From: Kristaps Dzonsons <kristaps@bsd.lv>
To: tech@mdocml.bsd.lv
Subject: Re: roff conditional recursion
Date: Thu, 07 Apr 2011 12:02:21 +0200 [thread overview]
Message-ID: <4D9D8BAD.6010809@bsd.lv> (raw)
In-Reply-To: <20110407003900.GA20175@britannica.bec.de>
[-- Attachment #1: Type: text/plain, Size: 672 bytes --]
On 07/04/2011 02:39, Joerg Sonnenberger wrote:
> Hi Ingo,
> at the moment, as soon as ROFF_ie is seen, r->stackpos is incremented.
> This happens in roff.c 1028. But the corresponding code for handling
> ROFF_el in 995 doesn't decrement it again. Shouldn't it be
> rule = r->rstack[r->rstackpos--];
> in line 998?
>
> Attached is a trivial example.
Joerg,
Enclosed is a patch that follows this logic, plus some clean-up and
documentation (it works on test.1 and doesn't seem to cause problems
anywhere else). The rstackpos, yes, can be safely "consumed" when the
`el' is called, and need not wait until the nodes are popped.
Ingo, any comments?
Thanks,
Kristaps
[-- Attachment #2: patch.roff.txt --]
[-- Type: text/plain, Size: 2209 bytes --]
Index: roff.c
===================================================================
RCS file: /usr/vhosts/mdocml.bsd.lv/cvs/mdocml/roff.c,v
retrieving revision 1.131
diff -u -r1.131 roff.c
--- roff.c 5 Apr 2011 22:22:33 -0000 1.131
+++ roff.c 7 Apr 2011 09:57:11 -0000
@@ -277,10 +277,6 @@
assert(r->last);
p = r->last;
- if (ROFF_el == p->tok)
- if (r->rstackpos > -1)
- r->rstackpos--;
-
r->last = r->last->parent;
free(p->name);
free(p->end);
@@ -976,29 +972,20 @@
int sv;
enum roffrule rule;
- /* Stack overflow! */
+ /*
+ * An `.el' has no conditional body: it will consume the value
+ * of the current rstack entry set in prior `ie' calls or
+ * defaults to DENY.
+ *
+ * If we're not an `el', however, then evaluate the conditional.
+ */
- if (ROFF_ie == tok && r->rstackpos == RSTACK_MAX - 1) {
- mandoc_msg(MANDOCERR_MEM, r->parse, ln, ppos, NULL);
- return(ROFF_ERR);
- }
-
- /* First, evaluate the conditional. */
-
- if (ROFF_el == tok) {
- /*
- * An `.el' will get the value of the current rstack
- * entry set in prior `ie' calls or defaults to DENY.
- */
- if (r->rstackpos < 0)
- rule = ROFFRULE_DENY;
- else
- rule = r->rstack[r->rstackpos];
- } else
- rule = roff_evalcond(*bufp, &pos);
+ rule = ROFF_el == tok ?
+ (r->rstackpos < 0 ?
+ ROFFRULE_DENY : r->rstack[r->rstackpos--]) :
+ roff_evalcond(*bufp, &pos);
sv = pos;
-
while (' ' == (*bufp)[pos])
pos++;
@@ -1018,16 +1005,20 @@
r->last->rule = rule;
+ /*
+ * An if-else will put the NEGATION of the current evaluated
+ * conditional into the stack of rules.
+ */
+
if (ROFF_ie == tok) {
- /*
- * An if-else will put the NEGATION of the current
- * evaluated conditional into the stack.
- */
- r->rstackpos++;
- if (ROFFRULE_DENY == r->last->rule)
- r->rstack[r->rstackpos] = ROFFRULE_ALLOW;
- else
- r->rstack[r->rstackpos] = ROFFRULE_DENY;
+ if (r->rstackpos == RSTACK_MAX - 1) {
+ mandoc_msg(MANDOCERR_MEM,
+ r->parse, ln, ppos, NULL);
+ return(ROFF_ERR);
+ }
+ r->rstack[++r->rstackpos] =
+ ROFFRULE_DENY == r->last->rule ?
+ ROFFRULE_ALLOW : ROFFRULE_DENY;
}
/* If the parent has false as its rule, then so do we. */
prev parent reply other threads:[~2011-04-07 10:02 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-07 0:39 Joerg Sonnenberger
2011-04-07 10:02 ` Kristaps Dzonsons [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=4D9D8BAD.6010809@bsd.lv \
--to=kristaps@bsd.lv \
--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).