May 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: [BOGUS PATCH] Re: bug setting scrollbar height w/ X resources? 26 years, 6 months

Hrvoje Niksic

"Kirill M. Katsnelson" <kkm(a)kis.ru> writes: > Some time ago, Hrvoje Niksic wrote... > |+ > | "Kirill M. Katsnelson" <kkm(a)kis.ru> writes: > | > Huh, me too... What a fun! > | > | First patch wins! :-) > | > | I first enjoyed crashes, now I fixed things up so that the new > | specifiers "only" don't work. > > Ok, my change eliminates the need for all those foo_changed_in_frame > functions, which call boring device methods... So, hopefully, there > will be no need for "scrollbar visible p changed in frame" device > method, unless I prove my desing fundamentally flawed. :) You sound like you know what you're doing, so I leave the floor to you. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- Thou Who might be our Father Who perhaps may be in Heaven... -- Roger Zelazny

2 participants
1 comments

{vertical,horizontal}-scrollbar-visible-p progris riport 26 years, 6 months

Hrvoje Niksic

Here is what I've done so far. The idea was to create a general x_scrollbar_size_changed_in_frame_1() function, and make it do all the usual stunts, except for one: when scrollbar is supposed to be invisible, pretend that the size is 0. >From the user's perspective, this means that setting vertical-scrollbar-visible-p to nil turns it off. However, (specifier-instance scrollbar-width) would still return 15 or whatever. For horizontal scrollbar, having horizontal-scrollbar-visible-p to nil means that the scrollbar is not there. Having it t means that the scrollbar *might* be there, depending on truncate-lines and whatnot. However, for some reason, this is not working. I assume that the problem is with the geometry calculations. I'm not sure. Anyway, here is the patch I've come up with. I'll keep hacking on this, but if someone has an idea, I'll appreciate it... 1998-05-04 Hrvoje Niksic <hniksic(a)srce.hr> * scrollbar.c (vertical_scrollbar_visible_p_changed_in_frame): New function. (horizontal_scrollbar_visible_p_changed_in_frame): Ditto. * window.h (struct window): Ditto. * frameslots.h: New slots. * scrollbar.c (specifier_vars_of_scrollbar): New specifiers horizontal-scrollbar-visible-p and vertical-scrollbar-visible-p. * scrollbar-x.c (x_scrollbar_size_changed_in_frame_1): New function. (x_scrollbar_width_changed_in_frame): Use it. (x_scrollbar_height_changed_in_frame): Ditto. * toolbar-x.c (x_toolbar_size_changed_in_frame_1): Use the cached value instead of checking through Fspecifier_instance. --- src/scrollbar.c.orig Mon May 4 03:04:59 1998 +++ src/scrollbar.c Mon May 4 04:16:39 1998 ＠＠ -63,6 +63,8 ＠＠ /* Height of the scrollbar. */ Lisp_Object Vscrollbar_height; +Lisp_Object Vhorizontal_scrollbar_visible_p, Vvertical_scrollbar_visible_p; + Lisp_Object Vscrollbar_pointer_glyph; EXFUN (Fcenter_to_window_line, 2); ＠＠ -603,6 +605,28 ＠＠ } /* This function is called as a result of a change to the + `vertical-scrollbar-visible-p' specifier. */ +static void +vertical_scrollbar_visible_p_changed_in_frame (Lisp_Object specifier, + struct frame *f, + Lisp_Object oldval) +{ + MAYBE_FRAMEMETH (f, vertical_scrollbar_visible_p_changed_in_frame, + (specifier, f, oldval)); +} + +/* This function is called as a result of a change to the + `horizontal-scrollbar-visible-p' specifier. */ +static void +horizontal_scrollbar_visible_p_changed_in_frame (Lisp_Object specifier, + struct frame *f, + Lisp_Object oldval) +{ + MAYBE_FRAMEMETH (f, horizontal_scrollbar_visible_p_changed_in_frame, + (specifier, f, oldval)); +} + +/* This function is called as a result of a change to the `scrollbar-pointer' glyph. */ static void scrollbar_pointer_changed_in_window (Lisp_Object specifier, struct window *w, ＠＠ -963,6 +987,36 ＠＠ slot_offset (struct frame, scrollbar_height), scrollbar_height_changed_in_frame); + + DEFVAR_SPECIFIER ("vertical-scrollbar-visible-p", &Vvertical_scrollbar_visible_p /* +*Whether vertical scrollbar is visible. +This is a specifier; use `set-specifier' to change it. +*/ ); + Vvertical_scrollbar_visible_p = Fmake_specifier (Qboolean); + set_specifier_fallback (Vvertical_scrollbar_visible_p, + list1 (Fcons (Qnil, Qt))); + set_specifier_caching (Vvertical_scrollbar_visible_p, + slot_offset (struct window, + vertical_scrollbar_visible_p), + some_window_value_changed, + slot_offset (struct frame, + vertical_scrollbar_visible_p), + vertical_scrollbar_visible_p_changed_in_frame); + + DEFVAR_SPECIFIER ("horizontal-scrollbar-visible-p", &Vhorizontal_scrollbar_visible_p /* +*Whether horizontal scrollbar is visible. +This is a specifier; use `set-specifier' to change it. +*/ ); + Vhorizontal_scrollbar_visible_p = Fmake_specifier (Qboolean); + set_specifier_fallback (Vhorizontal_scrollbar_visible_p, + list1 (Fcons (Qnil, Qt))); + set_specifier_caching (Vhorizontal_scrollbar_visible_p, + slot_offset (struct window, + horizontal_scrollbar_visible_p), + some_window_value_changed, + slot_offset (struct frame, + horizontal_scrollbar_visible_p), + horizontal_scrollbar_visible_p_changed_in_frame); } void --- src/scrollbar-x.c.orig Mon May 4 04:05:56 1998 +++ src/scrollbar-x.c Tue May 5 02:48:14 1998 ＠＠ -294,16 +294,39 ＠＠ } } -/* A device method. */ +/* This shares code for changing scrollbar size and visibility. */ static void -x_scrollbar_width_changed_in_frame (Lisp_Object specifier, struct frame *f, - Lisp_Object oldval) +x_scrollbar_size_changed_in_frame_1 (struct frame *f, + int verticalp, int change_size_p, + Lisp_Object oldval) { - XtWidgetGeometry req, repl; - Lisp_Object newval = f->scrollbar_width; + Lisp_Object *sizeptr = (verticalp ? + &f->scrollbar_width : &f->scrollbar_height); + Lisp_Object newvisiblep = (verticalp ? + f->vertical_scrollbar_visible_p + : f->horizontal_scrollbar_visible_p); + Lisp_Object oldsize, newsize; + + /* If no change is to be done, do nothing. */ + if ((change_size_p && EQ (oldval, *sizeptr)) + || (!change_size_p && EQ (oldval, newvisiblep))) + return; in_specifier_change_function++; + if (NILP (newvisiblep)) + newsize = Qzero; + else if (!INTP (*sizeptr)) + /* Happens during initialization. */ + newsize = Qzero; + else + newsize = *sizeptr; + + if (change_size_p) + oldsize = oldval; + else + oldsize = NILP (newvisiblep) ? *sizeptr : Qzero; + /* We want the text area to stay the same size. So, we query the current size and then adjust it for the change in the scrollbar width. */ ＠＠ -311,20 +334,25 ＠＠ /* mirror the value in the frame resources, unless it was already done. */ if (!in_resource_setting) - Xt_SET_VALUE (FRAME_X_TEXT_WIDGET (f), XtNscrollBarWidth, XINT (newval)); + Xt_SET_VALUE (FRAME_X_TEXT_WIDGET (f), + verticalp ? XtNscrollBarWidth : XtNscrollBarHeight, + XINT (newsize)); if (XtIsRealized (FRAME_X_CONTAINER_WIDGET (f))) { + XtWidgetGeometry req, repl; req.request_mode = 0; /* the query-geometry method looks at the current value of f->scrollbar_width, so temporarily set it back to the old one. */ - f->scrollbar_width = oldval; + if (change_size_p) + *sizeptr = oldsize; XtQueryGeometry (FRAME_X_CONTAINER_WIDGET (f), &req, &repl); - f->scrollbar_width = newval; + if (change_size_p) + *sizeptr = newsize; - repl.width += XINT (newval) - XINT (oldval); + repl.width += XINT (newsize) - XINT (oldsize); EmacsManagerChangeSize (FRAME_X_CONTAINER_WIDGET (f), repl.width, repl.height); } ＠＠ -334,43 +362,36 ＠＠ /* A device method. */ static void +x_scrollbar_width_changed_in_frame (Lisp_Object specifier, struct frame *f, + Lisp_Object oldval) +{ + x_scrollbar_size_changed_in_frame_1 (f, 1, 1, oldval); +} + +/* A device method. */ +static void x_scrollbar_height_changed_in_frame (Lisp_Object specifier, struct frame *f, Lisp_Object oldval) { - XtWidgetGeometry req, repl; - Lisp_Object newval = f->scrollbar_height; - - in_specifier_change_function++; - - /* We want the text area to stay the same size. So, we query the - current size and then adjust it for the change in the scrollbar - height. */ - - /* mirror the value in the frame resources, unless it was already - done. Also don't do it if this is the when the frame is being - created -- the widgets don't even exist yet, and even if they - did, we wouldn't want to overwrite the resource information - (which might specify a user preference). */ - if (!in_resource_setting) - Xt_SET_VALUE (FRAME_X_TEXT_WIDGET (f), XtNscrollBarHeight, XINT (newval)); - - if (XtIsRealized (FRAME_X_CONTAINER_WIDGET (f))) - { - req.request_mode = 0; - - /* the query-geometry method looks at the current value of - f->scrollbar_height, so temporarily set it back to the old - one. */ - f->scrollbar_height = oldval; - XtQueryGeometry (FRAME_X_CONTAINER_WIDGET (f), &req, &repl); - f->scrollbar_height = newval; + x_scrollbar_size_changed_in_frame_1 (f, 0, 1, oldval); +} - repl.height += XINT (newval) - XINT (oldval); - EmacsManagerChangeSize (FRAME_X_CONTAINER_WIDGET (f), repl.width, - repl.height); - } +/* A device method. */ +static void +x_vertical_scrollbar_visible_p_changed_in_frame (Lisp_Object specifier, + struct frame *f, + Lisp_Object oldval) +{ + x_scrollbar_size_changed_in_frame_1 (f, 1, 0, oldval); +} - in_specifier_change_function--; +/* A device method. */ +static void +x_horizontal_scrollbar_visible_p_changed_in_frame (Lisp_Object specifier, + struct frame *f, + Lisp_Object oldval) +{ + x_scrollbar_size_changed_in_frame_1 (f, 0, 0, oldval); } enum x_scrollbar_loop ＠＠ -819,6 +840,8 ＠＠ CONSOLE_HAS_METHOD (x, update_scrollbar_instance_status); CONSOLE_HAS_METHOD (x, scrollbar_width_changed_in_frame); CONSOLE_HAS_METHOD (x, scrollbar_height_changed_in_frame); + CONSOLE_HAS_METHOD (x, vertical_scrollbar_visible_p_changed_in_frame); + CONSOLE_HAS_METHOD (x, horizontal_scrollbar_visible_p_changed_in_frame); CONSOLE_HAS_METHOD (x, scrollbar_pointer_changed_in_window); #ifdef MEMORY_USAGE_STATS CONSOLE_HAS_METHOD (x, compute_scrollbar_instance_usage); --- src/frameslots.h.orig Mon May 4 03:33:50 1998 +++ src/frameslots.h Mon May 4 03:34:14 1998 ＠＠ -99,9 +99,12 ＠＠ #endif #ifdef HAVE_SCROLLBARS - /* Width and height of the scrollbars. */ + /* Size and visibility of the scrollbars. */ MARKED_SLOT (scrollbar_width); MARKED_SLOT (scrollbar_height); + + MARKED_SLOT (vertical_scrollbar_visible_p); + MARKED_SLOT (horizontal_scrollbar_visible_p); #endif #ifdef HAVE_TOOLBARS --- src/window.h.orig Mon May 4 03:34:23 1998 +++ src/window.h Mon May 4 03:34:58 1998 ＠＠ -194,6 +194,9 ＠＠ Lisp_Object scrollbar_width; /* Height of horizontal scrollbars. */ Lisp_Object scrollbar_height; + /* Visibility of scrollbars. */ + Lisp_Object vertical_scrollbar_visible_p; + Lisp_Object horizontal_scrollbar_visible_p; /* Pointer to use for vertical and horizontal scrollbars. */ Lisp_Object scrollbar_pointer; #endif /* HAVE_SCROLLBARS */ -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- Try to use "ad nauseam" at least once per flame. It doesn't mean anything; but it gives that polished feel to your postings.

2 participants
1 comments

XEmacs 20.4: forward-comment spuriously returns t at end of buffer 26 years, 6 months

Michael Ernst

At the end of a buffer, (forward-comment 1) or (forward-comment 999) returns t. These forms should return nil: ... If N comments are found as expected, with nothing except whitespace between them, return t; otherwise return nil. ... FSF Emacs's version of forward-comment gets this right. Here's a workaround until a change makes its way into the C source: ;; In XEmacs 20.4, (forward-comment 9999) at the end of a buffer returns t! (defadvice forward-comment (after forward-comment-eob activate) (if (and ad-return-value (eobp)) (setq ad-return-value nil))) -Michael Ernst mernst(a)cs.washington.edu

1 participants
0 comments

Re: [BOGUS PATCH] Re: bug setting scrollbar height w/ X resources? 26 years, 6 months

Hrvoje Niksic

"Kirill M. Katsnelson" <kkm(a)kis.ru> writes: > Some time ago, Hrvoje Niksic wrote... > |+ > | Didier, note that I'm working on this too (this time for real). > |- > > Huh, me too... What a fun! First patch wins! :-) I first enjoyed crashes, now I fixed things up so that the new specifiers "only" don't work. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- I'm a Lisp variable -- bind me!

2 participants
1 comments

Adding sound into IBM 26 years, 6 months

Rick

Hello! Could i ask you a question? How can i add sound into IBM, because i made my webpage, and adding sound into it, i used<embed src=http://......sound.mid>, and it doesn't play on IBM. Could you send me an e-mail, and help me to do that My e-mail: bancan(a)hotmail.com

1 participants
0 comments

Re: FAILURE - b38 - linux 26 years, 6 months

SL Baur

Jeff Miller <jmiller(a)smart.net> writes: > using > ./configure '--prefix=/usr/local' '--cflags=' '--package-path=/usr/local/lib/xemacs/packages-20.5-2' > the compile failed with: > cc -lc -lgcc /usr/lib/crtn.o `gcc -print-libgcc-file-name` > rangetab.o: In function `mark_range_table': > rangetab.o(.text+0xc): undefined reference to `error_check_range_table' ... > make[1]: Leaving directory `/mnt4/jmiller/xemacs-20.5-b15/src' Jeff, I can easily imagine failure of this sort if the New And Improved(TM) Makefile dependencies are out of date and you've recently mixed builds with and without error-checking. As far as I can tell, I did remember to rebuild the dependencies between betas. Do you have any .rej files or zero length files from patching?

4 participants
5 comments

Re: DND for XEmacs 21 26 years, 6 months

Oliver Graf

Jonathan Harris <jhar(a)tardis.ed.ac.uk> writes: > > By the way: how does windos handle a drag (i.e. when is a press a drag)? > > It's up to the application to initiate a drag operation. The win32 user > interface guidlines recommend that the application should initiate a > drag operation when the user clicks and holds the left-button (with or > without modifiers) within selected text. Most applications also initiate > a drag when the right button (button3 in XEmacs) is used in the same > way. Ah, we're on our own... Do you know Xdnd (the protocol used by KDE)? It only uses MIME data in drags. If you drag some files, the mime type url/* is used, where * is some action verb (i.e. move or copy). > Because the state of the Ctrl modifier is examined at _drop_ time, but > affects whether the _dragged_ text should be deleted, dragging is a > two-stage operation and needs both a start-drag and a stop-drag > function. I pass the modifier into the misc-user event as you can see in event-Xt.c (x_event_to_emacs_event). So we can check this also... The new functions are in a file called dragdrop.el. A patch will follow soon (next two hours). > I imagine that you're aiming for the lisp-level dispatch functions to be > independant of the dnd system in use. Does the desired mswindows > behaviour that I've described sound consistent with that goal? The independent drop function is already here: dragdrop-drop-dispatch (also defined in dragdrop.el -- look into tests/Dnd/droptest.el for an example). Drag *should* work semilar. I will create a misc-user-event if the button is held down for a specified time (some ms). This will have dragdrop-drag-dispatch as function. This function can do what it wants with the thing, for example calling the right low-level-protocol function. > Actually, dnd for text uses a different and more complicated win32 API > than dnd for files. I don't plan to implement dnd for _text_ for XEmacs > 21.0. This means that for XEmacs 21.0, the mswindows code won't be able > to initiate drags and will only have to handle file drops. Then you should be able to complete the changes after I have send the next patch... But don't forget to convert filepaths to URLs ('file:/usr/tmp/test.txt' is converted to '/usr/tmp/test.txt' by the default dragdrop-drop-dispatch function). Regards, Oliver.

2 participants
1 comments

Re: [Success] XEmacs 21.0b37 Repartida on Linux (GLIBC) 26 years, 6 months

SL Baur

Oliver Graf <ograf(a)fga.de> writes: > Is the first patch I sent already merged in? No. > Should I sent an incremental patch based on the first, or the > complete diff between 21.0b38 and my current tree today? Either way is fine at this point, but if you do a complete diff, please be sure to identify the obsoleted patch. If the previous patch had been merged, a diff between the first patch and the next one would have been preferable. > Do I get a 'Yes, patch is ok and merged' when I have submitted a > patch to xemacs-patches (or submitted it to xemacs-reviewer)? 1/2 Yes. You'll get a "patch is merged, thanks" message when I apply the patch. You won't receive a further message unless something went badly wrong.

1 participants
0 comments

Re: [Success] XEmacs 21.0b37 Repartida on Linux (GLIBC) 26 years, 6 months

Oliver Graf

SL Baur <steve(a)xemacs.org> writes: > Oliver Graf <ograf(a)fga.de> writes: > > > Whoops, seems like I'm in some root account. > > > Advice: don't drink lot's of beer and still patch and compile XEmacs. > > :-) > > Do we need to add this to the installation notes or the FAQ? :-) Hmmm, I think both would be better ;-) Is the first patch I sent already merged in? Should I sent an incremental patch based on the first, or the complete diff between 21.0b38 and my current tree today? Do I get a 'Yes, patch is ok and merged' when I have submitted a patch to xemacs-patches (or submitted it to xemacs-reviewer)? Regards, Oliver.

1 participants
0 comments

Re: [Success] XEmacs 21.0b37 Repartida on Linux (GLIBC) 26 years, 6 months

SL Baur

Oliver Graf <ograf(a)fga.de> writes: > Whoops, seems like I'm in some root account. > Advice: don't drink lot's of beer and still patch and compile XEmacs. :-) Do we need to add this to the installation notes or the FAQ? :-)

1 participants
0 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta May 1998