Described libisofs MD5 recording and implemented checksum array checksum

and stream detectable session checksum tag.

doc/checksums.txt

@@ -0,0 +1,101 @@
+
+               Description of libisofs MD5 checksumming
+
+          by Thomas Schmitt    - mailto:scdbackup@gmx.net
+          Libburnia project    - mailto:libburn-hackers@pykix.org
+
+
+MD5 is a 128 bit message digest with a very low probability to be the same for
+any pair of differing data files. It is described in RFC 1321. and can be
+computed e.g. by program md5sum.
+
+libisofs can equip its images with MD5 checksums for the whole session and
+for each single data file. See libisofs.h, iso_write_opts_set_record_md5().
+The checksums get loaded together with the directory tree if this is enabled by
+iso_read_opts_set_no_md5(). Loaded checksums can be inquired by
+iso_image_get_session_md5() and iso_file_get_md5().
+libisofs has own MD5 computation functions: iso_md5_start(), iso_md5_compute(),
+iso_md5_clone(), iso_md5_end().
+See iso_file_get_stream(), iso_stream_open() et.al. for reading file content
+from the loaded image.
+
+
+                      Representation in the Image
+
+The checksums are stored in an area at the end of the session, in order to
+allow quick loading from media with slow random access.
+There is an array of MD5 entries and a single block with a checksum tag.
+
+Location and layout of the checksum area is recorded as AAIP attribute
+"isofs.ca" of the root node.
+See doc/susp_aaip_2_0.txt for a general description of AAIP and
+doc/susp_aaip_isofs_names.txt for the layout of "isofs.ca".
+
+Because the inquiry of this attribute demands loading of the image tree,
+there is also a checksum tag after the checksum area.
+This tag can be detected on the fly when reading and checksumming the session
+from the start point as learned from a media table-of-content. It covers not
+only the payload of the session but also the checksum area.
+
+The single data files hold an index to their MD5 checksum in individual AAIP
+attributes "isofs.cx". Index I means: array base address + 16 * I. 
+
+The checksums cover the data content as it was actually written into the ISO
+image stream, not necessarily as it was on hard disk before or afterwards.
+This implies that content filtered files bear the MD5 of the filtered data
+and not of the original files on disk. When checkreading, one has to avoid
+any filtering. Dig out the stream which directly reads image data by calling 
+iso_stream_get_input_stream() until it returns NULL and use
+iso_stream_get_size() rather than iso_file_get_size().
+
+
+                              The MD5 array
+
+If there are N checksummed data files then the array consists of N + 2 entries
+with 16 bytes each.
+
+Entry number 0 holds a session checksum which covers the range from the session
+start block up to (but not including) the start block of the checksum area.
+This range is described by attribute "isofs.ca" of the root node.
+
+Entries 1 to N hold the checksums of individual data files.
+
+Entry number N + 1 holds the MD5 checksum of entries 0 to N.
+
+
+                             The Checksum Tag
+
+The next block after the array begins with the checksum tag and is padded
+by 0-bytes. The tag is a single line of printable text and has the following
+format:
+
+ libisofs_checksum_tag_v1 pos=... range_start=... range_size=... md5=... \n
+
+Example:
+  libisofs_checksum_tag_v1 pos=81552 range_start=32 range_size=81520 md5=f172b994e8eb565a011d220b2a8b7a19
+
+There are four parameters:
+
+  pos=
+  gives the block address where the tag supposes itself to be stored.
+  If this does not match the block address where the tag is found then this
+  either indicates that the tag is payload of the image or that the image has
+  been relocated. (The latter makes the image unuable.)
+
+  range_start=
+  The block address where the session is supposed to start. If this does not
+  match the session start on media then the image volume descriptors of the
+  image have been been relocated. (The latter can happen with the overwriteable
+  media.)
+
+  range_size=
+  The number of blocks beginning at range_start which are covered by the
+  checksum of the tag.  
+
+  md5=
+  The checksum of the tag. Encoded as 32 hex digits.
+  
+  The newline character at the end is mandatory. For now all bytes of the
+  block after that newline shall be zero. There may arise future extensions.
+
+ 

libisofs/ecma119.c

@@ -1057,6 +1057,9 @@ int ecma119_image_new(IsoImage *src, IsoWriteOpts *opts, Ecma119Image **img)
     target->checksum_ctx = NULL;
     target->checksum_counter = 0;
     target->checksum_buffer = NULL;
+    target->checksum_array_pos = 0;
+    target->checksum_range_start = 0;
+    target->checksum_range_size = 0;
 
 #endif
 

libisofs/ecma119.h

@@ -456,6 +456,9 @@ struct ecma119_image
     off_t checksum_counter;
     char image_md5[16];
     char *checksum_buffer;
+    uint32_t checksum_array_pos;
+    uint32_t checksum_range_start;
+    uint32_t checksum_range_size;
 
 #endif /* Libisofs_with_checksumS */
 

libisofs/md5.c

@@ -10,6 +10,7 @@
 #include <stdint.h>
 #include <stdlib.h>
 #include <string.h>
+#include <stdio.h>
 
 #include "writer.h"
 #include "messages.h"
@@ -364,7 +365,7 @@ int iso_md5_clone(void *old_md5_context, void **new_md5_context)
 {
     int ret;
 
-    ret = libisofs_md5(new_md5_context, old_md5_context, 0, NULL, 4);
+    ret = libisofs_md5(new_md5_context, old_md5_context, 0, NULL, 1 | 4);
     if (ret < 0)
         return ISO_OUT_OF_MEM;
     if (ret == 0)
@@ -474,18 +475,22 @@ int checksum_writer_compute_data_blocks(IsoImageWriter *writer)
     size_t size;
     Ecma119Image *t;
     int ret;
-    unsigned int lba;
 
     if (writer == NULL) {
         return ISO_ASSERT_FAILURE;
     }
     t = writer->target;
 
-    lba = t->curblock; /* (t->curblock already contains t->ms_block) */ 
+    t->checksum_array_pos = t->curblock;
-    size = (t->checksum_idx_counter + 2) / 128 + 1;
+                               /* (t->curblock already contains t->ms_block) */ 
-    t->curblock += size;
+    t->checksum_range_start = t->ms_block;
-
+    size = (t->checksum_idx_counter + 2) / 128;
-    /* >>> ??? reserve extra block for stream detectable checksum */;
+    if (size * 128 < t->checksum_idx_counter + 2)
+        size++;
+    t->curblock += size + 1;
+                     /* + 1 = extra block for stream detectable checksum tag */
+    t->checksum_range_size = t->checksum_array_pos + size
+                             - t->checksum_range_start;
 
     /* Allocate array of MD5 sums */
     t->checksum_buffer = calloc(size, 2048);
@@ -499,7 +504,8 @@ int checksum_writer_compute_data_blocks(IsoImageWriter *writer)
 
     /* Record lba,count,size,cecksum_type in "isofs.ca" of root node */
     ret = iso_root_set_isofsca((IsoNode *) t->image->root,
-                               (unsigned int) t->ms_block, lba,
+                               t->checksum_range_start, 
+                               t->checksum_array_pos,
                                t->checksum_idx_counter + 2, 16, "MD5", 0);
     if (ret < 0)
         return ret;
@@ -524,9 +530,11 @@ int checksum_writer_write_data(IsoImageWriter *writer)
 
 #ifdef Libisofs_with_checksumS
 
-    int wres, res;
+    int wres, res, l;
     size_t i, size;
     Ecma119Image *t;
+    void *ctx = NULL;
+    char md5[16], tag_block[2048];
 
     if (writer == NULL) {
         return ISO_ASSERT_FAILURE;
@@ -537,30 +545,69 @@ int checksum_writer_write_data(IsoImageWriter *writer)
 
     /* Write image checksum to index 0 */
     if (t->checksum_ctx != NULL) {
-
+        res = iso_md5_clone(t->checksum_ctx, &ctx);
-        /* >>> rather clone a result than killing t->checksum_ctx */;
+        if (res > 0) {
-
+            res = iso_md5_end(&ctx, t->image_md5);
-        res = iso_md5_end(&(t->checksum_ctx), t->image_md5);
+            if (res > 0)
-        if (res > 0)
+                memcpy(t->checksum_buffer + 0 * 16, t->image_md5, 16);
-            memcpy(t->checksum_buffer + 0, t->image_md5, 16);
+        }
     }
 
-    size = (t->checksum_idx_counter + 2) / 128 + 1;
+    size = (t->checksum_idx_counter + 2) / 128;
+    if (size * 128 < t->checksum_idx_counter + 2)
+        size++;
  
-    /* >>> write overall checksum as index t->checksum_idx_counter + 1 */;
+    /* Write checksum of checksum array as index t->checksum_idx_counter + 1 */
+    res = iso_md5_start(&ctx);
+    if (res > 0) {
+        for (i = 0; i < t->checksum_idx_counter + 1; i++)
+            iso_md5_compute(ctx,
+                          t->checksum_buffer + ((size_t) i) * (size_t) 16, 16);
+        res = iso_md5_end(&ctx, md5);
+        if (res > 0)
+           memcpy(t->checksum_buffer + (t->checksum_idx_counter + 1) * 16,
+                  md5, 16);
+    }
 
     for (i = 0; i < size; i++) {
-        wres = iso_write(t, t->checksum_buffer + ((size_t) 2048) * i,
+        wres = iso_write(t, t->checksum_buffer + ((size_t) 2048) * i, 2048);
-                         BLOCK_SIZE);
+        if (wres < 0) {
-        if (wres < 0)
+            res = wres;
-            return wres;
+            goto ex;
+        }
+    }
+    if (t->checksum_ctx == NULL) {
+        res = ISO_SUCCESS;
+        goto ex;
     }
 
-    /* >>> write scdbackup checksum tag to an extra block */;
+    /* Write stream detectable checksum tag to extra block */;
+    memset(tag_block, 0, 2048);
+    res = iso_md5_end(&(t->checksum_ctx), md5);
+    if (res > 0) {
+        sprintf(tag_block,
+           "libisofs_checksum_tag_v1 pos=%u range_start=%u range_size=%u md5=",
+           t->checksum_array_pos + (unsigned int) size,
+           t->checksum_range_start, t->checksum_range_size);
+        l = strlen(tag_block);
+        for (i = 0; i < 16; i++)
+            sprintf(tag_block + l + 2 * i, "%2.2x",
+                    ((unsigned char *) md5)[i]);
+        tag_block[l + 32] = '\n';
+    }
+    wres = iso_write(t, tag_block, 2048);
+    if (wres < 0) {
+        res = wres;
+        goto ex;
+    }
 
 #endif /* Libisofs_with_checksumS */
 
-    return ISO_SUCCESS;
+    res = ISO_SUCCESS;
+ex:;
+    if (ctx != NULL)
+        iso_md5_end(&ctx, md5);
+    return(res);
 }