You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Daniel Sahlberg <da...@gmail.com> on 2026/01/26 21:27:20 UTC

Re: svn commit: r1808139 - in /subversion/branches/better-pristines/subversion: include/svn_client.h libsvn_client/checkout.c libsvn_client/deprecated.c svn/checkout-cmd.c tests/libsvn_wc/utils.c

Forwarding to dev@... I couldn't set the recipient in Ponymail.

Den mån 26 jan. 2026 kl 22:26 skrev Daniel Sahlberg <ds...@apache.org>:

> I stumbled upon something while reviewing the docstring fixes and I don't
> quite understand. I had to do some acheology to find the original commit,
> this has subsequently been merged through several branches and the code now
> exist in trunk.
>
> On 2017/09/12 17:50:05 brane@apache.org wrote:
> > Author: brane
> > Date: Tue Sep 12 17:50:05 2017
> > New Revision: 1808139
> >
> > URL: http://svn.apache.org/viewvc?rev=1808139&view=rev
> > Log:
> > On the better-pristines branch: Add WC format parameter to
> svn_client_checkout.
> ...
> >
> ==============================================================================
> > ---
> subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
> (original)
> > +++
> subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
> Tue Sep 12 17:50:05 2017
> > @@ -2531,6 +2531,24 @@ svn_client_cat(svn_stream_t *out,
> >
> >  /*** From checkout.c ***/
> >  svn_error_t *
> > +svn_client_checkout3(svn_revnum_t *result_rev,
> > +                     const char *URL,
> > +                     const char *path,
> > +                     const svn_opt_revision_t *peg_revision,
> > +                     const svn_opt_revision_t *revision,
> > +                     svn_depth_t depth,
> > +                     svn_boolean_t ignore_externals,
> > +                     svn_boolean_t allow_unver_obstructions,
> > +                     svn_client_ctx_t *ctx,
> > +                     apr_pool_t *pool)
> > +{
> > +  return svn_error_trace(svn_client_checkout4(
> > +                             result_rev, URL, path,
> > +                             peg_revision, revision, depth,
> > +                             ignore_externals, FALSE, NULL, ctx, pool));
>
> Is there a reason why allow_unver_obstructions is not used here (it should
> be in place of FALSE) or is this just an oversight?
>
> > +}
> > +
> > +svn_error_t *
> >  svn_client_checkout2(svn_revnum_t *result_rev,
> >                       const char *URL,
> >                       const char *path,
>
> Thanks,
> Daniel
>

Re: svn commit: r1808139 - in /subversion/branches/better-pristines/subversion: include/svn_client.h libsvn_client/checkout.c libsvn_client/deprecated.c svn/checkout-cmd.c tests/libsvn_wc/utils.c

Posted by Daniel Sahlberg <da...@gmail.com>.
Den tis 27 jan. 2026 kl 00:34 skrev Branko Čibej <br...@apache.org>:

> On 26. 1. 26 22:27, Daniel Sahlberg wrote:
>
> Forwarding to dev@... I couldn't set the recipient in Ponymail.
>
> Den mån 26 jan. 2026 kl 22:26 skrev Daniel Sahlberg <dsahlberg@apache.org
> >:
>
>> I stumbled upon something while reviewing the docstring fixes and I don't
>> quite understand. I had to do some acheology to find the original commit,
>> this has subsequently been merged through several branches and the code now
>> exist in trunk.
>>
>> On 2017/09/12 17:50:05 brane@apache.org wrote:
>> > Author: brane
>> > Date: Tue Sep 12 17:50:05 2017
>> > New Revision: 1808139
>> >
>> > URL: http://svn.apache.org/viewvc?rev=1808139&view=rev
>> > Log:
>> > On the better-pristines branch: Add WC format parameter to
>> svn_client_checkout.
>> ...
>> >
>> ==============================================================================
>> > ---
>> subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
>> (original)
>> > +++
>> subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
>> Tue Sep 12 17:50:05 2017
>> > @@ -2531,6 +2531,24 @@ svn_client_cat(svn_stream_t *out,
>> >
>> >  /*** From checkout.c ***/
>> >  svn_error_t *
>> > +svn_client_checkout3(svn_revnum_t *result_rev,
>> > +                     const char *URL,
>> > +                     const char *path,
>> > +                     const svn_opt_revision_t *peg_revision,
>> > +                     const svn_opt_revision_t *revision,
>> > +                     svn_depth_t depth,
>> > +                     svn_boolean_t ignore_externals,
>> > +                     svn_boolean_t allow_unver_obstructions,
>> > +                     svn_client_ctx_t *ctx,
>> > +                     apr_pool_t *pool)
>> > +{
>> > +  return svn_error_trace(svn_client_checkout4(
>> > +                             result_rev, URL, path,
>> > +                             peg_revision, revision, depth,
>> > +                             ignore_externals, FALSE, NULL, ctx,
>> pool));
>>
>> Is there a reason why allow_unver_obstructions is not used here (it
>> should be in place of FALSE) or is this just an oversight?
>>
>
>
> Gosh. I really don't remember what I was thinking 8+ years ago. But
> looking at this now, it sure looks like an oversight. Most likely a
> copy-paste error from the implementation of svn_client_checkout2().
>
> -- Brane
>
>
Thanks, I have reviewed the use of svn_client_checkout3 in our codebase and
there are none (except for some API glue logic) and considering Brane's
reply above I can't see a reason why it should NOT use the actual argument.
I've committed a fix in r1931570.

With this, I will change my -0 to +1 on the r1931298 group and although I
haven't moved it to approved yet - I think we need r1931570 as well.

Cheers,
Daniel

Re: svn commit: r1808139 - in /subversion/branches/better-pristines/subversion: include/svn_client.h libsvn_client/checkout.c libsvn_client/deprecated.c svn/checkout-cmd.c tests/libsvn_wc/utils.c

Posted by Branko Čibej <br...@apache.org>.
On 26. 1. 26 22:27, Daniel Sahlberg wrote:
> Forwarding to dev@... I couldn't set the recipient in Ponymail.
>
> Den mån 26 jan. 2026 kl 22:26 skrev Daniel Sahlberg 
> <ds...@apache.org>:
>
>     I stumbled upon something while reviewing the docstring fixes and
>     I don't quite understand. I had to do some acheology to find the
>     original commit, this has subsequently been merged through several
>     branches and the code now exist in trunk.
>
>     On 2017/09/12 17:50:05 brane@apache.org wrote:
>     > Author: brane
>     > Date: Tue Sep 12 17:50:05 2017
>     > New Revision: 1808139
>     >
>     > URL: http://svn.apache.org/viewvc?rev=1808139&view=rev
>     <http://svn.apache.org/viewvc?rev=1808139&view=rev>
>     > Log:
>     > On the better-pristines branch: Add WC format parameter to
>     svn_client_checkout.
>     ...
>     >
>     ==============================================================================
>     > ---
>     subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
>     (original)
>     > +++
>     subversion/branches/better-pristines/subversion/libsvn_client/deprecated.c
>     Tue Sep 12 17:50:05 2017
>     > @@ -2531,6 +2531,24 @@ svn_client_cat(svn_stream_t *out,
>     >
>     >  /*** From checkout.c ***/
>     >  svn_error_t *
>     > +svn_client_checkout3(svn_revnum_t *result_rev,
>     > +                     const char *URL,
>     > +                     const char *path,
>     > +                     const svn_opt_revision_t *peg_revision,
>     > +                     const svn_opt_revision_t *revision,
>     > +                     svn_depth_t depth,
>     > +                     svn_boolean_t ignore_externals,
>     > +                     svn_boolean_t allow_unver_obstructions,
>     > +                     svn_client_ctx_t *ctx,
>     > +                     apr_pool_t *pool)
>     > +{
>     > +  return svn_error_trace(svn_client_checkout4(
>     > +                             result_rev, URL, path,
>     > +                             peg_revision, revision, depth,
>     > +                             ignore_externals, FALSE, NULL,
>     ctx, pool));
>
>     Is there a reason why allow_unver_obstructions is not used here
>     (it should be in place of FALSE) or is this just an oversight?
>


Gosh. I really don't remember what I was thinking 8+ years ago. But 
looking at this now, it sure looks like an oversight. Most likely a 
copy-paste error from the implementation of svn_client_checkout2().

-- Brane