- Dillo-dev - lists.mailman3.com

major commit upcoming

by Johannes.Hofmann＠gmx.de

On Thu, Oct 21, 2010 at 02:49:00PM +0200, Johannes Hofmann wrote: > -----Urspr?ngliche Nachricht----- > Von: Jorge Arellano Cid <jcid(a)dillo.org> > Gesendet: Do 21.10.2010 13:49 > An: dillo-dev(a)dillo.org; > Betreff: Re: [Dillo-dev] major commit upcoming > > > On Wed, Oct 20, 2010 at 10:05:28PM +0200, Johannes Hofmann wrote: > > > On Wed, Oct 20, 2010 at 12:39:37PM -0300, Jorge Arellano Cid wrote: > > > > On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > > > > > Hi, > > > > > > > > > > just a headsup, that I plan to commit some changes for the CSS on > > > > > <HTML> element issue that involve quite a few files. > > > > > Basically it is a minor API change for StyleEngine, but as > > > > > StyleEngine is called from quite a few places, the patch is rather > > > > > big. > > > > > Please let me know in case this is inconvenient atm. > > > > > > > > On my part, go ahead. > > > > I'm still working on the collapsing border code. > > > > > > Ok then - here you go.... > > > > > > Basically the changes are to fix the problem that CSS for the <HTML> > > > element was not evaluated (reported by Jeremy). > > > But the changes in StyleEngine will also help to implement dynamic > > > style changes in the future (as needed for :visited :hover :focus > > > etc). > > > > > > As always please report any regressions... > > > > It looks like it will help me! (inherit nonCSS stuff) > > Right, I moved this functionality from html.cc / table.cc into styleengine.cc. > > > > > --- > > > > BTW, I had a hard time with border collapsing, because after > > adapting dillo to render my test cases as Firefox, other tests > > showed big differences. It was after digging into cell_props, > > resetValues, {pre, post}processAttrs, parser, etc... that I found > > the problem: dillo is not discarding CSS with syntax errors: > > > > http://www.dillo.org/test/discard-syntax-errors.html > > > > Generally dillo should discard CSS with errors, but there are some cases > where it doesn't. I will look at your test case when I'm at home. Ah now I see what you mean... Declaration dropping does not work with shortcuts like margin:, border: and so on. It should work for single properties like border-bottom-style: and so on. I will look into it. Cheers, Johannes

Oct. 21, 2010

major commit upcoming

by Johannes.Hofmann＠gmx.de

On Thu, Oct 21, 2010 at 08:48:19AM -0300, Jorge Arellano Cid wrote: > On Wed, Oct 20, 2010 at 10:05:28PM +0200, Johannes Hofmann wrote: > > On Wed, Oct 20, 2010 at 12:39:37PM -0300, Jorge Arellano Cid wrote: > > > On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > > > > Hi, > > > > > > > > just a headsup, that I plan to commit some changes for the CSS on > > > > <HTML> element issue that involve quite a few files. > > > > Basically it is a minor API change for StyleEngine, but as > > > > StyleEngine is called from quite a few places, the patch is rather > > > > big. > > > > Please let me know in case this is inconvenient atm. > > > > > > On my part, go ahead. > > > I'm still working on the collapsing border code. > > > > Ok then - here you go.... > > > > Basically the changes are to fix the problem that CSS for the <HTML> > > element was not evaluated (reported by Jeremy). > > But the changes in StyleEngine will also help to implement dynamic > > style changes in the future (as needed for :visited :hover :focus > > etc). > > > > As always please report any regressions... > > It looks like it will help me! (inherit nonCSS stuff) > > --- > > BTW, I had a hard time with border collapsing, because after > adapting dillo to render my test cases as Firefox, other tests > showed big differences. It was after digging into cell_props, > resetValues, {pre, post}processAttrs, parser, etc... that I found > the problem: dillo is not discarding CSS with syntax errors: > > http://www.dillo.org/test/discard-syntax-errors.html > > Note1: I tried pushing to hg css_compat, but it didn't work. My > local repo behaved as the push was successful but no changes were > made in the remote one (setting hofmann as trusted in my local > hgrc didn't help either). > > Note2: index.html has 644 permissions Can you please retry - I hope it's fixed now. Cheers, Johannes

Oct. 21, 2010

major commit upcoming

by Johannes.Hofmann＠gmx.de

On Thu, Oct 21, 2010 at 11:01:42AM -0300, Jorge Arellano Cid wrote: > On Thu, Oct 21, 2010 at 02:49:00PM +0200, Johannes Hofmann wrote: > > > > [...] > > > > > > Note4: Considering the extensive use of "html->styleEngine", > > > I'd use something similar to HT2TB, S_TOP macros. This is: > > > SE(html) <=> html->styleEngine > > > SE(html)->setNonCssHint <=> html->styleEngine->setNonCssHint > > > I can quickly make a patch if you agree. > > > > > > > Yes, sure. Maybe an inlined method in DilloHtml like > > > > DilloHtml::setNonCssHint(...) { > > styleEngine->setNonCssHint(...); > > } > > > > would result in even shorter lines: > > > > html->setNonCssHint(...) > > DISCLAIMER: I'm by no mean a C++ guru! > > gcc says: > > html_common.hh:213: error: invalid use of incomplete type 'struct > StyleEngine' Bleh, that's what you get if you play tricks and make circular depending includes (as I did with styleengine.hh). > > AFAIU, the original setNonCssHint is also inlined, and it uses > the private "struct Node", which upon expansion is not defined. That shouldn't be a problem, inlining is not done by the preprocessor. > > I'd go with "SE(html)" ;-) Let me first try whether I can disentangle the includes. If that doesn't work we can use the macro. Cheers, Johannes

Oct. 21, 2010

major commit upcoming

by jcid＠dillo.org

On Thu, Oct 21, 2010 at 02:49:00PM +0200, Johannes Hofmann wrote: > > [...] > > > > Note4: Considering the extensive use of "html->styleEngine", > > I'd use something similar to HT2TB, S_TOP macros. This is: > > SE(html) <=> html->styleEngine > > SE(html)->setNonCssHint <=> html->styleEngine->setNonCssHint > > I can quickly make a patch if you agree. > > > > Yes, sure. Maybe an inlined method in DilloHtml like > > DilloHtml::setNonCssHint(...) { > styleEngine->setNonCssHint(...); > } > > would result in even shorter lines: > > html->setNonCssHint(...) DISCLAIMER: I'm by no mean a C++ guru! gcc says: html_common.hh:213: error: invalid use of incomplete type 'struct StyleEngine' AFAIU, the original setNonCssHint is also inlined, and it uses the private "struct Node", which upon expansion is not defined. I'd go with "SE(html)" ;-) -- Cheers Jorge.-

Oct. 21, 2010

AW: [Dillo-dev] major commit upcoming

by johannes.hofmann＠gmx.de

-----Urspr?ngliche Nachricht----- Von: Jorge Arellano Cid <jcid(a)dillo.org> Gesendet: Do 21.10.2010 13:49 An: dillo-dev(a)dillo.org; Betreff: Re: [Dillo-dev] major commit upcoming > On Wed, Oct 20, 2010 at 10:05:28PM +0200, Johannes Hofmann wrote: > > On Wed, Oct 20, 2010 at 12:39:37PM -0300, Jorge Arellano Cid wrote: > > > On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > > > > Hi, > > > > > > > > just a headsup, that I plan to commit some changes for the CSS on > > > > <HTML> element issue that involve quite a few files. > > > > Basically it is a minor API change for StyleEngine, but as > > > > StyleEngine is called from quite a few places, the patch is rather > > > > big. > > > > Please let me know in case this is inconvenient atm. > > > > > > On my part, go ahead. > > > I'm still working on the collapsing border code. > > > > Ok then - here you go.... > > > > Basically the changes are to fix the problem that CSS for the <HTML> > > element was not evaluated (reported by Jeremy). > > But the changes in StyleEngine will also help to implement dynamic > > style changes in the future (as needed for :visited :hover :focus > > etc). > > > > As always please report any regressions... > > It looks like it will help me! (inherit nonCSS stuff) Right, I moved this functionality from html.cc / table.cc into styleengine.cc. > > --- > > BTW, I had a hard time with border collapsing, because after > adapting dillo to render my test cases as Firefox, other tests > showed big differences. It was after digging into cell_props, > resetValues, {pre, post}processAttrs, parser, etc... that I found > the problem: dillo is not discarding CSS with syntax errors: > > http://www.dillo.org/test/discard-syntax-errors.html > Generally dillo should discard CSS with errors, but there are some cases where it doesn't. I will look at your test case when I'm at home. > Note1: I tried pushing to hg css_compat, but it didn't work. My > local repo behaved as the push was successful but no changes were > made in the remote one (setting hofmann as trusted in my local > hgrc didn't help either). > > Note2: index.html has 644 permissions Ok, I will check that. > > Note3: I'll work a bit more on the collapse patches and get > back to you before committing. > > Note4: Considering the extensive use of "html->styleEngine", > I'd use something similar to HT2TB, S_TOP macros. This is: > SE(html) <=> html->styleEngine > SE(html)->setNonCssHint <=> html->styleEngine->setNonCssHint > I can quickly make a patch if you agree. > Yes, sure. Maybe an inlined method in DilloHtml like DilloHtml::setNonCssHint(...) { styleEngine->setNonCssHint(...); } would result in even shorter lines: html->setNonCssHint(...) Cheers, Johannes

Oct. 21, 2010

major commit upcoming

by jcid＠dillo.org

On Wed, Oct 20, 2010 at 10:05:28PM +0200, Johannes Hofmann wrote: > On Wed, Oct 20, 2010 at 12:39:37PM -0300, Jorge Arellano Cid wrote: > > On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > > > Hi, > > > > > > just a headsup, that I plan to commit some changes for the CSS on > > > <HTML> element issue that involve quite a few files. > > > Basically it is a minor API change for StyleEngine, but as > > > StyleEngine is called from quite a few places, the patch is rather > > > big. > > > Please let me know in case this is inconvenient atm. > > > > On my part, go ahead. > > I'm still working on the collapsing border code. > > Ok then - here you go.... > > Basically the changes are to fix the problem that CSS for the <HTML> > element was not evaluated (reported by Jeremy). > But the changes in StyleEngine will also help to implement dynamic > style changes in the future (as needed for :visited :hover :focus > etc). > > As always please report any regressions... It looks like it will help me! (inherit nonCSS stuff) --- BTW, I had a hard time with border collapsing, because after adapting dillo to render my test cases as Firefox, other tests showed big differences. It was after digging into cell_props, resetValues, {pre, post}processAttrs, parser, etc... that I found the problem: dillo is not discarding CSS with syntax errors: http://www.dillo.org/test/discard-syntax-errors.html Note1: I tried pushing to hg css_compat, but it didn't work. My local repo behaved as the push was successful but no changes were made in the remote one (setting hofmann as trusted in my local hgrc didn't help either). Note2: index.html has 644 permissions Note3: I'll work a bit more on the collapse patches and get back to you before committing. Note4: Considering the extensive use of "html->styleEngine", I'd use something similar to HT2TB, S_TOP macros. This is: SE(html) <=> html->styleEngine SE(html)->setNonCssHint <=> html->styleEngine->setNonCssHint I can quickly make a patch if you agree. -- Cheers Jorge.-

Oct. 21, 2010

major commit upcoming

by Johannes.Hofmann＠gmx.de

On Wed, Oct 20, 2010 at 12:39:37PM -0300, Jorge Arellano Cid wrote: > On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > > Hi, > > > > just a headsup, that I plan to commit some changes for the CSS on > > <HTML> element issue that involve quite a few files. > > Basically it is a minor API change for StyleEngine, but as > > StyleEngine is called from quite a few places, the patch is rather > > big. > > Please let me know in case this is inconvenient atm. > > On my part, go ahead. > I'm still working on the collapsing border code. Ok then - here you go.... Basically the changes are to fix the problem that CSS for the <HTML> element was not evaluated (reported by Jeremy). But the changes in StyleEngine will also help to implement dynamic style changes in the future (as needed for :visited :hover :focus etc). As always please report any regressions... Cheers, Johannes

Oct. 20, 2010

major commit upcoming

by jcid＠dillo.org

On Tue, Oct 19, 2010 at 09:11:41PM +0200, Johannes Hofmann wrote: > Hi, > > just a headsup, that I plan to commit some changes for the CSS on > <HTML> element issue that involve quite a few files. > Basically it is a minor API change for StyleEngine, but as > StyleEngine is called from quite a few places, the patch is rather > big. > Please let me know in case this is inconvenient atm. On my part, go ahead. I'm still working on the collapsing border code. -- Cheers Jorge.-

Oct. 20, 2010

major commit upcoming

by Johannes.Hofmann＠gmx.de

Hi, just a headsup, that I plan to commit some changes for the CSS on <HTML> element issue that involve quite a few files. Basically it is a minor API change for StyleEngine, but as StyleEngine is called from quite a few places, the patch is rather big. Please let me know in case this is inconvenient atm. Cheers, Johannes

Oct. 19, 2010

Offtopic: big miners rescue

by vic＠smartqsys.com

Indeed! Hi there, A few days have passed since the big miners rescue finished as a complete success. Locally, we chileans, followed the events since the beginning, and specially moved since a small paper note emerged to earth level saying they were all alive. These days I've realized the press coverage was humongous. It's said that a thousand million saw a part of the rescue around the world. It's one of those few moments a lifetime gifts our souls with, where events, so powerful an real as these, make us look inwards and see the real value of things, instead of the things we usually entertain ourselves in. -- Cheers Jorge.- ------------------------------ _______________________________________________ Dillo-dev mailing list Dillo-dev(a)dillo.org http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev End of Dillo-dev Digest, Vol 62, Issue 14 *****************************************

Oct. 19, 2010