Based on https://bugzilla.altlinux.org/show_bug.cgi?format=multiple&id=4871 Note: For UTF-8 locale this patch uses Cyrillic CP866 by default. See CHARSET_MAP below. diff -ur unzip60.old/unix/unix.c unzip60/unix/unix.c --- unzip60.old/unix/unix.c 2010-06-24 13:09:57.000000000 +0300 +++ unzip60/unix/unix.c 2010-06-24 13:14:28.000000000 +0300 @@ -30,6 +30,9 @@ #define UNZIP_INTERNAL #include "unzip.h" +#include +#include + #ifdef SCO_XENIX # define SYSNDIR #else /* SCO Unix, AIX, DNIX, TI SysV, Coherent 4.x, ... */ @@ -1907,3 +1910,93 @@ } } #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; + buflen = 4*slen; + if (buflen > OUTBUFSIZ) // this should prevent crashes on some VERY long names + buflen = OUTBUFSIZ; + dlen = buflen; + 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 -ur unzip60.old/unix/unxcfg.h unzip60/unix/unxcfg.h --- unzip60.old/unix/unxcfg.h 2009-04-16 21:36:12.000000000 +0300 +++ unzip60/unix/unxcfg.h 2010-06-24 13:14:28.000000000 +0300 @@ -227,4 +227,30 @@ /* 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 + +#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 -ur unzip60.old/unzip.c unzip60/unzip.c --- unzip60.old/unzip.c 2009-04-16 21:26:52.000000000 +0300 +++ unzip60/unzip.c 2010-06-24 13:20:11.000000000 +0300 @@ -327,11 +327,21 @@ -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 -T print file times in sortable decimal format\ \n -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 -T print file times in sortable decimal format\ +\n -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 static ZCONST char Far ZipInfoUsageLine4[] = " -M page output through built-in \"more\"\n"; @@ -666,6 +676,7 @@ -C match filenames case-insensitively -L make (some) names \ lowercase\n %-42s -V retain VMS version numbers\n%s"; #else /* !VMS */ +#ifndef UNIX static ZCONST char Far UnzipUsageLine4[] = "\ modifiers:\n\ -n never overwrite existing files -q quiet mode (-qq => quieter)\n\ @@ -674,6 +685,18 @@ -U use escapes for all non-ASCII Unicode -UU ignore any Unicode fields\n\ -C match filenames case-insensitively -L make (some) names \ lowercase\n %-42s -V retain VMS version numbers\n%s"; +#else /* UNIX */ +static ZCONST char Far UnzipUsageLine4[] = "\ +modifiers:\n\ + -n never overwrite existing files -q quiet mode (-qq => quieter)\n\ + -o overwrite files WITHOUT prompting -a auto-convert any text files\n\ + -j junk paths (do not make directories) -aa treat ALL files as text\n\ + -U use escapes for all non-ASCII Unicode -UU ignore any Unicode fields\n\ + -C match filenames case-insensitively -L make (some) names \ +lowercase\n %-42s -V 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 */ #endif /* ?VMS */ #else /* !UNICODE_SUPPORT */ #ifdef VMS @@ -803,6 +826,10 @@ #endif /* UNICODE_SUPPORT */ +#ifdef UNIX + init_conversion_charsets(); +#endif + #if (defined(__IBMC__) && defined(__DEBUG_ALLOC__)) extern void DebugMalloc(void); @@ -1336,6 +1363,11 @@ 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 */ @@ -1517,6 +1549,35 @@ } 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; @@ -1592,6 +1653,35 @@ } 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 -ur unzip60.old/unzpriv.h unzip60/unzpriv.h --- unzip60.old/unzpriv.h 2010-06-24 13:09:57.000000000 +0300 +++ unzip60/unzpriv.h 2010-06-24 13:23:06.000000000 +0300 @@ -3008,7 +3008,7 @@ !(((islochdr) || (isuxatt)) && \ ((hostver) == 25 || (hostver) == 26 || (hostver) == 40))) || \ (hostnum) == FS_HPFS_ || \ - ((hostnum) == FS_NTFS_ && (hostver) == 50)) { \ + ((hostnum) == FS_NTFS_/* && (hostver) == 50*/)) { \ _OEM_INTERN((string)); \ } else { \ _ISO_INTERN((string)); \ diff -ur unzip60.old/zipinfo.c unzip60/zipinfo.c --- unzip60.old/zipinfo.c 2009-02-08 19:04:30.000000000 +0200 +++ unzip60/zipinfo.c 2010-06-24 13:14:31.000000000 +0300 @@ -457,6 +457,10 @@ 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 */ @@ -501,6 +505,35 @@ 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; @@ -521,6 +554,35 @@ 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;