You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Mark Phippard <ma...@gmail.com> on 2007/08/18 14:45:03 UTC
Trunk JavaHL is not building after merge from blame branch
I get this error:
/Users/mphippard/work/svn-trunk/subversion/bindings/javahl/native/SVNClient.cpp:1419:
error: initializing argument 8 of 'svn_error_t*
svn_client_blame4(const char*, const svn_opt_revision_t*, const
svn_opt_revision_t*, const svn_opt_revision_t*, const
svn_diff_file_options_t*, svn_boolean_t, svn_boolean_t, svn_error_t*
(*)(void*, apr_int64_t, svn_revnum_t, const char*, const char*,
svn_revnum_t, const char*, const char*, const char*, const char*,
apr_pool_t*), void*, svn_client_ctx_t*, apr_pool_t*)'
Also, Hyrum, I thought you were going to make BlameCallbackImpl
implement both the old and new interface?
Finally, in the new interface, which fields have the info from the
"original" change? I assume the field names with "merged" in them
contain the info from when the line was merged, and the other fields
from when it originated?
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Trunk JavaHL is not building after merge from blame branch
Posted by Mark Phippard <ma...@gmail.com>.
On 8/18/07, Mark Phippard <ma...@gmail.com> wrote:
> I get this error:
>
> /Users/mphippard/work/svn-trunk/subversion/bindings/javahl/native/SVNClient.cpp:1419:
> error: initializing argument 8 of 'svn_error_t*
> svn_client_blame4(const char*, const svn_opt_revision_t*, const
> svn_opt_revision_t*, const svn_opt_revision_t*, const
> svn_diff_file_options_t*, svn_boolean_t, svn_boolean_t, svn_error_t*
> (*)(void*, apr_int64_t, svn_revnum_t, const char*, const char*,
> svn_revnum_t, const char*, const char*, const char*, const char*,
> apr_pool_t*), void*, svn_client_ctx_t*, apr_pool_t*)'
>
> Also, Hyrum, I thought you were going to make BlameCallbackImpl
> implement both the old and new interface?
>
> Finally, in the new interface, which fields have the info from the
> "original" change? I assume the field names with "merged" in them
> contain the info from when the line was merged, and the other fields
> from when it originated?
The attached patch fixes the problem. I am not sure it is really
right though. Should this "merged_path" field even be part of the
structure? If it should, then should it also be passed through to the
Java layer? Right now, I just "swallow" that field in the C++ code.
I suspect it should not be in the struct to begin with.
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
Re: Trunk JavaHL is not building after merge from blame branch
Posted by Mark Phippard <ma...@gmail.com>.
On 8/18/07, Mark Phippard <ma...@gmail.com> wrote:
> On 8/18/07, Hyrum K. Wright <hy...@mail.utexas.edu> wrote:
> > > Also, Hyrum, I thought you were going to make BlameCallbackImpl
> > > implement both the old and new interface?
> >
> > I remember talking about that, but now it makes sense to have
> > BlameCallbackImpl just implement the BlameCallback2 interface. Both of
> > them are new with 1.5, so it doesn't make much sense to have
> > BlameCallbackImpl implement a now obsolete interface when it doesn't
> > have any compatibility needs.
>
> I agree, except it isn't doing that. It is only implementing BlameCallback.
I think I see. I am talking about BlameCallbackImpl and I think you
are talking about BlameCallbackWrapper -- the private class in
SVNClient.
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Trunk JavaHL is not building after merge from blame branch
Posted by Mark Phippard <ma...@gmail.com>.
On 8/18/07, Hyrum K. Wright <hy...@mail.utexas.edu> wrote:
> > Correct. I was thinking about something like the attached patch, do you
> > see any problems with that?
>
> Heh, just saw r26165. Nothing to see here, move along.
You should probably take a look at what I did as well as decide if
SVNClient.java ought to be changed at all. I think my approach to the
callback makes sense. That is basically how I implemented it in
Subclipse.
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Trunk JavaHL is not building after merge from blame branch
Posted by "Hyrum K. Wright" <hy...@mail.utexas.edu>.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hyrum K. Wright wrote:
> Mark Phippard wrote:
>> On 8/18/07, Hyrum K. Wright <hy...@mail.utexas.edu> wrote:
>>>> Also, Hyrum, I thought you were going to make BlameCallbackImpl
>>>> implement both the old and new interface?
>>> I remember talking about that, but now it makes sense to have
>>> BlameCallbackImpl just implement the BlameCallback2 interface. Both of
>>> them are new with 1.5, so it doesn't make much sense to have
>>> BlameCallbackImpl implement a now obsolete interface when it doesn't
>>> have any compatibility needs.
>> I agree, except it isn't doing that. It is only implementing BlameCallback.
>
> Correct. I was thinking about something like the attached patch, do you
> see any problems with that?
Heh, just saw r26165. Nothing to see here, move along.
- -Hyrum
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFGx0D2CwOubk4kUXwRAgNfAKDsAOYOk+LCg0YWu3+Lh6YjwiwPEQCg9LvU
LoJjhFh4p+MyiSmd/yGplRk=
=uiCN
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Trunk JavaHL is not building after merge from blame branch
Posted by "Hyrum K. Wright" <hy...@mail.utexas.edu>.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Mark Phippard wrote:
> On 8/18/07, Hyrum K. Wright <hy...@mail.utexas.edu> wrote:
>>> Also, Hyrum, I thought you were going to make BlameCallbackImpl
>>> implement both the old and new interface?
>> I remember talking about that, but now it makes sense to have
>> BlameCallbackImpl just implement the BlameCallback2 interface. Both of
>> them are new with 1.5, so it doesn't make much sense to have
>> BlameCallbackImpl implement a now obsolete interface when it doesn't
>> have any compatibility needs.
>
> I agree, except it isn't doing that. It is only implementing BlameCallback.
Correct. I was thinking about something like the attached patch, do you
see any problems with that?
- -Hyrum
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFGx0CNCwOubk4kUXwRAhaYAJ4+c+XaeZBivfhJfdoHfqkuuIpcLACfa3hS
obS7WHjTnlxTD+ig/VBRuHk=
=emDc
-----END PGP SIGNATURE-----
Re: Trunk JavaHL is not building after merge from blame branch
Posted by Mark Phippard <ma...@gmail.com>.
On 8/18/07, Hyrum K. Wright <hy...@mail.utexas.edu> wrote:
> > Also, Hyrum, I thought you were going to make BlameCallbackImpl
> > implement both the old and new interface?
>
> I remember talking about that, but now it makes sense to have
> BlameCallbackImpl just implement the BlameCallback2 interface. Both of
> them are new with 1.5, so it doesn't make much sense to have
> BlameCallbackImpl implement a now obsolete interface when it doesn't
> have any compatibility needs.
I agree, except it isn't doing that. It is only implementing BlameCallback.
> > Finally, in the new interface, which fields have the info from the
> > "original" change? I assume the field names with "merged" in them
> > contain the info from when the line was merged, and the other fields
> > from when it originated?
>
> The doc string in BlameCallback2 has been updated to reflect which
> arguments represent the merged revisions, and which ones are the
> original revisions. Does that help with the confusion?
I ran it in debug before you updated. I am all set.
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: Trunk JavaHL is not building after merge from blame branch
Posted by "Hyrum K. Wright" <hy...@mail.utexas.edu>.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Mark Phippard wrote:
> I get this error:
>
> /Users/mphippard/work/svn-trunk/subversion/bindings/javahl/native/SVNClient.cpp:1419:
> error: initializing argument 8 of 'svn_error_t*
> svn_client_blame4(const char*, const svn_opt_revision_t*, const
> svn_opt_revision_t*, const svn_opt_revision_t*, const
> svn_diff_file_options_t*, svn_boolean_t, svn_boolean_t, svn_error_t*
> (*)(void*, apr_int64_t, svn_revnum_t, const char*, const char*,
> svn_revnum_t, const char*, const char*, const char*, const char*,
> apr_pool_t*), void*, svn_client_ctx_t*, apr_pool_t*)'
Yup, remembered that change to the client interface this morning. As of
r26162 (hey, that's a palindrome!), JavaHL now compiles for me. This is
a single test failure, but it appears unrelated to my changes to blame.
> Also, Hyrum, I thought you were going to make BlameCallbackImpl
> implement both the old and new interface?
I remember talking about that, but now it makes sense to have
BlameCallbackImpl just implement the BlameCallback2 interface. Both of
them are new with 1.5, so it doesn't make much sense to have
BlameCallbackImpl implement a now obsolete interface when it doesn't
have any compatibility needs.
> Finally, in the new interface, which fields have the info from the
> "original" change? I assume the field names with "merged" in them
> contain the info from when the line was merged, and the other fields
> from when it originated?
The doc string in BlameCallback2 has been updated to reflect which
arguments represent the merged revisions, and which ones are the
original revisions. Does that help with the confusion?
- -Hyrum
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFGxzqACwOubk4kUXwRAl5DAKCOXvObMSF5fceK2S8oUcSxcsa9cQCeKd6R
cB5D8wPEQji35NS6YWuc0Qg=
=PGCG
-----END PGP SIGNATURE-----
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org