diff -ur unzip60-orig/fileio.c unzip60/fileio.c --- unzip60-orig/fileio.c 2009-04-20 02:03:44.000000000 +0200 +++ unzip60/fileio.c 2011-02-25 11:57:38.242056429 +0100 @@ -2126,9 +2126,16 @@ /* translate the text coded in the entry's host-dependent "extended ASCII" charset into the compiler's (system's) internal text code page */ +#ifdef UNIX + Ext_ASCII_TO_Native((char *)G.outbuf, G.pInfo->hostnum, + G.pInfo->hostver, G.pInfo->HasUxAtt, + FALSE, OUTBUFSIZ); +#else /* !UNIX */ Ext_ASCII_TO_Native((char *)G.outbuf, G.pInfo->hostnum, G.pInfo->hostver, G.pInfo->HasUxAtt, FALSE); +#endif /* UNIX */ + #ifdef WINDLL /* translate to ANSI (RTL internal codepage may be OEM) */ INTERN_TO_ISO((char *)G.outbuf, (char *)G.outbuf); @@ -2240,8 +2247,13 @@ /* translate the Zip entry filename coded in host-dependent "extended ASCII" into the compiler's (system's) internal text code page */ +#ifdef UNIX + Ext_ASCII_TO_Native(G.filename, G.pInfo->hostnum, G.pInfo->hostver, + G.pInfo->HasUxAtt, (option == DS_FN_L), FILNAMSIZ); +#else /* !UNIX */ Ext_ASCII_TO_Native(G.filename, G.pInfo->hostnum, G.pInfo->hostver, G.pInfo->HasUxAtt, (option == DS_FN_L)); +#endif /* UNIX */ if (G.pInfo->lcflag) /* replace with lowercase filename */ STRLOWER(G.filename, G.filename); Only in unzip60: fileio.c.orig diff -ur unzip60-orig/unix/unix.c unzip60/unix/unix.c --- unzip60-orig/unix/unix.c 2009-01-24 00:31:26.000000000 +0100 +++ unzip60/unix/unix.c 2011-02-25 11:57:38.259028876 +0100 @@ -30,6 +30,10 @@ #define UNZIP_INTERNAL #include "unzip.h" +#include +#include +#include + #ifdef SCO_XENIX # define SYSNDIR #else /* SCO Unix, AIX, DNIX, TI SysV, Coherent 4.x, ... */ @@ -1874,3 +1878,128 @@ } } #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. */ +const static CHARSET_MAP dos_charset_map[] = { + { "ANSI_X3.4-1968", "CP850" }, + { "ISO-8859-1", "CP850" }, + { "CP1252", "CP850" }, + { "KOI8-R", "CP866" }, + { "KOI8-U", "CP866" }, + { "ISO-8859-5", "CP866" } +}; + +typedef struct { + char *locale; + char *archive_charset; +} UTF8_CHARSET_MAP; + +/* In case local charset is UTF-8, lookup archive_charset + by locale. Currently very basic. */ +#define UTF8_CHARSET_MAP_DEFAULT "CP866" +const static UTF8_CHARSET_MAP utf8_charset_map[] = { + { "ja_JP", "CP932" }, + { "zh_CN", "GBK" }, + { "zh_TW", "BIG5" }, + { "ko_KR", "CP949" }, +}; + +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 *loc) +{ + 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); + if (!strcasecmp(local_charset, "UTF-8") || !strcasecmp(local_charset, +"UTF8") ) + { + strcpy(OEM_CP, UTF8_CHARSET_MAP_DEFAULT); + for(i = 0; i < sizeof(utf8_charset_map)/sizeof(UTF8_CHARSET_MAP); + i++) + { + if (!strncmp(loc, utf8_charset_map[i].locale, 5)) + { + strncpy(OEM_CP, utf8_charset_map[i].archive_charset, + sizeof(OEM_CP) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + break; + } + } + } + else + { + 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) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + 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, size_t sbuflen, char *from_charset) +{ + iconv_t cd; + char *s,*d, *buf; + size_t slen, dlen; + const char *local_charset; + + if(*from_charset == '\0') + return; + + local_charset = nl_langinfo(CODESET); + + if((cd = iconv_open(local_charset, from_charset)) == (iconv_t)-1) + return; + + slen = strlen(string); + s = string; + dlen = sbuflen - 1; + d = buf = malloc(sbuflen); + if(!d) + goto cleanup; + if(iconv(cd, (const char **)&s, &slen, &d, &dlen) != (size_t)-1) { + *d = '\0'; + strcpy(string, buf); + } + + free(buf); + cleanup: + iconv_close(cd); +} + +/* Convert a string from OEM_CP to the current locale charset. */ +void oem_intern(char *string, size_t sbuflen) +{ + charset_to_intern(string, sbuflen, OEM_CP); +} + +/* Convert a string from ISO_CP to the current locale charset. */ +void iso_intern(char *string, size_t sbuflen) +{ + charset_to_intern(string, sbuflen, ISO_CP); +} Only in unzip60/unix: unix.c.orig diff -ur unzip60-orig/unix/unxcfg.h unzip60/unix/unxcfg.h --- unzip60-orig/unix/unxcfg.h 2009-04-16 20:36:12.000000000 +0200 +++ unzip60/unix/unxcfg.h 2011-02-25 11:57:38.262941301 +0100 @@ -52,6 +52,7 @@ #include /* off_t, time_t, dev_t, ... */ #include +#include #ifdef NO_OFF_T typedef long zoff_t; @@ -227,4 +228,30 @@ /* 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, ssize) iso_intern(str1, ssize) + +#ifdef _OEM_INTERN +# undef _OEM_INTERN +#endif +#ifndef IZ_OEM2ISO_ARRAY +# define IZ_OEM2ISO_ARRAY +#endif +#define _OEM_INTERN(str1, ssize) oem_intern(str1, ssize) + +void iso_intern(char *, size_t); +void oem_intern(char *, size_t); +void init_conversion_charsets(const char *); + #endif /* !__unxcfg_h */ Only in unzip60/unix: unxcfg.h.orig diff -ur unzip60-orig/unzip.c unzip60/unzip.c --- unzip60-orig/unzip.c 2009-04-16 20:26:52.000000000 +0200 +++ unzip60/unzip.c 2011-02-26 00:03:41.011389301 +0100 @@ -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"; +#ifdef 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"; +#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"; +#endif /* UNIX */ #ifdef MORE static ZCONST char Far ZipInfoUsageLine4[] = " -M page output through built-in \"more\"\n"; @@ -666,6 +676,18 @@ -C match filenames case-insensitively -L make (some) names \ lowercase\n %-42s -V retain VMS version numbers\n%s"; #else /* !VMS */ +#ifdef 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"; +#else /* !UNIX */ static ZCONST char Far UnzipUsageLine4[] = "\ modifiers:\n\ -n never overwrite existing files -q quiet mode (-qq => quieter)\n\ @@ -674,6 +696,7 @@ -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"; +#endif /* UNIX */ #endif /* ?VMS */ #else /* !UNICODE_SUPPORT */ #ifdef VMS @@ -742,6 +765,9 @@ int i; #endif int retcode, error=FALSE; +#ifdef UNIX + const char *loc; +#endif #ifndef NO_EXCEPT_SIGNALS #ifdef REENTRANT savsigs_info *oldsighandlers = NULL; @@ -756,7 +782,12 @@ #endif /* NO_EXCEPT_SIGNALS */ /* initialize international char support to the current environment */ +#ifdef UNIX + loc = SETLOCALE(LC_CTYPE,""); + init_conversion_charsets(loc); +#else /* !UNIX */ SETLOCALE(LC_CTYPE, ""); +#endif /* UNIX */ #ifdef UNICODE_SUPPORT /* see if can use UTF-8 Unicode locale */ @@ -1336,6 +1367,11 @@ 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 +1553,37 @@ } 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) - 1); + ISO_CP[sizeof(ISO_CP) - 1] = '\0'; + } 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) - 1); + ISO_CP[sizeof(ISO_CP) - 1] = '\0'; + } + 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 +1659,37 @@ } 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) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + } 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) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + } + while(*(++s)); /* No params straight after charset name */ + } + break; +#endif /* ?UNIX */ case ('p'): /* pipes: extract to stdout, no messages */ if (negative) { uO.cflag = FALSE; Only in unzip60: unzip.c.orig diff -ur unzip60-orig/unzpriv.h unzip60/unzpriv.h --- unzip60-orig/unzpriv.h 2009-04-20 01:59:26.000000000 +0200 +++ unzip60/unzpriv.h 2011-02-25 11:57:38.275212165 +0100 @@ -3003,6 +3003,18 @@ * All other ports are assumed to code zip entry filenames in ISO 8859-1. */ #ifndef Ext_ASCII_TO_Native +#ifdef UNIX +# define Ext_ASCII_TO_Native(string, hostnum, hostver, isuxatt, islochdr, ssize) \ + if (((hostnum) == FS_FAT_ && \ + !(((islochdr) || (isuxatt)) && \ + ((hostver) == 25 || (hostver) == 26 || (hostver) == 40))) || \ + (hostnum) == FS_HPFS_ || \ + ((hostnum) == FS_NTFS_/* && (hostver) == 50*/)) { \ + _OEM_INTERN((string), (ssize)); \ + } else { \ + _ISO_INTERN((string), (ssize)); \ + } +#else /* !UNIX */ # define Ext_ASCII_TO_Native(string, hostnum, hostver, isuxatt, islochdr) \ if (((hostnum) == FS_FAT_ && \ !(((islochdr) || (isuxatt)) && \ @@ -3013,10 +3025,10 @@ } else { \ _ISO_INTERN((string)); \ } +#endif /* UNIX */ #endif - /**********************/ /* Global constants */ /**********************/ Only in unzip60: unzpriv.h.orig diff -ur unzip60-orig/zipinfo.c unzip60/zipinfo.c --- unzip60-orig/zipinfo.c 2009-02-08 18:04:30.000000000 +0100 +++ unzip60/zipinfo.c 2011-02-25 11:57:38.281586457 +0100 @@ -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 +505,37 @@ 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) - 1); + ISO_CP[sizeof(ISO_CP) - 1] = '\0'; + } 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) - 1); + ISO_CP[sizeof(ISO_CP) - 1] = '\0'; + } + 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 +556,37 @@ 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) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + } 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) - 1); + OEM_CP[sizeof(OEM_CP) - 1] = '\0'; + } + 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; Only in unzip60: zipinfo.c.orig