bugs.txt

REPORT BUGS TO kirkenda@cs.pdx.edu (STEVE KIRKENDALL)

NOTE: Each bug in this file is marked with either "*", "/", or "?".
The "*" means the bug is still pending.  The "/" means that I believe it has
been solved.  A "?" means that I've done some work on it, and *hope* it is
solved because it seems to work better but I never really understood what
was wrong in the first place.  A "?" will become a "/" if I don't receive
more reports of that bug.  More than 80% of this file lists *fixed* bugs.

NOTE: Not all of these are truly bugs.  This file also serves as my "to do"
list, so some items are just unimplemented features that sounded like a good
idea at one time or another.

* Add a new warpto value, "window", which moves the mouse pointer as little
  as possible.  Use XQueryPointer() and XTranslateCoordinates() to do the
  coordinate math.

	And maybe use XTranslateCoordinates() to choose the position of a
	pop-up dialog too.

* Something adds buffers with names that aren't full pathnames of files.
  Maybe the :make command?  This is a problem because it allows multiple
  buffers to be open for the same file.

* Running "elvis -Gtermcap README.html" starts with a scrambled display.

	This has something to do with the rcssince alias.  If I disable the
	autocmd that runs rcssince, it looks okay.

	It is apparently due to quirks of the guitcap->prgclose() function.
	That function tries to be clever about waiting for a keystroke with
	the cursor on the bottom row, which is basically what "scrambled"
	means in this context.  This is probably the nastiest, most complex
	aspect of the termcap interface -- it uses a couple of variables
	that are used in many functions, in subtle ways.  Ugh. (The purpose
	of all this cleverness is to allow programs to use the whole screen
	instead of just one window.  The portable part of Elvis doesn't know
	that guitcap.c is trying to do this, so guitcap.c needs to do some
	pretty bizarre things.)

	Current status is that it works okay, except that if you switch to
	another file, and that file has an RCS file, then Elvis will wait
	for you to hit <Enter> before switching to the other file.  Only
	in the termcap interface.

	What about vio?

* When the :gdb interface switches buffers, it doesn't always update the
  window name to reflect the name of the new buffer.  (This is a bug in
  Elvis' buffer-switching code, not the gdb interface.)

* Something weird with instman.sh, where it gives priority to "share"
  directories.  It isn't looking for the existence of directories correctly.

* The "html" display mode doesn't hide <style>...</style> contents.

* Is there some way I could make the persist.args flag avoid changing the
  args in some cases?  I'm specifically thinking of when Elvis is used as
  an external editor.  I don't want "~/.sylpheed/tmp/tmp.98S498DS" becoming
  my default argument string.

	Maybe add a -p flag to disable updates of the persistent information.
	Still allow it to be loaded, just not saved.  Though it might be nice
	to make saving/not-saving be buffer-specific, the args list isn't
	buffer-specific so that would be impractical.  (Vim uses "-iNONE",
	sort of.)

* In a search expression, ^X00 doesn't cause elvis to search for a NUL byte.

	The regbuild() function should probably convert any NUL bytes into
	\0 escape sequences.  The incsearch code may need special help.

-----------------------------------------------------------------------------
Fixed between 2.2.0 and 2.2.1

/ The "x11" interface can generate a BadMatch error for the X_SetInputFocus
  request.  This has been reported for both icewm and ratpoison WMs, though
  in different contexts.

/ When using incremental search, probably the search expression should be
  appended to history when user hits <Enter>.  Even if incsearch doesn't
  allow you to look back through history interactively, it should still be
  saved for the benefit of persistfile.

	I wanted to find a way to edit it, ideally by switching to the
	non-incremental version of a search since that supports all vi
	commands via a <Control-O> prefix.  But that quickly became very
	complex.  For now, incremental searches support ^Ok and ^Oj, and
	that's good enough to make the <Up> and <Down> arrow keys work.

/ Allow :source to read from filters.  That would be handy for programs that
  generate ex commands, such as the one used by the rcssince alias.

/ Describe the :load scripts in the Tips chapter.

/ The HTML tag <hr/> shows as "---->" instead of just "----".  More generally,
  it seems that some XML/HTML editors generate <tag/> instead of <tag> when
  there is no closing </tag>.  Elvis should handle this.

/ Many problems with the data/scripts/augz.ex script.

	/ When elvis sends the gzipped text out to the gunzip filter program,
	  it appends a newline.  There's no way around this, but it has the
	  side-effect of making gunzip output an "unexpected end of input"
	  error.

		Idea: If "partiallastline" is set, then omit the last newline
		when writing to the filter.  When reading back, if there is a
		final newline then clean up the "partiallastline" situation
		(delete the bogus newline and turn off "partiallastline").

	/ After gunzipping, the display mode and other options aren't set
	  to reflect the file's uncompressed contents.  For example, editing
	  a gzipped man page will leave the screen in "hex" display mode
	  instead of "man" display mode.

		This can be solved by temporarily changing the filename option
		and then running the "elvis.arf" script.  The only problem:
		the filename option is locked.  You can change it via the
		:file command, but that command produces output.  Maybe extend
		it so :file! (with ! suffix) produces no output?

/ Create a set of HTML buttons, loadable via ":load html".

/ The :eval command can get confused if the command that it runs also performs
  any calculations.  :eval should make a copy of the value returned by
  calculate(), and pass the copy to exstring().

/ At the end of an alias, elvis generates an AliasEnter event instead of
  an AliasLeave.

/ Backslash processing isn't quite right, in "simpler syntax" expressions.

	This is a little tricky because it is used for Windows file names,
	where backslashes are literal text.  The best compromise I could
	think of is to say \(, \), \$, and \\ are quoted characters anywhere
	except that at the front of the string \\ is two literal backslashes
	so that \\machine\dir\file names work.

/ The :suspend command doesn't work.  "Bad command name suspend".  Also,
  the shortest abbreviation should be ":su" according to POSIX, but Elvis
  requires ":sus".

/ After ":load since" and ":mkexrc", loading a file that has been changed
  since the last RCS checkin should have the changed lines highlighted, but
  they aren't always.  Similarly, ":au BufRead * 1d" fails on the first file,
  but works on any later files (loaded via ":e").

	This is due to the rcssince alias using some commands that aren't
	allowed during initialization.  The initialization is actually
	complete by the time BufRead events are executed, but ex.c doesn't
	know that because it detects initialization phase by testing for
	the existence of at least one window.

	The cleanest way to fix it is to add a new "initializing" option,
	which is set during initialization and and cleared when an edit
	buffer is created for the first text file.  ex.c should check that
	instead of the existence of a window.

/ The hlobject option should allow commas between object names.

/ In Unix, dirperm() doesn't return "directory" on directories.
  ':calc dirperm(".")' returns "readonly".

	This turned out to be due to the addition of the "dir:" pseudo-
	protocol, which was created so you could use ":alias readDIR ..."
	to create a directory browser.  Easy to fix.

/ The "ax" text object doesn't allow tag names to contain hyphens or colons.
  It should.

/ In security=safer mode, ":wq" fails even when invoked without a filename.
  According to the manual, ":wq" should be allowed.

	This is apparently because :wquit (and :edit and :ex) use the d_File
	flag so they always get the current file name by default.

	The :write command has special logic for this, but probably shouldn't.

/ The current security for ":e name" and ":sp name" is too strict -- even
  "elvis -client foo.c" fails.  I need to allow ":e name" from clients,
  without allowing ":w" from clients.  Is this yet another security level?
  security=client is exactly like security=normal except that all writing
  and !cmd is disabled.

	There are four things that must be avoided to maintain security:
	  1) You shouldn't be able to run external programs.
	  2) You shouldn't be able to write over any arbitrary file.
	  3) You shouldn't be able to load any arbitrary file, modify it, and
	     write it.
	  4) You shouldn't be able to create maps or aliases which trick the
	     user into doing any of the above.
	Items 1, 2, and 4 are pretty easy.  Item 3 requires some compromise
	and that's why there are two different secure modes.  "safer" allows
	you to load any file, and modify it, but not write it; this makes
	"safer" useless for interactive work.  "restricted" only allows you
	to load files named on the command line, but you can modify them
	and write them (back over the old version of the file, not out to
	any random file).

	So using "-S" to set security=safer is pointless; "-S" should set
	security=restricted.

	Some of the standard scripts in data/ need to be tweaked to work
	correctly with -S.

/ In "html" display mode, <li><p> looks ugly in Elvis, but many documents
  use that as a way to get gaps between list items.  Elvis should handle it
  better.

/ Running :mkexrc in one GUI can clobber options for another GUI.  For example,
  if you're running -Gtermcap and have some x11 options set, then :mkexrc will
  loose the x11 options.

	Just "store & ignore" options with unknown domains.  The :color command
	already does that, and it works well.  Basically, if ":set" detects
	that its args start with "gui." where "gui" is anything other than the
	current user interface, then it should store the args but not do
	anything else with them.  For the current user interface, "gui."
	should be skipped and the remainder processed normally.  Most of this
	logic is only needed if FEATURE_MKEXRC is defined.

	/ Change the name of the termcap GUI's options from "tcap" to "termcap"

	/ Some GUIs have window-specific options.  Although these are in the
	  "guiwin" domain (regardless of the GUI's name), they should still
	  be stored by :mkexrc as gui.option (where "gui" is the GUI's name).

/ The default "elvis.ini" creates some maps for keypad keys, without the
  "nosave" flag.  This means if you run ":mkexrc" while running in an xterm,
  you'll create maps which exist every time you run elvis.  Must add "nosave"!

/ <Shift-Tab> doesn't work in "x11" anymore.  The switch to ^S instead of
  ^K as the prefix character is tripping it up.  Also, <S-keyname> isn't
  recognized, or generated, and neither is #1s for shift-F1.

	/ The shift+tab key combination returns <S-ISO_Left_Tab>, not <S-Tab>
	  or <ISO_Left_Tab>.  Since <ISO_Left_Tab> is hard to type, Elvis
	  should standardize on using <S-Tab>.
	/ Function key labels in #x format can't use 'c' or 's' modifiers.
	/ Maps defined for function & cursor keys don't show any label.
	/ :mkexrc adds a bunch of flags to :map commands with the "visual"
	  keyword.

/ :%j only joins two lines, but :1,$j works correctly.

/ Sometimes ":%unr" doesn't remove all regions.  In particular, sometimes
  writing a file won't cause "unsaved" regions to become normal.  Going to
  the line and running ":unr" works though.  This also affects the :chregion
  command.

/ The :man alias doesn't handle backslashes in the man-page correctly.
  They interfere with the handling of \fX sequences.  See "man groff_man".

	Probably the simplest way to handle this is to change all backslashes
	into some harmless control character, such as ^\.  Then do the other
	stuff, and then go back and convert ^\ into a "\\" pair.

/ When a command is misspelled, the error message should give the whole name,
  not just the portion that had to be parsed for it to be recognized as bad.

/ When listing, I need to find a better way to distinguish between
  automatically installed keys and user-defined keys.  Maybe MAP_NOSAVE?

/ Make the "spellautoload" option be on by default.  It is harmless unless
  you use spell-checking, and in that case you almost certainly want it on.

/ The :map command needs to support a "noselect" keyword, which selects all
  of the command map states except "select".  Currently this is simply
  "command motion" but eventually I plan to implement "operator" and maybe
  others too.

/ The manual still shows ":gui icon" as the way to control elvis' icon.
  It should describe the "iconimage" option.

/ The :chregion command should not change the comment, unless either a new
  comment is supplied, or the old comment was identical to the old face.
  In the latter case, the comment should be changed to the new face name
  (which is what it does now).

	Done.  Also fixed a bug: The first char of the new comment was being
	skipped.

/ Near the end of the elvtags man-page, "Other versions of elvtags" should
  read "Other versions of ctags".

	The elvtags man-page is generated from the ctags man page.  Apparently
	the conversion rules aren't quite as smart as they should be.

/ The ":e +line file" command truly does limit the + argument to being a
  line address, not a real command.  That's bad!

/ Trouble entering the ^ character with a Norwegian keyboard in Windows.

/ ".IP \(bu 4" has a larger-than-expected indentation.  Is this because the
  indentation number has no suffix?  "4n" looks okay.

/ The command ":normal cwFOO^[" doesn't work.  Apparently the c operator
  doesn't work in a :normal command.
	  
	This turns out to be due to the way the "c" operator is implemented
	in oper().  It calls inputchange() to create the input state, but
	inputchange() doesn't create unless the state stack contains only
	a single vi command state.  During execution of a :normal command,
	there are at least 2 states (both vi command states) on the stack.
        The solution: Make inputchange() be smarter about when to reuse an
	existing input state.

/ When setting up a map in "x11", <b> is converted to ^K0062.  Single-letter
  names in brackets should always be left unchanged.  In fact, if the
  resulting string is a single ASCII character then the X11 name should
  be ignored; as it is now, you can't even use <less> to get the < character.
  (Though <lt> works.)

/ If you use ":e dirname" to edit the directory "dirname", then the links
  in the generated HTML aren't interpreted as being relative to dirname.
  They're treated as being relative to the current directory.

	A directory's filename should end with "/".  If the user gives a
	name without a "/" then elvis should add one.

/ The :dict alias (after ":load dict") doesn't handle single words correctly.

/ At ex.c:1605, the third argument to calculate() should be cast to
  CALCRULE, not ELVFNR.  Why doesn't GCC complain about that?

/ The :put command doesn't accept a cut buffer name.

/ In a file full of blank lines, "d)" causes a core dump.  So does  ")".

/ Lots of little changes to the OS/2 port.  This doesn't affect other ports.

/ Implement timeouts in -Gx11.

	Actually, timeouts were implemented for "usertime" (and aren't needed
	for "keytime").  The bug turned out to be that after a timeout, Elvis
	wasn't setting the flag that made it update the screen, so the screen
	would only be updated after another command or something like losing
	and regaining input focus.  I've fixed it so it sets the flag.

/ In visual command mode, <Esc> no longer generates a beep.  This is a change.
  Why was it changed?  Some people need to hear that beep.

	The behavior changed accidentally when I added <Esc> as a way to
	cancel visible selections for vim compatibility.  The selection
	cancelling function should fail for <Esc> if there is no selection
	pending.

/ In HTML mode, the default strings for <input type=submit> and
  <input type=reset> should be "Submit" and "Reset" respectively.  As it is
  now, those buttons disappear unless you supply an explcit value=... string.

/ In HTTP requests, elvis should send a User-Agent: header line.  Some web
  sites demand this.

/ The addition of filenamerules apparently broke the handling of newlines
  in the filename string.  They aren't treated as name delimiters anymore.
  This causes problems for things like ":n `grep -lw somesymbol *.c`"
  See line 1943 of ex.c.

/ The :andale alias was using the "luxi mono" font instead of "andale mono".
  (This alias is only defined for the "x11" user interface.)

/ Sometimes it would be nice if there was a way to set the application mode.
  For example, the GDB interface could use it to indicate whether the program
  was currently running or not.

	Maybe just a simple "state" option with no preset meaning, and a
	new "state" keyword in the "show" option which displays the state.

/ Implement an "elvgdb" program to act as a wrapper around the read GDB
  program.  Elvgdb will parse the output of GDB and send important information
  to Elvis.

/ Write a set of aliases to work eith "elvgdb".

/ Extend "elvgdb" to allow Elvis to send commands to GDB.  This is important
  because GDB doesn't output stack information unless you ask for it.

/ Add <F11> and <F12> to tinytcap.c entries.

/ The termcap interface should allow terminfo-emulating-termcap's key codes
  for shift-Fn and control-Fn keys.  Assume 12 keys so the 13th function key
  is really shift-F1, and the 25th function key is really control-F1.

	Terminfo-emulating-termcap uses :F1=: for <F11>, :F2=: for <F12>,
	and then continues through :F9=: for <F19> or <S-F7>, and then
	:FA=: for <F20> or <S-F8>, and continuing on through the letters.

/ You can compute a target buffer name by giving (=expression) in the
  address, but there's no easy way to compute a line number.  In aliases,
  it would often be nice if you could use something like =expression to
  compute a line number.

	Problem: How would the parser know where the line number ends and
	the command begins?  I'm tempted to use (expression) but that would
	look too much like a buffer name.

	Maybe just say =option fetches the value of the named option, and
	not allow expressions?  Or =(expr) for expressions?

/ It's almost impossible to combine the use of visual text selections and
  the :normal command.  This is important because it means you can't create
  a [Bold] GUI button which does ":normal cw<b>^P</b>" to embolden text.

	Created a new notation.  In :gui commands, you can now use
	":gui [label]" to create a button named "label" which, when pushed,
	causes the square brackets and label to be sent as though typed.
	You can then :map this to something useful.

/ Add a :phelp command, which is like :help except that it doesn't
  split.  Alternatively, add a help() function which returns the URL for
  the help for a given topic; then ":alias phelp push (help("!*"))" would
  achieve the same effect.

/ This is probably a good time to add the GuiEnter event.

/ Add persistent information, for storing the cursor position and maybe other
  information.  Add a "persistfile" option to store the filename, and a
  "persist" option to control what gets stored/loaded there.

	cursor		Move the cursor to its previous location
	change		Move the cursor to the previous change; if both
			"cursor" and "change" then "cursor" rules, and ''
			mark is set to "change".
	hours:hours	If entry is more than "hours" old, then don't move
			the cursor; instead, set '' to location where cursor
			would have gone.
	marks		Restore named marks that were in file.
	folds		Restore folds (and unfolds)
	regions		Restore highlight regions.
	search:lines	Save regexp history and current search
	ex:lines	Save ex history
	max:bytes	Approximate size limit on persistent information
	args		Default filenames

/ Need a way to remove folds.  This is different from merely unfolding, which
  remembers the attributes of the fold.  Add a :nofold command.

/ Add support for %< and #< to remove the extension from the current file name
  or alternate file name, respectively.  Vim has a similar feature.  (This is
  separate from Vim's fnamemodify() feature.)
  
-----------------------------------------------------------------------------

* Aliases shouldn't change the 'alternate file' (i.e., previousfile option).

* Add a real "note" facility.  Allow each note to have an icon, which is
  displayed in the left margin.  On text systems, a text symbol would be
  displayed instead.  The :note command would always use the text version.
  A separate command could be used to associate icon images with names.

	Use :note to associate a note with a given line, and :icon to
	associate an icon image with a name.  Splitting the association
	like this is good for themes.

	This could tie in nicely with the toolbar buttons.  If I work it
	right, toolbars could get icons at very little cost.  The names
	could correspond to the button names.

	Should a single line support multiple notes?  If :gdb uses notes to
	mark breakpoints and the current line, then what if the current line
	is a breakpoint?

	Maybe notes should be accessible via '"label" and definable via
	m"label" or m"label:comment".  Search via '">label" (or '">" for
	any symbol).  Notes would be drawn on the screen at their exact
	location, under the text.  On non-graphic GUIs, the notes would be
	revealed sort of like tags in showmarkup mode.

* The data/elvis.clr script should be able to use ":color!" (with a !) but
  can't because some of the compiled-in defaults are executed after that script
  is sourced, so the compiled-in defaults override ":color!".

* Idea for a trivial way to color herefiles: Use a :%s///x command to find
  them, and a region to highlight them.

	:color herefile black italic on white
	:try %s#<<\\\?\(\S\+\)#+,/^\s*\1$/-region herefile \1#x

	The elvis.syn file already supports color lines.  It doesn't allow
	you to run any ex commands, but maybe it should.  The above line
	should work perfectly well with security=safer and locked set.

* Maybe do something to control cursor motion around newlines.  Vim has
  an ugly 'whichwrap' option.

* In Windows, elvis.exe can accept filenames with spaces in them,
  but vi.exe can't.

	This sounds like a bug in the _spawnvp() system call.  Should
	32-bit programs even be calling _spawnvp()?

	This won't be fixed in time for 2.2.  If that's a problem for you,
	then copy elvis.exe to vi.exe.

* Sometimes it would be nice if there was a "flat" pseudocolor that was the
  average color of the background pixmap, after tinting.  Elvis already knows
  this color; it is used to find the best contrasting foreground color.

* When you give a command like ":dis s c++", elvis stores exactly that as
  the value of the display option.  This can confuse x11's toolbar buttons.
  The [Syntax] button will not be drawn pushed it, because it expects "syntax".

* See Luigi Mastrangelo's mail for some WinElvis bugs.

  * If :make encounters a large number of errors, then WinElvis can get stuck
    in an infinite loop.  This didn't happen in 2.1_4.

  * "WinElvis test*" will, if there is no file with the prefix "test", display
    an alias on the last line.

	I tried to reproduce this one.  For me, it's working correctly.
	I've cleaned up the wildcard matching code a lot since 2.2i (which
	he's using), so maybe that explains it.  I hope so.

* How about a gM operator command which stores the endpoints of the affected
  area and its paste-type, and a gA command which recalls that region as a
  visual selection (char or line style, as appropriate).

	Vim always remembers visual selections, and can use gv to recall them.
	But I want to support motion commands too, such as gM} to select text
	to the end of the current paragraph.  I believe this will be more
	useful in maps.

* If there was periodic autocmd event, then Elvis could use it to detect
  changes in files by other external programs, and reload the file.  This
  would be handy for a "more" program.

* I'm not sure, but I think that if you request the same URL with different
  parameters, elvis just uses the same copy every time.  The resource name
  is the same either way so they'd both be stored in the same buffer, and
  that's probably the cause of the problem.

	This isn't likely to be a problem very often since Elvis' forms are
	only cosmetic, not functional.  But it is possible to embed parameters
	in a link, and that's what triggers this problem.

	But there are other types of URLs which shouldn't be cached.
	Ideally Elvis should look at the reply header to detect these, and
	... maybe set a "nocacheable" option?

* Alias debugger would be nice -- maybe ":break alias foo" to stop at the
  :foo alias.  Note that :break already parses prefixes used by :map, so
  this would just need another prefix.  Plus code to make the prefix actually
  do something, of course.

* For "inputtab=identifier", the normal tag search doesn't look for tags
  in elvispath, so it doesn't find library functions.  Is that a good thing
  or a bad thing?

* MAYBE support "or color" for backgrounds.  Try to minimize the contrast with
  the normal background, instead of maximizing it, so the foreground colors
  still look good.  The "or color" backgrounds don't need to support images.

* In "man" mode, when markups are displayed the ^J is shown with a black ^
  and a highlighted (as "markup") J.

* Implement vim's "startofline" option.

* The showmatch option currently doesn't use matchchars.  Should it?

	Actually, it does use it to determine whether XML tag pairs should
	be highlighted by showmatch, but that's all.

* In "man" display mode, the .TS and .TE commands should not cause a blank
  line to be output.  If the user wants blank lines around a table, then 
  he/she should add .PP commands before .TS and after .TE.

	Actually, there seems to be a bug in the way multiple line breaks
	are handled.  .TS and .TE try to do just a line break, but if you
	use any macro that does a line break at the start of a line, then
	you get a blank line.  Elvis *should* ignore the line break request
	since the line has already been broken.

* The following doesn't look right in "html" display mode:

	<input type=submit value="Next day-->">

* Add commands for storing/retrieving a count argument.  This would be useful
  in maps.  For example, 33#! could store 33 in a register somewhere, and
  later #? could retrieve it for use by another command.

	##, #+	Add value to number
	#-	Subtract value from number
	#*	Multiply number by value
	#/	Divide number by value
	#=	Set number to value
	#<	If number is greater than value, then set to value
	#>	If number is less than value, then set to value
	#!	Set internal option to value
	#?	Retrieve internal option, use as count for next cmd.

	Alternatively, #a could store the value in cut buffer a, and then
	@a would naturally retrieve the value.

/ The tag stack doesn't seem to work right for user-defined protocols.
  Directories (the "readDIR" alias) fail completely when you hit ^T, and
  others such as readDICT move the cursor to the wrong place.

* Name completion for aliases isn't perfect.  If an alias has the same name
  as a built-in command, then Elvis counts both versions as being different
  possible completions.

* Move source into a src/ subdirectory.  This includes just about everything
  except data/, doc/, README.html, COPYING, INSTALL, HISTORY, and parts of
  configure and Makefile.in.

  * The "e2" script would also need to be changed.

* In URLs passed on the command line, anchors aren't handled correctly.
  Try "elvis 'README.html#2.2i'".  The same is also true of URLs passed to
  an ex command that expects a file name, such as ":e README.html#2.2i".

	There's a difference between URLs and filenames.  When elvis knows
	it is processing a URL it can be very smart about it.  But when given
	a local file name which happens to contain "#" how can elvis know if
	the "#" is part of the name, or an anchor?  And if it is an anchor,
	how can the generalized file loading code remember that it is supposed
	to search for that anchor?

	When following a link from another HTML file, URLs are handled like
	tags.  This won't work for URLs given on the command line, though,
	because there's no source document for the URL to be relative to,
	and also because the tags aren't going to be processed in "html"
	mode.

* Add support for vim-style jump lists.

	Jump lists are like the `` command, except that it maintains a history
	of jump locations.  Vim uses ^O to move backward through the jump list
	and <Tab> to move forward.  ^O is deliberately ignored in elvis because
	<Alt>+key and ":map visual" can both insert superfluous ^O keys before
	vi command keys.  The <Tab> key is used for moving to the next
	hypertext link in "html" mode, or for folding/unfolding text in
	"normal" and "syntax" modes.

	So elvis should use `+ to move forward and `- to move backward.
	Users who want vim's keys can map them.

* Should Undo turn off the "modified" flag, if it reverts to the most recently
  saved version?

	This would be another attribute in each buffer's Undo list.  Whenever
	an undo/redo version is added with "modified" off, any other options
	with "modified" off should be changed to show "modified" as being on.
	(Or better: whenever the "modified" option is turned off, all undo/redo
	versions should have their "modified" flag turned on.)

	It seems like the "edited" flag should come into play here too.

* Elvis' CHAR data type may clash with some standard ones.

* Add a help: protocol (readHELP alias) which looks up terms via :help.
  This way, the README.html file can have links into the manual.

	Tricky!  Too tricky for 2.2 at this late date.

* In syntax coloring mode, how about adding a text object for comments?

* For some users, the smartargs option is extremely slow and doesn't find
  the function's arguments anyway.  This has to be due to some quirk in
  the user's configuration, but what?

	An unsorted "tags" file?  This could be in the current directory,
	or any file or directory named in the "tags" or "elvispath" options.

	Or maybe it is sorted differently.  The "tags" file should always
	be sorted by the characters' ordinal values, regardless of the locale.
	Some versions of Exuberant Ctags use locale-specific sorting.

	See messages from Merijn Brand

	Would it be worth implementing the search code in elvis itself,
	instead of invoking ref, just for the sake of removing ref & bash
	from the scenario?

* Add a "rulerformat" option to control the appearance of the ruler.

	Vim has a "rulerformat" option with a very weird syntax.  I would
	prefer to use "simpler syntax" expressions.

	$1	current line number
	$2	current column number
	$3	offset from the beginning of the file
	$4	offset from the beginning of the line
	$5	the record/modify flag
	$6	the decimal code for the character under the cursor
	$7	the "list" version of the character under the cursor

	So the default format is equivalent to "$1($5?$5:",")$2".

? simoN reports that he gets intermittant "file's timestamp has changed"
  messages for no apparent reason.

? Sometimes "x11" doesn't refresh windows in response to an Expose event.

* If a visual selection is in effect and you use : to enter an ex command,
  then non-visual maps don't work while entering the ex command.  That's
  wrong.  This turned up as consequence of ":map! ^H ^OX" which should make
  the backspace key erase text immediately.  That mapping also has other
  problems -- it can't delete the first character in a line, for example.

* The :gui commands should be saved by :mkexrc.  Also, :gui! should make
  changes which aren't saved; the standard initialization scripts should
  use that.

* MAYBE add vim's glob(), globpath(), and expand() functions.  fnamemodify()
  too?

* Add a function to search for tags.  Its parameter string should be the
  same as the args of the :tag command.  It should return the tag as a
  comma-delimited list of name:attribute pairs; if there's more than one
  match then it should return a \n-delimited list.

	:"List the structs which have a "next" field
	:let t=tag("next")
	:for i = 1 .. t["\n",0]
	:do calc t["\n",i].struct

* Make the default array delimiter be '\n' if the string contains at least
  one '\n', or whitespace otherwise.  This will affect the :for command as
  well as arrays.  This will be handy when glob() is implemented.

	How does this relate to the idea of using {} to delimit nested
	arrays?  I guess I'd need to do the {} test before the \n test.
	One complication on the {} idea: The ex parser doesn't store the
	{ itself in the command, or the closing } either.  It does store
	the newline after the { though.  Maybe the parser could treat a
	newline as an opening quote for a string that extends to the end
	of the expression.

* Make the % operator concatenate strings with a newline between them.
  This will also allow you to say a[%,0] instead of a["\n",0].

* In elvis.syn, the "config.sys" entry reveals a problem: If a language
  has "ignorecase", and its keywords aren't given in lowercase, then
  elvis can dump core.

	I worked around this by changing those keywords to lowercase,
	but I'm not satisfied with that.  Elvis should never dump core.

* For the sake of vim compatibility, the X11 PRIMARY selection should be
  named "* instead of "^.  Also, it would be nice if "+ could be used to
  access the X11 CLIPBOARD selection.

* Allow the @ command to accept a count.  E.g., 42@x should run the
  contents of cut buffer x, 42 times.

* Find a better way for :map to distinguish predefined maps from those that
  should be saved by :mkexrc.  Currently, it omits any maps for keys that
  have a label, which isn't quite right.

* In Windows, the text-mode elvis doesn't switch screens anymore.

	This may only be in WindowsNT/2K.  There are some differences between
	consoles in WinNT/2K and Win95/98/ME/XP.

* Allow the search commands to accept a count.  E.g., 5/foo^M should
  search forward for the fifth foo.

* There are some quirks in incremental searching.  For example, the
  word completion macros don't work with incrementalsearch set.

* Add an "UnknownCommand" autocmd event.  It should be automatically
  invoked if an unrecognized ex command name is given.  After the autocmd
  completes, elvis should check again to see if a new alias has been
  defined with that name, and if so then no error is reported.  This
  will allow elvis to automatically load aliases as they're needed.

* Implement vim's "scrolloff" option.  It would be handy in the ("Eel) buffer.
  In vim this is a global option, but I'd like to make it a window option
  instead.  Also implement "sidescrolloff".

* Just a thought: If there was a "hidden" text attribute, then it could be
  used in man pages for text between .de and .., so macro definitions wouldn't
  be displayed.  Also, <script>...</script> and <style>...</style> could
  be hidden.

* If no normal background has been specified, or the background is set to
  "transparent", then changing the "background" option should cause a
  BgChanged event.  It should also re-evaluate all colors to possibly
  choose a different foreground.

* For smartargs, would it be possible to make the ',' character be overtyped
  only by another ','?  And maybe make any typed ',' delete text up to the
  next ',' in the doomed text.  That would be nice, because then as users
  type arguments, one typed argument will replace one smartargs hint.

	Modify syntax descriptions to store multiple function characters.
	For example, the line "function (,)" would tell elvis that '(' starts
	a function's args, ',' separates args, and ')' ends args.  If only
	two characters are given, assume the arg delimiter is space.  If
	only one is given, then don't treat any characters specially.

* Modelines set a buffer's options okay, but can't set the window's options
  as part of a :sp command because that window hasn't been created yet.

* It would be nice if ex/xbuf.ex could leave the other toolbar buttons
  unchanged.  Maybe allow multiple toolbars?

	I also want to support menu bars, and maybe a right-click menu.
	Maybe elvis should support generic widgets which can have an
	associated ex command (optionally with a form), and/or a list
	of other widgets to be displayed as menu items or toolbar buttons.
	The only difference between a menubar and toolbar would be their
	appearance -- the menu bar doesn't show icons.

	Widgets wouldn't necessarily be displayed anywhere.  You could
	define a large number of widgets that people *might* want to use,
	and then choose which ones serve as menu bars or tool bars, and
	which ones appear as menu items or toolbar buttons in those.
	New options would list the top-level menues, toolbars, and a
	right-click menu; some of those options could be buffer-dependent.

	Each widget should have the following attributes:
	* Label - the string used to identify it in elvis' config files.
	* Short - a text description of the item used in menus or button
		labels.  This defaults to the same as the label, with
		"_" changed to " ".
	* Long - a text description, displayed on the status line and in
		the dialog (if there is a dialog).  The default is the
		same as Short.
	* Sensitive - an expression which is TRUE if the widget can be
		clicked.  The default is TRUE.
	* Chosen - an expression which is TRUE if the widget should be
		shown as being chosen (i.e., a pushed-in button, or a
		checked menu item).  The default is FALSE.
	* Command - an ex command line to execute when clicked.
	* Dialog - A description of the dialog window to be created when
		the widget is clicked.  If the widget also has a Command
		then the Command is only executed when the Dialog's
		[Okay] or [Apply] button is clicked.
	* Menu - a list of widget labels, specifying the contents of a
		window to pop-up when the widget is clicked.  If there
		is also a Dialog and/or Command for this widget, then
		Dialog/Command is used for short clicks, and the Menu is
		used for longer clicks (or right-clicks).
	* Icon - a small image to use in toolbar buttons.  There is no
		default; the widget's short name is displayed instead.

* Implement the "redraw" option?

* If a URL contains percent signs or hash marks, it is hard to enter because
  elvis wants to replace those characters with the name of the current and
  alternate files, respectively.  At least, this is true with user-defined
  protocols.

* Make the ex/info.ex script use the "dir" page.

	See ex/info2.ex

* Traditionally, "-" has been a synonym for the "-s" flag, for running scripts.
  This is still part of POSIX, though is marked as being "obsolete" there.
  To make "-" be treated as a file name, you must give "--" before it, as in
  "elvis -- -".  Can I make elvis smarter about that?

	Maybe I could make "-" act like "-s" when stdin is a file, and read
	text from stdin otherwise?  Users could still use an explicit
	"-- -" to force the read-text behavior, or "-s" to force the script
	behavior.

	Sadly, there is no 100% portable way to detect whether stdin is a
	file or not.  The test would have to be done in an OS-dependent way.
	The isatty() function tests for character devices, not files.  The
	fstat() function uses funky names in Windows, and might not always
	work on non-files.

* Implement support for vim's CursorHold event.  (This occurs when the
  user doesn't cause any input events for a configurable amount of time,
  under control of the "updatetime" option, expressed in milliseconds.)

* htmltagatcursor() is still too slow in some contexts.

* For the sake of speed, dmmuadjust should probably not do any adjusting
  until the buffer is displayed.  That way, if there are *MANY* changes,
  we only need to adjust the display list once.

	Idea: Only regenerate when needed -- which mostly means while drawing
	a window.  For each window, keep track of the last offset that has
	been generated.  If a buffer is modified before that point, then
	reduce the window's offset to the changed point, and discard any
	saved formatting information after that.  When regenerating a window,
	start at the last saved point before-or-at the top of the screen,
	and work forward until the window is filled, saving points as they're
	generated.

	This should be at least as fast as the current method -- probably
	much faster.  More reliable too.

	The save points could either be the start of every line (which would
	make the points be dependent on the width of the window) or some
	specific tags which mostly force a new line to start (which would
	*NOT* depend on a window's attributes).  The "every line" method
	sounds simpler, but I'll need to remember to clobber all of the
	saved info if the display mode changes, or the width changes.

	Important comment:
		/* Start on the line BEFORE the one where changes begin.
		 * This is because if we insert a small word at the front of
		 * a line, it might fit at the end of the preceding line.
		 */

* The "syntax sh" display mode doesn't highlight instman.sh very well.  I
  think this is because dmsyntax.c:setup() allows strings to start at the
  end of a line.  If a string punctuation occurs at the end of a line, it is
  much more likely to be the *end* of a string.

* Make "x11" support a real bold+italic font.

* In X11, maybe use XTranslateCoordinates() to choose position of dialog window.

* Add a urlencode() function.  And maybe urldecode() too.

* Need a way to check options of other buffers.  For example, it would be
  nice if the toolbar in ex/xbuf.ex could test each buffer's "modified"
  option.

	Maybe use the -> operator.  The lhs would be a string identifying
	the buffer, and the rhs would be the label of an option to fetch.
	It would work by *TEMPORARILY* changing the bufoption value.

	This is mostly needed only in expressions, not the lhs of a :let
	command, because there you can use (bufname) to get a similar effect.

	local b
	let b = newbuffer()
	(=b) let bufdisplay = "syntax sh"
	calc b->bufdisplay

* It might be nice if the :man alias could search for the man source text,
  instead of recreating it from the nroff output.  Ideally, this would use
  a "manpath" global option (or just use $MANPATH, I guess) and a buffer-
  specific "mansection" option to store a list of sections and extensions.
  The mansection option is intended to address the problem of looking up
  write() in a C program -- it tends to find either the write(1) program,
  or the Python equivalent of the system call.

* Add an event when marks are set/reset.  This way users could...

	:au MarkSet * '[,']region marked '(aufile)
	:au MarkUnset * '[,']unregion marked

  ... to highlight the marked line.  Or maybe a generic "Normal" event
  which is triggered whenever a vi command is executed?  See vi.c:1089,
  and add a TWEAK flag to indicate which commands trigger events.

	:auevent MarkSet
	:au Normal m* '[,'] doau Markset

  I guess this would also require the :doau command to be extended to
  accept an address range.  That sounds useful all by itself!

* It would be nice if there was a way to stuff data into a cut buffer.  One
  example: Use it to copy the current file name into the display's selection.

  	Vim uses @x for this... and even uses the :let command!

  	:let "^ = current("file")

* Add stubs files for X extensions, OpenGL, etc.

* Maybe reimplement :g to work in two passes, like the real ex/vi.

* Add a buffer option for controlling the default face.  In addition to
  "normal" and "idle", some candidates would be "readonly" and "web".

* The '' command (go back to previous location) doesn't seem to work as well
  as it used to, especially after :s/// commands.  In Vim, ":help jumplist"
  says which Vim commands affect '', though I suspect that list is incomplete.
  It doesn't include "gD" for example.

	It would also be nice to support something like vim's jumplist,
	albeit with different commands. (Vim uses ^O and ^I.  I prefer
	Elvis' use for those keys.)  Maybe '+ to move forward through
	the jumplist and '- to move backward.  Maybe do something like
	viminfo to remember those marks between Elvis edit sessions.

	Some other useful marks: '< and '> for the ends of the current
	visible selection; '. for the position where the last change was
	make; [' and ]' to move to the previous/next line with a letter
	mark.

* Maybe implement something like viminfo.  It should store all the stuff
  that :mkexrc doesn't.  It should only be updated when a buffer is freed
  (which should certainly occur when Elvis is exiting, and maybe before
  that too).  At a minimum, it should store the most recent line number.
  It should ideally also store regions and folds.

	This might be possible with autocmds and scripts.  No changes
	to elvis necessary.  Check it out.  I know regions could be
	done this way.  Folds could be trickier, since they can nest.

* An idea: Support "invalidated" buffers.  Give each buffer a "valid" option.
  If it is invalid, then the next time that it is accessed, a BufInvalid
  event is generated, which should update the buffer.  This could be used
  (with a little extra magic, like a :internalsave command and :invalidate
  command) to update the "Elvis custom initalization" buffer.

	This is probably more trouble than it's worth.  My original goal
	was to provide a way to efficiently update the file-specific
	information for viminfo.  Then I tried to make it generic since
	:mkexrc's "Elvis custom initialization" buffer works in a similar
	way.  But this is all low-level stuff that doesn't really correspond
	to ex commands; it would be a forced fit, not a natural fit.
  
* Apparently "elvis -Gx11" doesn't always exit when X does -- it can get
  stuck in a loop where it uses all available CPU.

	Xlib normally kills the application when the connection to the server
	goes down unexpectedly.  Elvis will sometimes disable that behavior
	during initialization -- perhaps it doesn't enable it afterward?

* Can the new "prefix" keyword thing be extended to denote comments?
  And the %rX...X regular expressions in ruby?  Maybe recognize the start
  of a comment/regex as a prefix token, and associate a terminator string
  and font with it.  Maybe it could even handle preprocessor directives!

* Windows problems:
  * In Options->Font, the "Font style" is ignored -- can we remove it?
  ? WinElvis died one time while resizing, when it tried to refresh the screen.
    I haven't been able to reproduce this.

? Is there a bug in Windows reading the initialization scripts?  In particular
  maps that use the ^M character?

	I can't reproduce this.  It works correctly for me.

* Potential for some minor problems with parsing regular expressions, when
  magicchar is set to something weird.  Specifically, some parts of regbuild()
  and gettoken() assume that '[', '^', and '$' are always in magicchar.  If
  they aren't, then odd (though non-fatal) things will happen.

* Add support for vim's \= in replacement text.  It treats the rest of the
  replacement text as an expression, evaluates it, and uses the result as
  literal replacement text.  E.g., :$s/^/\=current("line")/ inserts a line
  number at the start of every line.  Maybe use $1 - $9 for \1 - \9, but
  how to handle & ?

* Some of the special characters in "man" should be graphic, but aren't.

	This is somewhat improved in 2.2h, but still needs work.

* Add a "regexp" (or "magicsyntax" ?) option which has no built-in purpose
  but can be used with ":autocmd OptChanged regexp ..." to automatically
  adjust the syntax.

	auevent RegexpChanged
	au OptChanged regexp eval doau RegexpChanged (regexp)
	au RegexpChanged perl set magicchar=^$.[*+?(){\| magicname magicperl
	au RegexpChanged vi set magicchar=^$.[* nomagicname nomagicperl

	Actually, it might make more sense to allow users to create their
	own options.  "regexp" could be an instance of that.

* In regular expressions, maybe allow all of the \letter combinations
  have configurable meanings.

	One way to do this would be via 52 magic<letter> options, with the
	value of the option being substituted for any \<letter> sequence
	used in a regular expression.

	This could interfere with the use of \Q, \V, and \E.  I should
	leave those hardcoded, with \E selecting the enhanced mode that
	is sensitive the the magic<letter> options.

	If I'm going to do all this, then the regular expression options
	should probably all be collected into a "regexp" group.  There will
	be a lot of them!

* In regular expressions, allow [[:startword:]] and [[:inword:]] to be used
  for accessing the syntax-coloring concept of "word".

* Maybe add a way to prevent :w from saving some types of regions.  This would
  be handy for a visual diff -- you could show both the old and new versions
  of lines mixed in the same buffer, and then save only the new text.

	This could be done via a macro, but I'm worried that it'd be too easy
	to accidentally bypass the macro, causing the mixture of both versions
	to be written to the file.

	Also, it might be nice if the "no save" lines could be hidden by
	setting an option.

	I wonder... would it be adequate if I simply made elvis sensitive
	to the special face name "nosave"?

* When editing the source-code for man pages, there is a bug in the handling
  of { and } paragraph commands, which affects the "ap" and "ip" text objects:
  If the cursor is on the '.' of a .PP or .SH command, then the paragraph seems
  to include the PRECEDING text instead of the CURRENT AND FOLLOWING text.

* Compatibility issue: The `m command should move the cursor back to the
  same column, not the same character.  I.e., it shouldn't care whether any
  text has been inserted or deleted on that line before the mark.  To test,
  go onto a long line and type 40|mm20|10x`m  -- this should leave the cursor
  in column 40, but elvis moves to column 30.

* :10,20s/foo/bar/3 should change lines 20 through 22 -- i.e., 3 lines starting
  at the last line in the range.  Elvis ignores the count when given a range,
  which is wrong.

* :g/^/j should join pairs of lines (it does in the real vi and in vim), but
  instead it joins all lines together into a single line.

* The S command looses the last indentation character when autoindent is set.
  Also, c_ (which is what S does) works the first time, but . doesn't, if the
  new text was empty.  If the new text was not empty, then . looses indent.

* When pasting rectangles at the ends of lines, elvis doesn't try to pad
  the lines so the rectangle keeps its shape.  Vim does.

* <img src=...> tags don't work except in HTTP: URLs?

* If multiple <img src=...> tags occur in a line, *not* as a menu, then
  elvis should load the image when ^] is pressed on one of them.  Instead,
  it always seems to load the first image in the line.  (Though I'm not 100%
  sure I understand what's really happening.)

* Change the meanings of some commands when a visual selection is pending:
	* s	Like c (rect repeats the change for each line)
	* C	Change whole lines (except rect changes from left edge)
	* S	Change whole lines
	* D	Delete whole lines (except rect deletes from left edge)
	* X	Delete whole lines [:map select X :d^M -- but cut buffers?]
	* Y	Yank whole lines [:map select Y :y^M -- but cut buffers?]
	* I	Insert at left edge (for rect, do it to each line)
	* A	Append at right edge (for rect, do it to each line)

* Some useful 'g' commands that elvis doesn't support yet:
	* gv	reselect previously highlighted text
	* ge	go backward to end of previous word.
	* gE	go backward to end of previous WORD

* Some useful 'z' commands that elvis doesn't support yet:
	* zh	sidescroll leftward.  If cursor forced off right edge, move it.
	* zl	sidescroll rightward.  If cursor forced off left edge, move it.
	* zs	sidescroll to move cursor's column to left edge
	* ze	sidescroll to move cursor's column to right edge

* The :fold command seems to confuse the :g command, perhaps because :fold
  changes the meaning of "line".  The :unfold command is okay, though.  In
  particular, this means you can't do this...

	:g/^{/,/^}/fold (current("tag"))

  ... but this works...

	:g/^{/,/^}/unfold (current("tag"))
	:%fold!

? The spell option should probably look for tags in more places.  Currently
  it doesn't seem to check any tags file except the current directory.

	Actually, it was apparently checking only the directories listed in
	the "tags" option.  Perhaps it should also check "elvispath"?

	If there is no "tags" file in the current directory, then it doesn't
	look in other tags either.  I remember this was a design decision,
	to speed things up in non-source directories.  I might want to make
	this an option, or make it sensitive to knownsyntax(), or something.

	NEWSFLASH: I think this issue is moot, since the new taglibrary
	option was created.

* Extend the "ref" program to limit itself to looking at tags that are
  callable from a specific source file, by checking the extensions against
  the "extensions" or "foreign" set, just like tagkind/taglibrary.

* Make elvis give a higher priority to tags that are callable by the current
  source file, using the same filename sensitivity.

	This would be a little tricky, since the list of extensions is
	currently stored in a static variable in descr.c.

* Make "tagkind" option store a separate tag/keyword dictionary for each
  buffer.  This will handle static tags, and also allow the keywords to be
  reloaded when the "tags" file changes.  (As currently implemented, tagkind
  adds tags to the language's shared keyword dictionary.)

	One big challenge: Currently, the list of filename extensions which
	are callable by the language is stored in a static variable in descr.c
	This works when the search only takes place as a language is loaded,
	but if we need to reload later, then we'll need to store that list
	somewhere -- probably attached to the language description in descr.c

* When lpcolumns!=columns, printing will often use the wrong font attributes
  when printing text in a markup display mode.  Apparently the array of
  line attributes is tripping it up.  NEWSFLASH: It can cause core dumps too.
  Try setting lpcolumns=300, for example.  This is especially true when using
  antialiased fonts in "x11".

	This should go away with the new markup code (when written).
	Until then, just be sure to ":let lpcolumns=columns" before printing.

* Maybe implement a multi-level progress bar.  Search/replace commands would
  use it, as would network I/O.   Aliases could also use it via a ":progress
  name=percent" command.  The progress is reset (goes away) whenever elvis
  is idle, or when the progress hits 100%.  While reset, the first named
  progress grabs it, and holds it until reset again.

	This should probably wait until window-at-a-time idle is supported.

* hlsearch doesn't mark all matches within a markup display mode.  Its
  behavior is very consistent, in that it fails to highlight the same
  instances each time.

	This would be hard to fix -- probably not worth the effort at this
	time.  The ideal solution would involve modifying regexec() to search
	across line boundaries.  The problem is caused by the fact that
	markup display modes have a different definition of "end of line"
	than the regexp matcher, so any text between a newline character
	and the end of the "markup" line is unchecked.

* Giving too many "Hit <Enter> to continue" prompts.

* "spellsuffix" doesn't always work.  It seems to skip looking for base words
  defined via :words -- it only checks the "spelldict" dictionary.  Example:
  "GUI" is added via :words, and spellsuffix contains "'s", but "GUI's"
  is still flagged as an error.

	I believe there are two things happening here.  First, it doesn't
	look for the root word in the tags dictionary; this allows lowercase
	"digraphs" to be flagged as an error even though "digraph" is accepted
	because it is a tag.  This is not a bug.

	The second thing is that it looks up the root word in a case-
	insensitive way (converting it to lowercase) so it doesn't find
	the uppercase "GUI" when trying to check "GUI's".  This is a bug...
	how to fix?

? From Stephane Billiart:
  There is a bug in the "configure" script -- it doesn't know that the
  internet files are required even if the network protocols are not
  used, with the "x11" interface.  Here's a patch that fixes it for Solaris...

	529,531c529
	<       why "   to ensure that 'const' is ignored.  Also, if X11 is to be supported then"
	<       why "   we also need -lsocket -lnsl.  A -R flag may also be added so the X11"
	<       why "   shared libraries can be found at run time."
	---
	>       why "   to ensure that 'const' is ignored."
	535a534,535
	>               why "For Solaris2, X11 requires -lsocket -lnsl.  A -R flag may also"
	>               why "   be added so the X11 shared libraries can be found at run time."
	538a539,542
	>       elif [ X"$inet" != X ]
	>       then
	>               why "For Solaris2, any network access needs -lsocket -lnsl"
	>               TLIBS="$TLIBS -lsocket -lnsl"

  Okay, here's my question: How could the network files be present but not
  used?  The "configure" script always uses HTTP and FTP if the network files
  are present, so that should always cause those files to be included under
  Solaris, right?  Or does Solaris sometimes have the libraries but not the
  header files?

------------------------------------------------------------------------------
FIXED BETWEEN 2.2j and the official 2.2.0 release

/ Use x11's colors as the defaults for WinElvis too.

	Actually there are no defaults for XElvis either.  I should probably
	add a data/elvis.clr script which sets up default colors for both
	graphical elvis and text-mode elvis.  The ~/.elvisrc file can be a
	good starting point.

/ The data/stubs/unixstub.c file seems to be very incomplete.

/ "make install" should be forgiving about the /etc/elvis directory.
  If a user runs "configure --prefix=$HOME" and installs elvis as themselves
  (not "root"), then "make install" shouldn't fail because it can't write to
  the /etc directory.

/ The readDIR alias loads the new buffer into the old window.  This causes
  problems with the stack, or with :sp commands.

	This is similar to the problem I had earlier with the rcssince
	alias, which was run via an autocmd event.  I solved that by adding
	a rule that autocmd events can never switch buffers.  The same
	solution would probably work pretty well for protocol aliases.

/ In Windows, the standard COMMAND.COM shell does not expand wildcards,
  but Cygwin's bash.exe does.  Elvis shouldn't expand wildcards if they've
  already been expanded.

	Obviously elvis will need some way to tell the difference --
	perhaps by checking for a SHELL environment variable.  See the
	following.

/ Wildcard expansion is closely tied to handling of parentheses.  It would
  be nice if, during wildcard expansion of command-line arguments from the
  shell, we could skip that stage.  Maybe create a new option?  Set it during
  processing of the command-line args, and then turn it off before processing
  the configuration scripts.  This would also prevent parentheses from being
  eaten in file names such as "file (2).txt".

	A more general option might be nice -- one that lets you control
	the aspects of filename handling in detail.  Its value could be a
	subset of "tilde,dollar,paren,wildcard,special,space".  "special"
	allows interpretation of %, #, and (if supported) <word>, as well
	as the :x modifiers that can follow those symbols.  "space" indicates
	that commands which accept multiple names should divide the name at
	any unquoted space; without that, you could only separate the names
	with <tab> characters.  Call the option "filenamerules".

	So, for Unix/Linux, the command line args would be interpreted using
	filenamerules="".  For Windows arguments (without $SHELL)
	filenamerules="tilde,dollar,wildcard".  After that, the default is
	always filenamerules="tilde,dollar,paren,wildcard,special,space" --
	the full set -- but you can change it in ~/.elvisrc.  Windows users
	might want to take out "space".

	Name completion should insert a tab after a complete name instead of
	a space, if the "space" flag isn't given.

/ If the user tries to get help for an alias defined with a '!' then look
  in the elvistip.html file.  Otherwise just say it is an alias.

/ When told to read a directory, elvis should pretend it was told to read
  using the "dir:" protocol.  Users can define a readDIR alias to allow
  directory browsing.

/ In X11, :mkexrc can save italicfont="".  It shouldn't.

/ Elvis' BOOLEAN and data type may clash with a standard one.  Change its
  name, and the names of the True and False values.

/ Elvis' ctype macros clash with the standard ones.  This is a problem in
  some GUIs because they use headers which #include <ctype.h>.  Change the
  names of elvis' ctype macros.

/ In "x11", <Shift-Insert> adds ^Sff63.  Is this a general problem?

	<Ctrl-V><Shift-Arrow> starts marking in addition to sending ^Sff51.
	(It is normal for <Shift-Arrow> to start marking text.)  Probably
	<Shift-Arrow> shouldn't send anything.  Even weirder: It starts
	marking in the main buffer even if you type <Ctrl-V><Shift-Arrow>
	in an ex command line.

/ Use <key> notation for maps in elvis.ini, and :mkexrc.  Not only is this
  more readable, but it also avoids fileeol() confusion when reading maps
  that end with ^M.

/ I think the region bug may not be completely fixed.  The "since.ex"
  script eventually causes problems, sometimes, when it tries to add a
  region for the current line when the line is changed.

	I think this is related to "undo", which can move marks.
	It could leave regions unsorted.

	To reproduce it: Make the second paragraph if this file be a
	region.  Delete the its last line.  Hit <Undo>.  Hit <Shift-O>
	to insert a line above the last line.  Wham!

/ With "matchchar=ax showmatch", an empty tag pair does not get highlighted
  because the ix object would have zero length.

	A general solution for this might not be possible.  If matchchar=ab
	then typing {} won't cause the { to be highlighted either, for the
	same reason.  Is it worth implementing a special case just for ax?

/ Draw the screen-filling '~' lines in the "nontext" face.

/ In doc/elvistip.html, the old "spelltags" option is mentioned.  Update
  description there to say what elvis really uses now ("spell").

/ The :autocmd system's DisplayEnter and DisplayLeave events are quirky.
  If you use DisplayEnter to adjust the "list" option for various display
  modes, then sometimes the screen is refreshed before the option is
  changed, so the screen is wrong until you hit a key.

	The audisplay() function detects this type of change.  As
	currently implemented, the DisplayEnter/DisplayLeave events
	are oriented toward adjusting maps in a display-dependent way,
	so audisplay() is called before processing keystrokes, and
	detects global changes to the display mode, not changes to a
	specific window.  This is useful, but doesn't work well for
	adjusting options that affect the way text is displayed.

	/ Rename the current DisplayEnter and DisplayLeave to DispMapEnter
	  and DispMapLeave.  Also rename the audisplay() function that is
	  used to trigger it.

	/ Create new DisplayEnter and DisplayLeave events that are triggered
	  when a window's display mode changes.  The DisplayEnter event
	  should also be triggered when the window is first created.  Basically
	  this should be triggered when win.display is changed.

	/ Extend the listchars option to support a "markup" flag which allows
	  "list" to display markups.  The presense of "markup" indicates that
	  markups should be listed; its absence indicates that markups should
	  be hidden by "list".

/ Making :s always succeed in a :g turns out to have been a bad idea.
  Some scripts and aliases, such as :makehtml, depend on errors from
  :s when there is no match.  I should implement vim's "e" flag.

/ Need a good way to display sets.  Write a :wrapset alias to do that.

/ Modify name subscripts to support sets.  L.name and L["name"] should
  return "True" if L contains name without a value; else it should return
  the value.  For missing names it should return "", which is a false value.

	When assigning to L.name, always store name:value, never just name.
	If the option contains the name without a value, then the name should
	be deleted from the value.

/ Add set operators that act on comma-delimited lists of names.

	L & R	Return the names that appear in both sets.  If values are
		preset, then keep the value from R.
	L | R	For each name in b, add it to a if it isn't already in a.
		If values are present, then keep the value from R.
		When adding values, try to insert them to keep the a sorted.
	L ^ R	Remove items from a if they appear in R.  (Can't use L&~R
		because there is no way to implement ~R.  Can't use L-R
		because we need ""-R as a way to handle negation.)

/ How about adding support for ":let <option> <operator>= <expr>" be
  interpreted as ":let <option> = <option> <operator> (<expr>)" ?

/ In :autocmd, the file pattern field can't be a comma-delimited list of
  patterns anymore.  I think I broke this when I modified it to use the
  more versatile OS-specific wildcard functions.

/ Allow dirperm() to return DIR_DIRECTORY for directories.  This should go
  between DIR_NOTFILE and DIR_NEW.

/ With ":set list listchars=trail:_", the trailing space thing looks
  really distracting while you're entering text.  If the cursor is
  located in the trailing space or at the end of the line, then the
  spaces before the cursor should not be highlighted.

/ Now that "html" hides comments better, it can't handle <!DOCTYPE>.
  It mistakes the <! for the start of a comment, which extends to the
  end of the file if there is no ->.

/ In WinElvis, f[/,2...] doesn't strip off the leading \ character as it
  should.  As a simpler case, ":calc /" returns "/" in Linux, but "" in
  Windows.  I think this is a bug in dirpath().
  
/ There's another bug in WinElvis which is triggered by ":e theme:".
  Something is converting the forward slashes into backslashes, which is
  a VERY BAD IDEA for URLs.

	The backslash conversion is being done by dirnormalize() and
	dirpath().  Unfortunately, external commands MUST use backslashes
	and protocols MUST use forward slashes.  Elvis' built-in file
	operations can use either style.

	If I try to make dirnormalize() and dirpath() be smart about when
	to use slashes or backslashes, I'm going to run into problems.
	In scripts, if nowhere else.  The readTHEME alias, for example,
	modifies the resource name, which doesn't look like a URL so it
	could easily be converted to use backslashes, accidentally.

	Probably URLs should accept either normal slashes or backslashes,
	by converting backslashes to slashes at a very low level.  The
	dirXXX() functions should recognize either type of slash, and output
	backslashes.

/ The readTHEME alias doesn't work because dirperm() and related functions
  don't recognize a leading ~.

	I modified the alias to explicitly convert a leading ~ into the
	value of the home option.

/ Add an option to make <backspace> get rid of backspaced-over text.

	I did.  It is named "cleantext", and allows a large degree of control
	over when elvis erases old text.

/ Is it time to add a substitution (or general ex command) function?

	For a PERL-like syntax, using "=~" instead of just "=" with a :let
	command should perform the ex command.  So we could replace a leading
	~ with the user's home directory like this:

		local magicname
		let a = elvispath[:,1]
		let a =~ s/^~/${home}/

	Note that applying an ex command to a value is a separate step,
	not part of the normal expression.  I'm not sure I like that, but
	it doesn't seem to have held Perl back.

-------------------------------------------------------------------------------
Fixed in 2.2j-beta:

/ In "x11", if you have two windows, with one iconified, and you close the
  non-iconified one, then elvis dies with an X11 error: BadDrawable for an
  XCopyArea call while trying to change the cursor in the iconified window.

/ In the "Tips" chapter, the URL for vile is out of date.

/ Clean up data/elvis.rc -- in particular the "maroon" thing.

/ ftp.false.com is no longer mirroring ftp.cs.pdx.edu.  The README.html
  file shouldn't mention it.

/ Implement text objects for XML: ax/ix.

/ Make "showmatch" match up XML tags.

/ Weirdness: If you edit a file with [a-z] at the end of the name, then one
  character is often highlighted until you make the first change to the file.
  Starting a visible selection also makes it go a way; canceling the selection
  makes it come back.  This doesn't seem to happen when you give the whole name
  without [a-z], or use [a-z] elsewhere in the name.

	This seems to happen only when showmatch is enabled.  I think elvis
	is computing offset of the [ that matches the ], and storing its
	highlight offset for the window.

/ ":themes" with no argument can't list themes in Windows.

	It's probably because shell("echo *.ex") doesn't expand filenames
	in Windows.  The quickest solution would be to make "for i in *.ex"
	smart enough to treat each match as a separate value, instead of
	evaluating an expression.

/ If a filename pattern ends with [a-z] then there is no NUL character to
  mark the end of it, so garbage is appended to the end of the name.  This
  happens under both Windows and Linux.

	This turned out to be a bug in the expression evaluator, not the
	wildcard matcher.  In calc.c, the simpler syntax version of ']'
	wasn't being handled correctly.

/ Support Vim's <key> notation in maps. (But not in the :normal command?)

	This is now mostly implemented.  the GUI-specific key name translation
	need more work.

/ Increase the whitespace around printouts when lptype=ps.

/ Extend matchchars to support text objects.  For example, if matchchars
  contains "ax" then % should bounce between the endpoints of an XML tag pair.

	First, handle the #if/#else/#endif case.  This doesn't use matchchars.

	Then, prescan the matchchars value, looking for any text objects.
	If any are found, find their endpoints by calling vitextobj().
	Remember the one that has an end point with the lowest offset.
	Collect the non-textobject pairs in a different string.

	Then scan forward for a matchchar, as usual.  The only difference
	is that if our scan reaches the end of the text object without
	finding a matchchar, then we'll use the object instead.

	When using an object, the cursor should move to start point,
	unless it was already at the start point in which case it should
	move to the character before the endpoint.

/ Replace lppaper with a more general lpoptions option.  Its value is a
  comma-delimited list of name:value pairs.  For lptype=ps and lptype=ps2
  (the only lptypes that use lppaper), the values are used to insert some
  definitions into the PostScript output.  Perhaps other drivers can benefit
  from it too.

	lpoptions=paper:letter,frame:false,bar:true,clip:true

/ Add a generic option to both "buf" and "win".  Describe how to use
  subscripts and field names, as a cheap way to store extra buffer-specific
  and window-specific options.  Maybe use "b" and "w"?  Or "B" and "W"?
  Or "bb" and "ww"?

/ Mark the "initialstate" option as having no default, so it won't be saved
  by :mkexrc.

/ The "html" display mode should do a better job of parsing comments.  HTML
  comments are often used to temporarily hide text which may include other
  HTML tags.

	This is now improved in "html" mode, but not "syntax html" mode.

/ Sometimes the :sp command loads the new file into both the new window
  and the old one.  The old window's previous buffer is hidden but not
  destroyed; you can use the :buffer command to switch that window back
  to the original buffer.

	This is known to happen when the "rcssince" alias is run automatically
	via ":au BufReadPost * rcssince".  Maybe not always, though.  I can't
	see anything in the rcssince alias that would change buffers.  Maybe
	it is some other autocmd command?

	Specifically, if you split to a file that has an RCS "*,v" file,
	but hasn't been modified since its last checkin, then the rcssince
	alias will execute an empty list of :region commands.  An empty
	list is treated like a :goto command to the current buffer and line.
	But the current buffer is the new buffer, in the old window, and
	that's where the problems happen.

	New rule: autocmd commands can never switch buffers.

/ Regions can cause core dumps.  Either a region struct with bogus data is
  added somehow, or a good region struct is overwritten with garbage.

	I have also seen instances where regions near the end of the file
	are freed or truncated when they shouldn't be.  Maybe the bogus
	region struct is really one whose struct has been freed but not
	removed from the struct list?

	Fixed!  There were some problems in :chregion, but I think the
	biggie was in an argumentless :region command which could overrun
	an output buffer, clobbering the "r" variable that it used for
	scanning.

/ Sometimes elvis fails to find tags, even though the tags clearly are in
  the "tags" file.  Try searching for colorfind() from within region.c.

	This turned out to be a bug in the dmsyntax.c:tagatcursor() function.

/ Allow hllayers to be set to higher values.  Note that it already is smart
  enough to recycle the "hllayers<n>" fonts.

/ Improve wildcard matching in :au commands.  Move the wildcard matcher into
  os*/osprg.c, so it can be OS-specific.  Note that the unix version already
  contains a wildcard matcher.

  / The [] wildcard matcher is slightly improved.  It now supports negation
    and ranges.  For example, [^0-9] matches anything but a digit.
  / This also allows Windows to support the [] wildcard.

/ Make name completion work for aliases.

/ Add buf.bb and win.ww options.  These should be generic string options
  with no hardcoded meaning.  Users can pack multiple values into them
  using named subscripts.

/ Modify the named subscript notation to use a comma-delimited list of
  name:value pairs instead of a space-delimited list of of name=value pairs.
  This will make elvis' named subscript notation work well with vim-style
  multivalued options such as listchars.

/ When assigning a value via a named subscript, if the value doesn't exist
  then append it using a comma separator.

/ Add vim's listchars option.  This affects the appearance of both the "list"
  and "wrap" options.

/ The exists(filename) function assumes that any filename with a protocol
  already exists.  This confuses the rcssince alias.

/ The "inputtab=identifier" feature (for tag name completion while editing
  source code) doesn't seem to work.

	This appears to have something to do with storing the names of
	library functions and tags as pseudo keywords, for the sake of
	highlighting.  Anyway, the dictionary seems to contain a lot of
	words that it shouldn't, and that's a problem.

	I fixed something extremely similar between revisions 2.91 and 2.92
	of dmsyntax.c.  This is what made me add the ISKEYWORD flag.  This
	fix affected smartargs, not inputtab=identifier.

/ Patch "configure" for NetBSD.  See message from Peter Bex.

/ Remove the :safer command.  It can be emulated via ":safely source".
  Add an alias to do that, to maintain backward compatibility.

/ Replace "safer" and "safest" with a single "security" option which
  can be set to multiple levels: normal, safer, restricted.

/ Split the "lib" directory into two separate directories: "doc" for
  documentation including the help files, and "data" for other runtime
  support files.

  / Move the files and delete "lib".
  / Edit "configure" to support --docdir=DIR, and to make the default
    elvispath include both the datadir and the docdir, and also /etc/elvis.
  / Edit "Makefile.in" to look for doc and data files in the appropriate
    directories, and to install them into the right directories.  Also,
    "make install" should create /etc/elvis if it doesn't exist already.
  / Edit elvis.spec.in as necessary.  In the RPM worldview, online help
    files are like data files, so "doc" is merely a subdirectory of "data",
    which limits the changes.  Also add /etc/elvis to the list of files
    owned by elvis.
  / Modify the "e2" script to set ELVISPATH=data:doc instead of just lib.
  / Update the manual

	This is basically done, except that the RPM package should ideally
	install the contents of the doc/ directory into RPM's standard
	location for documentation, and the default ELVISPATH should be
	adjusted accordingly.  How can I do that?

/ Rename data/ex to data/scripts.  I considered renaming the :load alias to
  :runtime, but decided against it since it really isn't like vim's :runtime.

/ Clean up the icons.
  / Move the XPM icons from guix11/ to data/icons.
  / Remove the "insticon.sh" script.
  / Make Xelvis search for icons in data/icons.
  / Create an "iconimage" option for specifying the option image, instead of
    ":gui icon filename".  This is because options are stored by :mkexrc
    but :gui commands aren't.
  / Make WinElvis support XPM icons too, via the "iconimage" option.
  / For the Windows and OS/2 binary distributions, convert the XPM icons
    to ICO format.
  / Convert the "Gray Presley" Windows icon to an XPM file.  Don't compile it
    into WinElvis anymore.  Remove instructions for switching between icons.

/ Using "-c'1d|wq'" doesn't exit until you hit a keystroke.

	Most of the user interfaces are pretty dumb about "quit" commands
	during initialization.  They should be smarter.  The problem is
	that "quit" commands don't really quit the editor -- they merely
	close the window.  This shouldn't be an insurmountable problem,
	though, since the "-c" commands are executed after the window has
	been created.

/ The "lpcontrast" option doesn't affect ptype=html or ptype=ansi output.
  This is a good thing since elvis doesn't know what the background color
  will be, but the manual should explain this.

/ The > operator and :> command should not affect preprocessor lines.

/ In syntax display mode, strings should have a configurable quote character.
  It isn't always backslash!   Some don't have one at all.

/ Make syntax description lookups be less case-sensitive.  They should first
  look for a case-sensitive match, and if that fails then try converting the
  filename to lowercase and repeating the search.  This way, "extension .sys"
  would work for the CONFIG.SYS file.

	This turned out to be trickier than I expected, because elvis stores
	previously used descriptions in a cache.  (The cache allows you to
	bounce between two C source files without reloading the C syntax
	each time.)  If this was case insensitive then the c syntax would
	be reused for ".C" files, which are really c++ source code.

	Also, reading the file twice would have been difficult, so I
	simply read it once, and arranged for the c++ entry to appear
	before the c entry, so ".C" will always use the C++ syntax.

	All extensions in the elvis.syn file should now be lowercase,
	unless an uppercase version is intended to be different from the
	lowercase version (as with .c for c source and .C for c++ source);
	in that situation, the extension should be given in uppercase form
	and its language should be described before any language with a
	lowercase version of the same extension.

/ If the spell option is off but the show option contains "spell", then
  the statusline just shows "?" for any tagged word in source files.
  The manual should warn people about this.

/ Implement a "scrollbgimage" option, which allows the background image to
  scroll in XElvis and WinElvis.

/ Move the "BUGS" file into doc/bugs.txt.  Check for any reference to "BUGS"
  in the manual or README.html.

/ Change the user-specific config directory from "~/.elvislib/" to "~/.elvis/".

  / Edit the "configure" script.
  / Edit the Win32, MSDOS, and OS/2 osconfig.h files to use "dotelvis" instead
    of "elvislib".
  / Edit the aliases in data/elvis.ali and data/ex/* to use the new directory.
  / Update the manual.

/ Modify the :mktheme alias to store the scrollbgimage option.

/ When recognizing HTML documents, elvis should allow whitespace before the
  "<".  This is easy for spaces and tabs, but could be a problem for newlines.

/ The :display command should strip trailing whitespace.  The way it is now,
  elvis -c 'dis s email | se bd="syntax email"' doesn't work because of the
  space before the '|'.

/ Add a "safest" option.  Just plain "safer" is good for protecting against
  malicious writing by a Trojan horse, but not against malicious reading by a
  hacker.  On the command line, -SS should set both options.  If both "safer"
  and "safest" are set, then...
  / The :read, :source, :safer, commands should fail
  / The :errlist and :wordfile commands should refuse a filename argument.
  / Filename completion should be disabled
  / Any :calc functions that examine files should be disabled.  This includes
    dirperm(), exists(), fileeol(), and time(filename).  In addition, the
    shell() function might as well be disabled so we can get a better error
    message.
  / User-specific set up files should be skipped.  In particular, elvispath
    should remove any element that starts with "~", and the elvis.ini script
    should skip $EXINIT and all the .exrc files.

/ The "safer" option should disable more...
  / The wordfile option should be locked.
  / Suspending via ^Z should be disabled.
  / The :snext, :split, :visual, :open, :push, :wquit, :xit commands should
    refuse any filename argument.
  / The :ex, :edit, and :file commands were disabled by safer, but should now
    only refuse an filename argument.  ":f" and ":e!" should work.
	No, ":e filename" should work too.  It's too useful!  Disable it
	in the paranoid version though.
  / The elvis.ini script should be run with "safer" turned off temporarily,
    even if "-S" is given on the command line.
/ HTML display mode should handle &euro;

/ The "timeout" option isn't documented.

/ "configure" still isn't setting elvispath correctly.  It has $(PREFIX)
  in the literal string.

/ Modify Makefile.in to incorporate $VERSION into the documentation files.

/ In "html" display mode, comments should allow ">" in the comment, and
  only end the comment at "-->".  Also, if the comment is too long to fit
  in the tag buffer then the comment contents should be clipped but the
  whole comment should still be skipped.

-------------------------------------------------------------------------------
Fixed in 2.2i-beta:

/ Bug in syntax highlighting: In Pascal code, if { comment } is a line
  two lines above the top of the screen, then the whole screen is formatted
  as a comment.

/ Add a typical ~/.elvisrc file to the distribution.

	Added it, as lib/elvis.rc

/ For WinElvis, offer an alternative title format -- the current hard-coded
  format is "WinElvis - [filename.ext]", which doesn't look so great in
  the task bar.

	I added a "titleformat" option.  I still need to test this,
	and if it works then I should document it too.

/ The MS "nmake" program allows forward slashes in filenames, which is nice,
  but if there's an error then the compiler will output the character as /\
  instead of just / or \, and that can confuse elvis.

/ In guitcap.c, allow colors to be used if :up=\E[A: *or* :AF=\E[3%dm:
  This is necessary for the "screen" program to get colors.

/ The -= operator doesn't work right in :color commands.

/ Make the "warpto" option move the pointer when a window is created.  This,
  together with "warpback", should avoid the focusing bug.

/ Elvis/x11 shouldn't always raise the window when changing focus.
  In particular, when [un]mapping windows.  This is important because some
  window managers (such as IceWM) iconify all windows when switching to
  a different workspace and then deiconify all of them when switching back.

	It seems like this should be related to the xw->warp flag --
	If a window shouldn't be raised, then the cursor shouldn't
	be warped to it either.

/ In the "termcap" interface, when scrolling forward the showmode box is
  highlighted; when scrolling backward it is not highlighted.

	I fixed this by adding a new "showmode" face, and using it to draw
	the showmode value.  You can now use ":color showmode green boxed" 
	to change its attributes.

	While I was at it, I also added a new "ruler" face for changing the
	color of the ruler.

/ In input mode, make ^X^X lock hex input mode until the next non-hex.  If
  the non-hex is ^X then it should be ignored; any other character should
  be interpreted normally.  Or maybe Backspace should remain in hex mode?

/ Rename the :w alias to :ww, so people who don't use RCS aren't baffled
  by unexpected error messages.

/ The smartcase option can break some aliases.  To get around it: add
  "local noignorecase" to the alias.

/ Blended fonts aren't always right.  In particular, a :region's font
  isn't combined with non-default fonts correctly.

/ Make WinElvis support background images.

/ WinElvis has some resource leaks.  This shows when you resize the window
  or do a lot of scrolling.

/ The "ref" program doesn't like it when text files have CR-LF newlines on
  a Unix system.

/ The "x11" interface could do a better job of compressing expose events in
  the text window.  This matters more for "html" display mode than for
  "normal".  Try opening multiple help windows, and then dragging some other
  window over them.

/ The BufDelete event can cause core dumps while exiting.  Try the ex/xbuf.ex
  script, for example.

/ If tagprg is unset, and an alias uses "local tagprg=foo", then when tagprg
  is restored at the end of the alias, it is restored to "" instead of NULL.
  This breaks tag expansion via the <Tab> key.  Doubtless other things are
  broken for other options, too.

/ When wrapscan occurs, the "wrapped" message doesn't show.

	Apparently the behavior of MSG_STATUS messages has changed.  Those
	messages go away after the screen updates.  I changed the "wrapped"
	message to use MSG_INFO instead.

	One problem: Attempting to XSetInputFocus() to an iconified window
	causes a BadMatch X protocol error.

/ Somehow the hllayers option has become "locked".

/ If ~/.netrc contains a "default" entry, then anonymous logins don't work.

/ Add a dummy "lisp" option.  Also, make sure elvisopt.html mentions the
  use of autocmd for making these dummy options do useful things.

/ In x11, ":se notoolbar" causes an X error in XCreateWindow.

/ It would be nice if 'b' or 'i' could be appended to a font name such as
  "courier*18" to select the bold/italic version of that font.  Italic
  versions are tough because they may be designated by -o- or -i- in the
  long font name.

/ It would also be nice if there was an option to specify a list of
  registry-encodings to try.  The default list should be iso8859-15,
  iso8859-1, *-*.  The list should probably be comma-delimited, like tabs.

	How are encodings handled in Xft?  There's a XFT_ENCODING parameter,
	but no XFT_REGISTRY.  My best guess after searching the web for
	example source code: The value combines both the registry and the
	encoding, with a hyphen between them -- which happens to be exactly
	like I intend to use for Xlib calls!

/ In a :g/re1/s/re2/text/ command, if the s/re2/text/ fails then the :g/re1/
  should still continue to look for other strings.  It appears that it is
  currently stopping on the first failed substitution.

/ The default value of the "background" option is often wrong.  That should
  be fixed, and even then it would be nice if $ELVISBG could override it.

/ Support an ELVISGUI environment variable, for specifying the default
  user interface.

/ Now that the "x11" GUI can read XPM files, maybe it's time to add support
  for color icons.

/ In some "x11" themes, the note in the [Save as] dialog is hard to read.
  It should be drawn in the toolbar fg color, not the tool fg color.
  
/ The 'c' flag for substitutions highlights the first match incorrectly.
  Also, after the last match it doesn't tell you how many changes occurred,
  which makes it hard for the user to recognize that they're back in normal
  command mode.

	Actually, it did report the number of changes if it exceeded the
	"report" option.  But since this is a slow interactive process
	anyway, and the user needs feedback more than usual, I think it
	should override "report" in this case.

/ The "syntax" display mode allows spaces between a function name and '(',
  which is good.  But it also allows tabs, which is bad because it makes
  some many macro definitions look odd.

/ Text selected in Elvis can't be pasted into Netscape's email composer.

/ Allow the "x11" user inferface to use color icons, if FEATURE_IMAGE is
  defined.  Maybe use ":gui icon filename.xpm" to load the icon.

/ ":safer ()" goes into an endless loop.

	Actually, I think it is reading from stdin, but that's still a
	bug.  The code has been treating missing or zero-length file names
	as synonyms for stdin, and that's wrong.

/ The BUGS file should specifically mention the names of the removed options,
  so that users can grep for the names when their .elvisrc scripts stop working.

	Relative to version 2.1_4, the following are replaced by the "show"
	option: showname, showtag, showcmd

	Relative to 2.1_4, the following are deleted: showstack, normalfont,
	otherfont, prepfont, keywordfont, functionfont, commentfont,
	normalstyle, boldstyle, italicstyle, fixedstyle, emphasizedstyle,
	underlinedstyle

/ Add .ZN to the "man" display mode.  It switches between fixed & normal
  fonts, in exactly the same way that .BR switches between bold & normal.

/ The .IP command should support an indentation parameter.  This is important
  when used for bullet lists.

/ Sometimes tag searches fail when they should succeed.  In at least some
  of these cases, the problem goes away if switch from "syntax" to "normal"
  display mode.

	This was failing for "smartargs" argument hints.  That bug is now
	fixed.  Were there others?

/ elvtags sometimes makes tags for reserved words such as for().  And what's
  worse is, elvis' smartargs is too stupid to realize that "for" is a keyword
  even though I specifically remember coding that test.

/ There appears to be a bug in the way subscripts are parsed by the built-in
  calculator.  ":calc a[1] a[2]" returns only a[1], but ":calc (a[1]) a[2]"
  works correctly.  I think the version without parentheses is using the value
  of a[2] as an index into a[1].

/ The :error command sometimes adds garbage to the end of the message.
  The following alias demonstrates this, when invoked without args.

	alias grep {
	    local b="this is a test"
	    if "!*" == ""
	    then error usage: grep text files...
	}

	Fixed!  It bug was in the handling of the "." operator, in simpler
	syntax (used for messages), outside of parentheses at the end of text.

/ The :local command can't restore buffer options if you switch buffers.
  This makes ":local ccprg=..." useless.

/ "make uninstall" chokes on the "themes" subdirectory.

/ Make ^W^W toggle between two windows, instead of cycling through all of them.
  Make ^Ww cycle through them all instead, and allow users who prefer the old
  behavior to :map ^W^W ^Ww

	Add a win.winsequence option, locked, which is set to an internal
	event counter whenever a window is created or receives a keyboard
	or mouse click event.  Make ^W^W go to the window with the highest
	value, other than the current window.

/ Don't spellcheck the library words (e.g., :check -lib libt libf libd libv)

/ Can't do tag searches on #include "file.h".  It works fine on <file.h> though.

/ Maybe make the rand() function take a space-delimited list, instead of a
  comma-delimited list?  That would be more consistent with the way arrays
  are handled.  Alternatively, maybe make "?" in a subscript refer to a
  random item in the array.

	I decided to support "?" as a subscript.  That seemed like the
	more versatile way to go.  It was also easier to implement.

/ "HOME=foo ELVISPATH=foo elvis -Gx11 -mono" starts with black-on-black text.
  In general, there are some weird things happening when -mono is used.

/ Cut/paste between elvis and KDE3 applications is broken.

	KDE3 no longer checks for the standard XA_STRING target format.
	Instead, it used COMPOUND_TEXT.  I fixed this by modifying elvis
	to support COMPOUND_TEXT.

/ In the "x11" interface, if no background color is specified for the cursor
  then the cursor should probably always show the foreground color.  (The
  cursor is normally displayed in the background color if elvis owns the
  current X selection, and the foreground otherwise.)

/ The prefersyntax option is interfering with protocol aliases' ability to
  choose a markup display mode.

	Probably elvis.arf should be skipped for user-defined protocols.
	If they want to run elvis.arf, then can run it themselves via an
	explicit "source (elvispath("elvis.arf"))" command.

	Fixed: I added a "userprotocol" buffer option, and made elvis.arf
	test it before doing anything that would interfere with a readPROTO
	alias.

/ The :info alias (ex/info.ex) replaces the current buffer with the info page
  buffer, if the current buffer has no name.  The window title is not updated.

	This was a bug in the :g function, now fixed.  (exedit.c, line 100)

/ Make the "configure" script support the --prefix flag.  It changes the
  default BINDIR and LIBDIR values to $prefix/bin and $prefix/lib/elvis,
  but they can still be overridden by --bindir and --datadir.  (Note that
  man pages are usually installed relative to the BINDIR, so the prefix will
  affect them too.)

/ Bug in "configure" script: For Solaris, "configure --without-x" will omit
  the TCP/IP libraries ("-lsocket -lnsl") but they're still needed for the
  HTTP and FTP protocols.

/ In elvis.syn, is "anchor ^" broken?  Ex scripts look bad, because all
  doublequote characters are interpreted as the start of a comment.

/ When parsing argv[], elvis should support the use "--" to mark the end of
  flags.

/ Add support for reading from stdin, when "-" is given as a file name.
  / Make the termcap interface smart enough to read from keyboard even if
    stdin is a file.  Verify that other interfaces don't care about stdin.
  / Make elvis smart about what it sends to stdout.  Again, this is mostly
    a change to the termcap interface.

/ Added a "ex/info.ex" script, which lets users view info pages in elvis.

/ For the "x11" interface, make <Shift-Insert> paste from the X selection.

/ Add stubs for the new Xft library.

/ Maybe make the "x11" interface support antialiased text, via the Xft
  library in XFree86.

/ ctags doesn't generate an entry for the last name in an enum list.

	Actually, it is scanning the comments, and generating tags for the
	last word in the last comment.  Other comment words get tags too,
	if they happen to be followed by a comma or semicolon.

/ ctags/elvtags doesn't handle "extern struct structname varname" correctly
  -- it treats varname as a typedef, not an extern variable.

/ The recent change to ctags regarding duplicate typedef tags is buggy.
  It can cause core dumps.

/ There's a bug in the "syntax" display mode, with column-sensitive keywords.
  They can be *EXREMELY* slow.

/ The :local command should not be allowed to use locked options such as
  bufchars.  Restoring such an option could cause a core dump.

/ When a buffer is modified for the first time, the equivalent of
  ":doau OptSet modified" should be run.  Also, the fact that internal
  changes like this don't always cause OptSet or OptChanged events
  should be documented.

/ Use the "Clarified" Artistic License.

/ The :lpr page header for "normal" and "syntax" display modes will be
  malformed, if the file name is too long to fit across the top of the page.

/ What font is used for the .IP tags?

	It was using font code 0 -- either "normal" or "idle".  I have
	changed it to use "formatted" instead.

/ In the "man" display mode, special characters such as \(bu aren't handled
  properly in argument of a .IP command.

	This is only fixed for \(bu, because the others just didn't seem
	worth the trouble.  Bullets are likely to be used as paragraph
	tags fairly often, but how often will anybody want to use a copyright
	symbol there?

/ In "man" .IP arguments, \sSIZE strings aren't being igored correctly.
  Part of the size is still displayed.

/ In man pages, unformatted text should be displayed in the "fixed" font.

-------------------------------------------------------------------------------
Fixed in 2.2h-beta:

/ Windows problems:
  / The "idle" color isn't used for the last elvis window when focus shifts
    to some other application.  My guess is that it handles WM_SETFOCUS but
    not WM_KILLFOCUS.
  / Whenever a new window is created the GUI should call eventfocus() for it,
    because Windows does give it input focus but doesn't send WM_SETFOCUS.

/ Until the first window is opened, any error messages should probably be
  written to stderr.  (I think it already does this for status & fatal
  messages, but not plain old errors.)

	Okay, they're written out but the script name/line number aren't.

/ Trouble in second "Normal" line, in samples/sample.tex

	Fixed.  This one could have caused problems for the start of any
	line in any of the markup display modes.

/ The "man" display mode can't parse/ignore relative sizes such as \s'-2'.

	It can handle size changes without quotes.  I just need to make it
	smart enough to handle quoted sizes too.

/ In man pages, \c should act sort of like a comment -- ignore everything
  up to the next newline -- but \c eats the newline while \" does not, in .nf
  mode.

/ Make sure the syntax description handles special chars such as \(bu and
  similar... and maybe show it correctly in "man" format.

/ Rename the "hlspell" color & option to "spell".  The similarity between
  "hlspell" and "hlsearch" makes both of those words hard to type in, even
  with the help of word completion.

/ After :lpr, the "# pages" message goes away too fast.  This only happens
  when printing to a pipe, not a file.

/ ":au OptChanged lines,columns" should work -- it'll allow users to define
  events that are sensitive to changes in window size.

	Hey!  I could delete the hard-coded message that shows the size of
	the window, and instead use...

		:au OptChanged lines,columns message (columns)x(lines)

/ After a search, the "/" remains on the bottom line, but the statusbar's
  text is hidden until the next keystroke.  The "/" doesn't go away until
  some other text comes along to replace it.

	It's worse than that.  If there's a statusbar then it clears the
	status text too.  So you can't see the results of the "show" option
	(e.g., the name of the function where the search succeeded).

/ Make syntax coloring keywords be more flexible.  For example, \fB should
  be a keyword in troff source, even if it is immediately followed by another
  alphanumeric character.

	Not all keywords should allow trailing alphanumerics, of course.
	Probably a new "prefix" syntax line should be used for marking the
	keywords that do allow trailing alphanumerics.

	The biggest challenge here will be rewriting the word classifier
	to do its work incrementally.  (Until now, it has worked by collecting
	the characters of the word and then looking it up.)

	/ Almost done!  Now if I can just get it to handle .\" comments...

	/ Remove FIRSTPUNCT and SECONDPUNCT.  They aren't needed anymore.

	/ Previously, punctuation keywords were always treated like prefix
	  keywords, I think.  This was handled when drawing the line image,
	  but now it should be handled by setting the ISPREFIX bit during
	  initialization.

	/ There's a bug in tab expansion after a keyword.  Apparently the
	  keyword's width isn't included in the column count.

	/ Document it!  Also, relocate the "set" description.

/ Modify the Makefile to allow MSDOS & Windows executables to have uppercase
  names, somehow.  Otherwise there's too strong of a chance of omitting the
  "*.EXE" files, when we go looking for "*.exe".

/ Don't spell-check :man pages.  They often contain valid words that become
  hyphenated at the end of a line, and elvis' spell-checker doesn't understand
  that.

	Done!  And I also made the bold/underline conversion be more
	forgiving, so it'll now handle Perl's quadruple-emboldened text.

/ Some fatal error messages don't shut down the user interface.  When using
  the "termcap" interface, this means the terminal is left in raw mode.

/ "elvis -r" can hang if there are no old session files waiting to be recovered.

/ Allow <Alt+Mouse1> to be used for selecting rectangles, at least under
  Windows.  Many other Windows editors use this convention.

/ After ":help :sbrowse", clicking on ":browse" follows the wrong link.  This
  is a bug in the htmltagatcursor() function.

/ The :for command kills elvis if given a loop expression which can't be
  evaluated.

/ Support at least two types of string quotes.  See Wim Delvaux's email.

	I started with Wim's patch, and tweaked it to treat the quotes
	as being distinct so a string can't start with " and end with '.

	This still isn't perfect -- sometimes it has a hard time figuring
	out whether the top line of the screen is a continuation of a string
	or not, which causes most of the screen to flash.  But for the
	particular case of sh syntax, it works good enough.

/ When a :local option reverts to its earlier value, this should trigger an
  OptChanged autocmd event.  Does it?

/ If <X11/xpm.h> can't be found, then FEATURE_IMAGE should be #undef'ed.
  (On a computer without the full development system installed, it's possible
  to have the library but not the header.)

/ The "config" alias is renamed to "features".

/ elvtags doesn't understand `extern "C"'.  Ideally, the contents of the
  block should be parsed normally.

/ elvtags should be smarter about typedef tags.  If a tag is known to be of
  kind:t then probably any duplicate tag is the result of a parsing error,
  and hence should not be stored.  (Unless in a different source file?)

	This works two ways: When adding a tag of kind:t, it should delete
	any existing tags.  And when adding a tag of any other kind, it
	should skip it if there's already one of kind:t.  (It helps that
	in the latter case, the kind:t tag will be the *only* tag with that
	name.)  This logic should be implemented in ctags.c, not the
	lower-level source files since we don't want to make elvis do this.

/ Running "configure --without-x" omits the word "undef" from FEATURE_IMAGE,
  so the file contains "#          FEATURE_IMAGE".

/ In elvis.syn, "anchor ^ keyword..." doesn't work anymore.  Comments in ex
  scripts aren't highlighted as such.

/ Proofread the "howto" chapter.  It has some old information.

/ Modify :theme to run the scripts via :safer instead of :source.

/ Modify dmsyntax to use a spelling dictionary for storing keywords.

/ The :bb command lists buffer names, which are now full pathnames.  It
  should list filenames, which are what the user actually entered, because
  the full pathnames are often too long.

/ Sometimes :mkexrc saves the bufdisplay option?  That should never happen!

/ When printing HTML, the page header shows the full pathname of the file/URL,
  I.e. the bufname.  It should probably show the filename instead, if there
  is one.

/ Aliases quit after running :lpr.  Apparently :lpr always returns an error
  code, even if it runs successfully.

/ Add support for a "make elvis.rpm" target.

/ Would it be possible to parse the tags file in a way that allows typedefs
  and functions to be highlighted as such, without modifying the syntax file?
  Maybe by checking the "kind:" attribute.

	I've added "taglibrary" and "tagkind" options.  If set, then when
	elvis loads a syntax description (first time only, thanks to syntax
	caching), it scans all tags files in elvispath for any tag which is
	defined in a file whose extension matches that language.  For each
	tag of kind "XXX", elvis adds the tag name as a keyword in font
	"kindXXX" or "libXXX".

	It is important to remember that C++ can call C functions, and that
	*.h files can be either C or C++.  Languages should be able to specify
	which extensions are callable as foreign functions; e.g., C++ can
	call "*.c" functions.  This will require adding a new line to the
	elvis.syn file.

/ When running "make install", hide the error message from "cp lib/* ...." so
  the user doesn't freak out when they see the message about lib/themes/.

/ In NT, the console doesn't get restored when elvis exits.  Also, while
  running elvis the ex history is displayed incorrectly.

	NT's console sucks.  Probably I should modify the tinytcap.c
	entry for the Windows console to make it move the cursor to the
	bottom of the screen and then clear that line, before trying to
	switch back to the original screen.  This would reduce the impact
	of Microsoft's bug.

	Actually it might be better to hard-code this into the Win32 version
	of tcaphelp.c, since finding the last line requires some math.

/ I've been told that Cygwin has changed the way it stores mount points.
  I should modify elvis to mimic the new behavior.  (I don't want to use
  Cygwin's libraries, because I want users to be able to run elvis without
  Cygwin.)

	The new method still stores SOME of the mounts in the registry, but
	some others appears to be hardcoded.  The only reliable way to get
	them will probably be to parse the output of Cygwin's "mount" command.

/ Using ex history to re-execute the command  :("a  generates an error
  message, apparently because it eats the newline after it.

/ Problem: The :au command doesn't work well with {...} notation, since the
  { doesn't appear at the start of the RHS.  Perhaps it should be modified
  to allow leading text.

	:au Edit * {
		if !modified && readonly
		then doau FileChangedRO
	}

	NOTE: When the :autocmd command lists autocmds, it should detect the
	presence of a newline in the command text, and add {...} around it.
	Perhaps it could borrow code from :alias?

/ elvtags can generate tags for numbers.  It shouldn't!

/ "make install" seems to mess up permissions on the "themes" directory.
  That directory must have execute permission!

/ elvtags doesn't find enum values that are set to explicit expressions,
  such as GDK_MOD1_MASK in lib/gdkstub.h

/ Something's wrong with ^] on #include files.  Try using ^] on "elvis.h"
  in an elvis source file.

/ In the stub files (lib/*.c), any typedefs are being marked as static, which
  prevents them from being found when they should be.

	Fixed by adding a "-g" flag to disable the "file:" hint.  This flag
	is now used during "make install".

/ "make install" uses ctags instead of elvtags, because of root's PATH.
  It should use elvtags, because ref and elvis work best with it.

/ When you mark a rectangle via ^V, and then run an ex command on the
  highlighted text, elvis should IDEALLY limit the changes to the highlighted
  columns.  This would be very difficult though, and it isn't at all clear
  what some ex commands should do -- the ex command set is very line oriented.

	How about this: Copy the rectangle into a temporary buffer, apply
	the ex command to that buffer (all lines of it), and then go back
	and replace the original rectangle with whatever the buffer has
	become.  (If the buffer is unchanged, the last step can be skipped.)

	This still wouldn't be quite perfect -- cut text would be marked
	as being "line" instead of "rectangle".  But :s would work perfectly
	and that's probably what matters most.

	This wouldn't even be particularly hard.  The temporary buffer could
	be a cut buffer.  The ex parser (or the : command) would cut the
	rectangle into that buffer, make it be the default buffer for the
	ex command, and then paste it afterward.

/ Elvis can dump core if you insert a hypertext reference into an HTML document
  while in the "html" display mode.  The partially-complete <a href=...> tag
  causes the problems.

/ Attempting to write an untitled buffer will give the error message "No file
  name for NULL".  It should simply say "No file name".  Or say the buffer
  name.

/ In X11, use "normalfont" as an alias for "font", for now.  This will allow
  users to update their configurations simply by starting elvis with their
  old normalfont setting, and then running :mkexrc to save the old setting
  under the new name.

/ For ANSI colors, add "default" which uses color 9 ("\E[39m" selects the
  default foreground color).

/ Increase the range of legal values for scrollwheelspeed.  It would be nice
  if users could do things like ":let sws=scroll".

/ When editing HTML *in syntax mode*, the spacebar and enter keys should not
  be mapped.  The mapmode option can't handle this.  Maybe add DisplayEnter
  and DisplayLeave autocmd events, and use those to enable the maps when
  entering/leaving "html" mode.

	/ Add the events to autocmd.h and autocmd.c.
	/ Write a new audisplay() function to test for changes.  It should
	  examine the state of the current window.  It should be sensitive
	  to the window's "display" option, and to the input state stack.
	/ Call that function at the top of mapdo() so any changed maps take
	  effect before the keys are processed.  Also call it again at the
	  top of each statekey() call, to react to maps that change the mode.
	/ Replace the current mapmode maps with autocmd maps, in elvis.ini.

/ Some bugs in Perl's POD lines (in syntax mode):
	/ POD lines are followed by empty lines; i.e., the formatting lines
	  are displayed with two newlines at the end instead of one.
	/ Sometimes POD lines cause elvis to dump core.

/ 'elvtags -x *.c' dumps core at the end of the list.

/ Need a way to prevent temporary maps from being saved by :mkexrc.  E.g.,
  HTML's <space> and <enter> maps should not be copied into ~/.elvisrc.

/ When generating "elvtags.man" from "ctags.man", protect the phrase
  "'exuberant' ctags".

/ The :augroup command doesn't accept the '!' suffix.

/ The ^Wd command doesn't work on samples/perl.pl

	This is probably because bufdisplay="syntax perl" instead of merely
	"syntax".

/ More vim autocmd events: BufAdd (synonym for BufCreate), BufNew, BufReadCmd,
  BufWinEnter, BufWinLeave, BufWipeout, BufWriteCmd, CmdWinEnter, CmdWinLeave,
  EncodingChanged, FileAppendCmd, FileChangedRO, FileReadCmd, FileWriteCmd,
  FuncUndefined, RemoteReply, TermResponse

	Most of these aren't worth implementing.  Just BufAdd, and maybe
	FileChangedRO.  Some of the others should probably be recognized,
	just so users can use :doau on them.  Maybe I should add a command
	for adding user-defined events?

-------------------------------------------------------------------------------
Fixed in 2.2g-beta:

/ The spell-checker didn't like words that start with '_', even though
  tag names often do.

/ Maybe allow lpheader font to be controlled via ":color header" ?  Still
  make it default to "fixed underlined", but some users might prefer boxed.

	Actually, it was already doing that for non-markup modes.  I tweaked
	the markup modes to do the same, and documented it.

/ Gnome patches from Klaus Elsbernd, needed for compiling the gnome interface
  with a non-GCC compiler, and run in SunOS

/ ctags *needs* support for "enum".  And probably "class" wouldn't be too hard.

/ ctags needs to reset the class hint when switching from one file to another.

/ The show=link option is slow if there are no <a ...> tags shortly before
  the cursor position.  It should probably stop looking if it sees a tag
  which is a paragraph/section marker.

	I thought I had this one fixed, but apparently the fix wasn't good
	enough.

/ ":map" doesn't list cursor key maps, and even some user maps such as "gw".
  Actually, I kind of like the fact that it omits cursor keys since the user
  probably already knows what they're mapped to, and it's nice to avoid clutter.
  But there should be *SOME* way to see them.

  	Fixed.  Also made ":map" list non-cursor keys, while ":map all" lists
  	all maps including cursor keys.

/ Use mapmode to make <Enter> follow a link in HTML.  Maybe also use <space>
  to page forward.

/ "map mode=html ^V  ^F" isn't saved correctly.

/ Don't set hlspell for the help files.  Why embarrass myself?

/ Elvis dumps core in a C++ file (or C header file) that starts with a comment
  if the "showlink" option is set.

/ "show=link" shouldn't affect any display modes other than "html".

/ The :r command displays the number of lines read, even if report=0

/ In "hex" mode, the line before the column headings is in a funny font.
  This only shows up when you do a ":sp file.bin" on a binary file; as soon
  as any event causes the screen to be redrawn, it looks okay.

/ The "modified, not saved" message uses filename, should show bufname.

/ There's a bug which causes a core dump when "theme:parchment" is loaded.

	Apparently this is in the "1;/^BEGIN /,$d" command, which deletes
	everything but the initial comment.  I suspect that it is failing
	because the last block is deleted, in a buffer which requires
	multiple blklist blocks.  (The "parchment" theme is the biggest
	one.)

	If I run that command manually it doesn't dump core...  until the
	third time!  This suggests that buffer cloning may be involved.

	SOLVED!  It took six hours, but I finally found bug in lowbuf.c's
	delblock() function, when a buffer is big enough to require a chain
	of BLKLIST blocks.  To delete a CHARS block in the n'th BLKLIST block,
	it was copying the first item from the n+1'th BLKLIST block and then
	recursively calling delblock() to shift the n+1'th block (and any later
	ones, the same way).  But delblock() frees the deleted CHARS block,
	so I have INCREMENT THE ALLOCATION COUNT FOR THAT BLOCK BEFORE THE
	RECURSIVE CALL!

/ For the termcap interface, the "italic" attribute can interfere with
  colors.  It is implemented as "half brightness", which Linux's console,
  for one, can't do in color.

	Added a "ttyitalic" option, similar to "ttyunderline".

/ Changing the 'background' option should cause colors to be reevaluated,
  if the actual background color is unknown.  This would mostly benefit
  the "termcap" interface, since the GUI interfaces always know exactly
  which background color they're using.

	Basically, this action could be done by calling recolor() for any
	font which has no "like" clause.  (The "like" fonts will automatically
	be reevaluated when the font that they're like is reevaluated.)
	Triggering this action will require writing a custom option-setting
	function.

/ The MS-DOS version has a hard time using bright colors for "normal" plus
  other things: color normal yellow|color comment bright green.

/ Modelines don't work?

	This was because the first "1,5s/ex:\(.*\):/\1/x" was failing, which
	prevented the elvis.arf script from continuing with the second
	"$-4,$s/ex:\(.*\):/\1/x" command.

/ "safer" mode is too strict.  Disallowing :alias and :autocmd causes problems
  in the initialization scripts.

  / Make elvis.ini check "safer" before running elvis.ali
  / Make :augroup be unsafe, so the script generated by :mkexrc will work
    whether safer is on or not.

/ Weird problem with PERL doc lines.  See Merijn's example.

	This was caused by a scanalloc() call without a matching scanfree(),
	in dmsyntax.c.  Each screen update left one more scan context lying
	around, and that caused elvis' buffer functions to get confused --
	there should never be any scan contexts active when the buffer is
	changed.  There are assert() statements to verify this, but those
	are normally disabled, so that's why this looked subtle.

/ Shouldn't draw guide lines on ex scrolled rows at bottom of window.

	Fixed, in an interesting way.  I added a new text face, "bottom",
	and made it immune to guide lines.  All ex commands and their output
	are drawn in this font, so they don't get guide lines.

/ Make "smartargs" be a buffer option?  Sometimes there will never be tags
  for functions; e.g., in HTML source where "functions" are really tag
  parameters.  Searching for a nonexistent tag can take a while!

	Done.  Also modified "elvis.arf" to turn off smartargs for html.

/ If you delete the last character from "boxed" text, the right edge of the
  box is not redrawn in its new location.  The box appears to be open-ended.

/ Add a BgChanged event for :autocmd.

/ add a "showface" option, which shows the name of the current font.  Maybe
  also add a current("face") function which returns the name of the face at
  the current cursor position.  If the face is a combination, then return "".

	Added...
		current("face")
		current("background")
		color("name")
		color("name","fg")
		color("name","bg")
		color("name","like")

	Replaced many showXXX options with a single versatile "show" option.
	file, tag, cmd, face, link, spell, region

/ The 'x' flag of the :s command should move the cursor to the current line
  before evaluating the command.  This affects the readTHEME alias.

	This is very quirky.  The bug mostly resides in parsewindowid(), but
	really it is just a manifestation of the fact that my ex parser doesn't
	allow ex commands to specify a cursor position -- it tries to use the
	window's cursor, or the top of bufdefault.  I think in readTHEME, it
	is using the latter.

	Fixed!  I just made sure that x's quirks matched parsewindowid()'s.

/ Some easy ideas for background images in X11:
  / Allow background color names to contain either a simple color name, or
    the name of an image file, or both.  When both are given, use the simple
    color to "tint" the image.  Return a pseudo-colorcode.  Also return the
    average RGB value of the image, so contrasting foregrounds can be chosen.
  / When an unscrollable image is being used, GUI->scroll() should always fail.

/ problems with partially specifying the "normal" color, and probably
  any of the special colors -- When you run :mkexrc it only remembers the
  most recent partial color specification.

/ The "or color" clause can be chosen before the background is set. E.g.,
  ":color normal black or white on black" chooses the new foreground based
  on the old background.

	Note that this is related to the "partial specify" problem -- the
	only reason for specifying "or" colors with the background is so
	that you'll be prepared when the next foregroundless :color command.

	Possible combined solution: Write a parser which returns the foreground
	and background colors AS STRINGS.  When doing this, treat the "or"
	word as plain text, so it is included in the foreground string for
	later processing.  Other attributes are returned as flags.  When
	doing a partial specification, parse both the old & new attribute
	strings; if the new one omitted a foreground or background, then
	use the foreground or background from the old string.  Process the
	background first, so the proper "or" color can be chosen.  Regenerate
	the combined text by using the foreground & background strings,
	with flags generated from the attribute bits.

/ In "x11", a bold 'm' leaves bits to the left it its character cell --
  maybe clip the cell?  It would make scrolling look better.  However,
  italic characters often slop to the right, which apparently produces far
  fewer problems, so the clipping should only be on the left.

/ Make WinElvis read its color names from a file, preferably in the same
  format as /usr/X11R6/lib/X11/rgb.txt

/ ":help ex" fails.  It tries to load "elvisex.html" in the current directory.

/ Describe the meaning of ";" in ex addresses.

/ The postscript print driver uses the video colors, not the carefully chosen
  lp colors.

/ It is so tempting to give only one foreground color, I really should add an
  "lpcontrast" option.

/ For the sake of uniformity, x11's "normalfont" option should be renamed to
  just plain "font", like in Windows.

/ ":help set ttyunderline" doesn't go to the ttyunderline message, unless
  you're using the "termcap" interface.  I wouldn't expect name completion
  to work, but you should at least be able to give the full name and find it.

/ "nottyunderline" shouldn't completely disable the use of underline --
  still try to use it if no background color has been set.  This will allow
  the window separator to be red.

	Actually, this is more of a personal preference.  But users who
	prefer to have no redness in their window borders could just say
	"color normal on black" so no harm is done.

/ Can't use "bold" attribute to distinguish between normal/idle.  Elvis
  appears to only be sensitive to the colors, not other attributes.

/ Windows problems:
  / The Options->Gui dialog can't change colors, and doesn't display the
    colors of that have been set elsewhere.
  / Text type names aren't sorted, and the label should probably be "face"
    instead of "text type" anyway.
  / Combo boxes should show more items.
  / The Options->Syntax menu item crashes elvis.  It is obsolete anyway,
    and should be deleted.
/ The :let command never triggers an OptChanged or OptSet autocmd event.

/ Add "kvt" to the list of terminals recognized by -Gx11.  Also, the termcap
  interface should assume "kvt" has a light background.

/ "make install" should create the /usr/local/lib/elvis/tags file.

/ The smartargs option doesn't allow a space before '('.

/ Make the toolbar's colors be configurable via resources (instead of the
  scrollbar), and then make the default settings of the scrollbar and
  statusbar be "like toolbar".

/ In the Windows, OS/2, and MS-DOS binary distributions, make the contents
  of the "lib" directory be read-only.  This is important for making
  ":set prefersyntax=writable" work right.

/ Writing an anonymous file should never change the name of an INTERNAL buffer.
  Currently, :mkexrc seems to do that.

/ The "elvis untitled #n" buffer names should correspond to bufid numbers.
  It's less confusing that way.

/ In "termcap" interface, sometimes the hlsearch highlighting doesn't show
  everything that it should.  This seems to depend on color settings.  See
  samples/hlsearch.cpp for an example.

	Not a bug.  This is due to the interaction of "bold white" text and
	"bright red" highlighting.  In the termcap interface, bold+bright
	always results in bright white, regardless of whatever color you're
	trying to use.

/ For HP, the default level of optimization doesn't work in elvis.  Need to
  make "configure" drop it to +O2.

	I tried changing the "configure" script to handle this, but I can't
	test it myself.

	Merijn Brand reports that it does indeed work.

/ Sometimes the cursor gets stuck at the bottom of the screen (not the status
  line where ex commands are entered, but the last one normally used in visual
  mode).  Hitting zz fixes this condition.

	This may be related to the "calling delimage() endlessly" bug, which
	is now fixed.  The "stuck at bottom" bug hasn't occurred since that
	fix.

-------------------------------------------------------------------------------
Fixed in 2.2f-beta:

/ Changing the "normal" background should cause all colors to be reevaluated,
  so that "or color" can be chosen properly.  Merely switching a window from
  "idle" to "normal" shouldn't do this -- only a ":color normal" command should.

/ Hey!  ":color normal gray" changes the background to black, in -Gx11?
  It should leave the color unchanged.

/ Need a way to make maps that only apply in certain display modes, or
  possibly some other factor.  Maybe add a "map=XXXX" flag, and a "map"
  buffer option?  Or a generic "option=value" flag?  Maybe add a :mapgroup
  command, like autocmd, and allow groups to be selectively enabled/disabled.

	I implemented this as a "mode=XXXX" flag, and a "mapmode" option.

/ After running an external command, the termcap interface restores colors
  BEFORE switching back to the edit screen, so the original screen colors
  are clobbered.

/ Make :alias list the user aliases, and :alias! list the system aliases.

/ Make :alias indent aliases to reflect the control flow.

/ Need a way for aliases to access the tail of a command.  The csh way to
  do this would be to put a * after the digit, as in !:2* to access everything
  except the first word.

/ Can the Edit autocmd event be partially dependent on bufdo()?  Should it be?
  This is desirable because it would reduce the number of events, but what if
  somebody "moves" the cursor while in input mode?

	Actually, moving while in input mode should probably call bufdo()
	anyway, so that individual changes can be undone.

	A more complex problem: global changes, via either :g or :%s.  There
	should probably be flag added to each buffer, to indicate that Edit
	events should occur more often than "undo" saves.

/ Make the window title show the filename if there is one; else the buffer name.

/ Add a way to mark regions.

/ With "optimize" turned off, the : command causes hlobject text to be
  highlighted weirdly.  I think this happens only when there are multiple
  windows.  It tries to recolor the current window, using the temporary
  fonts from the other window.

  Exposure events can be similarly weird, regardless of optimization.

	This might be solvable by comparing the color number against the
	npermanent.  If >= then allocate a new temporary color with the
	required attributes... BUT WHY SHOULD THAT BE NECESSARY?

/ Make :mkexrc generate a script that uses :try instead of :if.  This is
  important if FEATURE_CALC is turned off.

/ Allow more features to be disabled via FEATURE_XXXX -- tags, regexp, etc.

/ Modify the [[ and ]] commands to recognize functions that start with '{'
  at the end of the function header line, instead of in column 1 of the
  following line.  Assume function header lines are either unindented.
  This notation is standard in GNU source code, Perl, and (oddly enough)
  in elvis' own alias notation.  This should also affect the gd command.

	Done.  Also added "tweaksection" option to control it.

/ The "locked" and "readonly" option should never be saved in ~/.elvisrc
  Maybe other options too.

/ Adding "tag=" to the status line had the unintended side-effect of adding
  it to the result of current("tag") too.  And it looks uglier than I expected
  anyway.  Since the real source of confusion is when both "showtag" and
  "showspell" are set, it would probably be better to add numbers before
  the showspell words.  This makes them distinctive, and also makes the gs
  command easier to use.

/ Add link anchors to example aliases in lib/elvistip.html.  Refer to them
  in other parts of the manual.

/ Bug in suffixes -- it searches random memory past the end of the suffix
  list.

/ Something is wrong with subscripts.  Try :calc "foo"[,2...];"ps".

	There were two bugs: The precedence of [] was too low, and non-numeric
	arguments were being interpreted as an infinite list of 0's.

/ The "x11" user interface dumps core.  Try "e2 exmake.c", ":sp ex.c", resize
  the ex.c window to be 4 lines high, run ":window" or ":set".  This does not
  dump core in the "termcap" interface.  It may be related to spell checking.

	I suspect this has something to do with static variables being set
	with the expectation that one window will be used next, and then the
	other one is winds up being used instead.  I say this because in at
	least one occasion, the status line (below the text window) showed
	the same tag name for both windows even though they were for different
	buffers.

	Unfortunately, one side-effect of this is typically that the stack
	is clobbered, so it is difficult to tell what's going wrong.

	FIXED!  It turns out that some important variables were being left
	unset by drawopenedit().  The variables are normally set by the
	drawupdate() function which calls updateimage().  But drawopenedit()
	also calls updateimage(), so it had to set those variables too.

/ It would be nice if ":make" didn't switch buffers if there's already a
  buffer showing the erroneous file.

/ There should be a way to disable spelling for each buffer.  This is partially
  dependent upon whether the window is readonly or not, but it should be more
  versatile than that.  Maybe replace the spelltags option.

	I simply made "hlspell" be a buffer option instead of a global one.
	I also deleted the "spelltags" option since I no longer consider it
	to be useful... or not useful enough to justify making users read
	about it and understand it.

/ Sometimes elvis gets stuck calling delimage() endlessly.  In the "termcap"
  interface, if you start on an HTML file and then use :sp to load some other
  file, then elvis gets stuck in an endless loop trying to update the HTML
  window.

	Fixed!  This was happening because eventfocus() was calling draw()
	for the original window, which in turn tried to draw it using the
	"html" functions -- even though the cursor was still in the ex history
	buffer at the time.

/ Make the absolute() function be smart about ".." in path names.

/ Elvis should recognize different paths to the same file.  This is important
  because some compilers mess up names (javac, for one).

	The best solution here is probably related to the idea of giving each
	window its own virtual current working directory.  But that's more
	than I was planning to do for 2.2.  Instead, I can just make the bufname
	option store the full pathname; then buffind() will find the existing
	version of a buffer no matter how you express its filename.

/ Add g<Tab> -- move backward to previous link.  Make GUIs map <Shift-Tab>
  to g<Tab> if possible.

/ Include /usr/dict/words in the windows binary distribution.

/ Perl scripts aren't being recognized if they have no extension?

	Merijn Brand tells me this is due to the fact that he sets the
	magicchar and magicperl options to make elvis use Perl-like regexps,
	but the default elvis.arf script file tries to recognize Perl scripts
	by using traditional vi regexps.

	There are two ways to solve this.  One is by resetting the magicchar
	and magicperl options to their defaults at the start of each script
	or alias -- either hardcode the equivalent of ":lo mac=^$.[* nomap"
	or add a new OptLocal autocmd event to make it configurable.

	The other way is to add a special metacharacter which forces the
	remainder of the regexp to be parsed using traditional syntax, and
	then add that metacharacter to the start of the regexps in elvis.arf.
	I'm thinking \V would be nice.  Maybe also support \Q...\E quoting?

/ When pasting chars into an empty buffer, add a newline first.  This is
  particularly important when pasting via the GUI, but it may also be
  important for pasting between buffers.

/ When searching for tags: if there are both a macro and a function with the
  same tag name, use the function first.  This should probably be both an
  optional flag in ctags, and the sorting behavior or :tag and the ref program.

	What if I just initialize the heuristic tables with a higher
	preference for functions initially?  I'd need to do this in both
	"ref" and "elvis".

/ Add "color emphasized like bold" somewhere, for the benefit of syntax
  coloring.

	Fixed a different way: Since syntax coloring fonts should be given
	names that reflect their intended purpose, I modified elvis.syn to
	use other font names, and then added "color" lines to define the
	default appearance of those fonts.

/ Termcap "transparent" background doesn't work with Eterm.  If some text
  uses a different background, then the display code doesn't switch back
  to normal afterward.

-------------------------------------------------------------------------------
Fixed in 2.2d-alpha:

/ The "howto.html" file doesn't look good in Netscape.  This is due to the
  ugliness of the non-elvis <dt><dd> list.

/ The "showspell" option shouldn't suggest natural words where only tags are
  allowed.  For example, when editing source code, it shouldn't suggest natural
  words when the cursor is on a variable.

/ The newbuffer() function and :file command can both create different buffers
  with the same bufname.  So can writing an anonymous buffer out to a file.

/ Change the "showname" and "showtag" options' display formats to prefix
  their value with "file=" or "tag=" -- there are so many different things
  that can be shown down there that it can get confusing.  Oh, and maybe
  rename "showname" to "showfile".

/ Make :mkexrc save aliases.

? Occasionally, elvis will mess up the session file.  When this happens, the
  screen looks normal but the cursor doesn't move correctly.  I think this is
  a recent bug -- added after 2.2d.

	Maybe this has something to do with the message queue?  Elvis doesn't
	allow buffers to be modified while a scan operation is pending.  The
	(Elvis message queue) buffer is used for queuing messages, so it is
	modified each time a message is generated.

  FIXED? There was a bug in the indentation code, used by the shift commands
	and autoindent, among other things.  It was mixing scan operations
	with changes.  Also, wordatcursor(), used by the spell checker among
	other things, was sometimes neglecting to free a scan context when
	it was done, which eventually caused large numbers of scan contexts
	to accumulate.

/ Need a way to fire off commands when certain events happen.  Maybe vim's
  ":au" command, or something related to the :gui command which already runs
  events for GUI events.  Combine them somehow?  The immediate goal is to
  provide a way for a [buffer] menu to change when buffers are created or
  deleted.

/ In XFree86 4.0, the "x11" interface's toolbar hints have an extra character
  on the end -- maybe not trimming newline?

/ Maybe add "transparent" as a termcap background color -- it will just
  reset the colors via \E[m and then not set a background.  This will allow
  you to specify some background colors while still leaving others set to
  the default.  It won't always look good, but it is another option.

/ Under XFree86 4.0.3, the image under the cursor doesn't appear to be saved
  before the cursor is drawn.  On the next blink, it get screwed up sometimes.
  This also affects selections.

/ If multiple tags are found with equal sorting values, then sort them by
  the timestamp of the file where the tag is defined.  Do this in ref, in
  addition to the built-in tags searcher.  Use dirtime() to find the
  timestamp... but will that cause problems in ref?

/ osunix/osblock.c shouldn't trust the access() function, because broken
  links look like nonexistent files to it.  It should instead try calling
  open(..., O_RDWR|O_EXCL).  Also, think about what happens when blkopen()
  fails, because currently it leaves the tty in "raw" mode.

/ Windows patch from Mauro Persano: Use GetModuleFileName() to get the full
  pathname of the winelvis.exe file, if possible.  This should be more
  reliable than explicitly searching the path (though it will still search
  the path if GetModuleFileName() fails).

/ Spelling bug: If a word is known to be bad or incomplete, then any words
  that can be derived from it via suffixes will be considered good.  I.e.,
  the suffix test only checks whether the base word is known, not whether it
  is good.

/ The showspell option doesn't update the status line correctly when running
  in the termcap user interface.  It isn't updated as often as it should be.
  Specifically, it is only changed from empty to suggestions.  (Scrolling
  causes it to be empty, briefly.)

	This is exactly the way other status messages work, but other status
	messages aren't sensitive to the cursor position.  The genlastrow()
	function will need to be extended to clear the last row before the
	next update, if it is displaying alternative spellings.

/ The "gs" command complains about words that start with digits.

/ Spell checker bombs on words longer than 100 characters.  Any word that
  long should be assumed to be misspelled, and not have any suggested
  alternatives.

/ With incsearch set, elvis could dump code on some partially-entered
  regular expressions, such as "/[^[".

/ In expressions, allow $_ to be used as a synonym for line().  Allow $ before
  options, too.  This would make Perl users happy.

/ Can't yank from a locked buffer?

/ Change the description of makeprg and ccprg, to include a description of
  how to add double-quotes around $2.

/ Odd things happen when you edit a filename that looks like a file number.
  E.g., ":e 1".  It can clobber an existing buffer.

/ d/regexp/+0 retains the line containing regexp, but should delete it.

/ Spelling gives false errors when sidescrolling is enabled -- partial words
  along the left edge of the screen are flagged as being incorrect.

/ Speller should skip over words that begin with a digit.

/ Could incremental searches be smart enough to skip searching if more
  keys are in the type-ahead buffer?  That might avoid some useless search
  time.

? How to dump core: 1) start editing an existing file, 2) run ":make",
     3) run ":("Eel)sp", 4) run ":!ref ...".

? The :configure alias doesn't work when there are spaces in the
  directory names.

/ Create a script that map's vim's commands to the corresponding elvis commands
  including the following:
	rX	changes each non-newline to X [:map select r g=]
	x	Delete (like d) [:map select x d]
	U	make uppercase [:map select U gU]
	u	make lowercase [:map select u gu]
	~	toggle case [:map select ~ g~]
	J	Join whole lines (like :join) [:map select :j^M]
	^]	Tag search, for highlighted text [:map select ^] y:ta ^P^M]
	R	Change whole lines (same as S)
	o	Go to other end
	O	go to other edge of rectangle

/ Allow v/V/^V to change the type of a pending selection.

/ When a rectangular selection is pending, $ should cause it to extend all
  the way to the end of each line.

/ New motions for v/V/^V selections:
	g%	go to other end of selection  (same as vim's o)
	g^V	go to other vertical edge of rectangular selection (vim's O)

/ :>> doesn't shift twice, but should.

/ The :join command doesn't work correctly when multiple lines are visually
  selected -- it still joins just the current line & following line.

/ Wasn't there a window() function?  There should be!

/ Describe why "-Gx11" is needed when adding elvis to an X start menu in Linux.

/ ":help ^Wd" doesn't work.  Apparently multi-key vi commands aren't
  recognized by the viname() function.

/ Speller dumps core if a spell_t node has a max=255.  That's a valid Latin-1
  letter, but it messes up some loops that use a CHAR as a counter, since
  the counter can never be incremented past 255.

/ Support keyboard timeouts (the usertime option) in the "x11" user interface.

/ If the "find next misspelled word" command is used with a count, and the
  cursor is on a misspelled word, then use the count'th corrected spelling.
  For example, ^K moves to a misspelled word, and 1^K uses the first
  suggested correction.

/ Need to support 'g' commands, so we can standardize keystrokes for spelling.

	/ g0	go to leftmost char in current row (for wrapped lines)
	/ g^	go to first non-whitespace char in current row (for wrapped)
	/ g$	go to rightmost char in current row (for wrapped lines)
	/ gj	move down to same column of next row (next line, or next row
		of a long wrapped line)
	/ gk	move up to same column of previous row.
	/ gd	go to local definition of word that the cursor's on (first
		mention within the current function).
	/ gD	go to global definition of word that the cursor's on (first
		mention in the current file).
	/ gI	like "0i"
	/ gJ	join lines without space, like ":join!"
	/ gU	operator, makes text uppercase
	/ gu	operator, makes text lowercase
	/ g~	operator, makes text lowercase
	/ ga	show ascii code of current char in ascii, decimal, hex, octal
  and...
	/ gh    move left, skipping hidden characters (for markup modes)
	/ gl	move right, skipping hidden characters (for markup modes)
  	/ g=X	operator, replaces text characters with X
  	/ gs	go to next misspelled word.  If given a count, then before
  		moving, replace the current word with the count'th alternative
  		spelling.
	/ gS	go to end of current spelling word. (Useful for cgS, if none
  		of the alternative spellings are correct.)

  The first task is to create a means by which these commands can be parsed.
  Probably this should be a second table, similar to the vikeys[] array in
  vi.c.  An extra BOOLEAN flag in the vi input state will indicate whether
  the main table should be used, or the g table.

  Then I can start implementing the actual G commands.  Some will be easier
  than others, but none should be terribly hard... except gv.

  Tests:
  / Make sure invalid g commands are rejected safely.
  / Make sure valid ones are recognized.
  / Make sure ":help gs" works.
  / For operators, make sure "g~g~" and "g~~" both change the whole line.
  / Verify that g=x works for motions, text objects, and selections.
  / Verify that other operators work for motions, text objects, and selections.
  / Make sure changes can be undone.
  / Make sure . can repeat all of the commands that change text.

/ Elvis hangs on :%s/\( ,\)\+//g

	This turned out to be due to a recently introduced bug in non-greedy
	closure operators: They weren't checking the minimum number of
	repetitions.

/ When incsearch is set, ":map! >^[?<^Mye" dumps core.

	For backward incremental searches, the regexp's leavep field wasn't
	being saved.

/ Elvis hangs when hlsearch is set, for any regular expression which could
  match a zero-length string.

/ WinElvis doesn't update the screen correctly if it scrolls while part of
  the window is off-screen.

/ Allow :map to support more flags, in addition to "visual".  Some specific
  words that would be handy are:

    command	Map works in command mode
    input	Map works in input mode on a normal buffer
    history	Map works in ex and regexp history buffers
    motion	Map works when a motion is expected after an operator
    select	Map works when a visible selection is pending
    noremap	Don't expand maps in in rhs

  :map doesn't imply any of the "Map works..." modes, but if no modes are
  specified then it'll default to "command history motion".  :map! implies
  "input history".  Note that these are compatible with traditional vi,
  and aliases can be used to make :map act like vim's mode-specific mapping
  commands.

  To keep the map table clean, there are two ways a new entry can affect
  an existing entry:
   1.If the LHS matches an existing entry and some of the contexts are the
     same, then the existing entry should have those contexts removed.  If
     this means the existing map doesn't have any contexts left, then it
     should be deleted.
   2.If the LHS matches an existing entry, and the RHS also matches, then
     the existing entry's contexts should be merged into the new entry's
     contexts, and the existing entry should be deleted.
  The new entry is always added, regardless.

  Also, make sure that if the rhs starts with the lhs, then that portion
  is never subjected to remapping.

  The keywords should be allowed before either the lhs or the rhs.  Elvis
  has always supported "visual" before the rhs, but putting these before
  the lhs is more convenient in an alias.  To support this, the :map command
  can't depend on the ex parser to separate the lhs from the rhs, which seems
  odd.

  Abbreviations are similar to maps.  Although the flags used with maps
  don't really apply to abbreviations, there is a chance that the map changes
  could break the abbreviation table, so abbreviations should be tested too.

-------------------------------------------------------------------------------
Fixed in 2.2c-alpha:

/ The command ":map! X visual J" doesn't work.  ":map! X visual :normal J^M"
  sort of works, but it looses the whitespace.

	The first map should work now, but it'll work exactly like the
	second one -- without any whitespace between them.  I figure
	that's probably close enough.

/ One user reported that ^V{motion}d deletes text okay, but then pastes it
  all on one line.

	This happens when you're near the end of the buffer.  The pasted rows
	from a rectangle can only be pasted into existing lines.

/ Either the "fold" color or "spell" color isn't being allocated correctly.
  Elvis mixes them up.

	The "fold" color was being allocated the first time a fold was
	displayed -- which could be after some spelling errors have been
	highlighted via a temporary color, hence the confusion.

/ Should autowrite write files before :make, like vim does?  Yeah, I guess so.

/ Allow text object to work with the V/v/^V commands.  This is important!
  For example, "Va{d" deletes a block using whole-lines.  There is no other
  way to do that; "da{" and "%d%" both do character deletions, not line
  deletions.

/ Modify the behavior of ":set! option=value" so it changes the default value.
  I.e., store the new value as the default, and ONLY if the options wasn't
  explicitly set before it should change the actual value.  This should never
  set the option's OPT_SET flag, but may reset it if the new default is the
  same as the current actual value.

	This is about half-done now.  In its present state, there is a
	bug in the way it decides which options to specify in .elvisrc.

	Keep the desc->dflt pointer.  Leave it NULL for most options,
	indicating that they've never been set.  When an option is set
	via ":set" (no !) then store the old version as desc->dflt.  When
	set via ":set!", then store the new version as desc->dflt.  There
	is never any time when we take a snapshot of all options' values
	and call those the default.

/ In html mode, $ moves to the start of the following line, but looks like
  the start of the current line.  BOTH ARE WRONG!  It should move to the
  last character on the current line.

/ In visual mode, typing :<enter> should do nothing.  Currently it prints
  the line and advances.  (Patch supplied by Derek Peschel.)

/ ":vi +line" ignores the line when there's no filename.

/ Regexp changes:
  / Add a new "magicchar" option which lists the chars that don't need to
    be preceded by a backslash in order to be considered metacharacters.
  / Modify the !\x notation (backslash in an alias parameter) to support
    the new magicchar option.
  / Make the regexp error messages use \ only where needed.
  / Add support for alternation, the \| metacharacter.
  / Add support for using closure on complex subexpressions.
  / Add metacharacters for edge of a word: \B for not-an-edge, and either
    \b or \h for edge.
  / Add an option to choose the meaning of \b.  And eventually \uXXXX.
  / Add an option that disables Perl-style $name substitution.
  / Support $name substitution in replacement text of :s/re/text/ command.

/ There should be a way to handle Perl POD text as formatted text.  Basically
  this would require everything from =head1 through =cut to be treated as a
  big comment in the "documentation" font.  It probably isn't worth the effort
  to do anything special with markups within POD text, since elvis won't be
  used as the formatter, just the editor.  There should be a line in the
  elvis.syn file to indicate that the language uses POD, and what starts &
  ends it -- "documentation =cut", where "=cut" ends it, and anything that
  starts with "=" starts it.

/ Modify PostScript to automatically detect size of paper.

	Use "clippath pathbbox" to obtain the drawing boundaries.

/ Link error when config.h contains "#undef FEATURE_TEXTOBJ"

	This is because it also used "#define FEATURE_HLOBJECT", which
	requires text objects.  I fixed it by making "elvis.h" define
	FEATURE_TEXTOBJ whenever FEATURE_HLOBJECT is defined.

/ Allow <Shift-Arrow> to select text, like in most Windows programs.
  / For WinElvis
  / For XElvis

/ Need a way to distinguish between variable names and other unclassified
  text, in a syntax-colored buffer.  Perhaps this is another use of the
  "initialtoken" option.  Or a "mostly" option, which allows individual
  languages to be configured to use something other than "variable" for
  unclassified words.

  This is more important now, because spell-checking of variables is normally
  disabled, but in "syntax html" mode all words look like variable names.

/ The "syntax" display mode allows single characters to be declared as
  keywords (for example, to change their font), but if you do that and also
  use the character in a "useregexp" list, then the *character* is marked as
  preceding regexps, but the *keyword* is not.

/ The "showlink" option has the weird side-effect of making the current word
  be displayed on the status line, while in "man" display mode.  Harmless,
  but very odd.

/ The :browse and :sbrowse commands are broken -- the links don't work.
  This also affects the :kwic alias.  Apparently, somehow elvis lost the
  ability to handle file:...?command URLs.

/ Can't use incsearch as the target of an operator.

	Maybe disable "incsearch" when searching is used as the target of
	an operator.  That should be easy to do, and would probably be
	satisfactory to most users.

/ Incsearch doesn't set "wantcol" -- after a successful search, hitting j or k
  will cause the cursor to jump to its previous column, instead of staying in
  the same column.

/ In incsearch mode, the new $name variable interpolation doesn't work.

	The fix for this will probably involve calling regbuild() function,
	since that's where $name is implemented.  And if we're going to be
	calling that, then we might as well support the search flags too.

/ The visual command "/$" moves the cursor to the newline character, instead
  of the last character in the line.

/ Need a clean way to allow HTML files to be displayed initially in either
  "syntax" or "html" mode (whichever the user prefers).  Currently, all the
  user can do is modify elvis.arf to set bufdisplay differently.  This isn't
  good enough because setting it to "syntax" inhibits the ^Wd command, and
  may also affect some other commands.

	Added an "initialsyntax" buffer option, and a global "prefersyntax"
	option which is used to set "initialsyntax" automatically via the
	elvis.arf script.

/ Maybe add a "mkexrcfile" option.  String, defaults to "".  If set to some
  other value, then that's where :mkexrc writes by default, and :mkexrc is
  allowed to overwrite that file if it exists.  Whenever :mkexrc writes a
  file anywhere, it first sets mkexrcfile to the file name, so that the same
  file can be overwritten easily.

  The :cd command makes this dangerous.

/ In incsearch, ^V can't be used to escape control characters.

/ "hlsearch" and "selection" fonts don't mix well.  The foreground color
  is lost, at least with the "x11" interface.

/ Change the way "Elvis custom initialization" (for :mkexrc) is updated.
  Only consider updating it if the event counter has changed, so that complex
  initialization scripts only update it once.  The :mkexrc command regenerates
  the buffer before writing, just to be sure it is up to date.

  Ideally, it should be updated from event.c, near the end of each event
  handler function.  The ex.c file should simply set a flag indicating that
  the update is needed.

/ Need an easy way to store regexp text in variables.  Suggest Perl's $name
  or ${name} notation.

/ In expressions, make ":" operator concatenate path elements, using the
  OS-specific path delimiter (either ":" or ";").  This shouldn't interfere
  with its use in the "?:" ternary operator.

/ The :normal command has a bug.  The second invocation causes a core dump,
  at least when used inside an alias.  This prevents the :spell alias from
  working.

/ Modify the "x11" gui to support font names of the form "font*size", like
  the "windows" gui already does.  Rename "normalfont" to "font".

/ The cfmt alias is broken.  It seems to think all comments are C++ comments.
  Perhaps the error status of :s/// has changed?  I know I fixed the :g//s///
  behavior recently.

/ Folding interferes with <PgUp>/<PgDn>.

	This is probably because the dmnormal.move() function doesn't
	handle folds when the absolute value of the linedelta is > 1.

/ The :help command should allow = at the end of n option name.

/ It would be nice if the "aL" text object ignored preprocessor lines, the
  same way that it ignores blank lines.

/ In arithmetic expressions, using a regular expression as a function argument
  sometimes causes core dumps.

/ In incsearch, /<Enter> and ?<Enter> don't repeat the previous search.

/ The "autoselect" option was recently broken.

/ hlobject updates are too optimized.  If multiple windows, it won't always
  update the highlight range even after the cursor has moved outside of the
  highlight region.  Setting nooptimize hides this bug.  It seems to go
  away if there is only one window.  Maybe it is confused about the "current
  window"?

  Serious quirks in screen updates while editing, too.

/ Add an error message when a script contains an unterminated {...} block.
  Also, warn users to enclose the hlobject value in quotes so "a{" doesn't
  cause the parser to become confused.

/ Does hlobject sometimes cause foreground colors to be lost?

	Merijn Brand:
	If hlo/hll is on and the block (on the first level) is rather
	large, the block is backgrounded with the appropriate color, but
	the syntax highlighting of the bottom half of the screen
	disappears. I realize this a weak  which is rather
	hard to verify, but I've not come up yet with a 
	example.

	Here's one way to reproduce it: ":man pcre".

	This is almost certainly an example of using up all of the colors,
	so any attempt to merge two fonts returns just one -- the hlobject<N>
	font.  Why are so many fonts being used, in such simple contexts?

/ <Enter> doesn't advance the cursor.

/ In incsearch, /\ will hang elvis.

/ After an incsearch is cancelled, hide the status info.

/ To support highlighting of whole lines, allow the hlobject value to use
  a 'V' prefix, as in `:set hlobject="Va{"'

/ Some objects cause core dumps, if the buffer is empty.  This is most critical
  when you have "set hlobject=a{" in your .exrc file.

/ Shouldn't draw guide lines on ~ rows after last line of file.

-------------------------------------------------------------------------------
Fixed in 2.2c-alpha:

/ Still need to implement "boxed" text in "windows" printer.  Also, the screen
  needs to support COLOR_LEFTBOX and COLOR_RIGHTBOX, plus the "guide" color.

/ :g/re1/ s/re2/text/ stops at the first line which contains re1 but not re2.

/ Syntax descriptions aren't being cached correctly.

/ "/\@" only looks for the first letter (anywhere in the found word), even
  though ^A still searches for the whole word.

/ Add vim's :normal command.  When invoked without arguments, it can still
  switch to "normal" mode for backward compatibility.

/ The "syntax" display mode should allow tabs between the function name
  and the ( character.

/ Here's a weird bug: When running "elvis +make", sometimes the screen
  is cleared to a funny color during compilation, and then when the cursor
  has moved to the first error and is responding to keystrokes, it is
  stuck on the second-to-last line (just above the status line).  This is
  *not* open mode -- Hitting <z><z> fixes the scrolling.  One big clue: screen
  updates are very slow, as though it was redrawing the whole buffer from the
  top each time.

/ ":color selection" can only set attributes, not colors.

/ In the manual, the <title> sections all say "Elvis 2.1".

/ Backspacing while in indentation with autoindent is very quirky.  Apparently
  it moves the cursor back too far, when the shifting operation doesn't delete
  as much as it might in some other context.  It only occurs when smarttab is
  set.

/ The showlink option is slow if there are no <a ...> tags shortly before
  the cursor position.  It should probably stop looking if it sees a tag
  which is a paragraph/section marker.

/ Many web pages uses non-Latin1 characters for quote marks and such.  Perhaps
  ":set nonascii=hack" should translate the illegal characters into legal
  ones.  It might also be nice if nonascii characters could be flagged with
  a distinctive font, too.
	0x85	:
	0x91	`
	0x92	'
	0x93	"
	0x94	"
	0x96	-
	
/ The command "/[[:space]]" (without the second ":") causes a core dump.

/ The "termcap" interface's graphic chars don't work?  Since when?

	Actually, it works but rxvt doesn't support graphic characters when
	using the Sony "12x24" font, because that font doesn't have graphic
	characters in the usual places.  xterm works around that, but rxvt
	apparently does not.

/ Regular expressions should support \x notation inside character lists
  (the [] notation).

/ Revise the way the "draw.c" module compares the old screen image to the
  new one.  Instead of storing the old font code, it should store the old
  font's attributes: foreground RGB, background RGB, and bits.  The new
  screen image should continue to store the font code though; we need it
  to find the color code for the RGB values.  This solves the following
  problems:

  o Temporary font codes are recycled after the screen image has been
    updated, so we don't know what attributes it had.

  o Some attributes (such as guides and box-ends) are synthesized inside
    "draw.c", and are not stored in any font code (temporary or permanent).
    Synthesizing them this way requires code to be scattered throughout
    "draw.c" instead of collected in reasonable places, and it also limits
    the efficiency of updates.

  This requires replacing the di->curfont array or 1-byte codes with a
  di->curattr array of 8-byte attribute bundles.  It also requires some
  changes inside draw.c, because the way fonts are stored and compared
  will be changed.

  HOWEVER, THIS WILL NOT FIX ANOTHER TEMPORARY FONT PROBLEM: The markup
  display modes store the font code of the start of each line.  That code
  could be temporary.  A separate solution should be found for that.

/ Another good idea: Add guide lines -- vertical lines superimposed at
  certain column numbers, configurable for each window (or for each buffer?).
  Multiple guide lines should be supported.

  This should be implemented as a new "guide" option, whose value is a
  comma-delimited list of widths.  Ideally, the "tabstop" and "shiftwidth"
  options should be modified to use the same type of values.

/ Modify the way "boxed" text is drawn.  In addition to the basic "boxed"
  attribute, it should also pass bits indicating whether the top, bottom,
  left, and right edges should be drawn.

	This would be more complicated than it sounds.  Characters could
	be forced to be redrawn even if they haven't changed.  Strings of
	characters that are all in the same font could be forced to be
	drawn in different gui->text() calls, because some use the top/bottom
	edges and some don't.

	A simpler alternative might be to pass in separate bits for the
	left & right edges (of the entire chunk of text, not each character),
	but always draw the top & bottom whenever boxed text is drawn.  This
	would also give an easy, portable way to draw the guide lines.

/ Extend the :color command to support an "or <colorname>" clause.  If the
  <colorname> has a higher contrast against the background than the current
  color, then then use it; else ignore it.  This would allow commands such
  as ":color prep light green or dark green" -- if the background is light,
  preprocessor directives would be dark green; if the background is dark,
  they'll be light green.

  This *might* not need to store the color that isn't used; the command
  string would be enough.  The only time this would cause problems would
  be if the foreground was inherited but the background was specified, and
  the specified background happened to be close to the foreground.  Elvis
  couldn't choose the alternate color then, because it would have forgotten
  what it is.  Perhaps that's a good thing though -- easier to understand.

  Storing only the chosen foreground and background would also allow more
  than one alternate foreground to be specified.  For example, on a green
  background, "light green or dark green or pink" would select pink.

  This would work best if the background color was always parsed first, so
  "white or black on white" would always resolve to black on white, even if
  the inherited background would have been black.  THIS PROBABLY ISN'T
  IMPORTANT since the "or <colorname>" clause is unnecessary if you're
  forcing it to use a known background anyway.

  The :lpr command would always choose colors with the assumption that the
  background is white.  This is actually trickier than it sounds -- colors
  are only selected when a :color command is given, not when some other
  factor causes backgrounds to change.  So :color would need to compute
  both the video foreground and the lpr background, and store them both.

/ "ref write" hangs.

/ The guimoveto() function calls (*gui->moveto)() even if it is NULL.

/ The readeol option is recalculated by lib/elvis.brf even if the buffer
  already exists.  That's wrong; it should leave the option unchanged.
  Also, maybe make -b force binary I/O.  Suggested by Alexandros M Manoussakis
  <amm2@andrew.cmu.edu>

/ Some changes to :color

    / Allow "+=" to merge new attributes into old attributes.

    / Maybe allow "-=" to support removal of attributes.  This is a little
      tricky for colors and fixed/proportional.
    
    / After "+=" or "-=", synthesize a new string which configures all of
      the set attributes in a single statement.  This is required for :mkexrc
      to work reliably, and for ":color" to list complete attributes.  Doing
      this will require foreground & background color names to be stored.

/ "3ofoo^[" will insert 3 foo's, but they'll all be on the same line.  They
  should each be on a separate line.  (This also affects the . command.)

	Actually, the standard "o" command doesn't accept a count.  But if
	elvis' version does, its behavior should be more intuitive that it
	currently is.

/ When :a or :i has text on the command line (instead of following), it
  leaves the cursor on the line below the inserted text, instead of on the
  last line of the inserted text.

/ Create a "showlink" option which displays (on the status line) the value
  that tagatcursor() returns.  This would be nice in "html" mode but a pain
  in normal mode, so perhaps it should be a display option.  Either that, or
  make it sensitive to the same thing that changes the <Enter> key's behavior.

	This option would be tested in the draw.c module, so making it a
	display option would be awkward.  Make it global and check for the
	<Enter> key thing (i.e., check whether display->tagnext is defined).

-----------------------------------------------------------------------------
Fixed in 2.2b-alpha:

/ In "man" display mode, any blank lines before a .P or similar command will
  confuse elvis, preventing you from moving the cursor upward over a paragraph
  boundary.

/ The :browse and :sbrowse commands need to be tweaked.  They contain a
  work-around for a minor bug in the "html" display mode, but that bug has
  now been fixed, so the workaround (an extra space after <pre graphic>)
  should be removed.

/ Modify lowline() and lowoffset() so that if looking for a line near the end
  of a block, the newlines are counted from the end of the block instead of
  the beginning.  That should be faster.

/ Add an "%%Orientation: Landscape" comment to ps2 output after "%%CreationDate"

/ UNC ("\\host\dir\file") names still don't work correctly under Windows.

	This turned out to be due to a bug in the way elvis was checking
	Cygwin's mount table.  The bug only affected Cygwin users who had
	something mounted as "/".  For example, if "C:\" was mounted as "/"
	then "\\host\dir\file" was being interpreted as "C:\host\dir\file."

/ On terminals with "brain damaged newlines" (the :xn: termcap flag),
  modifying the last column on a line would have unpredictable results.

/ Add color support to the "html" lptype. <font color="#rrggbb">...</font>

/ Can't print HTML documents via "-c" anymore?  It dumps core.

/ lptype=html produces bad output.  ">" is output as "&gt" without a ";".
  Also, since 2.2a-alpha, it has been changing the font for each character,
  instead of only between fonts.

/ Bug: Can't interrupt elvis with ^C or anything else, with any interface.

/ New features for searches:
  / 'i' flag for ignorecase, 'c' for noignorecase, 's' for smartcase.
  / 'w' flag to imply \< \> around regexp.
  / 'x' flag to imply ^ $ around regexp.
  / Vim's smartcase.  Regexps that contain any uppercase letters force off
    the "ignorecase" option temporarily.
  / Support for \i and \I character classes. \I is any char that can start
    an identifier, and \i is any character that can appear later in an
    identifier.
  / Vim's hlsearch option -- highlight *ALL* matches, without selecting them.

/ From Merijn Brand:
  Something is broke in syntax highlighting.  Try this on an excerpt from
  the ColourChooser module [using ":display syntax perl"]: 

    $colour[1] =~ s/\s+//g ;        # Colourname
    $colour[2] = lc $colour ;       # colour name
    $colour[3] = $colour[2] ;
    $colour[3] =~ s/\b(\w)/\u$1/g ; # Colour Name
    $colour[4] = $colour[3] ;
    $colour[4] =~ s/\s+//g ;        # ColourName
    for my $i ( 5..9 ) {
        $colour[$i] =~ s/\d+$// ;   # Remove trailing digits
        } 

  and see what happens :-( to the first line.
  now change the s/\s+//g to s/\d+//g. Ooooops! 

/ ":color role1 like roll2 etc" includes all following text in the "roll2"
  name.  The "roll2" name *SHOULD* be exactly one word long every time.

/ On last line of a file, do "$v^=" to evaluate an expression.  The expression
  is stored on the last line, but the final newline is missing.

/ Add a :safely command.  It temporarily sets the safer flag (like :safer)
  while executing the rest of the line as an ex command (like :try or :then).
  This would be handy if the "ex" protocol was implemented via a :readex
  alias -- the alias could be smart enough to make "ex" links safe.

/ Add the :push command already!  Like :e, except that it saves previous
  position on the tag stack.

/ In "windows" GUI, after ":set number" you can't highlight whole lines via
  the mouse.  (The strip along the left edge of the window goes away?)

/ In "x11", the scrollwheel only works right in the text area.  Other windows
  treat it like any other mouse click.  All wheel events should be treated
  as though they occurred in the text area.

	Fixed... but the statusbar never requests mouse events, so that area
	is dead to scrollwheel events too.  That's okay.  I mostly wanted
	to make sure scrollwheel events wouldn't do anything disastrous in
	the toolbar.

/ Add a new() function, which creates a buffer but not a window, and returns
  the name of the buffer.  Optional argument is a desired buffer name, else
  it uses "Elvis untitled %d".

  This is the first time that functions will have side-effects.  If you want
  to evaluate an expression without displaying the result, just to get the
  side-effect, then you could use the :if command without a :then.

/ "make clean" should remove the executables.

-----------------------------------------------------------------------------
Fixed in 2.2a-alpha:

/ Font/color changes:
  / Support the new :color syntax
  / :color can add other attributes: bold, italic, underlined, boxed, fixed.
  / This only specifies *CHANGES* to colors or attributes.  Unspecified
    colors are retained from previous font.  Attributes are ORed.
  / Limits: 128 font names (so we can have standout versions).
  / Each font is stored as a struct containing a pointer to its name,
    fgcolor index, bgcolor index, attribute bitmap.  The name-to-index
    conversion can be slow -- it is only used when a display mode initializes
    itself.
  / When constructing a new screen image, for each character we store a font
    index, optionally with a bit set to indicate a selection.  This is compact
    and should be good enough when comparing images.
  / After an image is generated, it is combined with the "normal" font to
    inherit from it any colors/attributes which are not explicitly set.
    elsewhere.  The can also be an "idle" font.
  / Change all display modes to use new font convention.
  / Change all GUIs to use new font convention.
	(Except "vio", for now.)
  / Change all printer drivers to use new font convention.
	(Except "windows" is untested.)
  / Update the manual.

/ In input mode, make ^Z move rightward across any backspaced text, keeping
  it instead of deleting it.

/ Change the version number to 2.2a-alpha

/ Add smartargs option for displaying a function's arguments.

/ Modify "ref" to look through elvispath.  Put a file of ANSI C stubs into
  elvis' lib directory and generate a "tags" file from it.  This will allow
  "ref" to act as a reference for the standard library.