From: "Đoàn Trần Công Danh" <congdanhqx@gmail.com>
To: David Bremner <david@tethera.net>
Cc: notmuch@notmuchmail.org
Subject: Re: [PATCH] compat: expose canonicalize_file_name to C++
Date: Sun, 18 Apr 2021 11:13:38 +0700 [thread overview]
Message-ID: <YHux8vY+2+cexMyL@danh.dev> (raw)
In-Reply-To: <87zgxxasao.fsf@tethera.net>
On 2021-04-17 11:39:59-0300, David Bremner <david@tethera.net> wrote:
> Đoàn Trần Công Danh <congdanhqx@gmail.com> writes:
>
> >
> > However, I see that lib/open.cc uses g_key_file_get_value from GLib
> > already, we may switch to g_canonicalize_file_name then?
> >
>
> Yes that could work. I think the treatment of NULL input might need some
> extra care with g_canonicalize_file_name; at least my 5 minute attempt
> to do a replacement causes many test failures. Do you want to make a
> patch that uses g_canonicalize_file_name? The one other place we use
> canonicalize_file_name (not totally coincidentally) also uses glib, so
> in principle we could completely eliminate the compat function.
Now, I'm thinking more about it and digging into GLib history,
I think using g_canonicalize_filename would be problem, since it's
available from 2.57.1/2.58 only. I think we're requiring glib-2.0 2.22
as of it's now. A big jump in dependencies isn't worth it.
I think below patch would be better?
Anyway, I see some failure in the testsuite due to:
- *My* hostname(1) (from coreutils) doesn't understand "-f"
- All emacs tests depend on dtach(1) but the
test_require_external_prereq dtach is missing. Would you want
to see those patches for test_require_external_prereq dtach?
It's pretty trivial.
-----8<-----
Subject: [PATCH] compat: rename {,notmuch_}canonicalize_file_name
When compat canonicalize_file_name was introduced, it was limited to
C code only because it was used by C code only during that time.
From 5ec6fd4d, (lib/open: check for split configuration when creating
database., 2021-02-16), lib/open.cc, which is C++, relies on the
existent of canonicalize_file_name.
However, we can't blindly enable canonicalize_file_name for C++ code,
because different implementation has different additional signature for
C++ and users can arbitrarily add -DHAVE_CANONICALIZE_FILE_NAME=0 to
{C,CXX}FLAGS.
Let's put our implementation of canonicalize_file_name to our namespace,
and prefer system's canonicalize_file_name if it's existed via a macro.
---
compat/canonicalize_file_name.c | 6 +++++-
compat/compat.h | 7 +++----
lib/open.cc | 4 ++--
notmuch-config.c | 2 +-
4 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/compat/canonicalize_file_name.c b/compat/canonicalize_file_name.c
index 000f9e78..ba003268 100644
--- a/compat/canonicalize_file_name.c
+++ b/compat/canonicalize_file_name.c
@@ -3,8 +3,12 @@
#undef _GNU_SOURCE
#include <stdlib.h>
+#ifdef notmuch_canonicalize_file_name
+#undef notmuch_canonicalize_file_name
+#endif
+
char *
-canonicalize_file_name (const char *path)
+notmuch_canonicalize_file_name (const char *path)
{
#ifdef PATH_MAX
char *resolved_path = malloc (PATH_MAX + 1);
diff --git a/compat/compat.h b/compat/compat.h
index 8f15e585..a38bc460 100644
--- a/compat/compat.h
+++ b/compat/compat.h
@@ -38,11 +38,10 @@ extern "C" {
#endif
#if ! HAVE_CANONICALIZE_FILE_NAME
-/* we only call this function from C, and this makes testing easier */
-#ifndef __cplusplus
char *
-canonicalize_file_name (const char *path);
-#endif
+notmuch_canonicalize_file_name (const char *path);
+#else
+#define notmuch_canonicalize_file_name canonicalize_file_name
#endif
#if ! HAVE_GETLINE
diff --git a/lib/open.cc b/lib/open.cc
index 5d80a884..7454ffae 100644
--- a/lib/open.cc
+++ b/lib/open.cc
@@ -612,9 +612,9 @@ notmuch_database_create_with_config (const char *database_path,
_set_database_path (notmuch, database_path);
if (key_file && ! split) {
- char *mail_root = canonicalize_file_name (
+ char *mail_root = notmuch_canonicalize_file_name (
g_key_file_get_value (key_file, "database", "mail_root", NULL));
- char *db_path = canonicalize_file_name (database_path);
+ char *db_path = notmuch_canonicalize_file_name (database_path);
split = (mail_root && (0 != strcmp (mail_root, db_path)));
diff --git a/notmuch-config.c b/notmuch-config.c
index 16e86916..d8d47768 100644
--- a/notmuch-config.c
+++ b/notmuch-config.c
@@ -327,7 +327,7 @@ notmuch_conffile_save (notmuch_conffile_t *config)
}
/* Try not to overwrite symlinks. */
- filename = canonicalize_file_name (config->filename);
+ filename = notmuch_canonicalize_file_name (config->filename);
if (! filename) {
if (errno == ENOENT) {
filename = strdup (config->filename);
--
2.31.1
--
Danh
next prev parent reply other threads:[~2021-04-18 4:13 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-17 0:18 [PATCH] compat: expose canonicalize_file_name to C++ Đoàn Trần Công Danh
2021-04-17 12:13 ` David Bremner
2021-04-17 12:49 ` Đoàn Trần Công Danh
2021-04-17 14:39 ` David Bremner
2021-04-18 4:13 ` Đoàn Trần Công Danh [this message]
2021-04-18 7:08 ` Tomi Ollila
2021-04-18 10:38 ` Đoàn Trần Công Danh
2021-04-23 17:31 ` David Bremner
2021-04-24 0:29 ` Đoàn Trần Công Danh
2021-04-18 12:48 ` David Bremner
2021-04-18 16:19 ` Tomi Ollila
2021-04-18 16:47 ` Tomi Ollila
2021-04-24 0:57 ` [PATCH v3] compat: rename {,notmuch_}canonicalize_file_name Đoàn Trần Công Danh
2021-04-24 1:01 ` Đoàn Trần Công Danh
2021-04-24 1:05 ` [PATCH v4] " Đoàn Trần Công Danh
2021-04-24 11:43 ` David Bremner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://notmuchmail.org/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YHux8vY+2+cexMyL@danh.dev \
--to=congdanhqx@gmail.com \
--cc=david@tethera.net \
--cc=notmuch@notmuchmail.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://yhetil.org/notmuch.git/
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).