Alex Malinovich wrote:
> This is all assuming that the portion of the file that got corrupted
> ISN'T the portion storing the checksum data. If it is, the file is of no
> use to you.
*sigh* As you said, it doesn't matter what is corrupt if there is
corruption. But this is a far sight better than nothing at all.
> I seem to recall someone saying something about being responsible and
> educating one's self on a piece of software?
Yes, Alex, and you're showing your ignorance again.
> You don't "use up" a PAR.
> It's not a bandage, or a tissue. It's perfectly possible to restore more
> than one file with one PAR. (Depending on the total number of parts in
> the archive set and the total number of PAR's included.)
No, it's not. PAR (v1) had a fixed size and was used to recreate 1 file
*per PAR*. That file was generally an archive though it was possible to have
PAR applied against several files of varying sizes. The size of the PAR file
was set to the largest file in the set but the rule applied. 1 PAR file for
each file corrupt.
PAR (v2) changed this so that one had a number of blocks one could use to
recover corrupted portions of a file. So it was possible to use only a small
portion of a PAR file to recover only a small portion of a corrupted file. Of
course this did not change the underlying rule; they just changed the
measurement. Instead of 1 PAR file to recover 1 protected file they now
narrowed it down to 1 PAR block to recovered 1 proteted block.
You cannot, as you assert above, take a single PAR file and recover, say,
3 files from a set.
BTW, Alex, little background footnote. For a while I was pushing to
package par2 for Debian. I just wasn't pleased with the method of getting the
package into the distribution. So, unlike you and RAR, I know quite well what
PAR 1 & 2 can and cannot do.
> I don't believe I ever began "blasting" a piece of software. What I did
> was refute YOUR assertion that RAR is superior to a combination of free
> software tools that performs a nearly-identical function.
By claiming that the piece of software in question was incapable of doing
things it most certainly is capable of doing. IE, blasting.
> If you want to
> use a non-free archiving utility, you are quite welcome to do so. Just
> don't state AS FACT your OPINION that an equivalent result cannot be
> obtained using different tools.
I'm not stating my opinion. I'm stating fact. The fact is that tar + gz
does not = rar. There are structures within the rar archive which are missing
from tar + gz. Futhermore tar + gz + split does not even remotely begin to
approach a multi-volume rar. If you don't believe me, read the following and
try... *try* to understand... My comments prefaced with "---"
--- I've created a multivolume rar set of some pictures. The file,
Lflying.jpg exists in one and only one volume. I do an md5sum of the original
file for comparison.
{grey@mania:~/t/t} md5sum ../Lflying.jpg
050f14ee2bacf23aa571a81c336ba4e5 ../Lflying.jpg
--- Delete all other volumes but the one that it exists in.
{grey@mania:~/t/t} rm test.part[12356].rar
--- List the volume to show you the file is there. This is the first thing a
split tar.gz cannot do.
{grey@mania:~/t/t} rar l test.part4.rar
RAR 3.30 Copyright (c) 1993-2004 Eugene Roshal 22 Jan 2004
Shareware version Type RAR -? for help
Volume test.part4.rar
Name Size Packed Ratio Date Time Attr CRC Meth Ver
-------------------------------------------------------------------------------
Ascendance_tabard.png 145615 18247 <-- 20-09-05 23:50 -rw-r--r--
E9E546BC m3c 2.9
Lflying.jpg 49115 48521 98% 15-04-05 23:29 -rwxr--r-- C0111A15 m3c 2.9
rava_face.png 162619 14973 --> 15-04-05 23:29 -rwxr--r-- 3301E1E0 m3c
2.9-------------------------------------------------------------------------------
2 211734 81741 38%
--- Note the CRC right there in the volume. Gee, no need for my external
md5sum except for this exercise. Now to extract the file.
{grey@mania:~/t/t} rar x test.part4.rar Lflying.jpg
RAR 3.30 Copyright (c) 1993-2004 Eugene Roshal 22 Jan 2004
Shareware version Type RAR -? for help
Extracting from test.part4.rar
Extracting Lflying.jpg OK
All OK
--- md5sum again to prove this file is the same as the above file.
{grey@mania:~/t/t} md5sum Lflying.jpg
050f14ee2bacf23aa571a81c336ba4e5 Lflying.jpg
--- And for comparison, here's the previons md5sum
050f14ee2bacf23aa571a81c336ba4e5 ../Lflying.jpg
So, you show me tar.gz being able to extract a file from the middle of a
split archive when none of the other volumes *even exist* and you'll have one
leg to stand on about what is my opinion vs. what is *fact*.
For your edification...
{grey@mania:~/t/t} split -b 80k test.tgz
{grey@mania:~/t/t} ls -l
total 1705
-rw-r--r-- 1 grey grey 868500 2005-12-12 22:46 test.tgz
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xaa
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xab
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xac
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xad
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xae
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xaf
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xag
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xah
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xai
-rw-r--r-- 1 grey grey 81920 2005-12-12 22:46 xaj
-rw-r--r-- 1 grey grey 49300 2005-12-12 22:46 xak
{grey@mania:~/t/t} rm test.tgz
{grey@mania:~/t/t} rm xa[abcdefijk]
{grey@mania:~/t/t} tar xzf xag
gzip: stdin: not in gzip format
tar: Child returned status 1
tar: Error exit delayed from previous errors
I rest my case on your opinion versus my facts.
> I'm very familiar with the RAR program and its evolution over the last
> 10+ years, including the new features that 3.0 introduced.
And you expect me to believe that when you didn't even know about the
recovery volumes introduced in 3.0? Please.
> My question,
> as before, remains what practical benefit this software presents that
> CANNOT be achieved otherwise?
See above. Tired of educating you Alex. Really tired.
--
Steve C. Lamb | I'm your priest, I'm your shrink, I'm your
PGP Key: 8B6E99C5 | main connection to the switchboard of souls.
-------------------------------+---------------------------------------------
Attachment:
signature.asc
Description: OpenPGP digital signature