- Dillo-dev - lists.mailman3.com

Implementing </div> closing <span>, and friends
by james.from.wellington＠gmail.com Sept. 15, 2014

Sept. 15, 2014

Hello all, [Sorry if this is a dup; gmail may be playing tricks on me.] This page displays, with distortions, in dillo 3.0.4, but the image does not display after commit 317f010: https://medium.com/@verynicetweets/getting-out-of-the-house-is-so-hard-thes… A partially cleaned-up and stripped version is attached, where the comment block shows a suppression of the problem. The page is illegal: two unclosed <span>s interfere with a </div> and the following <div>. There is code around html.cc:1355, which is intended to implement heuristics to close unclosed tags but, in the cases where it is triggered, no tags are closed. Here is a patch which I think implements the intent of the heuristics: ----------------- $ hg diff diff -r 26378f85c4ea src/html.cc --- a/src/html.cc Sun Sep 14 09:56:22 2014 +0000 +++ b/src/html.cc Mon Sep 15 18:13:21 2014 +1200 @@ -1355,7 +1355,9 @@ i_SELECT = a_Html_tag_index("select"), i_TEXTAREA = a_Html_tag_index("textarea"); int w3c_mode = !prefs.w3c_plus_heuristics; - int stack_idx, tag_idx, matched = 0, expected = 0; + // TODO: If there was a clean way of never having a tag index zero, + // (it's currently <a>), then expected_idx could be prettier. + int stack_idx, tag_idx, matched = 0, expected_idx = -1, skipped = 0; TagInfo new_tag = Tags[new_idx]; /* Look for the candidate tag to close */ @@ -1368,26 +1370,42 @@ break; } else if (Tags[tag_idx].EndTag == 'O') { /* skip an optional tag */ + ++skipped; continue; } else if ((new_idx == i_BUTTON && html->InFlags & IN_BUTTON) || (new_idx == i_SELECT && html->InFlags & IN_SELECT) || (new_idx == i_TEXTAREA && html->InFlags & IN_TEXTAREA)) { /* let these elements close tags inside them */ + ++skipped; continue; } else if (w3c_mode || Tags[tag_idx].TagLevel >= new_tag.TagLevel) { /* this is the tag that should have been closed */ - expected = 1; - break; + /* skip it as if it were optional, but complain */ + ++skipped; + if (-1 == expected_idx) { + // ie this is the first thing we're unwinding + expected_idx = tag_idx; + } + continue; } } + // Either or both matched and expected_idx could now be set + + // TODO: decide whether to carry line numbers in the parser state stack, + // in order to report them here. + if (matched) { + if (-1 != expected_idx) { + BUG_MSG("Unexpected closing tag: </%s> -- expected </%s>", + new_tag.name, Tags[expected_idx].name); + } + Html_tag_cleanup_to_idx(html, stack_idx); - } else if (expected) { - BUG_MSG("Unexpected closing tag: </%s> -- expected </%s>.", - new_tag.name, Tags[tag_idx].name); } else { - BUG_MSG("Unexpected closing tag: </%s>.", new_tag.name); + BUG_MSG("Unexpected closing tag: </%s>" + " -- explored %d layer(s) and not closed.", + new_tag.name, skipped); } } ----------------- Closing tags that match no opening tag (I used </area>, which is illegal) do not cause catastrophic failure. So far, it has not broken on my usual workload. Gmail is noisy, so I wonder if something's going wrong there, but it looks fairly normal. I have no good idea what a general test suite for this code looks like. If someone wants to point me toward related tests, then I will think about extending them. Regards, James.

1 0

dillo can't handle strings
by mail＠gabucino.hu Sept. 14, 2014

Sept. 14, 2014

See the issue on IRIX: http://lists.dillo.org/pipermail/dillo-dev/ 2014-April/010079.html Well, the same applies to PowerPC (OS X). "ftp://" works, "http://" doesn't. Somewhere there's a huge bug. Does anybody care? -- G

3 7

I'd like to add you to my professional network on LinkedIn
by acassis＠gmail.com Sept. 11, 2014

Sept. 11, 2014

Hi, I'd like to include you in my network to share updates and stay in touch. - Alan Accept: http://www.linkedin.com/blink?simpleRedirect=0Ue30PdjcNej0OdPgVe3sRcjARfkh9… You are receiving Invitation emails. Unsubscribe here: http://www.linkedin.com/blink?simpleRedirect=1DsCZ5cylLr6NFp30Q9npBp2RLr6NF… Learn why we included this at the following link: http://www.linkedin.com/blink?simpleRedirect=3wUdPgZp4BBr6dFt79x9DhPpnlDnSl… © 2014, LinkedIn Corporation. 2029 Stierlin Ct. Mountain View, CA 94043, USA

1 1

Problems tackling a bug
by eocene＠gmx.com Sept. 10, 2014

Sept. 10, 2014

Sebastian wrote: > I noticed that the current tip of dillo hangs in most cases when > displaying <http://www.dillo.org/>. I succeeded to reproduce this bug > in gdb, but get many "<optimized out>" values. So I tried to reduce > optimization, but unfortunately had problems reproducing this bug: > > - "-O2" (standard): reproducable both stand-alone and in gdb; > - "-O1": reproducable stand-alone, but not in gdb; > - "-O0": not reproducable at all. > > Versions envolved: > > $ gcc --version > gcc (Debian 4.7.2-5) 4.7.2 > $ gdb --version > GNU gdb (GDB) 7.4.1-debian > > Any ideas? I have the same gcc and gdb, and dillo doesn't seem to want to hang for me. Does it depend on your dillorc options?

3 2

duckduckgo
by eocene＠gmx.com Sept. 8, 2014

Sept. 8, 2014

Just tried searching for something using duckduckgo with midori, using the standard search, not the light one. I copied the link URL and saw that they're doing the same thing that google does -- the link URL shown at the bottom of the screen is what you expect, but the link is really a redirection through duckduckgo. Doesn't look like dillo does that -- maybe something to do with javascript, didn't matter why to me. Anyway, I didn't exactly like ddg behaving this way, and I believe I will try using startpage more often instead. (Although I have to turn off CSS to get a next button on startpage results...)

2 1

url display in location bar
by eocene＠gmx.com Sept. 7, 2014

Sept. 7, 2014

I often wish that dillo would show the beginning of the URL instead of the end. ("Where am I in this tab?") diff -r b28e4a47d4a1 src/ui.cc --- a/src/ui.cc Sat Sep 06 16:56:22 2014 +0000 +++ b/src/ui.cc Sat Sep 06 19:33:36 2014 +0000 @@ -816,7 +816,7 @@ { if (!str) str = ""; Location->value(str); - Location->position(strlen(str)); + Location->position((Fl::focus() == Location) ? strlen(str) : 0); } /*

2 1

configure.in errors
by csoren＠isd.net Aug. 30, 2014

Aug. 30, 2014

> Chris wrote: > > This is from a fresh hg pull, running autogen.sh (or > > aclocal): > > > > configure.in:472: error: AC_SUBST: `src doc' is not a > > valid shell variable name > > configure.in:472: the top level > > autom4te: /usr/bin/m4 failed with exit status: 1 > > aclocal: autom4te failed with exit status: 1 > > > > changing the line AC_SUBST(src doc) to two lines of > > AC_SUBST(src) and AC_SUBST(doc) fixes it. > > That's odd. configure.in moved to configure.ac in late > 2012... > Bad hg syntax on my part, sorry! I fixed it with 'hg update -C`

1 0

configure.in errors
by csoren＠isd.net Aug. 30, 2014

Aug. 30, 2014

This is from a fresh hg pull, running autogen.sh (or aclocal): configure.in:472: error: AC_SUBST: `src doc' is not a valid shell variable name configure.in:472: the top level autom4te: /usr/bin/m4 failed with exit status: 1 aclocal: autom4te failed with exit status: 1 changing the line AC_SUBST(src doc) to two lines of AC_SUBST(src) and AC_SUBST(doc) fixes it.

1 0

Last hg tip
by jcid＠dillo.org Aug. 30, 2014

Aug. 30, 2014

Hi Sebastian, I'm following closely our hg tip and some things are improving. OTOH the last tip sometimes drops the horizontal scroller bar leaving content out of reach. e.g. [1] [1] http://lwn.net/Articles/609892/#Comments -- Cheers Jorge.-

1 0

Rendering of illegal widths (incl "px", which used to be legal)
by james.from.wellington＠gmail.com Aug. 25, 2014

Aug. 25, 2014

Hello All, The row of three pictures on this web-page: http://www.avrprojects.net/index.php?view=article&id=61 renders badly in dillo, and differently from Chrome and firefox, because two of them have: width="180px" which the source-code for dillo 3.0.4 (html.cc at about line 1450) tells me used to be legal html, but isn't anymore. Dillo: - reports this as a web bug (with which I agree) - sets the length to auto (which I think means using the default width) To minimise the damage from illegal pages, without hiding the problem, I propose to assume that the numbers in all widths are sensible, and to complain about, but ignore, the trailing garbage. I also discovered that the bug message and another, suppressed, message had type problems, which I have addressed in a way that looks reasonable on osx. My proposed patch is below. Please be aware that Jorge and I started a conversation about whether this was the right thing to do, which wedged. I tested this proposed change by hacking up saved copies of the web-page in various ways. One of them, which makes a mess of the bug message, is attached, but they are not really a clean test-case. I am not aware of any automated test into which this change should be added. If someone can point me toward one, then I will think about how to test it cleanly. Regards, James. ----------------------- $ diff html.cc.orig html.cc 1483,1485c1483,1488 < if (*end && !isspace (*end)) { < BUG_MSG("Garbage after length: %s\n", attr); < l = CSS_CREATE_LENGTH(0.0, CSS_LENGTH_TYPE_AUTO); --- > while (end && isspace(*end)) > ++end; > if (end && *end) { > BUG_MSG("Garbage after length: '%s'; " > "ignoring garbage; using %g (possibly in pixels)\n", > attr, CSS_LENGTH_VALUE(l)); 1489c1492 < _MSG("a_Html_parse_length: \"%s\" %d\n", attr, CSS_LENGTH_VALUE(l)); --- > _MSG("a_Html_parse_length: \"%s\" %g\n", attr, CSS_LENGTH_VALUE(l)); ----------------------- ps I am contemplating getting involved in other trivial rendering problems, and thinking about the todos in dpid/main.c:204 and dpid/main.c:297. My interest in the todos is that dillo functions with gmail, but keeps (de facto) logging me out by discarding my session cookie. Could someone please point me to generic advice about: - which source tree I should be building (Jorge mentioned GROWS, but I don't know what that means) - how to use mercurial to develop multiple patches in parallel and keep them separate, while running a built program that includes all of them ?

1 0