- Dillo-dev - lists.mailman3.com

Re: patch: reorder headers in query string

by place＠gobigwest.com

Jorge wrote: > On Wed, Mar 12, 2008 at 03:51:24PM +0000, place wrote: > > My idea of what seems like a good ordering, for what that's worth :) > > When I watch the headers, I always wonder why I didn't put Accept-Encoding > > next to Accept-Charset, for one thing... > > AFAIR headers are ordered as they appear in the RFC. > I'm not sure whether this is relevant, please check. Oh! Let's see. The RFC says: The order in which header fields with differing field names are received is not significant. However, it is "good practice" to send general-header fields first, followed by request-header or response- header fields, and ending with the entity-header fields. Following that, placing them in order of appearance, yields: POST Connection Accept-Charset Accept-Encoding Host Proxy-Authorization Referer User-Agent Content-Length Content-Type Cookie GET Cache-Control Connection Pragma Accept-Charset Accept-Encoding Host Proxy-Authorization Referer User-Agent Cookie Which is pretty close to what dillo has already if you don't count the headers that I have added, since I never realized there was any ordering.

March 12, 2008

Re: patch: rm ^F shortcut from menu.cc

by jcid＠dillo.org

Committed. -- Cheers Jorge.-

March 12, 2008

Re: patch: reorder headers in query string

by place＠gobigwest.com

Jorge wrote: > On Wed, Mar 12, 2008 at 03:51:24PM +0000, place wrote: > > [...] > > BTW, I sent a User-Agent patch long ago and never heard anything. > > I could send another... > > I seem to remember that you finally didn't think of it as > a very helpful idea, but don't find any email saying so... It's useful on the whole, but a user does have to keep it in mind when things start to act differently. - If I pretend to use Firefox, google will finally use UTF-8 like the headers specify -- but I have to type in "&start=n" a lot. That's frustrating that they don't like to listen to Accept-Charset. - I think more sites listen to Accept-Encoding when I pretend to use a big browser. - It's hard to feel any degree of anonymity on the web when there are... how many dillo users? Not a lot. - There's some kind of blog thing that knows which browsers send an Accept header. If I pretend to use one of them, I get a meaningless error page. This is good because it saves me from wasting time looking at blogs. Hmm. Maybe I'm talking myself out of having it in the real dillo for general use -- and maybe that's what happened last time, too.

March 12, 2008

Re: patch: reorder headers in query string

by jcid＠dillo.org

On Wed, Mar 12, 2008 at 03:51:24PM +0000, place wrote: > [...] > BTW, I sent a User-Agent patch long ago and never heard anything. > I could send another... I seem to remember that you finally didn't think of it as a very helpful idea, but don't find any email saying so... -- Cheers Jorge.-

March 12, 2008

Re: patch: reorder headers in query string

by jcid＠dillo.org

On Wed, Mar 12, 2008 at 03:51:24PM +0000, place wrote: > My idea of what seems like a good ordering, for what that's worth :) > When I watch the headers, I always wonder why I didn't put Accept-Encoding > next to Accept-Charset, for one thing... AFAIR headers are ordered as they appear in the RFC. I'm not sure whether this is relevant, please check. -- Cheers Jorge.-

March 12, 2008

[patch] allow drag scrolling on images without link

by Johannes.Hofmann＠gmx.de

On Tue, Mar 11, 2008 at 06:15:24PM -0400, Jorge Arellano Cid wrote: > Hi, > > On Mon, Mar 10, 2008 at 10:09:29PM +0100, Johannes Hofmann wrote: > > Hi, > > > > the patch below allows drag scrolling on images that don't have a > > link. > > Oh, I committed a patch fix to let click-to-load over unloaded images > to work again (i.e. from img-off mode). Good point. Thanks, Johannes

March 11, 2008

[patch] allow drag scrolling on images without link

by jcid＠dillo.org

Hi, On Mon, Mar 10, 2008 at 10:09:29PM +0100, Johannes Hofmann wrote: > Hi, > > the patch below allows drag scrolling on images that don't have a > link. Oh, I committed a patch fix to let click-to-load over unloaded images to work again (i.e. from img-off mode). -- Cheers Jorge.-

March 11, 2008

patch: meta charset changes decoder

by jcid＠dillo.org

Hi, On Fri, Mar 07, 2008 at 04:21:16PM +0000, place wrote: > Jorge wrote: > > On Fri, Mar 07, 2008 at 06:17:17AM +0000, place wrote: > > > This is no sort of proper fix whatsoever, but I was tired > > > of seeing the "would set charset" message. It's actually > > > surprisingly useful for remote files, since they're generally > > > still at the javascript-and-css point at the end of the first packet. > > > > You mean, the decoder injection usually happens in the javascript-and-css > > section, right? > > > > Please post some test-case URLs. > > Ummm, okay. > I'll search for ?????????????????? and see what comes up, > looking for pages where meta is setting it to something other than utf-8. OK, committed as is. I tried to make a "clean" switch by re-starting parsing all over again, but it needs more work (prototype works but unstable). > http://www.nbu.bg/ works. > http://www.biforum.org/ works. > http://bgstudent.8m.com/ works. > http://www.newobjects.com/dict.asp works. > http://www.mastylo.net/ fails that content-type testing in misc.c. > http://bultext.tripod.com/ works. (tripod's still around?) > http://textove.com/ works. > http://bgstories.athost.net/ works. > http://www.bglekar.com/ works even though it sets charset to ISO-8859-1! > (it's all entities) > > Getting bored, but I'll keep going until one breaks by having body text in > the first packet. > > http://www.bds-bg.org/ works. > http://www.kursove-neg.com/ works. > http://avast.110mb.com/ fails that content-type testing in misc.c. > http://esperanto.vnvsoft.com/ works. > http://www.dfbulgaria.org/ works. > http://www.bghelsinki.org/ works. > http://www.bcnl.org/ works. > http://www.angelfire.com/ca/canbul/Bcacb.html works. > http://www.ibl.bas.bg/ works. > http://bgjedi.com/ breaks. So no Bulgarian jedi for dillo yet. Thanks for the examples. > And of course the titles are all broken, but my window manager can only > understand latin1 anyway. Oh, titles and local files work here (with the unstable prototype). -- Cheers Jorge.-

March 11, 2008

[patch] allow drag scrolling on images without link

by jcid＠dillo.org

On Mon, Mar 10, 2008 at 10:09:29PM +0100, Johannes Hofmann wrote: > Hi, > > the patch below allows drag scrolling on images that don't have a > link. Committed. -- Cheers Jorge.-

March 11, 2008

[patch] further reduce redraws from resizing tables

by Johannes.Hofmann＠gmx.de

Hi, below is another patch to reduce the redraws caused by resizing tables. A new method Layout::queueDrawExcept(), that requests the drawing of a rectangle except another rectangle is added. It is then used in widget.cc to redraw the area that was previously occupied by a widget, but is no longer occupied after a resize or move. This makes it possible to redraw less when a table resizes. All in all it should reduce the flicker a bit when loading table based pages. Cheers, Johannes diff --git a/dw/layout.cc b/dw/layout.cc --- a/dw/layout.cc +++ b/dw/layout.cc @@ -610,11 +610,36 @@ void Layout::queueDraw (int x, int y, in area.width = width; area.height = height; + if (area.isEmpty ()) return; + for (container::typed::Iterator <View> it = views->iterator (); it.hasNext (); ) { View *view = it.getNext (); view->queueDraw (&area); } +} + +void Layout::queueDrawExcept (int x, int y, int width, int height, + int ex, int ey, int ewidth, int eheight) { + + if (x == ex && y == ey && width == ewidth && height == eheight) + return; + + // queueDraw() the four rectangles within rectangle (x, y, width, height) + // around rectangle (ex, ey, ewidth, eheight). + // Some or all of these may be empty. + + // upper left corner of the intersection rectangle + int ix1 = misc::max (x, ex); + int iy1 = misc::max (y, ey); + // lower right corner of the intersection rectangle + int ix2 = misc::min (x + width, ex + ewidth); + int iy2 = misc::min (y + height, ey + eheight); + + queueDraw (x, y, width, iy1 - y); + queueDraw (x, iy2, width, y + height - iy2); + queueDraw (x, iy1, ix1 - x, iy2 - iy1); + queueDraw (ix2, iy1, x + width - ix2, iy2 - iy1); } void Layout::queueResize (bool asap) diff --git a/dw/layout.hh b/dw/layout.hh --- a/dw/layout.hh +++ b/dw/layout.hh @@ -133,6 +133,8 @@ private: void updateCursor (); void updateBgColor (); void queueDraw (int x, int y, int width, int height); + void queueDrawExcept (int x, int y, int width, int height, + int ex, int ey, int ewidth, int eheight); void queueResize (bool asap); void removeWidget (); diff --git a/dw/table.cc b/dw/table.cc --- a/dw/table.cc +++ b/dw/table.cc @@ -61,6 +61,7 @@ Table::Table(bool limitTextWidth) hasColPercent = 0; colPercents = new misc::SimpleVector <float> (8); + redrawX = 0; redrawY = 0; } @@ -191,10 +192,11 @@ void Table::sizeAllocateImpl (core::Allo void Table::resizeDrawImpl () { + queueDrawArea (redrawX, 0, allocation.width - redrawX, getHeight ()); queueDrawArea (0, redrawY, allocation.width, getHeight () - redrawY); + redrawX = allocation.width; redrawY = getHeight (); } - void Table::setWidth (int width) { diff --git a/dw/table.hh b/dw/table.hh --- a/dw/table.hh +++ b/dw/table.hh @@ -378,7 +378,7 @@ private: int numRows, numCols, curRow, curCol; misc::SimpleVector<Child*> *children; - int redrawY; + int redrawX, redrawY; /** * \brief The extremes of all columns. @@ -437,11 +437,7 @@ private: void setCumHeight (int row, int value) { if (value != cumHeight->get (row)) { - if (row > 0) { - redrawY = misc::min ( redrawY, cumHeight->get (row - 1) ); - } else { - redrawY = 0; - } + redrawY = misc::min ( redrawY, value ); cumHeight->set (row, value); } } @@ -449,7 +445,7 @@ private: inline void setColWidth (int col, int value) { if (value != colWidths->get (col)) { - redrawY = 0; + redrawX = misc::min (redrawX, value); colWidths->set (col, value); } } diff --git a/dw/types.hh b/dw/types.hh --- a/dw/types.hh +++ b/dw/types.hh @@ -71,6 +71,7 @@ public: bool intersectsWith (Rectangle *otherRect, Rectangle *dest); bool isSubsetOf (Rectangle *otherRect); bool isPointWithin (int x, int y); + bool isEmpty () { return width <= 0 || height <= 0; }; }; /** diff --git a/dw/widget.cc b/dw/widget.cc --- a/dw/widget.cc +++ b/dw/widget.cc @@ -418,6 +418,18 @@ void Widget::sizeAllocate (Allocation *a // widget->allocation.x, widget->allocation.y, // widget->allocation.width, widget->allocation.ascent, // widget->allocation.descent); + + if (wasAllocated ()) { + layout->queueDrawExcept ( + this->allocation.x, + this->allocation.y, + this->allocation.width, + this->allocation.ascent + this->allocation.descent, + allocation->x, + allocation->y, + allocation->width, + allocation->ascent + allocation->descent); + } sizeAllocateImpl (allocation);

March 11, 2008

[patch] allow drag scrolling on images without link

by Johannes.Hofmann＠gmx.de

Hi, the patch below allows drag scrolling on images that don't have a link. Cheers, Johannes diff -r 2d5370131f9f dw/image.cc --- a/dw/image.cc Sat Mar 08 16:56:13 2008 +0100 +++ b/dw/image.cc Mon Mar 10 22:06:20 2008 +0100 @@ -157,12 +157,15 @@ void Image::leaveNotifyImpl (core::Event bool Image::buttonPressImpl (core::EventButton *event) { - if (event->button == 3){ + if (event->button == 3) { (void)emitLinkPress(getStyle()->x_link, getStyle()->x_img, -1,-1,event); + return true; + } else if (getStyle()->x_link != -1) { + clicking = true; + return true; } else { - clicking = true; + return false; } - return true; } bool Image::buttonReleaseImpl (core::EventButton *event) @@ -170,8 +173,10 @@ bool Image::buttonReleaseImpl (core::Eve if (clicking) { clicking = false; emitLinkClick (getStyle()->x_link, getStyle()->x_img, -1, -1, event); + return true; + } else { + return false; } - return true; } void Image::draw (core::View *view, core::Rectangle *area)

March 10, 2008

From the FLTK devs: [dejan.lekic@gmail.com: Re: [fltk.development] This year GSOC]

by jcid＠dillo.org

Hi Jeremy, On Thu, Feb 28, 2008 at 05:32:55PM +0000, Jeremy Henty wrote: > > Just to make sure no-one misses this, some love from the FLTK guys! > > Jeremy Henty I've tried several times to contact Dejan (at his gmail and nu6 addresses with no luck. Do your emails get answers from those addresses? and if so, would you mind asking him to whitelist me or something? -- Cheers Jorge.- > > > ----- Forwarded message from Dejan Lekic <dejan.lekic(a)gmail.com> ----- > > From: Dejan Lekic <dejan.lekic(a)gmail.com> > To: fltk-dev(a)easysw.com > Subject: Re: [fltk.development] This year GSOC > Date: Thu, 28 Feb 2008 08:25:55 -0800 > > > Personally, the thing I most anticipate and would like to see improved, > which is related to FLTK, is the Dillo port to FLTK2. > > Seriously, if Dillo2 could be done as a library, and provide a Group > widget which is directly embeddable into any FLTK2 application, that > would be IMHO a huuuuuge thing. > > So, I hope someone would help Dillo guys either via GSoC, or other way(s). > > More information: http://www.dillo.org . Get the fresh FLTK2 port of > Dillo from CVS: :pserver:anonymous@auriga.wearlab.de:/sfhome/cvs/dillo > > Kind regards > _______________________________________________ > fltk-dev mailing list > fltk-dev(a)easysw.com > http://lists.easysw.com/mailman/listinfo/fltk-dev > > ----- End forwarded message -----

March 8, 2008

patch: meta charset changes decoder

by place＠gobigwest.com

Jorge wrote: > On Fri, Mar 07, 2008 at 06:17:17AM +0000, place wrote: > > This is no sort of proper fix whatsoever, but I was tired > > of seeing the "would set charset" message. It's actually > > surprisingly useful for remote files, since they're generally > > still at the javascript-and-css point at the end of the first packet. > > You mean, the decoder injection usually happens in the javascript-and-css > section, right? > > Please post some test-case URLs. Ummm, okay. I'll search for ????????? and see what comes up, looking for pages where meta is setting it to something other than utf-8. http://www.nbu.bg/ works. http://www.biforum.org/ works. http://bgstudent.8m.com/ works. http://www.newobjects.com/dict.asp works. http://www.mastylo.net/ fails that content-type testing in misc.c. http://bultext.tripod.com/ works. (tripod's still around?) http://textove.com/ works. http://bgstories.athost.net/ works. http://www.bglekar.com/ works even though it sets charset to ISO-8859-1! (it's all entities) Getting bored, but I'll keep going until one breaks by having body text in the first packet. http://www.bds-bg.org/ works. http://www.kursove-neg.com/ works. http://avast.110mb.com/ fails that content-type testing in misc.c. http://esperanto.vnvsoft.com/ works. http://www.dfbulgaria.org/ works. http://www.bghelsinki.org/ works. http://www.bcnl.org/ works. http://www.angelfire.com/ca/canbul/Bcacb.html works. http://www.ibl.bas.bg/ works. http://bgjedi.com/ breaks. So no Bulgarian jedi for dillo yet. And of course the titles are all broken, but my window manager can only understand latin1 anyway.

March 7, 2008

patch: meta charset changes decoder

by jcid＠dillo.org

On Fri, Mar 07, 2008 at 06:17:17AM +0000, place wrote: > This is no sort of proper fix whatsoever, but I was tired > of seeing the "would set charset" message. It's actually > surprisingly useful for remote files, since they're generally > still at the javascript-and-css point at the end of the first packet. You mean, the decoder injection usually happens in the javascript-and-css section, right? Please post some test-case URLs. -- Cheers Jorge.-

March 7, 2008

patch: entities in options

by jcid＠dillo.org

On Fri, Mar 07, 2008 at 06:15:33AM +0000, place wrote: > I was getting crashes on the dFree(tok) in Html_parse_entity() > on a page that I whittled down to: > > <html> <head> <title></title> </head> > <body> > <form> > <select> > <option>&Someword > </select> > </form> > </body> </html> > > It turns out that strchr(":_.-", '\0') finds the null terminator in the string. Good!, committed. -- Cheers Jorge.-

March 7, 2008

patch: meta charset changes decoder

by place＠gobigwest.com

This is no sort of proper fix whatsoever, but I was tired of seeing the "would set charset" message. It's actually surprisingly useful for remote files, since they're generally still at the javascript-and-css point at the end of the first packet.

March 7, 2008

patch: entities in options

by place＠gobigwest.com

I was getting crashes on the dFree(tok) in Html_parse_entity() on a page that I whittled down to: <html> <head> <title></title> </head> <body> <form> <select> <option>&Someword </select> </form> </body> </html> It turns out that strchr(":_.-", '\0') finds the null terminator in the string. diff -pur dillo2/src/html.cc dillo2-cur/src/html.cc --- dillo2/src/html.cc 2008-03-06 14:58:56.000000000 +0000 +++ dillo2-cur/src/html.cc 2008-03-07 05:46:52.000000000 +0000 @@ -1350,7 +1350,7 @@ static int Html_parse_entity(DilloHtml * } else if (isalpha(*s)) { /* character entity reference */ - while (isalnum(*++s) || strchr(":_.-", *s)); + while (*++s && (isalnum(*s) || strchr(":_.-", *s))); c = *s; *s = 0;

March 7, 2008

patch: form data character encoding

by akemnade＠tzi.de

Hi, On Thu, 6 Mar 2008 15:24:37 +0000 "place" <place(a)gobigwest.com> wrote: > Jorge wrote: > > Committed with this change: > > > > - dFree(input); > > + dStr_free(input, 1); > > oops :) > > I hope this goes well because next I have some experimental code > that encodes non-ascii mime parameters (input names). From > what I've read on the web, my suspicion is that it's still not > well-supported by servers and not going to work -- although there's > been some clamoring for it for filename parameters specifically. > > Speaking of files, would anybody have a use for file inputs? > It's the big remaining gap in form submission, but the only > site I use with file inputs is quite javascripty. > Yes, me! As far as I can remember, I had to switch browsers several times because of dillo's lack of file input support. At least one site I use quite often has such a file input. Greetings Andreas Kemnade

March 6, 2008

[patch] simplify Region::addRectangle ()

by Johannes.Hofmann＠gmx.de

On Wed, Mar 05, 2008 at 03:22:09PM -0300, Jorge Arellano Cid wrote: > On Wed, Mar 05, 2008 at 06:19:49PM +0100, Johannes Hofmann wrote: > > Hi, > > > > the patch below simplifies Region::addRectangle () quite a bit. It > > now simply tries to minimize the sum of the areas of all rectangles > > in the Region while still covering all rectangles that have been > > added with addRectangle (). For example if we have > > > > Thereby the total area that is actually redrawn is reduced > > and the code gets much simpler. > > However it may now happen that small parts of the > > screen are drawn that were not requested to be drawn. > > It looks like a part of the explanation got trimmed. > Would you mind resend it? Yeah, I had started some fancy ASCII art which got more complex than the code involved :-) Cheers, Johannes

March 6, 2008

patch: form data character encoding

by place＠gobigwest.com

Jorge wrote: > Committed with this change: > > - dFree(input); > + dStr_free(input, 1); oops :) I hope this goes well because next I have some experimental code that encodes non-ascii mime parameters (input names). From what I've read on the web, my suspicion is that it's still not well-supported by servers and not going to work -- although there's been some clamoring for it for filename parameters specifically. Speaking of files, would anybody have a use for file inputs? It's the big remaining gap in form submission, but the only site I use with file inputs is quite javascripty.

March 6, 2008

patch: form data character encoding

by jcid＠dillo.org

On Wed, Mar 05, 2008 at 08:23:30PM +0000, place wrote: > It seems to be working well. > For example, you can go to www.google.gr and paste in ????????????????. > Instead of 2530 hits of gibberish, you now get 57,400,000 of Greek. > > I ran it through various tests, but of course it'll get better testing > from those of you who routinely use ?'s and ?'s and so on. > It's an annoyance that fltk won't allow normal X character composing. > I don't want to learn their special rules, so I have to paste anything > non-ascii from an xterm. Committed with this change: - dFree(input); + dStr_free(input, 1); Please check... ;-) -- Cheers Jorge.-

March 6, 2008

[patch] simplify Region::addRectangle ()

by jcid＠dillo.org

On Wed, Mar 05, 2008 at 06:19:49PM +0100, Johannes Hofmann wrote: > Hi, > > the patch below simplifies Region::addRectangle () quite a bit. It > now simply tries to minimize the sum of the areas of all rectangles > in the Region while still covering all rectangles that have been > added with addRectangle (). For example if we have > > Thereby the total area that is actually redrawn is reduced > and the code gets much simpler. > However it may now happen that small parts of the > screen are drawn that were not requested to be drawn. OK, forget the extra explanation, I read and understood the code and it looks quite reasonable. Committed. -- Cheers Jorge.-

March 5, 2008

patch: form data character encoding

by place＠gobigwest.com

It seems to be working well. For example, you can go to www.google.gr and paste in ????????. Instead of 2530 hits of gibberish, you now get 57,400,000 of Greek. I ran it through various tests, but of course it'll get better testing from those of you who routinely use ?'s and ?'s and so on. It's an annoyance that fltk won't allow normal X character composing. I don't want to learn their special rules, so I have to paste anything non-ascii from an xterm.

March 5, 2008

[patch] simplify Region::addRectangle ()

by jcid＠dillo.org

On Wed, Mar 05, 2008 at 06:19:49PM +0100, Johannes Hofmann wrote: > Hi, > > the patch below simplifies Region::addRectangle () quite a bit. It > now simply tries to minimize the sum of the areas of all rectangles > in the Region while still covering all rectangles that have been > added with addRectangle (). For example if we have > > Thereby the total area that is actually redrawn is reduced > and the code gets much simpler. > However it may now happen that small parts of the > screen are drawn that were not requested to be drawn. It looks like a part of the explanation got trimmed. Would you mind resend it? -- Cheers Jorge.-

March 5, 2008

[patch] simplify Region::addRectangle ()

by Johannes.Hofmann＠gmx.de

Hi, the patch below simplifies Region::addRectangle () quite a bit. It now simply tries to minimize the sum of the areas of all rectangles in the Region while still covering all rectangles that have been added with addRectangle (). For example if we have Thereby the total area that is actually redrawn is reduced and the code gets much simpler. However it may now happen that small parts of the screen are drawn that were not requested to be drawn. This patch does not reduce any flickering. Cheers, Johannes diff --git a/dw/types.cc b/dw/types.cc --- a/dw/types.cc +++ b/dw/types.cc @@ -240,40 +240,22 @@ void Region::addRectangle (Rectangle *rP for (it = rectangleList->iterator (); it.hasNext (); ) { Rectangle *ownRect = it.getNext (); - /* check if r an ownRect can be combined vertically */ - if (r->x == ownRect->x && r->width == ownRect->width) { - int combinedHeight = - misc::max(r->y + r->height, ownRect->y + ownRect->height) - - misc::min(r->y, ownRect->y); + int combinedHeight = + misc::max(r->y + r->height, ownRect->y + ownRect->height) - + misc::min(r->y, ownRect->y); + int combinedWidth = + misc::max(r->x + r->width, ownRect->x + ownRect->width) - + misc::min(r->x, ownRect->x); - if (combinedHeight <= r->height + ownRect->height) { + if (combinedWidth * combinedHeight <= + ownRect->width * ownRect->height + r->width * r->height) { + + r->x = misc::min(r->x, ownRect->x); r->y = misc::min(r->y, ownRect->y); + r->width = combinedWidth; r->height = combinedHeight; + rectangleList->removeRef (ownRect); - } - - /* check if r an ownRect can be combined horizontally */ - } else if (r->y == ownRect->y && r->height == ownRect->height) { - int combinedWidth = - misc::max(r->x + r->width, ownRect->x + ownRect->width) - - misc::min(r->x, ownRect->x); - - if (combinedWidth <= r->width + ownRect->width) { - r->x = misc::min(r->x, ownRect->x); - r->width = combinedWidth; - rectangleList->removeRef (ownRect); - } - } - } - - for (it = rectangleList->iterator (); it.hasNext (); ) { - Rectangle *ownRect = it.getNext (); - - if (r->isSubsetOf (ownRect)) { - delete r; - return; - } else if (ownRect->isSubsetOf (r)) { - rectangleList->removeRef (ownRect); } } diff --git a/dw/types.hh b/dw/types.hh --- a/dw/types.hh +++ b/dw/types.hh @@ -121,6 +121,8 @@ public: * \brief dw::core::Shape implementation for a point set. * Currently represented as a set of rectangles not containing * each other. + * It is guaranteed that the rectangles returned by rectangles () + * cover all rectangles that were added with addRectangle (). */ class Region: public Shape {

March 5, 2008

patch: choose charset for form submission

by jcid＠dillo.org

On Tue, Mar 04, 2008 at 03:02:36PM +0000, place wrote: > Jorge wrote: > > On Tue, Mar 04, 2008 at 06:47:22AM +0000, place wrote: > > > It picks a charset, but doesn't actually do any translation. > > > > Committed with minor changes. > > But if you dFree(first) ahead of time, won't that be bad for > html->formNew(method, action, enc, charset);? Ooops, yes. Reversed to original. -- Cheers Jorge.-

March 4, 2008

patch: choose charset for form submission

by place＠gobigwest.com

Jorge wrote: > On Tue, Mar 04, 2008 at 06:47:22AM +0000, place wrote: > > It picks a charset, but doesn't actually do any translation. > > Committed with minor changes. But if you dFree(first) ahead of time, won't that be bad for html->formNew(method, action, enc, charset);?

March 4, 2008

patch: choose charset for form submission

by jcid＠dillo.org

Hi, On Tue, Mar 04, 2008 at 06:47:22AM +0000, place wrote: > It picks a charset, but doesn't actually do any translation. Committed with minor changes. > I have to decide whether to distort the charset decoder to handle encoding. > It's nice to have all of the iconv stuff in one place, but it's > not nice to have something that's shaped to decode dbufs reused for > just any purpose... I'm all for clarity. If you need to add a "encode.c" source file, go ahead. If a simple function is enough, let it be. If distortion comes in, something is wrong. :-) -- Cheers Jorge.-

March 4, 2008

patch: choose charset for form submission

by place＠gobigwest.com

It picks a charset, but doesn't actually do any translation. I have to decide whether to distort the charset decoder to handle encoding. It's nice to have all of the iconv stuff in one place, but it's not nice to have something that's shaped to decode dbufs reused for just any purpose...

March 4, 2008

two warnings in table.cc

by jcid＠dillo.org

On Mon, Mar 03, 2008 at 07:03:57PM +0100, Johannes Hofmann wrote: > Hi, > > here are two more warnings like this: > > table.cc: In member function 'void dw::Table::forceCalcColumnExtremes()': > table.cc:778: warning: 'wMin' may be used uninitialized in this function > table.cc: In member function 'void dw::Table::forceCalcCellSizes()': > table.cc:483: warning: 'totalWidth' may be used uninitialized in this function One bug, one false positive. Fixes committed. These warnings are really helpful. -- Cheers Jorge.-

March 3, 2008

two warnings in table.cc

by Johannes.Hofmann＠gmx.de

Hi, here are two more warnings like this: table.cc: In member function 'void dw::Table::forceCalcColumnExtremes()': table.cc:778: warning: 'wMin' may be used uninitialized in this function table.cc: In member function 'void dw::Table::forceCalcCellSizes()': table.cc:483: warning: 'totalWidth' may be used uninitialized in this function Cheers, Johannes

March 3, 2008

patch: Re: [Dillo-dev] segfault in Html_submit_form2()

by place＠gobigwest.com

Jorge wrote: > The question is, do you want me to commit the API change too? It doesn't matter to me.

March 3, 2008

segfault in Html_submit_form2()

by jcid＠dillo.org

On Mon, Mar 03, 2008 at 05:48:47PM +0000, place wrote: > Johannes wrote: > > It was crashing on every form submit I tried (e.g. www.google.com) > > Hmm. Too bad I'm still on old Linux. > (I looked at DragonflyBSD some months ago, by the way, and was disappointed > that I couldn't install from a file on the hard drive. I don't even have a > CD writer.) > > > I also got a compiler warning: > > > > html.cc:3945: warning: 'boundary' may be used uninitialized in this function > > I wasn't getting a warning, but this brings up something I've been meaning > to ask: Jorge, what do you think of -Werror? So far as I can recall, the > only warning on the dillo2/ side currently is Html_input_image. > It would save me the trouble of monitoring compilation... I use: make|grep : quite handy! -- Cheers Jorge.-

March 3, 2008

segfault in Html_submit_form2()

by jcid＠dillo.org

On Mon, Mar 03, 2008 at 06:14:29PM +0100, Johannes Hofmann wrote: > Hi Jorge, > > On Mon, Mar 03, 2008 at 02:04:25PM -0300, Jorge Arellano Cid wrote: > > Hi Johannes, > > > > On Mon, Mar 03, 2008 at 05:16:51PM +0100, Johannes Hofmann wrote: > > > Hi, > > > > > > after updating to latest CVS, I got the following segfault when > > > submitting a simple form: > > > > > [...] > > > I think the problem is, that boundary only gets initialized if > > > form->enc == DILLO_HTML_ENC_MULTIPART. > > > > It doesn't segfault for me. Anyway I committed a small fix. > > Please test and post the URL that was triggering the bug. > > Yes, that fixes it. Good! > It was crashing on every form submit I tried (e.g. www.google.com) > I also got a compiler warning: > > html.cc:3945: warning: 'boundary' may be used uninitialized in this function Funny how it worked OK here for google, linuxpackages and our bug track engine. Even more, I don't get a compiler warning with: gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) -- Cheers Jorge.-

March 3, 2008

segfault in Html_submit_form2()

by place＠gobigwest.com

Johannes wrote: > It was crashing on every form submit I tried (e.g. www.google.com) Hmm. Too bad I'm still on old Linux. (I looked at DragonflyBSD some months ago, by the way, and was disappointed that I couldn't install from a file on the hard drive. I don't even have a CD writer.) > I also got a compiler warning: > > html.cc:3945: warning: 'boundary' may be used uninitialized in this function I wasn't getting a warning, but this brings up something I've been meaning to ask: Jorge, what do you think of -Werror? So far as I can recall, the only warning on the dillo2/ side currently is Html_input_image. It would save me the trouble of monitoring compilation...

March 3, 2008

patch: Re: [Dillo-dev] segfault in Html_submit_form2()

by jcid＠dillo.org

Hi, On Mon, Mar 03, 2008 at 05:28:36PM +0000, place wrote: > > after updating to latest CVS, I got the following segfault when > > submitting a simple form: > > Oops! Thanks, Johannes. Patch attached. > I had the code doing the right thing at some point in the past after it > crashed once on me, but then I forgot why I'd made the change and undid it :( FWIW, this patch changes the API of Html_append_input: - Html_append_input(DilloHtmlEnc encoding, const char *boundary, Dstr *url, + Html_append_input(DilloHtmlEnc encoding, Dstr *boundary, Dstr *url, Current CVS has a different patch that fixes the segfault. The question is, do you want me to commit the API change too? -- Cheers Jorge.-

March 3, 2008

patch: Re: [Dillo-dev] segfault in Html_submit_form2()

by place＠gobigwest.com

> after updating to latest CVS, I got the following segfault when > submitting a simple form: Oops! Thanks, Johannes. Patch attached. I had the code doing the right thing at some point in the past after it crashed once on me, but then I forgot why I'd made the change and undid it :( What was the URL?

March 3, 2008

segfault in Html_submit_form2()

by Johannes.Hofmann＠gmx.de

Hi Jorge, On Mon, Mar 03, 2008 at 02:04:25PM -0300, Jorge Arellano Cid wrote: > Hi Johannes, > > On Mon, Mar 03, 2008 at 05:16:51PM +0100, Johannes Hofmann wrote: > > Hi, > > > > after updating to latest CVS, I got the following segfault when > > submitting a simple form: > > > [...] > > I think the problem is, that boundary only gets initialized if > > form->enc == DILLO_HTML_ENC_MULTIPART. > > It doesn't segfault for me. Anyway I committed a small fix. > Please test and post the URL that was triggering the bug. Yes, that fixes it. It was crashing on every form submit I tried (e.g. www.google.com) I also got a compiler warning: html.cc:3945: warning: 'boundary' may be used uninitialized in this function Cheers, Johannes > > TIA. > > -- > Cheers > Jorge.- > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

March 3, 2008

segfault in Html_submit_form2()

by jcid＠dillo.org

Hi Johannes, On Mon, Mar 03, 2008 at 05:16:51PM +0100, Johannes Hofmann wrote: > Hi, > > after updating to latest CVS, I got the following segfault when > submitting a simple form: > [...] > I think the problem is, that boundary only gets initialized if > form->enc == DILLO_HTML_ENC_MULTIPART. It doesn't segfault for me. Anyway I committed a small fix. Please test and post the URL that was triggering the bug. TIA. -- Cheers Jorge.-

March 3, 2008

segfault in Html_submit_form2()

by Johannes.Hofmann＠gmx.de

Hi, after updating to latest CVS, I got the following segfault when submitting a simple form: Program terminated with signal 11, Segmentation fault. #0 0x0805de09 in a_Html_form_event_handler (data=0x28794400, form_receiver=0x28835620, v_resource=0x28841000) at html.cc:4006 4006 input->name, val->str); (gdb) bt #0 0x0805de09 in a_Html_form_event_handler (data=0x28794400, form_receiver=0x28835620, v_resource=0x28841000) at html.cc:4006 #1 0x0805b301 in form::Form::activate (this=0x28835620, resource=0x28841000) at form.cc:100 #2 0x0808e6f7 in dw::core::ui::Resource::ActivateEmitter::emitToReceiver ( this=0x28841008, receiver=0x28835628, signalNo=0, argc=2, argv=0xbfbff15c) at ui.cc:101 #3 0x08093886 in loutsignal::Emitter::emitVoid (this=0x28841008, signalNo=0, argc=2, argv=0xbfbff15c) at signal.cc:81 #4 0x0808e727 in dw::core::ui::Resource::ActivateEmitter::emitActivate (this=0x28841008, resource=0x28841000) at ui.cc:109 #5 0x080aa945 in fltk::Input::handle_key () #6 0x080aadde in fltk::Input::handle () #7 0x080ab0d4 in fltk::Input::handle () #8 0x080ceddb in fltk::Widget::send () #9 0x080b67e4 in fltk::handle () #10 0x080b9bfb in fltk::handle () #11 0x080bb16b in do_queued_events () #12 0x080bb3ad in fltk::wait () #13 0x080bb55d in fltk::run () I think the problem is, that boundary only gets initialized if form->enc == DILLO_HTML_ENC_MULTIPART. Cheers, Johannes

March 3, 2008

dw::Table valgrind-friendliness

by jcid＠dillo.org

On Sun, Mar 02, 2008 at 09:23:53PM +0100, Johannes Hofmann wrote: > On Sun, Mar 02, 2008 at 03:45:09PM +0000, Jeremy Henty wrote: > > On Sun, Mar 02, 2008 at 04:14:00PM +0100, Johannes Hofmann wrote: > > > > > I think your proposed fix is fine. It's just that I don't like the > > > wrapper methods too much in the first place and now we need even > > > more code to support it... > > > > I'm not particularly attached to my proposed fix, my real concern is > > being valgrind clean because anything that makes valgrind harder to > > use is bad news. My patch was just the simplest way I could see to > > achieve that. If there's a better way to do it then that's great! > > > > ok, here is my proposal. It reduces the additional code in table.cc > a bit. Thanks to Jeremy for testing with valgrind. OK, committed. Please beware that this may cloak bugs. It was due to an assertion from the old code that the Table::draw() bug was found: http://lists.auriga.wearlab.de/pipermail/dillo-dev/2008-March/003837.html > Thanks to Jeremy for testing with valgrind. Ditto. -- Cheers Jorge.-

March 3, 2008

Re: patch: boundary for multipart/form-data submission with zero parts

by jcid＠dillo.org

Hi, On Sun, Mar 02, 2008 at 07:05:08PM +0000, place wrote: > Looking again at rfc 2046, I notice that an empty boundary string is not legal. > > As for what to do in the body in that case, it's conceivable that a lifetime > of digging through RFCs would yield a definitive answer for a weird > corner case, but it's far from being worth the trouble... > (Firefox sends a body consisting of an ending boundary line "--[boundary]--") Committed. -- Cheers Jorge.-

March 3, 2008

patch: multipart/form-data encoding for form submission

by jcid＠dillo.org

Hi, On Sun, Mar 02, 2008 at 09:20:53PM +0000, place wrote: > > For: > > > > > + // BUG: strstr stops at NULL. Cannot use for file inputs. > > > + success = !strstr(DataStr->str, boundary->str); > > > > please use memcmp(). > > How about if I make a separate patch for that at some point with > a Dstr_memmem() or something like that? OK. > > Some basic testing would be welcomed, but the main probe will > > be done by usage when committed. > > I added some comments, > changed the size of my dList_new() because the old value -- a holdover > from when I thought I'd do it differently -- could mislead the unwary, > and moved the dStr_free(boundary) down a little, > but this version should have no functional changes from the > version sent to the list yesterday. Committed. -- Cheers Jorge.-

March 3, 2008

dw::Table valgrind-friendliness

by Johannes.Hofmann＠gmx.de

On Sun, Mar 02, 2008 at 03:45:09PM +0000, Jeremy Henty wrote: > On Sun, Mar 02, 2008 at 04:14:00PM +0100, Johannes Hofmann wrote: > > > I think your proposed fix is fine. It's just that I don't like the > > wrapper methods too much in the first place and now we need even > > more code to support it... > > I'm not particularly attached to my proposed fix, my real concern is > being valgrind clean because anything that makes valgrind harder to > use is bad news. My patch was just the simplest way I could see to > achieve that. If there's a better way to do it then that's great! > ok, here is my proposal. It reduces the additional code in table.cc a bit. Thanks to Jeremy for testing with valgrind. Johannes diff -r 4582dc80e696 dw/table.cc --- a/dw/table.cc Sun Mar 02 13:40:23 2008 +0100 +++ b/dw/table.cc Sun Mar 02 21:21:17 2008 +0100 @@ -516,8 +516,8 @@ void Table::forceCalcCellSizes () printf(" totalWidth2 = %d curCol=%d\n", totalWidth,curCol); #endif - colWidths->setSize (numCols); - cumHeight->setSize (numRows + 1); + colWidths->setSize (numCols, 0); + cumHeight->setSize (numRows + 1, 0); rowSpanCells->setSize (0); baseline->setSize (numRows); #ifdef DBG @@ -857,7 +857,7 @@ void Table::apportion2 (int totalWidth, if (!maxAutoWidth) // no LEN_AUTO cols! return; - colWidths->setSize (colExtremes->size ()); + colWidths->setSize (colExtremes->size (), 0); if (!forceTotalWidth && maxAutoWidth < availAutoWidth) { // Enough space for the maximum table, don't widen past max. diff -r 4582dc80e696 lout/misc.hh --- a/lout/misc.hh Sun Mar 02 13:40:23 2008 +0100 +++ b/lout/misc.hh Sun Mar 02 21:21:17 2008 +0100 @@ -161,6 +161,18 @@ public: inline void setSize(int newSize) { this->num = newSize; this->resize (); } /** + * \brief Set the size explicitely and initialize new values. + * + * May be necessary before called before misc::SimpleVector::set. + */ + inline void setSize (int newSize, T t) { + int oldSize = this->num; + setSize (newSize); + for (int i = oldSize; i < newSize; i++) + set (i, t); + } + + /** * \brief Return the reference of one element. * * \sa misc::SimpleVector::get

March 2, 2008

dw::Table valgrind-friendliness

by onepoint＠starurchin.org

On Sun, Mar 02, 2008 at 03:45:57PM +0100, Johannes Hofmann wrote: > can you please send the valgrind log? I replied privately to you with it (since it's big enough to trigger list moderation). Jeremy Henty

March 2, 2008

dw::Table valgrind-friendliness

by Johannes.Hofmann＠gmx.de

On Sun, Mar 02, 2008 at 02:57:14PM +0000, Jeremy Henty wrote: > On Sun, Mar 02, 2008 at 02:43:56PM +0000, Jeremy Henty wrote: > > On Sun, Mar 02, 2008 at 11:31:16AM -0300, Jorge Arellano Cid wrote: > > > On Sun, Mar 02, 2008 at 02:15:39PM +0000, Jeremy Henty wrote: > > > > > > > > Dillo has been neglecting its valgrind cleanliness recently and > > > > it turns out the culprit is dw::Table, which resizes its > > > > internal vectors without initialising any new values that > > > > appear. > > > > > > AFAIU, this patch covers the symptoms of a BUG that current CVS > > > fixes. Please test current CVS against valgrind. > > > > No change. 121 "Conditional jump..." errors, reduced to 2 by my > > patch. > > To explain, these errors appear because the code updates the entries > of the cumHeight and colWidths vectors with wrapper methods > setCumHeight and setColWidth. These wrappers begin by comparing the > new value with old one and branching on the result, so you get the > "Conditional jump..." errors whenever the old value was not > initialized. That means these errors have nothing to do with any of > the Table algorithms, they're entirely due to the way the Table code > manages the cumHeight and colWidths vectors. These wrapper methods were introduced by my redraw patch. I will fix this. Johannes > > Jeremy Henty > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

March 2, 2008

dw::Table valgrind-friendliness

by onepoint＠starurchin.org

On Sun, Mar 02, 2008 at 02:43:56PM +0000, Jeremy Henty wrote: > On Sun, Mar 02, 2008 at 11:31:16AM -0300, Jorge Arellano Cid wrote: > > On Sun, Mar 02, 2008 at 02:15:39PM +0000, Jeremy Henty wrote: > > > > > > Dillo has been neglecting its valgrind cleanliness recently and > > > it turns out the culprit is dw::Table, which resizes its > > > internal vectors without initialising any new values that > > > appear. > > > > AFAIU, this patch covers the symptoms of a BUG that current CVS > > fixes. Please test current CVS against valgrind. > > No change. 121 "Conditional jump..." errors, reduced to 2 by my > patch. To explain, these errors appear because the code updates the entries of the cumHeight and colWidths vectors with wrapper methods setCumHeight and setColWidth. These wrappers begin by comparing the new value with old one and branching on the result, so you get the "Conditional jump..." errors whenever the old value was not initialized. That means these errors have nothing to do with any of the Table algorithms, they're entirely due to the way the Table code manages the cumHeight and colWidths vectors. Jeremy Henty

March 2, 2008

dw::Table valgrind-friendliness

by Johannes.Hofmann＠gmx.de

Hi Jeremy, can you please send the valgrind log? Johannes On Sun, Mar 02, 2008 at 02:43:56PM +0000, Jeremy Henty wrote: > On Sun, Mar 02, 2008 at 11:31:16AM -0300, Jorge Arellano Cid wrote: > > On Sun, Mar 02, 2008 at 02:15:39PM +0000, Jeremy Henty wrote: > > > > > > Dillo has been neglecting its valgrind cleanliness recently and it > > > turns out the culprit is dw::Table, which resizes its internal > > > vectors without initialising any new values that appear. > > > > AFAIU, this patch covers the symptoms of a BUG that current CVS > > fixes. Please test current CVS against valgrind. > > No change. 121 "Conditional jump..." errors, reduced to 2 by my > patch. > > Jeremy Henty > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

March 2, 2008

dw::Table valgrind-friendliness

by onepoint＠starurchin.org

On Sun, Mar 02, 2008 at 11:31:16AM -0300, Jorge Arellano Cid wrote: > On Sun, Mar 02, 2008 at 02:15:39PM +0000, Jeremy Henty wrote: > > > > Dillo has been neglecting its valgrind cleanliness recently and it > > turns out the culprit is dw::Table, which resizes its internal > > vectors without initialising any new values that appear. > > AFAIU, this patch covers the symptoms of a BUG that current CVS > fixes. Please test current CVS against valgrind. No change. 121 "Conditional jump..." errors, reduced to 2 by my patch. Jeremy Henty

March 2, 2008

Fixed a mistake in the CSS-box-model PNG image (style-box-model.png).

by jcid＠dillo.org

Hi, I just committed this fix. The wrong documentation image mislead me a lot while working on the redraw problems. Now it's accurate (CSS box model in namespacedw_1_1core_1_1style.html). -- Cheers Jorge.-

March 2, 2008