NameDateSize

..20-Dec-201670

build/H20-Dec-201619

catalog/H20-Dec-201639

cl/H20-Dec-201610

clib/H20-Dec-201619

common/H20-Dec-201635

docs/H20-Dec-201612

ex/H20-Dec-201656

FAQH A D08-Mar-20157.5 KiB

include/H20-Dec-201614

ip/H20-Dec-20169

ip_cl/H20-Dec-20164

LAYOUTH A D08-Mar-20153.8 KiB

LICENSEH A D08-Mar-20152.2 KiB

perl_api/H20-Dec-20167

perl_scripts/H20-Dec-20166

READMEH A D08-Mar-20154.7 KiB

tcl_api/H20-Dec-20163

tcl_scripts/H20-Dec-20166

tk/H20-Dec-201610

vi/H20-Dec-201642

README

1#	@(#)README	8.147 (Berkeley) 10/19/96
2
3This is the README for nex/nvi, a freely redistributable implementation
4of the ex/vi text editors originally distributed as part of the Fourth
5Berkeley Software Distribution (4BSD), by the University of California,
6Berkeley.
7
8The source code for nex/nvi can be retrieved by using anonymous ftp to
9ftp.cs.berkeley.edu.  The file ucb/4bsd/nvi.tar.gz is the gzip'd archive,
10of version 1.71 of nex/nvi.  This version is believed to be stable and
11problem free.  The file ucb/4bsd/nvi-###.ALPHA.tar.gz is a gzip'd archive
12of the current alpha-test release of nex/nvi.  This version reflects the
13current development tree, and will be more likely to have problems.
14
15See the file:
16	build/README	for information on building nvi.
17	LAYOUT		for a description of where everything is.
18	LICENSE		for the copyright and redistribution terms.
19
20If you have any questions about nex/nvi, problems with it, or concerns
21about the conditions for redistribution, please contact me:
22
23	Keith Bostic		+1-508-287-4781
24	394 E. Riding Dr.	bostic@bostic.com
25	Carlisle, MA 01741
26	USA
27
28Keith Bostic
29
30=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
31o This software is several years old and is the product of many folks' work.  
32
33	This software was originally derived from software contributed to
34	the University of California, Berkeley by Steve Kirkendall, the
35	author of the vi clone elvis.  Without his work, this work would
36	have been far more difficult.
37
38	IEEE POSIX 1003.2 style regular expression support is courtesy of
39	Henry Spencer, for which I am *very* grateful.
40
41	Elan Amir did the original 4BSD curses work that made it possible
42	to support a full-screen editor using curses.
43
44	George Neville-Neil added the Tcl interpreter, and the initial
45	interpreter design was his.
46
47	Sven Verdoolaege added the Perl interpreter.
48
49	Rob Mayoff provided the original Cscope support.
50
51o Many, many people suggested enhancements, and provided bug reports and
52  testing, far too many to individually thank.
53
54o From the original vi acknowledgements, by William Joy and Mark Horton:
55
56	Bruce Englar encouraged the early development of this display
57	editor.  Peter Kessler helped bring sanity to version 2's
58	command layout.  Bill Joy wrote versions 1 and 2.0 through 2.7,
59	and created the framework that users see in the present editor.
60	Mark Horton added macros and other features and made the editor
61	work on a large number of terminals and Unix systems.
62
63o And...
64	The financial support of UUNET Communications Services is gratefully
65	acknowledged.
66
67=-=-=-=-=-=-=-=-=-=-=
68o Status:
69
70This software is in beta test, and it's pretty stable.  Almost all of the
71historic functionality in ex/vi is there, the only major missing pieces
72are open mode and the lisp edit option.
73
74Nvi is largely 8-bit clean.  This isn't difficult to fix, and was left in
75during initial development to keep things simple.  Wide character support
76will be integrated at the same time that it is made fully 8-bit clean.
77
78There aren't a lot of new features in nex/nvi, but there are a few things
79you might like.  The "Additional Features" section of the reference work
80(docs/USD.doc/vi.ref/vi.ref.txt, docs/USD.doc/vi.ref/vi.ref.ps) has more
81information.
82
83=-=-=-=-=-=-=-=-=-=-=
84o Debugging:
85
86Code fixes are greatly appreciated, of course, but if you can't provide
87them, please email me as much information as you can as to how I might
88reproduce the bug, and I'll try to fix it locally.  Stack traces of core
89dumps are only rarely helpful -- an example file with a set of keystrokes
90that causes the problem is almost invariably necessary.  I know it's
91annoying, but simply playing with the bug until you can reproduce it at
92will, with minimal keystrokes, is immensely helpful to me.
93
94Please include the following in the bug report;
95
96	o The version of nvi you're running (use :version to get it).
97	o The row/column dimensions of the screen (80 x 32).
98	o Unless you're confident that they're not part of the problem,
99	  your startup files (.exrc, .nexrc) and the environment variable
100	  (EXINIT, NEXINIT) values.  (Cutting and pasting the output
101	  of ":set all" is usually sufficient.)
102
103If you want to do your own debugging, recompile the program with DEBUG
104defined.  (Configuring with --enable-debug will do this for you.)  This
105turns on the additional command-line option -D, that takes either s or w
106as an argument.  The option -Ds causes nvi to ignore the EXINIT and
107.exrc files on startup, and -Dw causes nvi to print out the process id
108and wait for you to enter a <carriage-return> to continue.
109
110If you're running a memory checker (e.g. Purify) on nvi, you will first
111want to  recompile everything with "-DPURIFY" set in the CFLAGS.  This
112initializes allocated pages in the DB code, and free's allocated memory
113at the end of the nvi execution.
114