Open GoogleCodeExporter opened 9 years ago
Hi, The Unarchiver is a great software, I know and appreciate it.
PeaZip backend for handling RAR format is Igor Pavlov's 7z, and its port p7zip
on *x systems.
I chose it because of good performances, stability, feature completeness, very
good syntax, and because it handles most of the mainstream archive formats, so
allowing me to reduce the number of dependencies on third parts component to
test, to keep updated and to distribute with the software package, in order
either to reduce the weight of the package - and to speed up development and
debugging as fewer different backend syntax are supported.
The code for extracting RAR files was entirely rewritten by Igor Pavlov and
released under LGPL, but RAR is a proprietary format and any work supporting
its format specifications must, as explicit said in the RAR license by the
original author of the format specifications Eugene Roshal, provide the
additional clause of not being reverse engineered for creation of RAR files.
Any third party work, either the one from 7-Zip or form The Unarchiver, to have
the right of using proprietary RAR file format specifications in order to
provide a working extractor, must bring this additional clause with it, or at
least it is what is required in RAR license - licensing is a very complex field
and if and how this rule can be enforced, and if a software with this
additional clause can still be considered Free Software is still matter of
debate.
The consensus (at least on SourceForge, that would otherwise not allow to
publish non-free software on its servers) is that 7-Zip's unrar code is Free
Software, anyway since this rules comes from the very file format
specifications of RAR (that both implementations use) 7-Zip's unrar and The
Unarchiver's unrar must be considered equally free or (non free).
As for the RARv3 support, can you please specify what part of RARv3
specifications 7z/p7zip's unrar code is not able to handle?
As for what I know it added support to RARv3 many years ago (i.e. during 2002 I
can see some resolved bugs about RARv3 on 7-Zip's tracker on SourceForge)
Original comment by giorgio.tani.software@gmail.com
on 26 May 2011 at 8:28
By the way, the *x version of unrar code (in p7zip) was, for those ongoing
concerns about its status as Free Software, separated on a standalone .so, so
it can be removed as the user / packager wishes.
Original comment by giorgio.tani.software@gmail.com
on 26 May 2011 at 8:32
Cf this release by the FSF:
http://www.fsf.org/blogs/licensing/free-rarv3-extraction
If the RAR code is obtained through clean room reverse engineering, rather than
from using Roshal's code, then it can be released under any license the author
chooses. I assume that's how The Unarchiver's rar implementation came about.
I mentioned RARv3 because until now, if you wanted to extract rar files, you
either had to use the unrar-free utility, which did not support RARv3, or go
with the proprietary unrar with Roshal's code you mentioned. So technically, I
don't feel there is anything wrong with 7z's implementation; it's just not
FOSS.
I admit I didn't browse through The Unarchiver code to look for the specific
license header for that rar implementation, and if you happen to know that The
Unarchiver's .rar implementation is also under Roshal's proprietary license,
then my bad, but I was going on the FSF's release. I guess I should have
mentioned that in my first post!
Original comment by vascofal...@gmail.com
on 26 May 2011 at 9:19
RAR license states
http://www.win-rar.com/index.php?id=24&kb=1&kb_article_id=27
"9. You may not use, copy, emulate, clone, rent, lease, sell, modify,
decompile, disassemble, otherwise reverse engineer, or transfer
the licensed program, or any subset of the licensed program, except
as provided for in this agreement. Any such unauthorized use shall
result in immediate and automatic termination of this license and
may result in criminal and/or civil prosecution."
So it is not even clear if a reverse engineering of any of the original RAR
codebase can be considered legitimate for any purpose - while the same license
says using the UNRAR specifications is fine unless you try to build a RAR
compressor.
Anyway, as I said before, copyright is a complex matter and it is not easy to
understand what clauses can an author legitimately put on a proprietary
software, and what makes a software free software.
Copyright laws can differ widely on different countries, and anyway FSF, OSI
and other notable subject's views on copyright and copyleft topics can vary a
lot too.
SourceForge consider 7z/p7zip Free Software: SF only allows on their server
software that is released under OSI compliant licenses, so they either consider
the Pavlov's unrar code compliant with OSI definition of Open Source or we must
think they didn't noticed such an issue with a software that is in their top
ten since a decade, that would be absurd.
Original comment by giorgio.tani.software@gmail.com
on 26 May 2011 at 9:49
As I see this topic was quoted and linked on third party blogs and websites,
I'll clarify better my opinion - anyway the premise is that I'm not, in any
way, an expert on copyright laws.
The Unarchiver is a great piece of software, so it is 7-Zip.
I'm glad those software are open rather than closed.
RAR is still a great archive format, even if it was superseeded in compression
ratio and speed by other formats like .7z and FreeArc's .arc (that is really
astounding in terms of compression/speed ratio), it is still a one of the best
general puropose compressors, and having good error correction and strong
encryption makes it one of the best formats around.
But the primary reason RAR support is crucial for free software community is
that it is extremely popular, and while generally the free software community
recommends against relying on closed formats (so generally does not endorse
creation of RAR files), being able to open RAR files (possibly to convert them
in an open format) is a just too important feature the free software community
cannot miss.
I'm not glad it is closed, but it simply too important to be ignored.
All the problem here comes from the closed license of RAR: I have deep respect
for the great work RAR author did, but all those issues are one more reason to
believe in copyleft, that's why I chose to work with (and to release my work
as) Open Source.
Original comment by giorgio.tani.software@gmail.com
on 27 May 2011 at 1:56
Hi,
I think you cited the WinRAR/RAR license text that covers WinRAR/RAR
implementations ("the licensed program, or any subset of the licensed
program"), and this has nothing to do with The Unarchiver, an independent RAR3
implementation.
7-zip unrar code is currently licensed under GPL + unrar restrictions because
is based on Alexander Roshal's unrar original code. See
http://www.7-zip.org/license.txt
Original comment by gotru...@gmail.com
on 2 Aug 2011 at 12:00
[deleted comment]
For what it's worth, The Unarchive was recently accepted into Debian Main. I
think this should clear any doubts at whether it's really a free implementation
or not. See here: http://packages.debian.org/sid/theunarchiver
Original comment by vascofal...@gmail.com
on 5 Oct 2011 at 3:53
I'd tend to agree with the argument that cleanroom engineering is a valid,
legal way to achieve interoperability (in this case, extracting a file that
e.g. WinRAR created). Any licenses for the file format are then invalid - you
don't make use of other people's code and IP law (in e.g. US, Europe) grants
you exceptions to achieve interoperability (see e.g.
http://en.wikipedia.org/wiki/Clean_room_design)
If The Unarchiver is based on cleanroom engineering, I'd suggest considering
switching to it, especially if Debian - who is notoriously strict about IP
issues - has accepted it.
Original comment by reiniero...@gmail.com
on 28 May 2013 at 8:12
Original issue reported on code.google.com by
vascofal...@gmail.com
on 26 May 2011 at 7:47