summaryrefslogtreecommitdiffstats
path: root/delete_lz.cc
diff options
context:
space:
mode:
Diffstat (limited to 'delete_lz.cc')
-rw-r--r--delete_lz.cc167
1 files changed, 167 insertions, 0 deletions
diff --git a/delete_lz.cc b/delete_lz.cc
new file mode 100644
index 0000000..11c3a14
--- /dev/null
+++ b/delete_lz.cc
@@ -0,0 +1,167 @@
+/* Tarlz - Archiver with multimember lzip compression
+ Copyright (C) 2013-2019 Antonio Diaz Diaz.
+
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 2 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>.
+*/
+
+#define _FILE_OFFSET_BITS 64
+
+#include <cctype>
+#include <cerrno>
+#include <climits>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <string>
+#include <vector>
+#include <pthread.h>
+#include <stdint.h>
+#include <unistd.h>
+#include <lzlib.h>
+
+#include "arg_parser.h"
+#include "lzip_index.h"
+#include "tarlz.h"
+
+
+/* Deleting from a corrupt archive must not worsen the corruption. Stop and
+ tail-copy as soon as corruption is found. */
+int delete_members_lz( const char * const archive_namep,
+ const Arg_parser & parser,
+ std::vector< char > & name_pending,
+ const Lzip_index & lzip_index,
+ const int filenames, const int infd, const int outfd,
+ const bool missing_crc, const bool permissive )
+ {
+ Resizable_buffer rbuf;
+ LZ_Decoder * const decoder = LZ_decompress_open();
+ if( !rbuf.size() || !decoder || LZ_decompress_errno( decoder ) != LZ_ok )
+ { show_error( mem_msg ); return 1; }
+
+ long long istream_pos = 0; // source of next data move
+ const long long cdata_size = lzip_index.cdata_size();
+ int retval = 0;
+ for( long i = 0; i < lzip_index.members(); ++i )
+ {
+ const long long mdata_pos = lzip_index.dblock( i ).pos();
+ long long data_pos = mdata_pos;
+ const long long mdata_end = lzip_index.dblock( i ).end();
+ if( data_pos >= mdata_end ) continue; // empty lzip member
+ const long long member_pos = lzip_index.mblock( i ).pos();
+ long long file_pos = member_pos;
+ const long long member_end = lzip_index.mblock( i ).end();
+
+ long long member_begin = 0; // first pos of current tar member
+ Extended extended; // metadata from extended records
+ bool prev_extended = false; // prev header was extended
+ LZ_decompress_reset( decoder ); // prepare for new member
+ if( !safe_seek( infd, member_pos ) ) { retval = 1; break; }
+ while( true ) // process one tar header per iteration
+ {
+ if( data_pos >= mdata_end )
+ {
+ if( data_pos == mdata_end && !prev_extended ) break;
+ // member end exceeded or ends in extended
+ show_file_error( archive_namep, "Member misalignment found." );
+ retval = 2; goto done;
+ }
+ if( !prev_extended ) member_begin = data_pos;
+ Tar_header header;
+ const char * msg = 0;
+ retval = archive_read_lz( decoder, infd, file_pos, member_end,
+ cdata_size, header, header_size, &msg );
+ if( retval != 0 ) { show_file_error( archive_namep, msg ); goto done; }
+ data_pos += header_size;
+ if( !verify_ustar_chksum( header ) )
+ {
+ if( block_is_zero( header, header_size ) ) // EOF
+ {
+ if( prev_extended && !permissive )
+ { show_file_error( archive_namep, fv_msg1 ); retval = 2; }
+ goto done;
+ }
+ show_file_error( archive_namep, ( data_pos > header_size ) ?
+ bad_hdr_msg : posix_lz_msg );
+ retval = 2;
+ goto done;
+ }
+
+ const Typeflag typeflag = (Typeflag)header[typeflag_o];
+ if( typeflag == tf_global )
+ {
+ if( prev_extended && !permissive )
+ { show_file_error( archive_namep, fv_msg2 ); retval = 2; goto done; }
+ Extended dummy; // global headers are parsed and ignored
+ retval = parse_records_lz( decoder, infd, file_pos, member_end,
+ cdata_size, data_pos, dummy, header,
+ rbuf, &msg, true );
+ if( retval == 0 ) continue;
+ show_file_error( archive_namep, msg ? msg : gblrec_msg );
+ goto done;
+ }
+ if( typeflag == tf_extended )
+ {
+ if( prev_extended && !permissive ) { msg = fv_msg3; retval = 2; }
+ else retval = parse_records_lz( decoder, infd, file_pos, member_end,
+ cdata_size, data_pos, extended, header,
+ rbuf, &msg, permissive );
+ if( retval == 0 && !extended.crc_present() && missing_crc )
+ { msg = mcrc_msg; retval = 2; }
+ if( retval == 0 ) { prev_extended = true; continue; }
+ show_file_error( archive_namep, msg ? msg : extrec_msg );
+ goto done;
+ }
+ prev_extended = false;
+
+ extended.fill_from_ustar( header ); // copy metadata from header
+
+ long long rest = extended.file_size();
+ const int rem = rest % header_size;
+ if( rem ) rest += header_size - rem; // padding
+ if( data_pos + rest >= mdata_end ) data_pos += rest;
+ else // skip tar member
+ if( ( retval = skip_member_lz( decoder, infd, file_pos, member_end,
+ cdata_size, data_pos, rest, &msg ) ) != 0 )
+ goto done;
+
+ if( !check_skip_filename( parser, name_pending, extended.path().c_str(),
+ filenames ) ) // delete tar member
+ {
+ // verify that members match
+ if( member_begin != mdata_pos || data_pos != mdata_end )
+ { show_file_error( extended.path().c_str(),
+ "Can't delete: not individually compressed." );
+ retval = 2; extended.reset(); continue; }
+ if( !show_member_name( extended, header, 1, rbuf ) )
+ { retval = 1; goto done; }
+ const long long size = member_pos - istream_pos;
+ if( size > 0 ) // move pending data each time a member is deleted
+ {
+ if( istream_pos == 0 )
+ { if( !safe_seek( outfd, size ) ) { retval = 1; break; } }
+ else if( !safe_seek( infd, istream_pos ) ||
+ !copy_file( infd, outfd, size ) ) { retval = 1; break; }
+ }
+ istream_pos = member_end;
+ }
+ extended.reset();
+ }
+ }
+done:
+ if( LZ_decompress_close( decoder ) < 0 && !retval )
+ { show_error( "LZ_decompress_close failed." ); retval = 1; }
+ // tail copy keeps trailing data
+ return tail_copy( archive_namep, parser, name_pending, lzip_index,
+ istream_pos, infd, outfd, retval );
+ }