2 archives\n\ + -I CHARSET specify a character encoding for UNIX and other archives\n\n"; +#endif /* !UNIX */ static ZCONST char Far UnzipUsageLine5[] = "\ Examples (see unzip.txt for more info):\n\ @@ -667,6 +689,10 @@ SETLOCALE(LC_CTYPE,""); +#ifdef UNIX + init_conversion_charsets(); +#endif + #if (defined(__IBMC__) && defined(__DEBUG_ALLOC__)) extern void DebugMalloc(void); @@ -1088,6 +1114,11 @@ argc = *pargc; argv = *pargv; +#ifdef UNIX + extern char OEM_CP[MAX_CP_NAME]; + extern char ISO_CP[MAX_CP_NAME]; +#endif + while (++argv, (--argc > 0 && *argv != NULL && **argv == '-')) { s = *argv + 1; while ((c = *s++) != 0) { /* "!= 0": prevent Turbo C warning */ @@ -1251,6 +1282,35 @@ } break; #endif /* MACOS */ +#ifdef UNIX + case ('I'): + if (negative) { + Info(slide, 0x401, ((char *)slide, + "error: encodings can't be negated")); + return(PK_PARAM); + } else { + if(*s) { /* Handle the -Icharset case */ + /* Assume that charsets can't start with a dash to spot arguments misuse */ + if(*s == '-') { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + strncpy(ISO_CP, s, sizeof(ISO_CP)); + } else { /* -I charset */ + ++argv; + if(!(--argc > 0 && *argv != NULL && **argv != '-')) { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + s = *argv; + strncpy(ISO_CP, s, sizeof(ISO_CP)); + } + while(*(++s)); /* No params straight after charset name */ + } + break; +#endif /* ?UNIX */ case ('j'): /* junk pathnames/directory structure */ if (negative) uO.jflag = FALSE, negative = 0; @@ -1326,6 +1386,35 @@ } else ++uO.overwrite_all; break; +#ifdef UNIX + case ('O'): + if (negative) { + Info(slide, 0x401, ((char *)slide, + "error: encodings can't be negated")); + return(PK_PARAM); + } else { + if(*s) { /* Handle the -Ocharset case */ + /* Assume that charsets can't start with a dash to spot arguments misuse */ + if(*s == '-') { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + strncpy(OEM_CP, s, sizeof(OEM_CP)); + } else { /* -O charset */ + ++argv; + if(!(--argc > 0 && *argv != NULL && **argv != '-')) { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -O argument")); + return(PK_PARAM); + } + s = *argv; + strncpy(OEM_CP, s, sizeof(OEM_CP)); + } + while(*(++s)); /* No params straight after charset name */ + } + break; +#endif /* ?UNIX */ case ('p'): /* pipes: extract to stdout, no messages */ if (negative) { uO.cflag = FALSE; diff -urN ./zipinfo.c ../../unzip-5.52/zipinfo.c --- a/zipinfo.c 2008-04-23 22:52:13.000000000 +0900 +++ b/zipinfo.c 2008-04-23 21:47:24.000000000 +0900 @@ -447,6 +447,10 @@ int tflag_slm=TRUE, tflag_2v=FALSE; int explicit_h=FALSE, explicit_t=FALSE; +#ifdef UNIX + extern char OEM_CP[MAX_CP_NAME]; + extern char ISO_CP[MAX_CP_NAME]; +#endif #ifdef MACOS uO.lflag = LFLAG; /* reset default on each call */ @@ -491,6 +495,35 @@ uO.lflag = 0; } break; +#ifdef UNIX + case ('I'): + if (negative) { + Info(slide, 0x401, ((char *)slide, + "error: encodings can't be negated")); + return(PK_PARAM); + } else { + if(*s) { /* Handle the -Icharset case */ + /* Assume that charsets can't start with a dash to spot arguments misuse */ + if(*s == '-') { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + strncpy(ISO_CP, s, sizeof(ISO_CP)); + } else { /* -I charset */ + ++argv; + if(!(--argc > 0 && *argv != NULL && **argv != '-')) { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + s = *argv; + strncpy(ISO_CP, s, sizeof(ISO_CP)); + } + while(*(++s)); /* No params straight after charset name */ + } + break; +#endif /* ?UNIX */ case 'l': /* longer form of "ls -l" type listing */ if (negative) uO.lflag = -2, negative = 0; @@ -511,6 +544,35 @@ G.M_flag = TRUE; break; #endif +#ifdef UNIX + case ('O'): + if (negative) { + Info(slide, 0x401, ((char *)slide, + "error: encodings can't be negated")); + return(PK_PARAM); + } else { + if(*s) { /* Handle the -Ocharset case */ + /* Assume that charsets can't start with a dash to spot arguments misuse */ + if(*s == '-') { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -I argument")); + return(PK_PARAM); + } + strncpy(OEM_CP, s, sizeof(OEM_CP)); + } else { /* -O charset */ + ++argv; + if(!(--argc > 0 && *argv != NULL && **argv != '-')) { + Info(slide, 0x401, ((char *)slide, + "error: a valid character encoding should follow the -O argument")); + return(PK_PARAM); + } + s = *argv; + strncpy(OEM_CP, s, sizeof(OEM_CP)); + } + while(*(++s)); /* No params straight after charset name */ + } + break; +#endif /* ?UNIX */ case 's': /* default: shorter "ls -l" type listing */ if (negative) uO.lflag = -2, negative = 0;