@iftex
@chapter Killing and Moving Text

  @dfn{Killing} means erasing text and copying it into the @dfn{kill ring},
from which it can be retrieved by @dfn{yanking} it.  Some other systems
that have recently become popular use the terms ``cutting'' and ``pasting''
for these operations.

  The most common way of moving or copying text with Emacs is to kill it
and later yank it in one or more places.  This is safe because all the
text killed recently is stored in the kill ring, and it is versatile,
because you can use the same commands for killing syntactic units and
for moving those units.  There are other ways of copying text for
special purposes.

  Emacs has only one kill ring, so you can kill text in one buffer and yank
it in another buffer. If you are using XEmacs under X, you can
also use the X selection mechanism to copy text from one buffer to
another, or between applications. @xref{Using X Selections}.

@end iftex

@node Killing, Yanking, Additional Mouse Operations, Top
@section Deletion and Killing
@findex delete-char
@findex delete-backward-char

@cindex killing
@cindex cutting
@cindex deletion
@kindex C-d
@kindex DEL
  Most commands that erase text from the buffer save it. You can get
the text back if you change your mind, or you can move or copy it to
other parts of the buffer.  Commands which erase text and save it in the
kill ring are known as @dfn{kill} commands.  Some other commands erase
text but do not save it; they are known as @dfn{delete} commands.  (This
distinction is made only for erasing text in the buffer.)

The commands' names and individual descriptions use the words
@samp{kill} and @samp{delete} to indicate what they do.  If you perform
a kill or delete command by mistake, use the @kbd{C-x u} (@code{undo})
command to undo it (@pxref{Undo}). The delete commands include @kbd{C-d}
(@code{delete-char}) and @key{DEL} (@code{delete-backward-char}), which
delete only one character at a time, and those commands that delete only
spaces or newlines.  Commands that can destroy significant amounts of
nontrivial data usually kill.@refill

@subsection Deletion

@table @kbd
@item C-d
Delete next character (@code{delete-char}).
@item @key{DEL}
Delete previous character (@code{delete-backward-char}).
@item M-\
Delete spaces and tabs around point (@code{delete-horizontal-space}).
@item M-@key{SPC}
Delete spaces and tabs around point, leaving one space
(@code{just-one-space}).
@item C-x C-o
Delete blank lines around the current line (@code{delete-blank-lines}).
@item M-^
Join two lines by deleting the intervening newline, and any indentation
following it (@code{delete-indentation}).
@end table

  The most basic delete commands are @kbd{C-d} (@code{delete-char}) and
@key{DEL} (@code{delete-backward-char}).  @kbd{C-d} deletes the
character after point, the one the cursor is ``on top of''.  Point
doesn't move.  @key{DEL} deletes the character before the cursor, and
moves point back.  You can delete newlines like any other characters in
the buffer; deleting a newline joins two lines.  Actually, @kbd{C-d} and
@key{DEL} aren't always delete commands; if you give them an argument,
they kill instead, since they can erase more than one character this
way.

@kindex M-\
@findex delete-horizontal-space
@kindex M-SPC
@findex just-one-space
@kindex C-x C-o
@findex delete-blank-lines
@kindex M-^
@findex delete-indentation
  The other delete commands delete only formatting characters: spaces,
tabs and newlines.  @kbd{M-\} (@code{delete-horizontal-space}) deletes
all spaces and tab characters before and after point.
@kbd{M-@key{SPC}} (@code{just-one-space}) does the same but leaves a
single space after point, regardless of the number of spaces that
existed previously (even zero).

  @kbd{C-x C-o} (@code{delete-blank-lines}) deletes all blank lines after
the current line. If the current line is blank, it deletes all blank lines
preceding the current line as well as leaving one blank line, the current
line.  @kbd{M-^} (@code{delete-indentation}) joins the current line and
the previous line, or, if given an argument, joins the current line and 
the next line by deleting a newline and all surrounding spaces, possibly
leaving a single space.  @xref{Indentation,M-^}.

@subsection Killing by Lines

@table @kbd
@item C-k
Kill rest of line or one or more lines (@code{kill-line}).
@end table

@kindex C-k
@findex kill-line
  The simplest kill command is @kbd{C-k}.  If given at the beginning of
a line, it kills all the text on the line, leaving the line blank.  If
given on a blank line, the blank line disappears.  As a consequence, a
line disappears completely if you go to the front of a non-blank line
and type @kbd{C-k} twice.

  More generally, @kbd{C-k} kills from point up to the end of the line,
unless it is at the end of a line.  In that case, it kills the newline
following the line, thus merging the next line into the current one.
Emacs ignores invisible spaces and tabs at the end of the line when deciding
which case applies: if point appears to be at the end of the line, you
can be sure the newline will be killed.

  If you give @kbd{C-k} a positive argument, it kills that many lines
and the newlines that follow them (however, text on the current line
before point is not killed).  With a negative argument, @kbd{C-k} kills
back to a number of line beginnings.  An argument of @minus{}2 means
kill back to the second line beginning.  If point is at the beginning of
a line, that line beginning doesn't count, so @kbd{C-u - 2 C-k} with
point at the front of a line kills the two previous lines.

  @kbd{C-k} with an argument of zero kills all the text before point on the
current line.

@subsection Other Kill Commands
@findex kill-region
@findex kill-word
@findex backward-kill-word
@findex kill-sexp
@findex kill-sentence
@findex backward-kill-sentence
@kindex M-d
@kindex M-DEL
@kindex C-M-k
@kindex C-x DEL
@kindex M-k
@kindex C-w

@c DoubleWideCommands
@table @kbd
@item C-w
Kill region (from point to the mark) (@code{kill-region}).
@xref{Words}.
@item M-d
Kill word (@code{kill-word}).
@item M-@key{DEL}
Kill word backwards (@code{backward-kill-word}).
@item C-x @key{DEL}
Kill back to beginning of sentence (@code{backward-kill-sentence}).
@xref{Sentences}.
@item M-k
Kill to end of sentence (@code{kill-sentence}).
@item C-M-k
Kill sexp (@code{kill-sexp}).  @xref{Lists}.
@item M-z @var{char}
Kill up to next occurrence of @var{char} (@code{zap-to-char}).
@end table

   @kbd{C-w} (@code{kill-region}) is a very general kill command; it
kills everything between point and the mark. You can use this command to
kill any contiguous sequence of characters by first setting the mark at
one end of a sequence of characters, then going to the other end and
typing @kbd{C-w}.

@kindex M-z
@findex zap-to-char
  A convenient way of killing is combined with searching: @kbd{M-z}
(@code{zap-to-char}) reads a character and kills from point up to (but not
including) the next occurrence of that character in the buffer.  If there
is no next occurrence, killing goes to the end of the buffer.  A numeric
argument acts as a repeat count.  A negative argument means to search
backward and kill text before point.

  Other syntactic units can be killed: words, with @kbd{M-@key{DEL}} and
@kbd{M-d} (@pxref{Words}); sexps, with @kbd{C-M-k} (@pxref{Lists}); and
sentences, with @kbd{C-x @key{DEL}} and @kbd{M-k}
(@pxref{Sentences}).@refill

@node Yanking, Using X Selections, Killing, Top
@section Yanking
@cindex moving text
@cindex copying text
@cindex kill ring
@cindex yanking
@cindex pasting

  @dfn{Yanking} means getting back text which was killed. Some systems
call this ``pasting''.  The usual way to move or copy text is to kill it
and then yank it one or more times.

@table @kbd
@item C-y
Yank last killed text (@code{yank}).
@item M-y
Replace re-inserted killed text with the previously killed text
(@code{yank-pop}).
@item M-w
Save region as last killed text without actually killing it
(@code{copy-region-as-kill}).
@item C-M-w
Append next kill to last batch of killed text (@code{append-next-kill}).
@end table

@menu
* Kill Ring::       Where killed text is stored.  Basic yanking.
* Appending Kills:: Several kills in a row all yank together.
* Earlier Kills::   Yanking something killed some time ago.
@end menu

@node Kill Ring, Appending Kills, Yanking, Yanking
@subsection The Kill Ring

@kindex C-y
@findex Yank
  All killed text is recorded in the @dfn{kill ring}, a list of blocks of
text that have been killed.  There is only one kill ring, used in all
buffers, so you can kill text in one buffer and yank it in another buffer.
This is the usual way to move text from one file to another.
(@xref{Accumulating Text}, for some other ways.)

  If you have two separate Emacs processes, you cannot use the kill ring
to move text. If you are using XEmacs under X, however, you can
use the X selection mechanism to move text from one to another.

If you are using XEmacs under X and have one Emacs process with
multiple frames, they do share the same kill ring.  You can kill or
copy text in one Emacs frame, then yank it in the other frame
belonging to the same process.

  The command @kbd{C-y} (@code{yank}) reinserts the text of the most recent
kill.  It leaves the cursor at the end of the text and sets the mark at
the beginning of the text.  @xref{Mark}.

  @kbd{C-u C-y} yanks the text, leaves the cursor in front of the text,
and sets the mark after it, if the argument is with just a @kbd{C-u}.
Any other argument, including @kbd{C-u} and digits, has different
results, described below, under ``Yanking Earlier Kills''.

@kindex M-w
@findex copy-region-as-kill
 To copy a block of text, you can also use @kbd{M-w}
(@code{copy-region-as-kill}), which copies the region into the kill ring
without removing it from the buffer. @kbd{M-w} is similar to @kbd{C-w}
followed by @kbd{C-y} but does not mark the buffer as ``modified'' and
does not actually cut anything.

@node Appending Kills, Earlier Kills, Kill Ring, Yanking
@subsection Appending Kills

@cindex television
  Normally, each kill command pushes a new block onto the kill ring.
However, two or more kill commands in a row combine their text into a
single entry, so that a single @kbd{C-y} yanks it all back. This means
you don't have to kill all the text you want to yank in one command; you
can kill line after line, or word after word, until you have killed what
you want, then get it all back at once using @kbd{C-y}. (Thus we join
television in leading people to kill thoughtlessly.)

  Commands that kill forward from point add onto the end of the previous
killed text.  Commands that kill backward from point add onto the
beginning.  This way, any sequence of mixed forward and backward kill
commands puts all the killed text into one entry without rearrangement.
Numeric arguments do not break the sequence of appending kills.  For
example, suppose the buffer contains:

@example
This is the first
line of sample text
and here is the third.
@end example

@noindent
with point at the beginning of the second line.  If you type @kbd{C-k C-u 2
M-@key{DEL} C-k}, the first @kbd{C-k} kills the text @samp{line of sample
text}, @kbd{C-u 2 M-@key{DEL}} kills @samp{the first} with the newline that
followed it, and the second @kbd{C-k} kills the newline after the second
line.  The result is that the buffer contains @samp{This is and here is the
third.} and a single kill entry contains @samp{the first@key{RET}line of
sample text@key{RET}}---all the killed text, in its original order.

@kindex C-M-w
@findex append-next-kill
  If a kill command is separated from the last kill command by other
commands (not just numeric arguments), it starts a new entry on the kill
ring.  To force a kill command to append, first type the command @kbd{C-M-w}
(@code{append-next-kill}). @kbd{C-M-w} tells the following command,
if it is a kill command, to append the text it kills to the last killed
text, instead of starting a new entry.  With @kbd{C-M-w}, you can kill
several separated pieces of text and accumulate them to be yanked back
in one place.@refill

@node Earlier Kills,, Appending Kills, Yanking
@subsection Yanking Earlier Kills

@kindex M-y
@findex yank-pop
  To recover killed text that is no longer the most recent kill, you need
the @kbd{Meta-y} (@code{yank-pop}) command.  You can use @kbd{M-y} only
after a @kbd{C-y} or another @kbd{M-y}.  It takes the text previously
yanked and replaces it with the text from an earlier kill.  To recover
the text of the next-to-the-last kill, first use @kbd{C-y} to recover
the last kill, then @kbd{M-y} to replace it with the previous
kill.@refill

  You can think in terms of a ``last yank'' pointer which points at an item
in the kill ring.  Each time you kill, the ``last yank'' pointer moves to
the new item at the front of the ring.  @kbd{C-y} yanks the item
which the ``last yank'' pointer points to.  @kbd{M-y} moves the ``last
yank'' pointer to a different item, and the text in the buffer changes to
match.  Enough @kbd{M-y} commands can move the pointer to any item in the
ring, so you can get any item into the buffer.  Eventually the pointer
reaches the end of the ring; the next @kbd{M-y} moves it to the first item
again.

  Yanking moves the ``last yank'' pointer around the ring, but does not
change the order of the entries in the ring, which always runs from the
most recent kill at the front to the oldest one still remembered.

  Use @kbd{M-y} with a numeric argument to advance the ``last
yank'' pointer by the specified number of items.  A negative argument
moves the pointer toward the front of the ring; from the front of the
ring, it moves to the last entry and starts moving forward from there.

  Once the text you are looking for is brought into the buffer, you can
stop doing @kbd{M-y} commands and the text will stay there. Since the
text is just a copy of the kill ring item, editing it in the buffer does
not change what's in the ring.  As long you don't kill additional text,
the ``last yank'' pointer remains at the same place in the kill ring:
repeating @kbd{C-y} will yank another copy of the same old kill.

  If you know how many @kbd{M-y} commands it would take to find the
text you want, you can yank that text in one step using @kbd{C-y} with
a numeric argument.  @kbd{C-y} with an argument greater than one
restores the text the specified number of entries back in the kill
ring.  Thus, @kbd{C-u 2 C-y} gets the next to the last block of killed
text.  It is equivalent to @kbd{C-y M-y}.  @kbd{C-y} with a numeric
argument starts counting from the ``last yank'' pointer, and sets the
``last yank'' pointer to the entry that it yanks.

@vindex kill-ring-max
  The variable @code{kill-ring-max} controls the length of the kill
ring; no more than that many blocks of killed text are saved.

@node Using X Selections, Accumulating Text, Yanking, Top
@section Using X Selections
@comment  node-name,  next,  previous,  up

In the X window system, mouse selections provide a simple mechanism for
text transfer between different applications.  In a typical X
application, you can select text by pressing the left mouse button and
dragging the cursor over the text you want to copy.  The text becomes the
primary X selection and is highlighted.  The highlighted region is also
the Emacs selected region.

@itemize @bullet
@item
Since the region is the primary X selection, you can go to a different X
application and click the middle mouse button: the text that you selected in
the previous application is pasted into the current application.
@item
Since the region is the Emacs selected region, you can use all region
commands (@kbd{C-w, M-w} etc.) as well as the options of the @b{Edit}
menu to manipulate the selected text.
@end itemize

@menu
* X Clipboard Selection::     	Pasting to the X clipboard.
* X Selection Commands::	Other operations on the selection.
* X Cut Buffers::       	X cut buffers are available for compatibility.
* Active Regions::      	Using zmacs-style highlighting of the
                        	 selected region.
@end menu

@node X Clipboard Selection, X Selection Commands, Using X Selections, Using X Selections
@comment  node-name,  next,  previous,  up
@subsection The Clipboard Selection
@cindex clipboard selections

There are other kinds of X selections besides the @b{Primary} selection; one
common one is the @b{Clipboard} selection.  Some applications prefer to
transfer data using this selection in preference to the @b{Primary}.
One can transfer text from the @b{Primary} selection to the  @b{Clipboard}
selection with the @b{Copy} command under the @b{Edit} menu in the menubar.

Usually, the clipboard selection is not visible.  However, if you run the
@file{xclipboard} application, the text most recently copied to the clipboard
(with the @b{Copy} command) is displayed in a window.  Any time new text is
thus copied, the @file{xclipboard} application makes a copy of it and displays
it in its window.  The value of the clipboard can survive the lifetime of the
running Emacs process.  The @code{xclipboard} man page provides more details.

Warning: If you use the @file{xclipboard} application, remember that it
maintains a list of all things that have been pasted to the clipboard (that
is, copied with the @b{Copy} command).  If you don't manually delete elements
from this list by clicking on the @b{Delete} button in the @code{xclipboard}
window, the clipboard will eventually consume a lot of memory.

In summary, some X applications (such as @file{xterm}) allow one to paste
text in them from XEmacs in the following way:

@itemize @bullet
@item
Drag out a region of text in Emacs with the left mouse button,
making that text be the @b{Primary} selection.

@item
Click the middle button in the other application, pasting the @b{Primary}
selection. 
@end itemize

With some other applications (notably, the OpenWindows and Motif tools) you
must use this method instead:

@itemize @bullet
@item
Drag out a region of text in Emacs with the left mouse button,
making that text be the @b{Primary} selection.

@item
Copy the selected text to the @b{Clipboard} selection by selecting the
@b{Copy} menu item from the @b{Edit} menu, or by hitting the @b{Copy}
key on your keyboard.

@item
Paste the text in the other application by selecting @b{Paste} from its
menu, or by hitting the @b{Paste} key on your keyboard.
@end itemize


@node X Selection Commands, X Cut Buffers, X Clipboard Selection, Using X Selections
@subsection Miscellaneous X Selection Commands
@comment  node-name,  next,  previous,  up
@cindex cut buffers
@cindex primary selections

@findex x-copy-primary-selection
@findex x-delete-primary-selection
@findex x-insert-selection
@findex x-kill-primary-selection
@findex x-mouse-kill
@findex x-own-secondary-selection
@findex x-own-selection
@findex x-set-point-and-insert-selection
@table @kbd
@item M-x x-copy-primary-selection
Copy the primary selection to both the kill ring and the Clipboard.
@item M-x x-insert-selection
Insert the current selection into the buffer at point.
@item M-x x-delete-primary-selection
Deletes the text in the primary selection without copying it to the kill
ring or the Clipboard.
@item M-x x-kill-primary-selection
Deletes the text in the primary selection and copies it to 
both the kill ring and the Clipboard.
@item M-x x-mouse-kill
Kill the text between point and the mouse and copy it to 
the clipboard and to the cut buffer.
@item M-x x-own-secondary-selection
Make a secondary X selection of the given argument. 
@item M-x x-own-selection
Make a primary X selection of the given argument.  
@item M-x x-set-point-and-insert-selection
Set point where clicked and insert the primary selection or the
cut buffer.
@end table

@node X Cut Buffers, Active Regions, X Selection Commands, Using X Selections
@subsection X Cut Buffers
@comment  node-name,  next,  previous,  up

X cut buffers are a different, older way of transferring text between
applications.  XEmacs supports cut buffers for compatibility
with older programs, even though selections are now the preferred way of
transferring text.

X has a concept of applications "owning" selections.  When you select
text by clicking and dragging inside an application, the application
tells the X server that it owns the selection.  When another
application asks the X server for the value of the selection, the X
server requests the information from the owner. When you use
selections, the selection data is not actually transferred unless
someone wants it; the act of making a selection doesn't transfer data.
Cut buffers are different: when you "own" a cut buffer, the data is
actually transferred to the X server immediately, and survives the
lifetime of the application.

Any time a region of text becomes the primary selection in Emacs,
Emacs also copies that text to the cut buffer.  This makes it possible
to copy text from an XEmacs buffer and paste it into an older,
non-selection-based application (such as Emacs 18).

Note: Older versions of Emacs could not access the X selections, only
the X cut buffers.

@node Active Regions, , X Cut Buffers, Using X Selections
@subsection Active Regions
@comment  node-name,  next,  previous,  up
@cindex active regions

  By default, both the text you select in an Emacs buffer using the
click-and-drag mechanism and text you select by setting point and the
mark is highlighted. You can use Emacs region commands as well as the
@b{Cut} and @b{Copy} commands on the highlighted region you selected
with the mouse.

If you prefer, you can make a distinction between text selected with the
mouse and text selected with point and the mark by setting the variable
@code{zmacs-regions} to @code{nil}.  In that case:

@itemize @bullet
@item
The text selected with the mouse becomes both the X selection and the
Emacs selected region. You can use menu-bar commands as well as Emacs
region commands on it. 
@item
The text selected with point and the mark is not highlighted. You can
only use Emacs region commands on it, not the menu-bar items. 
@end itemize

  Active regions originally come from Zmacs, the Lisp Machine editor.
The idea behind them is that commands can only operate on a region when
the region is in an "active" state.  Put simply, you can only operate on
a region that is highlighted.

@vindex zmacs-regions
The variable @code{zmacs-regions} checks whether LISPM-style active
regions should be used.  This means that commands that operate on the
region (the area between point and the mark) only work while
the region is in the active state, which is indicated by highlighting.
Most commands causes the region to not be in the active state;
for example, @kbd{C-w} only works immediately after activating the
region.

More specifically:
@itemize @bullet
@item
Commands that operate on the region only work if the region is active.
@item
Only a very small set of commands causes the region to become active---
those commands whose semantics are to mark an area, such as @code{mark-defun}.
@item
The region is deactivated after each command that is executed, except that
motion commands do not change whether the region is active or not.
@end itemize 

@code{set-mark-command} (@kbd{C-SPC}) pushes a mark and activates the
region.  Moving the cursor with normal motion commands (@kbd{C-n},
@kbd{C-p}, etc.) will cause the region between point and the
recently-pushed mark to be highlighted.  It will remain highlighted
until some non-motion command is executed.

@code{exchange-point-and-mark} (@kbd{C-x C-x}) activates the region.
So if you mark a region and execute a command that operates on it, you
can reactivate the same region with @kbd{C-x C-x} (or perhaps @kbd{C-x
C-x C-x C-x}) to operate on it again.

Generally, commands that push marks as a means of navigation, such as
@code{beginning-of-buffer} (@kbd{M-<}) and @code{end-of-buffer}
(@kbd{M->}), do not activate the region.  However, commands that push
marks as a means of marking an area of text, such as @code{mark-defun}
(@kbd{M-C-h}), @code{mark-word} (@kbd{M-@@}), and @code{mark-whole-buffer}
(@kbd{C-x h}), do activate the region.

When @code{zmacs-regions} is @code{t}, there is no distinction between
the primary X selection and the active region selected by point and the
mark.  To see this, set the mark (@key{C-SPC}) and move the cursor
with any cursor-motion command: the region between point and mark is
highlighted, and you can watch it grow and shrink as you move the
cursor.

Any other commands besides cursor-motion commands (such as inserting or
deleting text) will cause the region to no longer be active; it will no
longer be highlighted, and will no longer be the primary selection.
Errors also remove highlighting from a region.

Commands that require a region (such as @kbd{C-w}) signal an error if
the region is not active.  Certain commands cause the region to be in
its active state.  The most common ones are @code{push-mark}
(@key{C-SPC}) and @code{exchange-point-and-mark} (@kbd{C-x C-x}).

@vindex zmacs-region-stays
When @code{zmacs-regions} is @code{t}, programs can be non-intrusive
on the state of the region by setting the variable @code{zmacs-region-stays}
to a non-@code{nil} value.  If you are writing a new Emacs command that
is conceptually a ``motion'' command and should not interfere with the
current highlightedness of the region, then you may set this variable.
It is reset to @code{nil} after each user command is executed.

@findex zmacs-activate-region
When @code{zmacs-regions} is @code{t}, programs can make the region between
point and mark go into the active (highlighted) state by using the
function @code{zmacs-activate-region}. Only a small number of commands
should ever do this. 

@findex zmacs-deactivate-region
When @code{zmacs-regions} is @code{t}, programs can deactivate the region
between point and the mark by using @code{zmacs-deactivate-region}.
Note: you should not have to call this function; the command loop calls
it when appropriate. 

@node Accumulating Text, Rectangles, Using X Selections, Top
@section Accumulating Text
@findex append-to-buffer
@findex prepend-to-buffer
@findex copy-to-buffer
@findex append-to-file
@cindex copying text
@cindex accumulating text

  Usually you copy or move text by killing it and yanking it, but there are
other ways that are useful for copying one block of text in many places, or
for copying many scattered blocks of text into one place.

  If you like, you can accumulate blocks of text from scattered
locations either into a buffer or into a file.  The relevant commands
are described here.  You can also use Emacs registers for storing and
accumulating text.  @xref{Registers}.

@table @kbd
@item M-x append-to-buffer
Append region to contents of specified buffer (@code{append-to-buffer}).
@item M-x prepend-to-buffer
Prepend region to contents of specified buffer.
@item M-x copy-to-buffer
Copy region into specified buffer, deleting that buffer's old contents.
@item M-x insert-buffer
Insert contents of specified buffer into current buffer at point.
@item M-x append-to-file
Append region to the end of the contents of specified file.
@end table

  To accumulate text into a buffer, use the command @kbd{M-x
append-to-buffer}, which inserts a copy of the region into the buffer
@var{buffername}, at the location of point in that buffer.  If there is
no buffer with the given name, one is created.

  If you append text to a buffer that has been used for editing, the
copied text goes to the place where point is.  Point in that buffer is
left at the end of the copied text, so successive uses of
@code{append-to-buffer} accumulate the text in the specified buffer in
the same order as they were copied.  Strictly speaking, this command does
not always append to the text already in the buffer; but if this command
is the only command used to alter a buffer, it does always append to the
existing text because point is always at the end.

  @kbd{M-x prepend-to-buffer} is similar to @code{append-to-buffer}, but
point in the other buffer is left before the copied text, so successive
prependings add text in reverse order.  @kbd{M-x copy-to-buffer} is
similar, except that any existing text in the other buffer is deleted,
so the buffer is left containing just the text newly copied into it.

  You can retrieve the accumulated text from that buffer with @kbd{M-x
insert-buffer}, which takes @var{buffername} as an argument.  It inserts
a copy of the text in buffer @var{buffername} into the selected buffer.
You could alternatively select the other buffer for editing, perhaps moving
text from it by killing or with @code{append-to-buffer}.  @xref{Buffers}, for
background information on buffers.

  Instead of accumulating text within Emacs in a buffer, you can append
text directly into a file with @kbd{M-x append-to-file}, which takes
@var{file-name} as an argument.  It adds the text of the region to the
end of the specified file.  The file is changed immediately on disk.
This command is normally used with files that are @i{not} being visited
in Emacs.  Using it on a file that Emacs is visiting can produce
confusing results, because the file's text inside Emacs does not change
while the file itself changes.

@node Rectangles, Registers, Accumulating Text, Top
@section Rectangles
@cindex rectangles

  The rectangle commands affect rectangular areas of text: all
characters between a certain pair of columns, in a certain range of lines.
Commands are provided to kill rectangles, yank killed rectangles, clear
them out, or delete them.  Rectangle commands are useful with text in
multicolumnar formats, like code with comments at the right,
or for changing text into or out of such formats.

  To specify the rectangle a command should work on, put the mark at one
corner and point at the opposite corner.  The specified rectangle is
called the @dfn{region-rectangle} because it is controlled about the
same way the region is controlled.  Remember that a given
combination of point and mark values can be interpreted either as
specifying a region or as specifying a rectangle; it is up to the
command that uses them to choose the interpretation.

@table @kbd
@item M-x delete-rectangle
Delete the text of the region-rectangle, moving any following text on
each line leftward to the left edge of the region-rectangle.
@item M-x kill-rectangle
Similar, but also save the contents of the region-rectangle as the
``last killed rectangle''.
@item M-x yank-rectangle
Yank the last killed rectangle with its upper left corner at point.
@item M-x open-rectangle
Insert blank space to fill the space of the region-rectangle.
The previous contents of the region-rectangle are pushed rightward.
@item M-x clear-rectangle
Clear the region-rectangle by replacing its contents with spaces.
@end table

  The rectangle operations fall into two classes: commands deleting and
moving rectangles, and commands for blank rectangles.

@findex delete-rectangle
@findex kill-rectangle
  There are two ways to get rid of the text in a rectangle: you can discard
the text (delete it) or save it as the ``last killed'' rectangle.  The
commands for these two ways are @kbd{M-x delete-rectangle} and @kbd{M-x
kill-rectangle}.  In either case, the portion of each line that falls inside
the rectangle's boundaries is deleted, causing following text (if any) on
the line to move left.

  Note that ``killing'' a rectangle is not killing in the usual sense; the
rectangle is not stored in the kill ring, but in a special place that
only records the most recently killed rectangle (that is, does not
append to a killed rectangle).  Different yank commands
have to be used and only one rectangle is stored, because yanking
a rectangle is quite different from yanking linear text and yank-popping
commands are difficult to make sense of.

  Inserting a rectangle is the opposite of deleting one.  You specify
where to put the upper left corner by putting point there.  The
rectangle's first line is inserted at point, the rectangle's second line
is inserted at a point one line vertically down, and so on.  The number
of lines affected is determined by the height of the saved rectangle.

@findex yank-rectangle
  To insert the last killed rectangle, type @kbd{M-x yank-rectangle}.
This can be used to convert single-column lists into double-column
lists; kill the second half of the list as a rectangle and then
yank it beside the first line of the list.

@findex open-rectangle
@findex clear-rectangle
  There are two commands for working with blank rectangles: @kbd{M-x
clear-rectangle} erases existing text, and @kbd{M-x open-rectangle}
inserts a blank rectangle.  Clearing a rectangle is equivalent to
deleting it and then inserting a blank rectangle of the same size.

  Rectangles can also be copied into and out of registers.
@xref{RegRect,,Rectangle Registers}.