* [Edbrowse-dev] Frameset crash
@ 2015-09-07 0:45 Tyler Spivey
2015-09-07 14:53 ` Karl Dahlke
0 siblings, 1 reply; 3+ messages in thread
From: Tyler Spivey @ 2015-09-07 0:45 UTC (permalink / raw)
To: edbrowse-dev
http://sprunge.us/MiNL
#0 0x00007ffff7497ec0 in tidyGetChild () from /usr/lib/libtidy-0.99.so.0
#1 0x00000000004403dd in traverseNode (node=0x0, level=0) at html-tidy.c:35
#2 0x0000000000440448 in traverseBody () at html-tidy.c:43
#3 0x0000000000440561 in html2nodes (
htmltext=0x6c3a70 "<frameset cols=\"25%,*,25%\">\n<frame
src=\"frame_a.htm\">\n<frame src=\"frame_b.htm\">\n<frame
src=\"frame_c.htm\">\n</frameset>\n\n") at html-tidy.c:82
#4 0x000000000042b808 in encodeTags (
html=0x6c3a70 "<frameset cols=\"25%,*,25%\">\n<frame
src=\"frame_a.htm\">\n<frame src=\"frame_b.htm\">\n<frame
src=\"frame_c.htm\">\n</frameset>\n\n", fromSource=1 '\001') at html.c:1734
#5 0x000000000042eb03 in htmlParse (
buf=0x6c3a70 "<frameset cols=\"25%,*,25%\">\n<frame
src=\"frame_a.htm\">\n<frame src=\"frame_b.htm\">\n<frame
src=\"frame_c.htm\">\n</frameset>\n\n", remote=1) at html.c:2735
#6 0x00000000004141df in browseCurrentBuffer () at buffers.c:4898
#7 0x000000000041355e in runCommand (line=0x690742 <line+2>
"http://sprunge.us/MiNL") at buffers.c:4567
#8 0x0000000000413c9d in edbrowseCommand (line=0x690740 <line> "b
http://sprunge.us/MiNL", script=0 '\000')
at buffers.c:4738
#9 0x0000000000406969 in main (argc=0, argv=0x7fffffffe110) at main.c:1325
^ permalink raw reply [flat|nested] 3+ messages in thread* [Edbrowse-dev] Frameset crash
2015-09-07 0:45 [Edbrowse-dev] Frameset crash Tyler Spivey
@ 2015-09-07 14:53 ` Karl Dahlke
2015-09-07 23:40 ` Kevin Carhart
0 siblings, 1 reply; 3+ messages in thread
From: Karl Dahlke @ 2015-09-07 14:53 UTC (permalink / raw)
To: edbrowse-dev
Yeah the problem here is there's no <body> and I was traversing the body.
Kevin showed me a way around it, if body is null then traverse the root,
which contains the frameset, and that fixed the seg fault,
but the head was scanned twice, as it is under the root,
which leads to the slap-your-forehead moment of saying,
"Gee, let's start at the root."
Then everything works fine, all the time.
I have made this push.
By the way, with js turned off, and other variables out of the way,
the tidy5 parser is ten times as fast as mine.
It probably uses lex and yacc,
or techniques therefrom, which are honed for efficiency.
Karl Dahlke
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Edbrowse-dev] Frameset crash
2015-09-07 14:53 ` Karl Dahlke
@ 2015-09-07 23:40 ` Kevin Carhart
0 siblings, 0 replies; 3+ messages in thread
From: Kevin Carhart @ 2015-09-07 23:40 UTC (permalink / raw)
To: Karl Dahlke; +Cc: edbrowse-dev
Oh, of course, no need to switch on (isFrameset) or anything.
Thanks!
On Mon, 7 Sep 2015, Karl Dahlke wrote:
> Yeah the problem here is there's no <body> and I was traversing the body.
> Kevin showed me a way around it, if body is null then traverse the root,
> which contains the frameset, and that fixed the seg fault,
> but the head was scanned twice, as it is under the root,
> which leads to the slap-your-forehead moment of saying,
> "Gee, let's start at the root."
> Then everything works fine, all the time.
> I have made this push.
>
> By the way, with js turned off, and other variables out of the way,
> the tidy5 parser is ten times as fast as mine.
> It probably uses lex and yacc,
> or techniques therefrom, which are honed for efficiency.
>
> Karl Dahlke
> _______________________________________________
> Edbrowse-dev mailing list
> Edbrowse-dev@lists.the-brannons.com
> http://lists.the-brannons.com/mailman/listinfo/edbrowse-dev
>
--------
Kevin Carhart * 415 225 5306 * The Ten Ninety Nihilists
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2015-09-07 23:37 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-09-07 0:45 [Edbrowse-dev] Frameset crash Tyler Spivey
2015-09-07 14:53 ` Karl Dahlke
2015-09-07 23:40 ` Kevin Carhart
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).