You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@subversion.apache.org by Carl Gherardi <Ca...@nautronix.com.au> on 2004/02/06 06:57:19 UTC
Valid cvs symbols
Hey all,
I'm in the process of evaluating svn, and attempting to import some RCS
repositories.
To do this I did as the cvs manual says and copied the rcs tree into the
rcs tree and edited the CVSROOT/modules file.
I've attempted to run cvs2svn and got the following
Error while parsing cvs/src/dataflow/transform_client/test_runner.cpp,v:
'28_05_02_2_46pm' is not a valid tag or branch name.
I had a quick look at the python script, and the problem is in pythons
rcsparse, not cvs2svn.
Q:
1) Is this an illegal tag? (Couldn't find a definition in the cvs
manual)
2) Any quick and dirty hacks to get this imported anyway? This is one of
many tags beginning with a number. (NOTE: The repository has been
developed in rcs for several years, there are no branches just symbols
to mark releases so using trunk only is not an option)
Any help or suggestions will be much appreciated.
Cheers
Carl
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Carl Gherardi <Ca...@nautronix.com.au>.
Apologies for sending this twice Karl - Didn't cc list.
On Fri, 2004-02-13 at 22:15, kfogel@collab.net wrote:
> Hmmm.
>
> If you create a repository containing just that file in a subdir, and
> convert the repository, do you still get the error? (I.e., what's the
> narrowest reproduction case we can trim this down to?)
>
> Once we have that, I'd like to try reproducing it myself.
>
> Thanks,
> -Karl
Ok the archive attached is the entire repository.
The error i'm getting is:
# ./cvs2svn.py -n -s /spare/general /spare/cvs
----- pass 1 -----
/spare/cvs/src/publisher/subscription/subscription_message.cpp,v
Traceback (most recent call last):
File "./cvs2svn.py", line 2277, in ?
main()
File "./cvs2svn.py", line 2273, in main
convert(ctx, start_pass=start_pass)
File "./cvs2svn.py", line 2129, in convert
_passes[i](ctx)
File "./cvs2svn.py", line 1959, in pass1
os.path.walk(ctx.cvsroot, visit_file, (cd, p, stats))
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 279, in walk
func(arg, top, names)
File "./cvs2svn.py", line 336, in visit_file
p.parse(open(pathname, 'rb'), cd)
File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 239,
in parse
File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 225,
in parse_rcs_deltatext
File "./cvs2svn.py", line 231, in set_revision_info
timestamp, author, op, old_ts = self.rev_data[revision]
KeyError: 1.21.1.1
I appreciate the effort.
Thanks
Carl Gherardi
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
Re: [w] Valid cvs symbols
Posted by kf...@collab.net.
Carl Gherardi <Ca...@nautronix.com.au> writes:
> Nope I wasn't. I am now, and I appear to be getting a little further.
>
> Now getting to (trimmed a little)
>
> cvs/src/publisher/subscription/subscription_message.h,v
> Traceback (most recent call last):
> File "./cvs2svn.py", line 1959, in pass1
> os.path.walk(ctx.cvsroot, visit_file, (cd, p, stats))
> File "/usr/lib/python2.2/posixpath.py", line 287, in walk
> walk(name, func, arg)
> File "/usr/lib/python2.2/posixpath.py", line 287, in walk
> walk(name, func, arg)
> File "/usr/lib/python2.2/posixpath.py", line 287, in walk
> walk(name, func, arg)
> File "/usr/lib/python2.2/posixpath.py", line 279, in walk
> func(arg, top, names)
> File "./cvs2svn.py", line 336, in visit_file
> p.parse(open(pathname, 'rb'), cd)
> File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 239,
> in parse
> File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 225,
> in parse_rcs_deltatext
> File "./cvs2svn.py", line 231, in set_revision_info
> timestamp, author, op, old_ts = self.rev_data[revision]
> KeyError: 1.13.1
>
> (I was in subversion-0.37.0/tools/cvs2svn/)
>
> Looking into the rcs file, there is a 1.13.1.1 branch and the section
Hmmm.
If you create a repository containing just that file in a subdir, and
convert the repository, do you still get the error? (I.e., what's the
narrowest reproduction case we can trim this down to?)
Once we have that, I'd like to try reproducing it myself.
Thanks,
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Carl Gherardi <Ca...@nautronix.com.au>.
On Tue, 2004-02-10 at 20:48, kfogel@collab.net wrote
> Did you using the raw cvs2svn distributed with Subversion as I
> described, instead of the prepackaged /usr/bin/cvs2svn?
>
Nope I wasn't. I am now, and I appear to be getting a little further.
Now getting to (trimmed a little)
cvs/src/publisher/subscription/subscription_message.h,v
Traceback (most recent call last):
File "./cvs2svn.py", line 1959, in pass1
os.path.walk(ctx.cvsroot, visit_file, (cd, p, stats))
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 287, in walk
walk(name, func, arg)
File "/usr/lib/python2.2/posixpath.py", line 279, in walk
func(arg, top, names)
File "./cvs2svn.py", line 336, in visit_file
p.parse(open(pathname, 'rb'), cd)
File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 239,
in parse
File "subversion-0.37.0/tools/cvs2svn/rcsparse/common.py", line 225,
in parse_rcs_deltatext
File "./cvs2svn.py", line 231, in set_revision_info
timestamp, author, op, old_ts = self.rev_data[revision]
KeyError: 1.13.1
(I was in subversion-0.37.0/tools/cvs2svn/)
Looking into the rcs file, there is a 1.13.1.1 branch and the section
looks like:
1.13.1.1
log
@Changed messages to inherit from ValueObject.
@
text
@d4 1
a4 1
"$Header:
/project/general/user/philip/src/publisher/subscription/RCS/subscription_message.h,v 1.13 2001/09/13 07:10:13 winter Exp philip $ Nautronix";
d11 1
d24 2
a43 5
...
Which all looks pretty good. So i'm stuck again.
Any ideas?
Carl
--
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by kf...@collab.net.
Carl Gherardi <Ca...@nautronix.com.au> writes:
> I misread the error message(never used python before) what language
> structure does python use to have "common" as an attribute? I've grepped
> rcsparse for the word common with no results.
That's because "common" is the name of the file providing the
functionality:
.../tools/cvs2svn/rcsparse/common.py
> I dont think thats the problem anyway, or the import statement at the
> top of cvs2svn would fail (right?) and the call trace would be from line
> 8, not 338 as above.
Did you using the raw cvs2svn distributed with Subversion as I
described, instead of the prepackaged /usr/bin/cvs2svn?
> I am a bit suspect about the version of rcsparse being used though.
>
> I've just been hunting for rcsparse on the viewcvs site, and its been
> atticed with the comment
>
> "This module is no longer needed. Use the vclib.ccvs.rcsparse module
> instead." on Sep 5 2002.
There's nothing to be suspicious about: rcsparse is a library.
Formerly, softwares A and B (ViewCVS and cvs2svn) both depended on the
library. Then software A lost its dependency, in favor of something
newer. But software B still depends on it. Softwares A and B do not
depend on each other.
This is why we distribute the required version of rcsparse with
cvs2svn. That way we know cvs2svn will get the right version.
(Of course, I'm not saying that cvs2svn should never upgrade its
rcsparse dependency, merely that we *haven't* upgraded yet.)
> The version I have installed on my machine is quite different from the
> version in attic on their site.
>
> Any suggestions are welcome.
Same suggestion I made before :-).
Best,
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Carl Gherardi <Ca...@nautronix.com.au>.
On Sat, 2004-02-07 at 00:44, kfogel@collab.net wrote:
> Carl Gherardi <Ca...@nautronix.com.au> writes:
> > File "/usr/bin/cvs2svn", line 338, in visit_file
> > except rcsparse.common.RCSExpected:
> > AttributeError: 'module' object has no attribute 'common'
> >
> > Any ideas? - i cant find where the error message is comming from.
>
> "/usr/bin/cvs2svn" ?
> Hmm. What version of cvs2svn.py are you using (look near the top of
> the file for the $Revision keyword), and where did you get it?
>
These are the packages for Redhat 9 from David Summers at:
http://subversion.tigris.org/project_packages.html
$Revision 7921
> See tools/cvs2svn/README in the subversion dist. cvs2svn.py needs to
> find the rcsparse module, which ships as a subdir in the same dir that
> cvs2svn.py lives in. We have some packaging work to do, obviously,
> but for now, invoking .../tools/cvs2svn/cvs2svn.py is usually the
> easiest solution.
I misread the error message(never used python before) what language
structure does python use to have "common" as an attribute? I've grepped
rcsparse for the word common with no results.
I dont think thats the problem anyway, or the import statement at the
top of cvs2svn would fail (right?) and the call trace would be from line
8, not 338 as above.
I am a bit suspect about the version of rcsparse being used though.
I've just been hunting for rcsparse on the viewcvs site, and its been
atticed with the comment
"This module is no longer needed. Use the vclib.ccvs.rcsparse module
instead." on Sep 5 2002.
The version I have installed on my machine is quite different from the
version in attic on their site.
Any suggestions are welcome.
Thanks
--
Carl
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by kf...@collab.net.
Carl Gherardi <Ca...@nautronix.com.au> writes:
> File "/usr/bin/cvs2svn", line 338, in visit_file
> except rcsparse.common.RCSExpected:
> AttributeError: 'module' object has no attribute 'common'
>
> Any ideas? - i cant find where the error message is comming from.
"/usr/bin/cvs2svn" ?
Hmm. What version of cvs2svn.py are you using (look near the top of
the file for the $Revision keyword), and where did you get it?
See tools/cvs2svn/README in the subversion dist. cvs2svn.py needs to
find the rcsparse module, which ships as a subdir in the same dir that
cvs2svn.py lives in. We have some packaging work to do, obviously,
but for now, invoking .../tools/cvs2svn/cvs2svn.py is usually the
easiest solution.
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Carl Gherardi <Ca...@nautronix.com.au>.
On Fri, 2004-02-06 at 17:21, Carl Gherardi wrote:
> On Fri, 2004-02-06 at 16:30, Kai Bartels wrote:
> > Hi Carl,
> >
> > I got the very same error when trying to migrate.
> > The "Open Source Development with CVS" sais that CVS tag names may
> > contain letters, numbers, hyphens and underscores and have to start
> > with a letter.
> *ugh* Anyone have a prewritten script?
Solution for anyone curious:
At the root of your cvs repository.
#Add legal tag to replace old one
find . -type f -name \*,v -exec \
rcs -NBORK20m_trial_18_09:20m_trial_18_09 {} \;
#Remove illegal tag.
find . -type f -name \*,v -exec rcs -N20m_trial_18_09_2002 {} \;
Now i've got:
File "/usr/bin/cvs2svn", line 338, in visit_file
except rcsparse.common.RCSExpected:
AttributeError: 'module' object has no attribute 'common'
Any ideas? - i cant find where the error message is comming from.
*sigh*
Carl
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Carl Gherardi <Ca...@nautronix.com.au>.
On Fri, 2004-02-06 at 16:30, Kai Bartels wrote:
> Hi Carl,
>
> I got the very same error when trying to migrate.
> The "Open Source Development with CVS" sais that CVS tag names may
> contain letters, numbers, hyphens and underscores and have to start
> with a letter.
I suspected as much...
> I also sais that cvs is very strict about it, which is obviously
> wrong. :^)
It might be - our developers haven't actually been using cvs only rcs
with a ton of wrapper scripts (hence the evaluation)
> I didn't have very much of them, so I just renamed the tags
> inside the cvs repository.
*ugh* Anyone have a prewritten script?
> When you successfully migrate your repository I'd be glad
> to hear about the statistics (size of repository, number of
> commits), because I still have problems with it. :^(
>
We have about 10 med size repositories ranging from 500 files to around
8000. I couldn't give you checkin stats.
Cheers
Carl
--
NB: Sorry for the crap to follow
--
This email is confidential and intended solely for the use of the individual to whom it is addressed.
Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
this email is strictly prohibited. If you have received this email in error please contact the sender.
Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
messages and attachments before reading them.
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Re: [w] Valid cvs symbols
Posted by Kai Bartels <ka...@picturesafe.de>.
Hi Carl,
I got the very same error when trying to migrate.
The "Open Source Development with CVS" sais that CVS tag names may contain
letters, numbers, hyphens and underscores and have to start with a letter.
I also sais that cvs is very strict about it, which is obviously wrong. :^)
I had tag names with a slash in them and I used a standard cvs version
to produce these tags (vendor tags in imports in my case).
I didn't have very much of them, so I just renamed the tags inside the cvs
repository.
When you successfully migrate your repository I'd be glad to hear about
the statistics (size of repository, number of commits), because I still
have problems with it. :^(
regards, Kai
On Fri, Feb 06, 2004 at 02:57:19PM +0800, Carl Gherardi wrote:
> Hey all,
>
> I'm in the process of evaluating svn, and attempting to import some RCS
> repositories.
>
> To do this I did as the cvs manual says and copied the rcs tree into the
> rcs tree and edited the CVSROOT/modules file.
>
> I've attempted to run cvs2svn and got the following
> Error while parsing cvs/src/dataflow/transform_client/test_runner.cpp,v:
> '28_05_02_2_46pm' is not a valid tag or branch name.
>
> I had a quick look at the python script, and the problem is in pythons
> rcsparse, not cvs2svn.
>
> Q:
> 1) Is this an illegal tag? (Couldn't find a definition in the cvs
> manual)
> 2) Any quick and dirty hacks to get this imported anyway? This is one of
> many tags beginning with a number. (NOTE: The repository has been
> developed in rcs for several years, there are no branches just symbols
> to mark releases so using trunk only is not an option)
>
> Any help or suggestions will be much appreciated.
>
> Cheers
>
> Carl
>
>
> --
> This email is confidential and intended solely for the use of the individual to whom it is addressed.
> Any views or opinions presented are solely those of the author and do not necessarily represent those of NAUTRONIX LTD.
>
> If you are not the intended recipient, you have received this email in error and use, dissemination, forwarding, printing, or copying of
> this email is strictly prohibited. If you have received this email in error please contact the sender.
>
> Although our computer systems use active virus protection software, and we take various measures to reduce the risk of viruses
> being transmitted in e-mail messages and attachments sent from this company, we cannot guarantee that such e-mail messages
> and attachments are free from viruses on receipt. It is a condition of our using e-mail to correspond with you, that any and all liability
> on our part arising directly or indirectly out of any virus is excluded. Please ensure that you run virus checking software on all e-mail
> messages and attachments before reading them.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
> For additional commands, e-mail: users-help@subversion.tigris.org
>
--
"Das Kriterium gesunder Vernunft sei auf die Menschheitsgeschichte
nicht anwendbar." <Stanislaw Lem>
KBartels@picturesafe.de + www.picturesafe.de
picturesafe GmbH, Lueerstr. 3, D-30175 Hannover fon:+49 511 85620 56
++ PGP Key fingerprint 1324 98A8 3BF8 4166 05E1 82D3 E168 6849 6022 5D32 ++
-----BEGIN GEEK CODE BLOCK-----
VERSION: 3.12
GCS d- s: a C++$ UL/S++$ P++ L++ E W-- N w--- !O !M PS+ PE Y+ PGP+ t R@ tv+>-
b+>++ DI++ G e+++ h* y?
------END GEEK CODE BLOCK------
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org