mirror/emacs
1
0
Fork 0
mirror of https://git.savannah.gnu.org/git/emacs.git synced 2024-11-23 07:19:15 +00:00
Code Issues Releases Activity

; * src/xdisp.c: Improve the introductory commentary.

Browse Source
This commit is contained in:
Eli Zaretskii 2020-04-24 18:09:20 +03:00
parent a92ca1f177
commit 369761b36d
1 changed files with 176 additions and 95 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

271
src/xdisp.c
View File

@ -30,8 +30,9 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
Updating the display is triggered by the Lisp interpreter when it
decides it's time to do it. This is done either automatically for
you as part of the interpreter's command loop or as the result of
calling Lisp functions like `sit-for'. The C function `redisplay'
in xdisp.c is the only entry into the inner redisplay code.
calling Lisp functions like `sit-for'. The C function
`redisplay_internal' in xdisp.c is the only entry into the inner
redisplay code.
The following diagram shows how redisplay code is invoked. As you
can see, Lisp calls redisplay and vice versa.
@ -89,7 +90,15 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
second glyph matrix is constructed, the so called `desired glyph
matrix' or short `desired matrix'. Current and desired matrix are
then compared to find a cheap way to update the display, e.g. by
reusing part of the display by scrolling lines.
reusing part of the display by scrolling lines. The actual update
of the display of each window by comparing the desired and the
current matrix is done by `update_window', which calls functions
which draw to the glass (those functions are specific to the type
of the window's frame: X, w32, NS, etc.).
Once the display of a window on the glass has been updated, its
desired matrix is used to update the corresponding rows of the
current matrix, and then the desired matrix is discarded.
You will find a lot of redisplay optimizations when you start
looking at the innards of redisplay. The overall goal of all these
@ -119,13 +128,13 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
. try_window
This function performs the full redisplay of a single window
assuming that its fonts were not changed and that the cursor
will not end up in the scroll margins. (Loading fonts requires
re-adjustment of dimensions of glyph matrices, which makes this
method impossible to use.)
This function performs the full, unoptimized, redisplay of a
single window assuming that its fonts were not changed and that
the cursor will not end up in the scroll margins. (Loading
fonts requires re-adjustment of dimensions of glyph matrices,
which makes this method impossible to use.)
These optimizations are tried in sequence (some can be skipped if
The optimizations are tried in sequence (some can be skipped if
it is known that they are not applicable). If none of the
optimizations were successful, redisplay calls redisplay_windows,
which performs a full redisplay of all windows.
@ -145,38 +154,62 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
Desired matrices.
Desired matrices are always built per Emacs window. The function
`display_line' is the central function to look at if you are
interested. It constructs one row in a desired matrix given an
Desired matrices are always built per Emacs window. It is
important to know that a desired matrix is in general "sparse": it
only has some of the glyph rows "enabled". This is because
redisplay tries to optimize its work, and thus only generates
glyphs for rows that need to be updated on the screen. Rows that
don't need to be updated are left "disabled", and their contents
should be ignored.
The function `display_line' is the central function to look at if
you are interested in how the rows of the desired matrix are
produced. It constructs one row in a desired matrix given an
iterator structure containing both a buffer position and a
description of the environment in which the text is to be
displayed. But this is too early, read on.
Glyph rows.
A glyph row is an array of `struct glyph', where each glyph element
describes a "display element" to be shown on the screen. More
accurately, a glyph row can have up to 3 different arrays of
glyphs: one each for every display margins, and one for the "text
area", where buffer text is displayed. The text-area glyph array
is always present, whereas the arrays for the marginal areas are
present (non-empty) only if the corresponding display margin is
shown in the window. If the glyph array for a marginal area is not
present its beginning and end coincide, i.e. such arrays are
actually empty (they contain no glyphs). Frame glyph matrics, used
on text-mode terminals (see below) never have marginal areas, they
treat the entire frame-wide row of glyphs as a single large "text
area".
Iteration over buffer and strings.
Characters and pixmaps displayed for a range of buffer text depend
on various settings of buffers and windows, on overlays and text
properties, on display tables, on selective display. The good news
is that all this hairy stuff is hidden behind a small set of
interface functions taking an iterator structure (struct it)
interface functions taking an iterator structure (`struct it')
argument.
Iteration over things to be displayed is then simple. It is
started by initializing an iterator with a call to init_iterator,
started by initializing an iterator with a call to `init_iterator',
passing it the buffer position where to start iteration. For
iteration over strings, pass -1 as the position to init_iterator,
and call reseat_to_string when the string is ready, to initialize
iteration over strings, pass -1 as the position to `init_iterator',
and call `reseat_to_string' when the string is ready, to initialize
the iterator for that string. Thereafter, calls to
get_next_display_element fill the iterator structure with relevant
information about the next thing to display. Calls to
set_iterator_to_next move the iterator to the next thing.
`get_next_display_element' fill the iterator structure with
relevant information about the next thing to display. Calls to
`set_iterator_to_next' move the iterator to the next thing.
Besides this, an iterator also contains information about the
display environment in which glyphs for display elements are to be
produced. It has fields for the width and height of the display,
the information whether long lines are truncated or continued, a
current X and Y position, and lots of other stuff you can better
see in dispextern.h.
current X and Y position, the face currently in effect, and lots of
other stuff you can better see in dispextern.h.
The "stop position".
@ -184,57 +217,62 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
infrequently. These include the face of the characters, whether
text is invisible, the object (buffer or display or overlay string)
being iterated, character composition info, etc. For any given
buffer or string position, the sources of information that
affects the display can be determined by calling the appropriate
primitives, such as Fnext_single_property_change, but both these
buffer or string position, the sources of information that affects
the display can be determined by calling the appropriate
primitives, such as `Fnext_single_property_change', but both these
calls and the processing of their return values is relatively
expensive. To optimize redisplay, the display engine checks these
sources of display information only when needed. To that end, it
always maintains the position of the next place where it must stop
and re-examine all those potential sources. This is called "stop
position" and is stored in the stop_charpos field of the iterator.
The stop position is updated by compute_stop_pos, which is called
whenever the iteration reaches the current stop position and
processes it. Processing a stop position is done by handle_stop,
which invokes a series of handlers, one each for every potential
source of display-related information; see the it_props array for
those handlers. For example, one handler is handle_face_prop,
which detects changes in face properties, and supplies the face ID
that the iterator will use for all the glyphs it generates up to
the next stop position; this face ID is the result of realizing the
face specified by the relevant text properties at this position.
Each handler called by handle_stop processes the sources of display
sources of display information only when needed, not for every
character. To that end, it always maintains the position of the
next place where it must stop and re-examine all those potential
sources. This is called "the stop position" and is stored in the
`stop_charpos' field of the iterator. The stop position is updated
by `compute_stop_pos', which is called whenever the iteration
reaches the current stop position and processes it. Processing a
stop position is done by `handle_stop', which invokes a series of
handlers, one each for every potential source of display-related
information; see the `it_props' array for those handlers. For
example, one handler is `handle_face_prop', which detects changes
in face properties, and supplies the face ID that the iterator will
use for all the glyphs it generates up to the next stop position;
this face ID is the result of "realizing" the face specified by the
relevant text properties at this position (see xfaces.c). Each
handler called by `handle_stop' processes the sources of display
information for which it is "responsible", and returns a value
which tells handle_stop what to do next.
which tells `handle_stop' what to do next.
Once handle_stop returns, the information it stores in the iterator
fields will not be refreshed until the iteration reaches the next
stop position, which is computed by compute_stop_pos called at the
end of handle_stop. compute_stop_pos examines the buffer's or
string's interval tree to determine where the text properties
change, finds the next position where overlays and character
composition can change, and stores in stop_charpos the closest
position where any of these factors should be reconsidered.
Once `handle_stop' returns, the information it stores in the
iterator fields will not be refreshed until the iteration reaches
the next stop position, which is computed by `compute_stop_pos'
called at the end of `handle_stop'. `compute_stop_pos' examines
the buffer's or string's interval tree to determine where the text
properties change, finds the next position where overlays and
character composition can change, and stores in `stop_charpos' the
closest position where any of these factors should be reconsidered.
Handling of the stop position is done as part of the code in
`get_next_display_element'.
Producing glyphs.
Glyphs in a desired matrix are normally constructed in a loop
calling get_next_display_element and then PRODUCE_GLYPHS. The call
to PRODUCE_GLYPHS will fill the iterator structure with pixel
information about the element being displayed and at the same time
produce glyphs for it. If the display element fits on the line
being displayed, set_iterator_to_next is called next, otherwise the
glyphs produced are discarded. The function display_line is the
workhorse of filling glyph rows in the desired matrix with glyphs.
In addition to producing glyphs, it also handles line truncation
and continuation, word wrap, and cursor positioning (for the
latter, see also set_cursor_from_row).
calling `get_next_display_element' and then `PRODUCE_GLYPHS'. The
call to `PRODUCE_GLYPHS' will fill the iterator structure with
pixel information about the element being displayed and at the same
time will produce glyphs for it. If the display element fits on
the line being displayed, `set_iterator_to_next' is called next,
otherwise the glyphs produced are discarded, and `display_line'
marks this glyph row as a "continued line". The function
`display_line' is the workhorse of filling glyph rows in the
desired matrix with glyphs. In addition to producing glyphs, it
also handles line truncation and continuation, word wrap, and
cursor positioning (for the latter, see `set_cursor_from_row').
Frame matrices.
That just couldn't be all, could it? What about terminal types not
supporting operations on sub-windows of the screen (a.k.a. "TTY" or
"text-mode terminal")? To update the display on such a terminal,
"text-mode terminals")? To update the display on such a terminal,
window-based glyph matrices are not well suited. To be able to
reuse part of the display (scrolling lines up and down), we must
instead have a view of the whole screen. This is what `frame
@ -252,19 +290,62 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
using the frame matrices, which allows frame-global optimization of
what is actually written to the glass.
To be honest, there is a little bit more done, but not much more.
If you plan to extend that code, take a look at dispnew.c. The
function build_frame_matrix is a good starting point.
Frame matrices don't have marginal areas, only a text area. That
is, the entire row of glyphs that spans the width of a text-mode
frame is treated as a single large "text area" for the purposes of
manipulating and updating a frame glyph matrix.
To be honest, there is a little bit more done for frame matrices,
but not much more. If you plan to extend that code, take a look at
dispnew.c. The function build_frame_matrix is a good starting
point.
Simulating display.
Some of Emacs commands and functions need to take display layout
into consideration. For example, C-n moves to the next screen
line, but to implement that, Emacs needs to find the buffer
position which is directly below the cursor position on display.
This is not trivial when buffer display includes variable-size
elements such as different fonts, tall images, etc.
To solve this problem, the display engine implements several
functions that can move through buffer text in the same manner as
`display_line' and `display_string' do, but without producing any
glyphs for the glyph matrices. The workhorse of this is
`move_it_in_display_line_to'. Its code and logic are very similar
to `display_line', but it differs in two important aspects: it
doesn't produce glyphs for any glyph matrix, and it returns a
status telling the caller how it ended the iteration: whether it
reached the required position, hit the end of line, arrived at the
window edge without exhausting the buffer's line, etc. Since the
glyphs are not produced, the layout information available to the
callers of this function is what is recorded in `struct it' by the
iteration process.
Several higher-level functions call `move_it_in_display_line_to' to
perform more complex tasks: `move_it_by_lines' can move N lines up
or down from a given buffer position and `move_it_to' can move to a
given buffer position or to a given X or Y pixel coordinate.
These functions are called by the display engine itself as well,
when it needs to make layout decisions before producing the glyphs.
For example, one of the first things to decide when redisplaying a
window is where to put the `window-start' position; if the window
is to be recentered (the default), Emacs makes that decision by
starting from the position of point, then moving up the number of
lines corresponding to half the window height using
`move_it_by_lines'.
Bidirectional display.
Bidirectional display adds quite some hair to this already complex
design. The good news are that a large portion of that hairy stuff
is hidden in bidi.c behind only 3 interfaces. bidi.c implements a
reordering engine which is called by set_iterator_to_next and
reordering engine which is called by `set_iterator_to_next' and
returns the next character to display in the visual order. See
commentary on bidi.c for more details. As far as redisplay is
concerned, the effect of calling bidi_move_to_visually_next, the
concerned, the effect of calling `bidi_move_to_visually_next', the
main interface of the reordering engine, is that the iterator gets
magically placed on the buffer or string position that is to be
displayed next in the visual order. In other words, a linear
@ -279,27 +360,27 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
monotonously changing with vertical positions. Also, accounting
for face changes, overlays, etc. becomes more complex because
non-linear iteration could potentially skip many positions with
changes, and then cross them again on the way back (see
handle_stop_backwards)...
such changes, and then cross them again on the way back (see
`handle_stop_backwards')...
One other prominent effect of bidirectional display is that some
paragraphs of text need to be displayed starting at the right
margin of the window---the so-called right-to-left, or R2L
paragraphs. R2L paragraphs are displayed with R2L glyph rows,
which have their reversed_p flag set. The bidi reordering engine
which have their `reversed_p' flag set. The bidi reordering engine
produces characters in such rows starting from the character which
should be the rightmost on display. PRODUCE_GLYPHS then reverses
the order, when it fills up the glyph row whose reversed_p flag is
set, by prepending each new glyph to what is already there, instead
of appending it. When the glyph row is complete, the function
extend_face_to_end_of_line fills the empty space to the left of the
leftmost character with special glyphs, which will display as,
well, empty. On text terminals, these special glyphs are simply
blank characters. On graphics terminals, there's a single stretch
glyph of a suitably computed width. Both the blanks and the
stretch glyph are given the face of the background of the line.
This way, the terminal-specific back-end can still draw the glyphs
left to right, even for R2L lines.
should be the rightmost on display. `PRODUCE_GLYPHS' then reverses
the order, when it fills up the glyph row whose `reversed_p' flag
is set, by prepending each new glyph to what is already there,
instead of appending it. When the glyph row is complete, the
function `extend_face_to_end_of_line' fills the empty space to the
left of the leftmost character with special glyphs, which will
display as, well, empty. On text terminals, these special glyphs
are simply blank characters. On graphics terminals, there's a
single stretch glyph of a suitably computed width. Both the blanks
and the stretch glyph are given the face of the background of the
line. This way, the terminal-specific back-end can still draw the
glyphs left to right, even for R2L lines.
Bidirectional display and character compositions.
@ -310,23 +391,23 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
Emacs display supports this by providing "character compositions",
most of which is implemented in composite.c. During the buffer
scan that delivers characters to PRODUCE_GLYPHS, if the next
scan that delivers characters to `PRODUCE_GLYPHS', if the next
character to be delivered is a composed character, the iteration
calls composition_reseat_it and next_element_from_composition. If
they succeed to compose the character with one or more of the
calls `composition_reseat_it' and `next_element_from_composition'.
If they succeed to compose the character with one or more of the
following characters, the whole sequence of characters that were
composed is recorded in the `struct composition_it' object that is
part of the buffer iterator. The composed sequence could produce
one or more font glyphs (called "grapheme clusters") on the screen.
Each of these grapheme clusters is then delivered to PRODUCE_GLYPHS
in the direction corresponding to the current bidi scan direction
(recorded in the scan_dir member of the `struct bidi_it' object
that is part of the iterator). In particular, if the bidi iterator
currently scans the buffer backwards, the grapheme clusters are
delivered back to front. This reorders the grapheme clusters as
appropriate for the current bidi context. Note that this means
that the grapheme clusters are always stored in the LGSTRING object
(see composite.c) in the logical order.
Each of these grapheme clusters is then delivered to
`PRODUCE_GLYPHS' in the direction corresponding to the current bidi
scan direction (recorded in the `scan_dir' member of the `struct
bidi_it' object that is part of the iterator). In particular, if
the bidi iterator currently scans the buffer backwards, the
grapheme clusters are delivered back to front. This reorders the
grapheme clusters as appropriate for the current bidi context.
Note that this means that the grapheme clusters are always stored
in the `LGSTRING' object (see composite.c) in the logical order.
Moving an iterator in bidirectional text
without producing glyphs.
@ -337,18 +418,18 @@ along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>. */
As far as the iterator is concerned, the geometry of such rows is
still left to right, i.e. the iterator "thinks" the first character
is at the leftmost pixel position. The iterator does not know that
PRODUCE_GLYPHS reverses the order of the glyphs that the iterator
delivers. This is important when functions from the move_it_*
`PRODUCE_GLYPHS' reverses the order of the glyphs that the iterator
delivers. This is important when functions from the `move_it_*'
family are used to get to certain screen position or to match
screen coordinates with buffer coordinates: these functions use the
iterator geometry, which is left to right even in R2L paragraphs.
This works well with most callers of move_it_*, because they need
This works well with most callers of `move_it_*', because they need
to get to a specific column, and columns are still numbered in the
reading order, i.e. the rightmost character in a R2L paragraph is
still column zero. But some callers do not get well with this; a
notable example is mouse clicks that need to find the character
that corresponds to certain pixel coordinates. See
buffer_posn_from_coords in dispnew.c for how this is handled. */
`buffer_posn_from_coords' in dispnew.c for how this is handled. */
#include <config.h>
#include <stdlib.h>