From: Ingo Schwarze <schwarze@usta.de>
To: tech@mdocml.bsd.lv
Subject: Re: [PATCH] Fix "Wait: No Child Process"
Date: Tue, 10 Mar 2015 04:23:41 +0100 [thread overview]
Message-ID: <20150310032341.GB18431@athene.usta.de> (raw)
In-Reply-To: <54FE0AFA.3070909@bsd.lv>
Hi Kristaps,
Kristaps Dzonsons wrote on Mon, Mar 09, 2015 at 10:04:58PM +0100:
> I'm still getting this error on Mac OS X, only for gzip'd files
> when invoked as man(1). Invoking directly as mandoc(1) does not
> cause the issue.
Gah, there was indeed a second issue causing this symptom, unrelated
to the one reported by bapt@. It was introduced as a side effect of
main.c rev. 1.212 ("do not spawn a pager when there is no output").
Fixed now...
Thanks for reporting,
Ingo
Log Message:
-----------
Fix a regression caused in rev. 1.212, reported by kristaps@:
When using a pager and the first manual shown is gzip'ed,
the gunzip(1) process ended up as a child of the pager process
such that the man(1) process couldn't wait for it, preventing
proper display of the manual.
Solve this by making the pager a child of the man(1) process
(instead of the other way round), which requires being a bit
more careful about properly closing file descriptors after use
and waiting for the pager before exiting man(1).
Modified Files:
--------------
mdocml:
main.c
Revision Data
-------------
Index: main.c
===================================================================
RCS file: /home/cvs/mdocml/mdocml/main.c,v
retrieving revision 1.223
retrieving revision 1.224
diff -Lmain.c -Lmain.c -u -p -r1.223 -r1.224
--- main.c
+++ main.c
@@ -20,6 +20,7 @@
#include <sys/types.h>
#include <sys/param.h> /* MACHINE */
+#include <sys/wait.h>
#include <assert.h>
#include <ctype.h>
@@ -481,6 +482,21 @@ out:
free(defos);
+ /*
+ * Flush the output and signal end of file.
+ * If a pager is attached, it allows browsing to the end.
+ * Otherwise, it does no harm, we are about to exit anyway.
+ */
+
+ fclose(stdout);
+
+ /*
+ * If we spawned a pager, wait for the user to close it.
+ * Otherwise, this call fails with no adverse effect.
+ */
+
+ wait(NULL);
+
return((int)rc);
}
@@ -951,18 +967,19 @@ spawn_pager(void)
progname, strerror(errno));
exit((int)MANDOCLEVEL_SYSERR);
case 0:
+ break;
+ default:
close(fildes[0]);
if (dup2(fildes[1], STDOUT_FILENO) == -1) {
fprintf(stderr, "%s: dup output: %s\n",
progname, strerror(errno));
exit((int)MANDOCLEVEL_SYSERR);
}
+ close(fildes[1]);
return;
- default:
- break;
}
- /* The original process becomes the pager. */
+ /* The child process becomes the pager. */
close(fildes[1]);
if (dup2(fildes[0], STDIN_FILENO) == -1) {
@@ -970,6 +987,7 @@ spawn_pager(void)
progname, strerror(errno));
exit((int)MANDOCLEVEL_SYSERR);
}
+ close(fildes[0]);
pager = getenv("MANPAGER");
if (pager == NULL || *pager == '\0')
--
To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv
prev parent reply other threads:[~2015-03-10 3:23 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-02 10:55 Baptiste Daroussin
2015-03-02 14:56 ` Ingo Schwarze
2015-03-09 21:04 ` Kristaps Dzonsons
2015-03-10 3:23 ` Ingo Schwarze [this message]
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=20150310032341.GB18431@athene.usta.de \
--to=schwarze@usta.de \
--cc=tech@mdocml.bsd.lv \
/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;
as well as URLs for NNTP newsgroup(s).