libextractor

GNU libextractor
Log | Files | Refs | Submodules | README | LICENSE

archive_extractor.c (3894B)

      1 /*
      2      This file is part of libextractor.
      3      Copyright (C) 2012 Christian Grothoff
      4 
      5      libextractor is free software; you can redistribute it and/or modify
      6      it under the terms of the GNU General Public License as published
      7      by the Free Software Foundation; either version 3, or (at your
      8      option) any later version.
      9 
     10      libextractor is distributed in the hope that it will be useful, but
     11      WITHOUT ANY WARRANTY; without even the implied warranty of
     12      MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
     13      General Public License for more details.
     14 
     15      You should have received a copy of the GNU General Public License
     16      along with libextractor; see the file COPYING.  If not, write to the
     17      Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
     18      Boston, MA 02110-1301, USA.
     19  */
     20 /**
     21  * @file plugins/archive_extractor.c
     22  * @brief plugin to support archives (such as TAR)
     23  * @author Christian Grothoff
     24  */
     25 #include "platform.h"
     26 #include "extractor.h"
     27 #include <archive.h>
     28 #include <archive_entry.h>
     29 
     30 /**
     31  * Callback for libarchive for 'reading'.
     32  *
     33  * @param a archive handle
     34  * @param client_data our 'struct EXTRACTOR_ExtractContext'
     35  * @param buff where to store data with pointer to data
     36  * @return number of bytes read
     37  */
     38 static ssize_t
     39 read_cb (struct archive *a,
     40          void *client_data,
     41          const void **buff)
     42 {
     43   struct EXTRACTOR_ExtractContext *ec = client_data;
     44   ssize_t ret;
     45 
     46   *buff = NULL;
     47   if (-1 == (ret = ec->read (ec->cls, (void **) buff, 16 * 1024)))
     48     return ARCHIVE_FATAL;
     49   return ret;
     50 }
     51 
     52 
     53 /**
     54  * Older versions of libarchive do not define __LA_INT64_T.
     55  */
     56 #if ARCHIVE_VERSION_NUMBER < 2000000
     57 #define __LA_INT64_T size_t
     58 #else
     59 #ifndef __LA_INT64_T
     60 #define __LA_INT64_T int64_t
     61 #endif
     62 #endif
     63 
     64 
     65 /**
     66  * Callback for libarchive for 'skipping'.
     67  *
     68  * @param a archive handle
     69  * @param client_data our 'struct EXTRACTOR_ExtractContext'
     70  * @param request number of bytes to skip
     71  * @return number of bytes skipped
     72  */
     73 static __LA_INT64_T
     74 skip_cb (struct archive *a,
     75          void *client_data,
     76          __LA_INT64_T request)
     77 {
     78   struct EXTRACTOR_ExtractContext *ec = client_data;
     79 
     80   if (-1 == ec->seek (ec->cls, request, SEEK_CUR))
     81     return 0;
     82   return request;
     83 }
     84 
     85 
     86 /**
     87  * Main entry method for the ARCHIVE extraction plugin.
     88  *
     89  * @param ec extraction context provided to the plugin
     90  */
     91 void
     92 EXTRACTOR_archive_extract_method (struct EXTRACTOR_ExtractContext *ec);
     93 
     94 void
     95 EXTRACTOR_archive_extract_method (struct EXTRACTOR_ExtractContext *ec)
     96 {
     97   struct archive *a;
     98   struct archive_entry *entry;
     99   const char *fname;
    100   const char *s;
    101   char *format;
    102 
    103   format = NULL;
    104   a = archive_read_new ();
    105 #if ARCHIVE_VERSION_NUMBER >= 3000000
    106   archive_read_support_filter_all (a);
    107 #else
    108   archive_read_support_compression_all (a);
    109 #endif
    110   archive_read_support_format_all (a);
    111   if (archive_read_open2 (a, ec, NULL, &read_cb, &skip_cb, NULL)!= ARCHIVE_OK)
    112     return;
    113 
    114   while (ARCHIVE_OK == archive_read_next_header (a, &entry))
    115   {
    116     if ( (NULL == format) &&
    117          (NULL != (fname = archive_format_name (a))) )
    118       format = strdup (fname);
    119     s = archive_entry_pathname (entry);
    120     if (0 != ec->proc (ec->cls,
    121                        "tar",
    122                        EXTRACTOR_METATYPE_FILENAME,
    123                        EXTRACTOR_METAFORMAT_UTF8,
    124                        "text/plain",
    125                        s, strlen (s) + 1))
    126       break;
    127   }
    128 #if ARCHIVE_VERSION_NUMBER >= 3000000
    129   archive_read_free (a);
    130 #else
    131   archive_read_finish (a);
    132 #endif
    133   if (NULL != format)
    134   {
    135     if (0 != ec->proc (ec->cls,
    136                        "tar",
    137                        EXTRACTOR_METATYPE_FORMAT,
    138                        EXTRACTOR_METAFORMAT_UTF8,
    139                        "text/plain", format, strlen (format) + 1))
    140     {
    141       free (format);
    142       return;
    143     }
    144     free (format);
    145   }
    146 }
    147 
    148 
    149 /* end of tar_extractor.c */