public inbox for linux-nfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Alice Mitchell <ajmitchell@redhat.com>
To: Steve Dickson <SteveD@RedHat.com>,
	Linux NFS Mailing list <linux-nfs@vger.kernel.org>
Subject: Re: [RFC PATCH 0/1] Enable config.d directory to be processed.
Date: Mon, 02 Nov 2020 15:57:02 +0000	[thread overview]
Message-ID: <1ac387a1ef608258b2e23e7923a1c4e2ec6b25b3.camel@redhat.com> (raw)
In-Reply-To: <4836616f-3aa6-d0bd-22db-cd7fecf4dce9@RedHat.com>

On Mon, 2020-11-02 at 09:23 -0500, Steve Dickson wrote:
> Hello,
> 
> On 11/2/20 8:24 AM, Steve Dickson wrote:
> > > You would need to write an equivalent of conf_load_file() that
> > > created a new transaction id and read in all the files before
> > > committing them to do it this way.
> > 

How about the following as an alternative...

It changes none of the past behaviour, but if you wanted to add an
optional directory structure to a config file then simply add this to
the default single config file that we ship.

/etc/nfsmount.conf:
[NFSMount_Global_Options]
include=-/etc/nfsmount.conf.d/*.conf


The leading minus tells it that it isnt an error if its empty, and it
will load all of the fragments it finds in there as well as the
existing single file.  Apply the same structure to any existing config
file that you want to also have a directory for.

-Alice



diff --git a/support/nfs/conffile.c b/support/nfs/conffile.c
index 58c03911..aade50c8 100644
--- a/support/nfs/conffile.c
+++ b/support/nfs/conffile.c
@@ -53,6 +53,7 @@
 #include <libgen.h>
 #include <sys/file.h>
 #include <time.h>
+#include <glob.h>
 
 #include "conffile.h"
 #include "xlog.h"
@@ -690,6 +691,7 @@ conf_parse_line(int trans, char *line, const char *filename, int lineno, char **
 	if (strcasecmp(line, "include")==0) {
 		/* load and parse subordinate config files */
 		_Bool optional = false;
+		glob_t globdata;
 
 		if (val && *val == '-') {
 			optional = true;
@@ -704,33 +706,46 @@ conf_parse_line(int trans, char *line, const char *filename, int lineno, char **
 			return;
 		}
 
-		subconf = conf_readfile(relpath);
-		if (subconf == NULL) {
-			if (!optional)
-				xlog_warn("config error at %s:%d: error loading included config",
-					  filename, lineno);
-			if (relpath)
-				free(relpath);
-			return;
-		}
+		if (glob(relpath, GLOB_MARK|GLOB_NOCHECK, NULL, &globdata)) {
+			xlog_warn("config error at %s:%d: error with matching pattern", filename, lineno);
+		} else 
+		{
+			for (size_t g=0; g<globdata.gl_pathc; g++) {
+				const char * subpath = globdata.gl_pathv[g];
 
-		/* copy the section data so the included file can inherit it
-		 * without accidentally changing it for us */
-		if (*section != NULL) {
-			inc_section = strdup(*section);
-			if (*subsection != NULL)
-				inc_subsection = strdup(*subsection);
-		}
+				if (subpath[strlen(subpath)-1] == '/') {
+					continue;
+				}
+				subconf = conf_readfile(subpath);
+				if (subconf == NULL) {
+					if (!optional)
+						xlog_warn("config error at %s:%d: error loading included config",
+							  filename, lineno);
+					if (relpath)
+						free(relpath);
+					return;
+				}
+
+				/* copy the section data so the included file can inherit it
+				 * without accidentally changing it for us */
+				if (*section != NULL) {
+					inc_section = strdup(*section);
+					if (*subsection != NULL)
+						inc_subsection = strdup(*subsection);
+				}
 
-		conf_parse(trans, subconf, &inc_section, &inc_subsection, relpath);
+				conf_parse(trans, subconf, &inc_section, &inc_subsection, relpath);
 
-		if (inc_section)
-			free(inc_section);
-		if (inc_subsection)
-			free(inc_subsection);
+				if (inc_section)
+					free(inc_section);
+				if (inc_subsection)
+					free(inc_subsection);
+				free(subconf);
+			}
+		}
 		if (relpath)
 			free(relpath);
-		free(subconf);
+		globfree(&globdata);
 	} else {
 		/* XXX Perhaps should we not ignore errors?  */
 		conf_set(trans, *section, *subsection, line, val, 0, 0);



  parent reply	other threads:[~2020-11-02 15:57 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-29 21:04 [RFC PATCH 0/1] Enable config.d directory to be processed Steve Dickson
2020-10-29 21:04 ` [PATCH 1/1] conffile: process config.d directory config files Steve Dickson
2020-11-02 13:03 ` [RFC PATCH 0/1] Enable config.d directory to be processed Alice Mitchell
2020-11-02 13:24   ` Steve Dickson
2020-11-02 14:23     ` Steve Dickson
2020-11-02 15:05       ` Alice Mitchell
2020-11-02 15:16         ` Chuck Lever
2020-11-02 16:37           ` Steve Dickson
2020-11-02 16:42         ` Steve Dickson
2020-11-02 15:57       ` Alice Mitchell [this message]
2020-11-02 19:42         ` Steve Dickson
2020-11-02 22:01           ` McIntyre, Vincent (CASS, Marsfield)
2020-11-03 10:14           ` Alice Mitchell
2020-11-03 17:16             ` Steve Dickson

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

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1ac387a1ef608258b2e23e7923a1c4e2ec6b25b3.camel@redhat.com \
    --to=ajmitchell@redhat.com \
    --cc=SteveD@RedHat.com \
    --cc=linux-nfs@vger.kernel.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.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox