[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn commit: r25253 - trunk/subversion/bindings/swig/ruby/svn

From: Daniel Rall <dlr_at_collab.net>
Date: 2007-06-08 21:35:08 CEST

On Thu, 07 Jun 2007, Joe Swatosh wrote:
...
> On 6/5/07, Kouhei Sutou <kou@cozmixng.org> wrote:
...
> > Daniel Rall <dlr@collab.net> wrote:
> >
> >> It's the same in Java and C and Python and pretty much every language
> >> (thus the original split between "mergeinfo" and "merge info"). I
> >> don't have a strong preference here, but in the core C API, we wanted
> >> to pick one and go with it, and the one chosen was "mergeinfo", to
> >> help illuminate the fact that it's a single term, and not verb_noun,
> >> which in some API names turned out to be quite confusing.
> >
> >The problem of the original is that some API names use
> >"mergeinfo" and others use "merge info". In the Ruby
> >bindings, we provide both "mergeinfo" and "merge_info".
> >
> >Users will not be confused because one is just alias of
> >other. They can use both of them what they like in their
> >script. And Ruby has many aliases. For example, collect and
> >map, size and length, and so on. I think those aliases help
> >users to write their script what they like. I hope the Ruby
> >bindings have Rubyish API because the Ruby bindings is used
> >by Rubyist.
>
> kou,
>
> I've been thinking about this a lot more than I probably should. I
> think what Dan is saying is that English is slippery, and that
> "mergeinfo" should be considered a "noun phrase" and therefore treated
> like a noun (a single entity).

Yeah, that's exactly what I was saying. I doubt I'm alone on this
one, but I haven't done a poll or anything to confirm my suspicions.

> I double checked and as of r25313 when I do a search on my WC, I do
> find some merge_info matches, but none appear to be in the public
> interface.

Yup, I've eliminated them from the public API.

> Since these are all new APIs(?), I've reconsidered and I now think it
> would be better to just use mergeinfo.

+1

> Dan, should the merge_info occurances below be replaced?
> What about merge-info?

I'd change all of them, including "merge-info".

> D:\SVN\src-trunk>findstr /c:merge_info /s /d:subversion *.* | findstr
> /v /r /c:^bindings /c:^tests /c:"\\.svn\\text"
> subversion:
> libsvn_fs_base\revs-txns.c:svn_fs_base__set_txn_merge_info(svn_fs_t *fs,
> libsvn_fs_base\revs-txns.h:svn_fs_base__set_txn_merge_info(svn_fs_t *fs,
> libsvn_fs_base\tree.c:/* The input for txn_body_change_merge_info(). */
> libsvn_fs_base\tree.c:struct change_merge_info_args
> libsvn_fs_base\tree.c: type "struct change_merge_info_args").
> Conforms to the callback
> libsvn_fs_base\tree.c:txn_body_change_merge_info(void *baton,
> libsvn_fs_base\tree.c: struct change_merge_info_args *args = baton;
> libsvn_fs_base\tree.c:
> SVN_ERR(svn_fs_base__set_txn_merge_info(args->root->fs,
> args->root->txn,
> libsvn_fs_base\tree.c:base_change_merge_info(svn_fs_root_t *root,
> libsvn_fs_base\tree.c: apr_hash_t *merge_info,
> libsvn_fs_base\tree.c: struct change_merge_info_args args;
> libsvn_fs_base\tree.c: SVN_ERR(svn_mergeinfo__to_string((svn_string_t
> **) &args.value, merge_info,
> libsvn_fs_base\tree.c: return svn_fs_base__retry_txn(root->fs,
> txn_body_change_merge_info, &args,
> libsvn_fs_base\tree.c:change_txn_merge_info(struct
> change_node_prop_args *args, trail_t *trail)
> libsvn_fs_base\tree.c:
> SVN_ERR(svn_fs_base__set_txn_merge_info(args->root->fs, txn_id,
> canon_path,
> libsvn_fs_base\tree.c: SVN_ERR(change_txn_merge_info(args, trail));
> libsvn_fs_base\tree.c: base_change_merge_info,
> libsvn_fs_fs\tree.c:fs_change_merge_info(svn_fs_root_t *root,
> libsvn_fs_fs\tree.c: fs_change_merge_info,
> libsvn_ra_dav\mergeinfo.c: { SVN_XML_NAMESPACE,
> "merge-info-report", ELEM_merge_info_report, 0 },
> libsvn_ra_dav\mergeinfo.c: { SVN_XML_NAMESPACE, "merge-info-item",
> ELEM_merge_info_item, 0 },
> libsvn_ra_dav\mergeinfo.c: { SVN_XML_NAMESPACE, "merge-info-path",
> ELEM_merge_info_path,
> libsvn_ra_dav\mergeinfo.c: { SVN_XML_NAMESPACE, "merge-info-info",
> ELEM_merge_info_info,
> libsvn_ra_dav\mergeinfo.c: if (elm->id != ELEM_merge_info_report)
> libsvn_ra_dav\mergeinfo.c: if (elm->id == ELEM_merge_info_item)
> libsvn_ra_dav\mergeinfo.c: if (elm->id == ELEM_merge_info_item)
> libsvn_ra_dav\mergeinfo.c: case ELEM_merge_info_path:
> libsvn_ra_dav\mergeinfo.c: case ELEM_merge_info_info:
> libsvn_ra_dav\ra_dav.h: ELEM_merge_info_report,
> libsvn_ra_dav\ra_dav.h: ELEM_merge_info_item,
> libsvn_ra_dav\ra_dav.h: ELEM_merge_info_path,
> libsvn_ra_dav\ra_dav.h: ELEM_merge_info_info,
> libsvn_ra_serf\merge.c:} merge_info_t;
> libsvn_ra_serf\merge.c:static merge_info_t *
> libsvn_ra_serf\merge.c: merge_info_t *info;
> libsvn_ra_serf\merge.c: merge_info_t *info;
> libsvn_ra_serf\merge.c: merge_info_t *info;
> libsvn_ra_serf\merge.c: merge_info_t *info;
> libsvn_ra_svn\client.c:static svn_error_t
> *ra_svn_get_merge_info(svn_ra_session_t *session,
> libsvn_ra_svn\client.c: ra_svn_get_merge_info,
> svnserve\serve.c:static svn_error_t *get_merge_info(svn_ra_svn_conn_t
> *conn, apr_pool_t *pool,
> svnserve\serve.c: { "get-merge-info", get_merge_info },

  • application/pgp-signature attachment: stored
Received on Fri Jun 8 21:35:23 2007

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.