- Dillo-dev - lists.mailman3.com

Re: patch: base url page menu

by jcid＠dillo.org

On Wed, Nov 14, 2007 at 11:18:17PM +0000, place wrote: > Jorge wrote: > > The "Load images" via page menu patch was committed. > > Are you seeing the menus popping up with the load images option > under the pointer? What I get looks suspiciously like the previous > option clicked plus the height of the control panel. > *resizes control panel* > Yes, _very_ suspiciously. I'm experiencing the same as you. With the offset being exactly the panel height. No problem at all when the whole panel is hidden. It's possible to workaround by adding an offset inside the a_Menu_*_popup functions, but probably the proper solution is to get the event in window coordinates instead of viewport-based ones. AFAIU the popup offset is OK with pure-fltk2 widgets, so maybe there's a problem in how FLTK2 perceives the viewport or the panel... I haven't investigated further along these lines though... -- Cheers Jorge.-

Nov. 17, 2007

[patch] avoid complete redraws for child widget updates

by jcid＠dillo.org

Hi, On Fri, Nov 16, 2007 at 04:19:42PM +0100, Johannes Hofmann wrote: > Hello, > > here comes a patch that avoids complete redrawing if only a child > widget (button or form input) requests redrawing. Committed. > I don't really like the fact, that fltk widgets are sometimes drawn > via Group::draw() and sometimes via theLayout->expose(), but this is > already the case without my patch, so I just leave it as it is for > now. Yes. AFAIR we talked about this with Sebastian. IIRC he also didn't like much the mixed approach, and if he found a simpler way to do it, it'd be changed. The complex button stuff, and the fltk embedding could be avoided by re-implementing those in dw, but it looks like overkill. > In the long run, I would propose to add a mechanism similar to > fltk DAMAGE_CHILD to dw. If you find a simpler/better way, it's more than welcomed. -- Cheers Jorge.-

Nov. 17, 2007

[PATCH]: fix occasional segfault in dw::Table::reallocChildren

by jcid＠dillo.org

Hi Jeremy, On Fri, Nov 16, 2007 at 03:02:37PM +0000, Jeremy Henty wrote: > > This patch fixes an occasional segfault when adding columns to a table > that contains cells that span multiple rows or columns. This was > reported to the list last month by Matthias Franz. I've been bitten > by this on several web sites (see the details below), so it's a useful > fix. Yes it is. Committed! BTW, thanks for the excellent explanation. > Analysis of the bug: > > A dillo table stores its children in a single vector that is the > concatenation of the rows. This means that when the number of columns > increases the contents of the vector must be rearranged. The code > does this in place, starting at the end and working backwards. When > the code reaches a point that corresponds to a multi-row/column cell, > it finds a placeholder that stores the row/column of the actual cell. > > The important point is that the actual cell is stored in the vector > *before* any placeholders that refer to it. Therefore the cell has > not yet been moved when the code reaches the placeholder (because it > works backwards from the end). The code must calculate the index of > the actual cell from the row/column in the placeholder based on the > *old* number of columns. > > Unfortunately the code uses the *new* number of columns, so it may > retrieve a cell from the wrong place. Sometimes it retrieves a NULL > and segfaults when it tries to dereference it. > > > Test cases: > > <html><head></head><body><table><tbody> > <tr><td colspan="2"></td></tr> > <tr><td rowspan="2"></td></tr> > <tr><td colspan="2"></td></tr> > </tbody></table></body></html> > > [NB: the next one is from Matthias Franz.] > > <html> > <table> > <tr></tr> > <tr><td colspan="3"></td></tr> > <td colspan="3"> > </table> > </html> > > > Online pages that trigger the bug: > > <a href="http://legacyeditorial.gettyimages.com/source/search/details_pop.aspx?iid=7…"> > <a href="http://www.eweek.com/article2/0,1759,2216435,00.asp"> > <a href="http://www.possible-area.org/bad_timing161107.php"> > <a href="http://www.eweek.com/article2/0,1895,2216978,00.asp"> > <a href="http://www.nostalgia.com/nf_moreinfo.html?sku=10576"> -- Cheers Jorge.-

Nov. 17, 2007

patch: text search

by place＠gobigwest.com

> On Fri, Nov 16, 2007 at 08:12:17PM +0000, place wrote: > > Johannes wrote: > > > On Fri, Nov 16, 2007 at 07:10:48PM +0000, place wrote: > > > > Johannes wrote: > > > > > Very nice! I was text search a lot. > > > > > What about incremental searching as you type? > > > > > I don't mind the title bar... > > > > > > > > Hmm... I'd have to modify findtext.cc some... > > > > I may get around to it at some point here, though I'm > > > > feeling the need to look into what's going on with > > > > redirection and gzip, so if anybody else gets there > > > > first, it'll be fine with me! :) > > > > > > I was thinking about something like: > > > > > > > > > diff -r f18e37a8e3ac src/dialog.cc > > > --- a/src/dialog.cc Fri Nov 16 20:25:11 2007 +0100 > > > +++ b/src/dialog.cc Fri Nov 16 20:27:18 2007 +0100 > > > @@ -141,14 +141,6 @@ static void findtext_search_cb(Widget *, > > > const char *key = tf->i->value(); > > > bool case_sens = tf->cb->value(); > > > > > > - /* > > > - * Somehow fltk even regards the first loss of focus for the > > > - * window as a WHEN_ENTER_KEY_ALWAYS event. > > > - */ > > > - int e = event_key(); > > > - if ((e != ReturnKey) && (e != LeftButton)) > > > - return; > > > - > > > if (key[0] != '\0') > > > a_UIcmd_findtext_search(tf->bw, key, case_sens); > > > > > > @@ -176,7 +168,7 @@ TextFinder::TextFinder(int ww, int wh, B > > > > > > begin(); > > > i = new Input(0, 0, ww, ih); > > > - i->when(WHEN_ENTER_KEY_ALWAYS); > > > + i->when(WHEN_CHANGED); > > > i->callback(findtext_search_cb, this); > > > > > > cb = new CheckButton(0, ih, ww, wh-ih-bh, "Case-sensitive"); > > > > > > > > > on top of your patch of course. Works for me. > > > > I had tried that before replying earlier, and I was happy at > > first, but then after I hit Enter a few times on the final > > string, I was disappointed when I found backspace made the > > search go to the top of the page again. To me it felt like it > > should've been relative to where I found myself the last time > > I hit enter. I don't make much use of anything with incremental > > search, though, so I don't know what the standard behavior is. > > Though I think the current behaviour is consistent, I understand > your concern. Firefox never goes back when you shorten the > search string even if you have not yet hit enter. But I think that's > bad if you just mistyped the search string and want to correct it. That's true. I guess it just caught me by surprise.

Nov. 16, 2007

patch: text search

by Johannes.Hofmann＠gmx.de

On Fri, Nov 16, 2007 at 08:12:17PM +0000, place wrote: > Johannes wrote: > > On Fri, Nov 16, 2007 at 07:10:48PM +0000, place wrote: > > > Johannes wrote: > > > > Very nice! I was text search a lot. > > > > What about incremental searching as you type? > > > > I don't mind the title bar... > > > > > > Hmm... I'd have to modify findtext.cc some... > > > I may get around to it at some point here, though I'm > > > feeling the need to look into what's going on with > > > redirection and gzip, so if anybody else gets there > > > first, it'll be fine with me! :) > > > > I was thinking about something like: > > > > > > diff -r f18e37a8e3ac src/dialog.cc > > --- a/src/dialog.cc Fri Nov 16 20:25:11 2007 +0100 > > +++ b/src/dialog.cc Fri Nov 16 20:27:18 2007 +0100 > > @@ -141,14 +141,6 @@ static void findtext_search_cb(Widget *, > > const char *key = tf->i->value(); > > bool case_sens = tf->cb->value(); > > > > - /* > > - * Somehow fltk even regards the first loss of focus for the > > - * window as a WHEN_ENTER_KEY_ALWAYS event. > > - */ > > - int e = event_key(); > > - if ((e != ReturnKey) && (e != LeftButton)) > > - return; > > - > > if (key[0] != '\0') > > a_UIcmd_findtext_search(tf->bw, key, case_sens); > > > > @@ -176,7 +168,7 @@ TextFinder::TextFinder(int ww, int wh, B > > > > begin(); > > i = new Input(0, 0, ww, ih); > > - i->when(WHEN_ENTER_KEY_ALWAYS); > > + i->when(WHEN_CHANGED); > > i->callback(findtext_search_cb, this); > > > > cb = new CheckButton(0, ih, ww, wh-ih-bh, "Case-sensitive"); > > > > > > on top of your patch of course. Works for me. > > I had tried that before replying earlier, and I was happy at > first, but then after I hit Enter a few times on the final > string, I was disappointed when I found backspace made the > search go to the top of the page again. To me it felt like it > should've been relative to where I found myself the last time > I hit enter. I don't make much use of anything with incremental > search, though, so I don't know what the standard behavior is. Though I think the current behaviour is consistent, I understand your concern. Firefox never goes back when you shorten the search string even if you have not yet hit enter. But I think that's bad if you just mistyped the search string and want to correct it. Johannes > > > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

Nov. 16, 2007

patch: text search

by place＠gobigwest.com

Johannes wrote: > On Fri, Nov 16, 2007 at 07:10:48PM +0000, place wrote: > > Johannes wrote: > > > Very nice! I was text search a lot. > > > What about incremental searching as you type? > > > I don't mind the title bar... > > > > Hmm... I'd have to modify findtext.cc some... > > I may get around to it at some point here, though I'm > > feeling the need to look into what's going on with > > redirection and gzip, so if anybody else gets there > > first, it'll be fine with me! :) > > I was thinking about something like: > > > diff -r f18e37a8e3ac src/dialog.cc > --- a/src/dialog.cc Fri Nov 16 20:25:11 2007 +0100 > +++ b/src/dialog.cc Fri Nov 16 20:27:18 2007 +0100 > @@ -141,14 +141,6 @@ static void findtext_search_cb(Widget *, > const char *key = tf->i->value(); > bool case_sens = tf->cb->value(); > > - /* > - * Somehow fltk even regards the first loss of focus for the > - * window as a WHEN_ENTER_KEY_ALWAYS event. > - */ > - int e = event_key(); > - if ((e != ReturnKey) && (e != LeftButton)) > - return; > - > if (key[0] != '\0') > a_UIcmd_findtext_search(tf->bw, key, case_sens); > > @@ -176,7 +168,7 @@ TextFinder::TextFinder(int ww, int wh, B > > begin(); > i = new Input(0, 0, ww, ih); > - i->when(WHEN_ENTER_KEY_ALWAYS); > + i->when(WHEN_CHANGED); > i->callback(findtext_search_cb, this); > > cb = new CheckButton(0, ih, ww, wh-ih-bh, "Case-sensitive"); > > > on top of your patch of course. Works for me. I had tried that before replying earlier, and I was happy at first, but then after I hit Enter a few times on the final string, I was disappointed when I found backspace made the search go to the top of the page again. To me it felt like it should've been relative to where I found myself the last time I hit enter. I don't make much use of anything with incremental search, though, so I don't know what the standard behavior is.

Nov. 16, 2007

patch: text search

by Johannes.Hofmann＠gmx.de

On Fri, Nov 16, 2007 at 07:10:48PM +0000, place wrote: > Johannes wrote: > > Very nice! I was text search a lot. > > What about incremental searching as you type? > > I don't mind the title bar... > > Hmm... I'd have to modify findtext.cc some... > I may get around to it at some point here, though I'm > feeling the need to look into what's going on with > redirection and gzip, so if anybody else gets there > first, it'll be fine with me! :) I was thinking about something like: diff -r f18e37a8e3ac src/dialog.cc --- a/src/dialog.cc Fri Nov 16 20:25:11 2007 +0100 +++ b/src/dialog.cc Fri Nov 16 20:27:18 2007 +0100 @@ -141,14 +141,6 @@ static void findtext_search_cb(Widget *, const char *key = tf->i->value(); bool case_sens = tf->cb->value(); - /* - * Somehow fltk even regards the first loss of focus for the - * window as a WHEN_ENTER_KEY_ALWAYS event. - */ - int e = event_key(); - if ((e != ReturnKey) && (e != LeftButton)) - return; - if (key[0] != '\0') a_UIcmd_findtext_search(tf->bw, key, case_sens); @@ -176,7 +168,7 @@ TextFinder::TextFinder(int ww, int wh, B begin(); i = new Input(0, 0, ww, ih); - i->when(WHEN_ENTER_KEY_ALWAYS); + i->when(WHEN_CHANGED); i->callback(findtext_search_cb, this); cb = new CheckButton(0, ih, ww, wh-ih-bh, "Case-sensitive"); on top of your patch of course. Works for me. Cheers, Johannes

Nov. 16, 2007

patch: text search

by place＠gobigwest.com

Johannes wrote: > Very nice! I was text search a lot. > What about incremental searching as you type? > I don't mind the title bar... Hmm... I'd have to modify findtext.cc some... I may get around to it at some point here, though I'm feeling the need to look into what's going on with redirection and gzip, so if anybody else gets there first, it'll be fine with me! :) If nothing else, it would be nice to change the behavior so that changing the search key doesn't restart the search from the top of the document.

Nov. 16, 2007

patch: text search

by Johannes.Hofmann＠gmx.de

Very nice! I was text search a lot. What about incremental searching as you type? I don't mind the title bar... Cheers, Johannes On Fri, Nov 16, 2007 at 04:22:01AM +0000, place wrote: > I'm not much of a drawing/gui person, but it's hard not being > able to search pages, so I gave it a try (attached). > > - If you hit ^Q on this window or the page source or page > bugs windows, dillo will crash because it assumes it's > closing a bw. > > - I didn't want a title bar, and I didn't want it to be able > to outlive its bw, but I didn't know what to do about it. > > - If searching scrolls the window, highlighting isn't undone > properly. (If you search for something like "a", the last > instance on each line will remain highlighted) > > > *plays with it a bit more* > > Maybe it's good to have a title bar after all. It's easier to > move out of the way than the 0.8.x one was... > > diff -pur dillo2/src/dialog.cc dillo2-cur/src/dialog.cc > --- dillo2/src/dialog.cc 2007-11-07 05:03:53.000000000 +0000 > +++ dillo2-cur/src/dialog.cc 2007-11-16 03:30:20.000000000 +0000 > @@ -11,22 +11,26 @@ > > // UI dialogs > > +#include <fltk/events.h> // for the unfortunate (temporary?) event key testing > #include <fltk/Window.h> > #include <fltk/ask.h> > #include <fltk/file_chooser.h> > #include <fltk/TextBuffer.h> > +#include <fltk/Input.h> > +#include <fltk/CheckButton.h> > #include <fltk/ReturnButton.h> > #include <fltk/TextDisplay.h> > > #include "dialog.hh" > #include "misc.h" > +#include "uicmd.hh" > > using namespace fltk; > > /* > - * Callback for the text window dialog. > + * Close dialog window. > */ > -void text_window_cb(Widget *, void *vwin) > +static void window_close_cb(Widget *, void *vwin) > { > Window *window = (Window*)vwin; > > @@ -107,10 +111,91 @@ void a_Dialog_text_window(const char *tx > td->buffer(text_buf); > > ReturnButton *b = new ReturnButton (0, wh-bh, ww, bh, "Close"); > - b->callback(text_window_cb, window); > + b->callback(window_close_cb, window); > > window->resizable(window); > window->end(); > window->show(); > } > > +/* > + * Dialog to find text in page. > + */ > +class TextFinder : public Window { > +public: > + TextFinder(int ww, int wh, BrowserWindow *bw); > + BrowserWindow *bw; > + Input *i; > + CheckButton *cb; > + ReturnButton *findb; > + Button *clrb; > + Button *clsb; > +}; > + > +/* > + * Find next occurrence of input key > + */ > +static void findtext_search_cb(Widget *, void *vtf) > +{ > + TextFinder *tf = (TextFinder *)vtf; > + const char *key = tf->i->value(); > + bool case_sens = tf->cb->value(); > + > + /* > + * Somehow fltk even regards the first loss of focus for the > + * window as a WHEN_ENTER_KEY_ALWAYS event. > + */ > + int e = event_key(); > + if ((e != ReturnKey) && (e != LeftButton)) > + return; > + > + if (key[0] != '\0') > + a_UIcmd_findtext_search(tf->bw, key, case_sens); > + > +} > + > +/* > + * Reset search state > + */ > +static void findtext_clear_cb(Widget *, void *vtf) > +{ > + TextFinder *tf = (TextFinder *)vtf; > + tf->i->value(""); > + a_UIcmd_findtext_reset(tf->bw); > +} > + > +/* > + * Construct text search window > + */ > +TextFinder::TextFinder(int ww, int wh, BrowserWindow *bw) : > + Window(ww, wh, "unwanted title") > +{ > + int button_width = 70, ih = 35, bh = 30, gap = 10; > + > + this->bw = bw; > + > + begin(); > + i = new Input(0, 0, ww, ih); > + i->when(WHEN_ENTER_KEY_ALWAYS); > + i->callback(findtext_search_cb, this); > + > + cb = new CheckButton(0, ih, ww, wh-ih-bh, "Case-sensitive"); > + > + findb = new ReturnButton(gap, wh-bh, button_width, bh, "Find"); > + findb->callback(findtext_search_cb, this); > + > + clrb = new Button(button_width+2*gap, wh-bh, button_width, bh, "Clear"); > + clrb->callback(findtext_clear_cb, this); > + > + clsb = new Button(2*button_width+3*gap, wh-bh, button_width, bh, "Close"); > + clsb->callback(window_close_cb, this); > + end(); > + > + hotspot(i); // place input widget beneath the cursor > +} > + > +void a_Dialog_findtext(BrowserWindow *bw) > +{ > + TextFinder *tf = new TextFinder(250, 90, bw); > + tf->show(); > +} > diff -pur dillo2/src/dialog.hh dillo2-cur/src/dialog.hh > --- dillo2/src/dialog.hh 2007-11-08 19:50:20.000000000 +0000 > +++ dillo2-cur/src/dialog.hh 2007-11-16 03:00:01.000000000 +0000 > @@ -1,6 +1,8 @@ > #ifndef __DIALOG_HH__ > #define __DIALOG_HH__ > > +#include "bw.h" > + > #ifdef __cplusplus > extern "C" { > #endif /* __cplusplus */ > @@ -14,7 +16,7 @@ const char *a_Dialog_save_file(const cha > char *a_Dialog_open_file(const char *msg, > const char *pattern, const char *fname); > void a_Dialog_text_window(const char *txt, const char *title); > - > +void a_Dialog_findtext(BrowserWindow *bw); > > #ifdef __cplusplus > } > diff -pur dillo2/src/menu.cc dillo2-cur/src/menu.cc > --- dillo2/src/menu.cc 2007-11-14 22:07:49.000000000 +0000 > +++ dillo2-cur/src/menu.cc 2007-11-16 02:09:24.000000000 +0000 > @@ -115,7 +115,8 @@ static void Menu_add_bookmark_cb(Widget* > */ > static void Menu_find_text_cb(Widget* ) > { > - a_UIcmd_fullscreen_toggle(popup_bw); > +// a_UIcmd_fullscreen_toggle(popup_bw); > + a_UIcmd_findtext_dialog(popup_bw); > } > > /* > diff -pur dillo2/src/uicmd.cc dillo2-cur/src/uicmd.cc > --- dillo2/src/uicmd.cc 2007-11-14 22:07:49.000000000 +0000 > +++ dillo2-cur/src/uicmd.cc 2007-11-16 03:03:19.000000000 +0000 > @@ -696,3 +696,43 @@ void a_UIcmd_fullscreen_toggle(BrowserWi > BW2UI(bw)->fullscreen_toggle(); > } > > +/* > + * Open the text search dialog. > + */ > +void a_UIcmd_findtext_dialog(BrowserWindow *bw) > +{ > + a_Dialog_findtext(bw); > +} > + > +/* > + * Search for next occurrence of key. > + */ > +void a_UIcmd_findtext_search(BrowserWindow *bw, const char *key, int case_sens) > +{ > + Layout *l = (Layout *)bw->render_layout; > + > + switch (l->search(key, case_sens)) { > + case FindtextState::RESTART: > + a_UIcmd_set_msg(bw, "No further occurrences of \"%s\". " > + "Restarting from the top.", key); > + break; > + case FindtextState::NOT_FOUND: > + a_UIcmd_set_msg(bw, "\"%s\" not found.", key); > + break; > + case FindtextState::SUCCESS: > + default: > + a_UIcmd_set_msg(bw, ""); > + } > +} > + > +/* > + * Reset text search state. > + */ > +void a_UIcmd_findtext_reset(BrowserWindow *bw) > +{ > + Layout *l = (Layout *)bw->render_layout; > + l->resetSearch(); > + > + a_UIcmd_set_msg(bw, ""); > +} > + > diff -pur dillo2/src/uicmd.hh dillo2-cur/src/uicmd.hh > --- dillo2/src/uicmd.hh 2007-11-14 22:07:49.000000000 +0000 > +++ dillo2-cur/src/uicmd.hh 2007-11-16 02:37:32.000000000 +0000 > @@ -27,6 +27,9 @@ void a_UIcmd_search_dialog(void *vbw); > void a_UIcmd_book(void *vbw); > void a_UIcmd_add_bookmark(BrowserWindow *bw, const DilloUrl *url); > void a_UIcmd_fullscreen_toggle(BrowserWindow *bw); > +void a_UIcmd_findtext_dialog(BrowserWindow *bw); > +void a_UIcmd_findtext_search(BrowserWindow *bw, const char *key, int case_sens); > +void a_UIcmd_findtext_reset(BrowserWindow *bw); > void a_UIcmd_page_popup(void *vbw, const DilloUrl *url, > const char *bugs_txt, int prefs_load_images); > void a_UIcmd_link_popup(void *vbw, const DilloUrl *url); > _______________________________________________ > Dillo-dev mailing list > Dillo-dev(a)dillo.org > http://lists.auriga.wearlab.de/cgi-bin/mailman/listinfo/dillo-dev

Nov. 16, 2007

[patch] avoid complete redraws for child widget updates

by Johannes.Hofmann＠gmx.de

Hello, here comes a patch that avoids complete redrawing if only a child widget (button or form input) requests redrawing. I don't really like the fact, that fltk widgets are sometimes drawn via Group::draw() and sometimes via theLayout->expose(), but this is already the case without my patch, so I just leave it as it is for now. In the long run, I would propose to add a mechanism similar to fltk DAMAGE_CHILD to dw. To test go to google and type in some text. Cheers, Johannes

Nov. 16, 2007