On Thu, Jul 8, 2010 at 10:27, <rhuij...@apache.org> wrote:
>...
> +++ subversion/trunk/subversion/libsvn_wc/wc-queries.sql Thu Jul 8 14:27:22
> 2010
> @@ -400,7 +400,7 @@ INSERT OR REPLACE INTO BASE_NODE (
> changed_author, properties, repos_id, repos_relpath, checksum, changed_date,
> depth, symlink_target, dav_cache)
> /* NOTE: ?6 is duplicated because we insert the same value in two columns.
> */
> -VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?6, ?7, ?8, ?9, ?10, ?11, ?12, ?13, ?14,
> ?15);
> +VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7, ?8, ?9, ?10, ?11, ?12, ?13, ?14, ?15,
> ?16);
A little awareness would be nice. You know... like that comment JUST
ABOVE the line you changed?
>...
> +++ subversion/trunk/subversion/libsvn_wc/wc_db.c Thu Jul 8 14:27:22 2010
> @@ -5353,9 +5353,10 @@ struct commit_baton {
> svn_wc__db_pdh_t *pdh;
> const char *local_relpath;
>
> - svn_revnum_t new_revision;
> - apr_time_t new_date;
> - const char *new_author;
> + apr_int64_t new_revision;
> + apr_int64_t changed_rev;
What? You mean svn_revnum_t.
Don't change the structure. Keep it semantics. The only point where it
needs to be an apr_int64_t is when you bind it. Change the semantics
*there*. Shoot... maybe we'll add an "r" code at some point to bind
revnums.
Using the wrong type for a given semantic... leads to problems.
>...
> @@ -5485,29 +5486,30 @@ commit_node(void *baton, svn_sqlite__db_
>
> SVN_ERR(svn_sqlite__get_statement(&stmt, cb->pdh->wcroot->sdb,
> STMT_APPLY_CHANGES_TO_BASE));
> - SVN_ERR(svn_sqlite__bindf(stmt, "issttisb",
> + SVN_ERR(svn_sqlite__bindf(stmt, "issttiisb",
> cb->pdh->wcroot->wc_id, cb->local_relpath,
> parent_relpath,
> presence_map, new_presence,
> kind_map, new_kind,
> - (apr_int64_t)cb->new_revision,
> - cb->new_author,
> + cb->new_revision,
> + cb->changed_rev,
Keep the cast.
>...
Cheers,
-g
