This file lists changes in Exploring Expect.  Note that some of these
changes are not "errors" per se but improvements suggested by style
and/or experience.  [LAME EXCUSE MODE OFF]

**********************************************************************
Changes to 2nd Printing
**********************************************************************

Where	Description

ii	Duplicates page i.  It should be entirely different.  Printer
	error?

p28	"addition" should read "subtraction" in the third line.

p32	switch command defaults to exact pattern matching, not glob.

p33	Last line is missing another "<space>".

p36	Procedure is misnamed.  It computes circumference, not area.

p39	First line of for loop should read:
	for {set index [expr [llength $list]-1]} {$index>=0} {incr index -1} {

p45	Delete "ASCII".

p47	Replaced "dictionary order" with "according to underlying hardware"

p62	"eof" returns 1 if an eof has already been encountered (e.g., by read
	or gets), 0 otherwise.

p103	Delete extraneous word "take".

p111	In 5th to last para, expect should be expect_out.

p116	In figure caption, expect should be expect_out.

p140	Should be expect_out(0,string) where it says expect_out(string,0).

p142	In 1st para, "are you sure" should be "for confirmation".

p153	Shell script should say "sleep", not "exec sleep".

p217	Deleted extraneous word "to".

p317	In interact command, "-reset" should come before "~032".

p328	At middle of page, the three interact patterns are missing closing ".

p329	3rd para from bottom, 1st line, expect_out should be in Courier.

p386	Last sentence, "the" is missing "e".

p517	At bottom of page, "-lexpect" should be "-lexpectk".

p524-5	Tcl 7.4 checks for overflow so replace
	233280, 9301, and 49297 with 259200, 7141, and 54773
	(this check was later removed, so this is no longer errata!!)


Several places:
	Rewrote "kill -STOP 0" as "kill -STOP [pid]" in several places
	after finding Linux doesn't support the former.

**********************************************************************
Changes to 3rd Printing
**********************************************************************

p41	2nd to last example should spell "Hello world!" like the other
	examples.

p44	First argument to linsert is a list.  The examples are right but
	the text erroneously says "name of a list".

p67	Line 13 should read "program writes" not "programs writes".	

p73	"I typed "Nice weather, eh" should read "..., eh?"

p76	2nd to last paragraph should read "...same line as their *patterns*".

p93	Last line before last code fragment should read "returns the
	string "n*w".  3rd line in code frag should read:

	expect "\\[XY]" ;# matches n followed by anything followed by w

	Third sentence in subsequent para should read "Tcl then
	produces "n*w" as the result of the XY command."

p113	last para, 1st line, should read "For instance, [a] paragraph".

p121	"Expect is much faster then" should be ".... than"

p124	2nd to last para "All of these flags follow this *model* is
	more understandable than *pattern*.

p131	There is no "gethostbyname" script.  It should be "gethostbyaddr".

p139	Line 4, English is improved by using "avoid" instead of "prevent".
	Line 6, end of line is missing a period.

p145	Regexps are missing "-re" prefix.
	Changed reference from page 135 to 133 to make clearer.

p151	Changed "Kb" to "K" for clarity.

p152	Code fragment is missing a closing brace after full_buffer line.

p170	"spawn id" references should've been removed.

p208	Remove "two" in last sentence in 3rd para.

p240	More understandable by swapping lines in first code example
	so that "set ftp" comes last.

p242-3	Refs to "cmd" are missing preceding "$" in body of procedure.

p251	Line 1 of 2nd to last para should read "is surround[ed] by".

p274	2nd para, last line, "preceding" spelled wrong.

p297	Inability of Expect to detect exec failures has been removed.
	(Alas, such a large change will have to wait for a new edition.
	However, this shouldn't cause any actual problems for anyone.)

p302	2nd para, last line, "initialized" spelled wrong.

p313	2nd to last para, middle, "the the" should be "the".

p334	2nd para, 1st line spells "preceding" wrong.

p342	The first example is vastly improved by adding "-o" before "eof".

p346	4th para, last line, replace "these kind" with "this kind".

p407	2nd to last para, "the value the" should be "the value of".

p446	Delete "t" at end of "Althought".

**********************************************************************
Changes to 4th Printing (no date set yet for this printing)
**********************************************************************

p67	Middle of page, "exec rm ..." should be "eval exec rm ...".

p191	Middle of page, line 3 of proc sendexpect should be: expect {

p252	Middle of page, "$id 2" should be "$id2".

p254	4th para, replace final period with "; host1, host2, and host3
	similarly are spawn ids."

p259	Examples should have "..." after patterns to indicate actions.

p260	action of last eof on page should be:

	eof {
	   puts "$program died unexpectedly?"
	   exit 1
	}

p266	In last example, "pat act" should be "pat1 act1".

p291	In 2nd sentence in 4th full para, "your" should be "you".

p344	In 1st full para, "effects" should be "effect".

p374	In last sentence of 1st paragraph, add "the" before "child".

p425	At bottom of page, status of breakpoint 4 should read ...:17
	to match earlier breakpoint definition.

p457	Change ":kf1=\EOP:" to "kf1=\EOP," for terminfo-style.

p477	Middle of page, exec should be in courier.

p 532	Words "before" and "after" are reversed in one-line
	descriptions of expect_after and expect_before.

**********************************************************************
For slogging through this entire file, you get to enjoy this humorous
"correction".
**********************************************************************

From: Bellave Jayaram <bellave.jayaram@tempe.vlsi.com>
To: libes@nist.gov
Subject: Exploring Expect
Date: Mon, 13 Jan 1997 17:41:53 -0700

Don,
	I was reading your book and noticed on pg. 176, the reference to
Steve Jobs.  Looks like you will have to change vrfy jobs@apple.com as
returning GOOD now :)
	I really like the book since it has so many examples.

Jayaram
 ____________________________________________________________________
| Bellave Jayaram                   | Staff CAE Engineer /DSD/DA     |
| Tel. (602) 752-6338               | VLSI Technology Inc            |
| Fax  (602) 752-6019               | 8375 S. River Pkwy. M/S 285    |
| bellave.jayaram@tempe.vlsi.com    | Tempe, AZ 85284                |
|____________________________________________________________________|