diff options
author | Glenn L McGrath | 2000-12-15 06:50:09 +0000 |
---|---|---|
committer | Glenn L McGrath | 2000-12-15 06:50:09 +0000 |
commit | 4f1b0124c73ba30ba6cd97840b6fff0df90ad40c (patch) | |
tree | 30f251739028ad317cc6ded61bbe656425a1b832 /archival/ar.c | |
parent | 8abc78aa84a88c12fca41ae10e84898ad5089cae (diff) | |
download | busybox-4f1b0124c73ba30ba6cd97840b6fff0df90ad40c.zip busybox-4f1b0124c73ba30ba6cd97840b6fff0df90ad40c.tar.gz |
Rewrite, fix a bug with multiple long filenames, simplify structure, remove cruft, make code conform closer to style guide, saves 350 Bytes.
Diffstat (limited to 'archival/ar.c')
-rw-r--r-- | archival/ar.c | 456 |
1 files changed, 142 insertions, 314 deletions
diff --git a/archival/ar.c b/archival/ar.c index 88cdd4f..bd54bb6 100644 --- a/archival/ar.c +++ b/archival/ar.c @@ -3,7 +3,7 @@ * Mini ar implementation for busybox * * Copyright (C) 2000 by Glenn McGrath - * Written by Glenn McGrath <bug1@netconnect.com.au> 1 June 2000 + * Written by Glenn McGrath <bug1@optushome.com.au> 1 June 2000 * * Based in part on BusyBox tar, Debian dpkg-deb and GNU ar. * @@ -21,317 +21,138 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * - * Last modified 20 September 2000 */ -#include <stdio.h> -#include <string.h> #include <fcntl.h> -#include <errno.h> -#include <ctype.h> -#include <time.h> -#include <utime.h> -#include <unistd.h> -#include <stdlib.h> -#include <sys/types.h> -#include <sys/stat.h> -#include <malloc.h> #include "busybox.h" -#define BLOCK_SIZE 60 -#define PRESERVE_DATE 1 /* preserve original dates */ -#define VERBOSE 2 /* be verbose */ -#define DISPLAY 4 /* display contents */ -#define EXT_TO_FILE 8 /* extract contents of archive */ -#define EXT_TO_STDOUT 16 /* extract to stdout */ -#define RECURSIVE 32 +typedef struct ar_headers_s { + char *name; + size_t size; + uid_t uid; + gid_t gid; + mode_t mode; + time_t mtime; + off_t offset; + struct ar_headers_s *next; +} ar_headers_t; -#define MAX_NAME_LENGTH 100 - -//#define BB_DECLARE_EXTERN -//#define bb_need_io_error -//#include "messages.c" - -//#define BB_AR_EXPERIMENTAL_UNTAR - -#if defined BB_AR_EXPERIMENTAL_UNTAR -typedef struct rawTarHeader { - char name[100]; /* 0-99 */ - char mode[8]; /* 100-107 */ - char uid[8]; /* 108-115 */ - char gid[8]; /* 116-123 */ - char size[12]; /* 124-135 */ - char mtime[12]; /* 136-147 */ - char chksum[8]; /* 148-155 */ - char typeflag; /* 156-156 */ - char linkname[100]; /* 157-256 */ - char magic[6]; /* 257-262 */ - char version[2]; /* 263-264 */ - char uname[32]; /* 265-296 */ - char gname[32]; /* 297-328 */ - char devmajor[8]; /* 329-336 */ - char devminor[8]; /* 337-344 */ - char prefix[155]; /* 345-499 */ - char padding[12]; /* 500-512 */ -} rawTarHeader_t; -#endif - -typedef struct rawArHeader { /* Byte Offset */ - char name[16]; /* 0-15 */ - char date[12]; /* 16-27 */ - char uid[6], gid[6]; /* 28-39 */ - char mode[8]; /* 40-47 */ - char size[10]; /* 48-57 */ - char fmag[2]; /* 58-59 */ -} rawArHeader_t; - -typedef struct headerL { - char name[MAX_NAME_LENGTH]; - size_t size; - uid_t uid; - gid_t gid; - mode_t mode; - time_t mtime; - off_t offset; - struct headerL *next; -} headerL_t; - -#if defined BB_AR_EXPERIMENTAL_UNTAR /* - * identify Tar header (magic field) and reset srcFd to entry position + * return the headerL_t struct for the filename descriptor */ -static int checkTarMagic(int srcFd) -{ - off_t headerStart; - char magic[6]; - - headerStart = lseek(srcFd, 0, SEEK_CUR); - lseek(srcFd, (off_t) 257, SEEK_CUR); - full_read(srcFd, magic, 6); - lseek(srcFd, headerStart, SEEK_SET); - if (strncmp(magic, "ustar", 5)!=0) - return(FALSE); - return(TRUE); -} - - -static int readTarHeader(int srcFd, headerL_t *current) +static ar_headers_t get_headers(int srcFd) { - rawTarHeader_t rawTarHeader; - unsigned char *temp = (unsigned char *) &rawTarHeader; - long sum = 0; - int i; - off_t initialOffset; - - initialOffset = lseek(srcFd, 0, SEEK_CUR); - if (full_read(srcFd, (char *) &rawTarHeader, 512) != 512) { - lseek(srcFd, initialOffset, SEEK_SET); - return(FALSE); - } - for (i = 0; i < 148 ; i++) - sum += temp[i]; - sum += ' ' * 8; - for (i = 156; i < 512 ; i++) - sum += temp[i]; - if (sum!= strtol(rawTarHeader.chksum, NULL, 8)) - return(FALSE); - sscanf(rawTarHeader.name, "%s", current->name); - current->size = strtol(rawTarHeader.size, NULL, 8); - current->uid = strtol(rawTarHeader.uid, NULL, 8); - current->gid = strtol(rawTarHeader.gid, NULL, 8); - current->mode = strtol(rawTarHeader.mode, NULL, 8); - current->mtime = strtol(rawTarHeader.mtime, NULL, 8); - current->offset = lseek(srcFd, 0 , SEEK_CUR); - - current->next = (headerL_t *) xmalloc(sizeof(headerL_t)); - current = current->next; - return(TRUE); -} -#endif - -/* - * identify Ar header (magic) and reset srcFd to entry position - */ -static int checkArMagic(int srcFd) -{ - off_t headerStart; - char arMagic[8]; - - headerStart = lseek(srcFd, 0, SEEK_CUR); - if (full_read(srcFd, arMagic, 8) != 8) { - error_msg("fatal error\n"); - return (FALSE); - } - lseek(srcFd, headerStart, SEEK_SET); - - if (strncmp(arMagic,"!<arch>",7) != 0) - return(FALSE); - return(TRUE); -} - -/* - * get, check and correct the converted header - */ -static int readArEntry(int srcFd, headerL_t *entry) -{ - size_t nameLength; - rawArHeader_t rawArHeader; - off_t initialOffset; - - initialOffset = lseek(srcFd, 0, SEEK_CUR); - if (full_read(srcFd, (char *) &rawArHeader, 60) != 60) { - lseek(srcFd, initialOffset, SEEK_SET); - return(FALSE); - } - if ((rawArHeader.fmag[0]!='`') || (rawArHeader.fmag[1]!='\n')) { - lseek(srcFd, initialOffset, SEEK_SET); - return(FALSE); - } - - strncpy(entry->name, rawArHeader.name, 16); - nameLength=strcspn(entry->name, " \\"); - entry->name[nameLength]='\0'; - parse_mode(rawArHeader.mode, &entry->mode); - entry->mtime = atoi(rawArHeader.date); - entry->uid = atoi(rawArHeader.uid); - entry->gid = atoi(rawArHeader.gid); - entry->size = (size_t) atoi(rawArHeader.size); - entry->offset = initialOffset + (off_t) 60; - - nameLength = strcspn(entry->name, "/"); + typedef struct raw_ar_header_s { /* Byte Offset */ + char name[16]; /* 0-15 */ + char date[12]; /* 16-27 */ + char uid[6]; + char gid[6]; /* 28-39 */ + char mode[8]; /* 40-47 */ + char size[10]; /* 48-57 */ + char fmag[2]; /* 58-59 */ + } raw_ar_header_t; + raw_ar_header_t raw_ar_header; + + ar_headers_t *head, *entry; + char ar_magic[8]; + char *long_name=NULL; - /* handle GNU style short filenames, strip trailing '/' */ - if (nameLength > 0) - entry->name[nameLength]='\0'; + head = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); + entry = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); - /* handle GNU style long filenames */ - if (nameLength == 0) { - /* escape from recursive call */ - if (entry->name[1]=='0') - return(TRUE); - - /* the data section contains the real filename */ - if (entry->name[1]=='/') { - char tempName[MAX_NAME_LENGTH]; - - if (entry->size > MAX_NAME_LENGTH) - entry->size = MAX_NAME_LENGTH; - full_read(srcFd, tempName, entry->size); - tempName[entry->size-3]='\0'; - - /* read the second header for this entry */ - /* be carefull, this is recursive */ - if (readArEntry(srcFd, entry)==FALSE) - return(FALSE); + /* check ar magic */ + if (full_read(srcFd, ar_magic, 8) != 8) + error_msg_and_die("cannot read magic\n"); + if (strncmp(ar_magic,"!<arch>",7) != 0) + error_msg_and_die("invalid magic\n"); + + while (full_read(srcFd, (char *) &raw_ar_header, 60)==60) { + /* check the end of header markers are valid */ + if ((raw_ar_header.fmag[0]!='`') || (raw_ar_header.fmag[1]!='\n')) { + char newline[1]; + if (raw_ar_header.fmag[1]!='`') { + break; + } + /* some version of ar, have an extra '\n' after each entry */ + read(srcFd, newline, 1); + if (newline[0]!='\n') { + break; + } + /* fix up the header, we started reading 1 byte too early due to a '\n' */ + memmove((char *) &raw_ar_header, (char *)&raw_ar_header+1, 59); + /* dont worry about adding the last '\n', we dont need it now */ + } - if ((entry->name[0]='/') && (entry->name[1]='0')) - strcpy(entry->name, tempName); + entry->size = (size_t) atoi(raw_ar_header.size); + /* long filenames have '/' as the first character */ + if (raw_ar_header.name[0] == '/') { + if (raw_ar_header.name[1] == '/') { + /* multiple long filenames are stored as data in one entry */ + long_name = (char *) xrealloc(long_name, entry->size); + full_read(srcFd, long_name, entry->size); + continue; + } else { - error_msg("Invalid long filename\n"); - return(FALSE); + /* The number after the '/' indicates the offset in the ar data section + (saved in variable long_name) that conatains the real filename */ + const int long_name_offset = (int) atoi((char *) &raw_ar_header.name[1]); + entry->name = xmalloc(strlen(&long_name[long_name_offset])); + strcpy(entry->name, &long_name[long_name_offset]); } } - } - return(TRUE); -} - -/* - * return the headerL_t struct for the specified filename - */ -static headerL_t *getHeaders(int srcFd, headerL_t *head, int funct) -{ -#if defined BB_AR_EXPERIMENTAL_UNTAR - int tar=FALSE; -#endif - int ar=FALSE; - headerL_t *list; - off_t initialOffset; - - list = (headerL_t *) xmalloc(sizeof(headerL_t)); - initialOffset=lseek(srcFd, 0, SEEK_CUR); - if (checkArMagic(srcFd)==TRUE) - ar=TRUE; - -#if defined BB_AR_EXPERIMENTAL_UNTAR - if (checkTarMagic(srcFd)==TRUE) - tar=TRUE; - - if (tar==TRUE) { - while(readTarHeader(srcFd, list)==TRUE) { - off_t tarOffset; - list->next = (headerL_t *) xmalloc(sizeof(headerL_t)); - *list->next = *head; - *head = *list; - - /* recursive check for sub-archives */ - if ((funct & RECURSIVE) == RECURSIVE) - head = getHeaders(srcFd, head, funct); - tarOffset = (off_t) head->size/512; - if ( head->size % 512 > 0) - tarOffset++; - tarOffset=tarOffset*512; - lseek(srcFd, head->offset + tarOffset, SEEK_SET); - } - } -#endif - - if (ar==TRUE) { - lseek(srcFd, 8, SEEK_CUR); - while(1) { - if (readArEntry(srcFd, list) == FALSE) { - lseek(srcFd, ++initialOffset, SEEK_CUR); - if (readArEntry(srcFd, list) == FALSE) - return(head); - } - list->next = (headerL_t *) xmalloc(sizeof(headerL_t)); - *list->next = *head; - *head = *list; - /* recursive check for sub-archives */ - if (funct & RECURSIVE) - head = getHeaders(srcFd, head, funct); - lseek(srcFd, head->offset + head->size, SEEK_SET); + else { + /* short filenames */ + entry->name = xmalloc(16); + strncpy(entry->name, raw_ar_header.name, 16); } + entry->name[strcspn(entry->name, " /")]='\0'; + + /* convert the rest of the now valid char header to its typed struct */ + parse_mode(raw_ar_header.mode, &entry->mode); + entry->mtime = atoi(raw_ar_header.date); + entry->uid = atoi(raw_ar_header.uid); + entry->gid = atoi(raw_ar_header.gid); + entry->offset = lseek(srcFd, 0, SEEK_CUR); + + /* add this entries header to our combined list */ + entry->next = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); + *entry->next = *head; + *head = *entry; + lseek(srcFd, (off_t) entry->size, SEEK_CUR); } - return(head); -} - -/* - * find an entry in the linked list matching the filename - */ -static headerL_t *findEntry(headerL_t *head, const char *filename) -{ - while(head->next != NULL) { - if (strcmp(filename, head->name)==0) - return(head); - head=head->next; - } - return(NULL); + return(*head); } extern int ar_main(int argc, char **argv) { - int funct = 0, opt=0; + const int preserve_date = 1; /* preserve original dates */ + const int verbose = 2; /* be verbose */ + const int display = 4; /* display contents */ + const int extract_to_file = 8; /* extract contents of archive */ + const int extract_to_stdout = 16; /* extract to stdout */ + + int funct = 0, opt=0; int srcFd=0, dstFd=0; - headerL_t *header, *entry, *extractList; - while ((opt = getopt(argc, argv, "ovtpxR")) != -1) { + ar_headers_t head, *extract_list=NULL; + + extract_list = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); + + while ((opt = getopt(argc, argv, "ovtpx")) != -1) { switch (opt) { case 'o': - funct |= PRESERVE_DATE; + funct |= preserve_date; break; case 'v': - funct |= VERBOSE; + funct |= verbose; break; case 't': - funct |= DISPLAY; - break; - case 'x': - funct |= EXT_TO_FILE; + funct |= display; break; case 'p': - funct |= EXT_TO_STDOUT; + funct |= extract_to_stdout; break; - case 'R': - funct |= RECURSIVE; + case 'x': + funct |= extract_to_file; break; default: usage(ar_usage); @@ -345,47 +166,54 @@ extern int ar_main(int argc, char **argv) if ( (srcFd = open(argv[optind], O_RDONLY)) < 0) error_msg_and_die("Cannot read %s\n", argv[optind]); - optind++; - entry = (headerL_t *) xmalloc(sizeof(headerL_t)); - header = (headerL_t *) xmalloc(sizeof(headerL_t)); - extractList = (headerL_t *) xmalloc(sizeof(headerL_t)); + optind++; + head = get_headers(srcFd); - header = getHeaders(srcFd, header, funct); /* find files to extract or display */ - if (optind<argc) { - /* only handle specified files */ - while(optind < argc) { - if ( (entry = findEntry(header, argv[optind])) != NULL) { - entry->next = (headerL_t *) xmalloc(sizeof(headerL_t)); - *entry->next = *extractList; - *extractList = *entry; + /* search through argv and build extract list */ + for (;optind<argc; optind++) { + ar_headers_t *ar_entry; + ar_entry = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); + ar_entry = &head; + while (ar_entry->next != NULL) { + if (strcmp(argv[optind], ar_entry->name) == 0) { + ar_headers_t *tmp; + tmp = (ar_headers_t *) xmalloc(sizeof(ar_headers_t)); + *tmp = *extract_list; + *extract_list = *ar_entry; + extract_list->next = tmp; + break; } - optind++; - } + ar_entry=ar_entry->next; + } + } + + /* if individual files not found extract all files */ + if (extract_list->next==NULL) { + extract_list = &head; } - else - extractList = header; - while(extractList->next != NULL) { - if (funct & EXT_TO_FILE) { - if (is_directory(extractList->name, TRUE, NULL)==FALSE) - create_path(extractList->name, 0666); - dstFd = open(extractList->name, O_WRONLY | O_CREAT, extractList->mode); - lseek(srcFd, extractList->offset, SEEK_SET); - copy_file_chunk(srcFd, dstFd, (size_t) extractList->size); + /* find files to extract or display */ + while (extract_list->next != NULL) { + if (funct & extract_to_file) { + dstFd = open(extract_list->name, O_WRONLY | O_CREAT, extract_list->mode); + } + else if (funct & extract_to_stdout) { + dstFd = fileno(stdout); + } + if ((funct & extract_to_file) || (funct & extract_to_stdout)) { + lseek(srcFd, extract_list->offset, SEEK_SET); + copy_file_chunk(srcFd, dstFd, (size_t) extract_list->size); } - if (funct & EXT_TO_STDOUT) { - lseek(srcFd, extractList->offset, SEEK_SET); - copy_file_chunk(srcFd, fileno(stdout), (size_t) extractList->size); + if (funct & verbose) { + printf("%s %d/%d %8d %s ", mode_string(extract_list->mode), + extract_list->uid, extract_list->gid, + extract_list->size, time_string(extract_list->mtime)); } - if ( (funct & DISPLAY) || (funct & VERBOSE)) { - if (funct & VERBOSE) - printf("%s %d/%d %8d %s ", mode_string(extractList->mode), - extractList->uid, extractList->gid, - extractList->size, time_string(extractList->mtime)); - printf("%s\n", extractList->name); + if ((funct & display) || (funct & verbose)){ + printf("%s\n", extract_list->name); } - extractList=extractList->next; + extract_list=extract_list->next; } return EXIT_SUCCESS; } |