« back to all changes in this revision
Viewing changes to ed/ed.1
- browse files at revision 11
- compare with another revision
- download diff
- download tarball
- view history from revision 11
- Committer: Bazaar Package Importer
- Author(s): Daniel Baumann
- Date: 2010-06-04 17:22:03 UTC
- mfrom: (1.1.6 upstream)
- Revision ID: james.westby@ubuntu.com-20100604172203-ei85j0da495sr8ut
Tags: 1:6-1
* Adding Kai as co-maintainer.
* Merging upstream version 6.
* Merging upstream version 6.
- files added:
- ascii
- cmp
- dd
- diff
- ed
- factor
- fmt
- join
- lib9/sec
- look
- pbd
- primes
- sam
- sha1sum
- split
- strings
- unicode
- unutf
- files removed:
- rc/x.tab.h
- files modified:
- Makefile
added
removed
ed/ed.1
1
.TH ED 1
2
.SH NAME
3
ed \- text editor
4
.SH SYNOPSIS
5
.B ed
6
[
7
.B -
8
]
9
[
10
.B -o
11
]
12
[
13
.I file
14
]
15
.SH DESCRIPTION
16
.I Ed
17
is a venerable text editor.
18
.PP
19
If a
20
.I file
21
argument is given,
22
.I ed
23
simulates an
24
.L e
25
command (see below) on that file:
26
it is read into
27
.I ed's
28
buffer so that it can be edited.
29
The options are
30
.TP
31
.B -
32
Suppress the printing
33
of character counts by
34
.LR e ,
35
.LR r ,
36
and
37
.L w
38
commands and of the confirming
39
.L !
40
by
41
.L !
42
commands.
43
.TP
44
.B -o
45
(for output piping)
46
Write all output to the standard error file except writing by
47
.L w
48
commands.
49
If no
50
.I file
51
is given, make
52
.B /dev/stdout
53
the remembered file; see the
54
.L e
55
command below.
56
.PP
57
.I Ed
58
operates on a `buffer', a copy of the file it is editing;
59
changes made
60
in the buffer have no effect on the file until a
61
.L w
62
(write)
63
command is given.
64
The copy of the text being edited resides
65
in a temporary file called the
66
.IR buffer .
67
.PP
68
Commands to
69
.I ed
70
have a simple and regular structure: zero, one, or
71
two
72
.I addresses
73
followed by a single character
74
.IR command ,
75
possibly
76
followed by parameters to the command.
77
These addresses specify one or more lines in the buffer.
78
Missing addresses are supplied by default.
79
.PP
80
In general, only one command may appear on a line.
81
Certain commands allow the
82
addition of text to the buffer.
83
While
84
.I ed
85
is accepting text, it is said
86
to be in
87
.I "input mode."
88
In this mode, no commands are recognized;
89
all input is merely collected.
90
Input mode is left by typing a period
91
.L .
92
alone at the
93
beginning of a line.
94
.PP
95
.I Ed
96
supports the
97
.I "regular expression"
98
notation described in
99
.IR regexp (7).
100
Regular expressions are used in addresses to specify
101
lines and in one command
102
(see
103
.I s
104
below)
105
to specify a portion of a line which is to be replaced.
106
If it is desired to use one of
107
the regular expression metacharacters as an ordinary
108
character, that character may be preceded by
109
.RB ` \e '.
110
This also applies to the character bounding the regular
111
expression (often
112
.LR / )
113
and to
114
.L \e
115
itself.
116
.PP
117
To understand addressing in
118
.I ed
119
it is necessary to know that at any time there is a
120
.I "current line."
121
Generally, the current line is
122
the last line affected by a command; however,
123
the exact effect on the current line
124
is discussed under the description of
125
each command.
126
Addresses are constructed as follows.
127
.TP
128
1.
129
The character
130
.LR . ,
131
customarily called `dot',
132
addresses the current line.
133
.TP
134
2.
135
The character
136
.L $
137
addresses the last line of the buffer.
138
.TP
139
3.
140
A decimal number
141
.I n
142
addresses the
143
.IR n -th
144
line of the buffer.
145
.TP
146
4.
147
.BI \'x
148
addresses the line marked with the name
149
.IR x ,
150
which must be a lower-case letter.
151
Lines are marked with the
152
.L k
153
command.
154
.TP
155
5.
156
A regular expression enclosed in slashes (
157
.LR / )
158
addresses
159
the line found by searching forward from the current line
160
and stopping at the first line containing a
161
string that matches the regular expression.
162
If necessary the search wraps around to the beginning of the
163
buffer.
164
.TP
165
6.
166
A regular expression enclosed in queries
167
.L ?
168
addresses
169
the line found by searching backward from the current line
170
and stopping at the first line containing
171
a string that matches the regular expression.
172
If necessary
173
the search wraps around to the end of the buffer.
174
.TP
175
7.
176
An address followed by a plus sign
177
.L +
178
or a minus sign
179
.L -
180
followed by a decimal number specifies that address plus
181
(resp. minus) the indicated number of lines.
182
The plus sign may be omitted.
183
.TP
184
8.
185
An address followed by
186
.L +
187
(or
188
.LR - )
189
followed by a
190
regular expression enclosed in slashes specifies the first
191
matching line following (or preceding) that address.
192
The search wraps around if necessary.
193
The
194
.L +
195
may be omitted, so
196
.L 0/x/
197
addresses the
198
.I first
199
line in the buffer with an
200
.LR x .
201
Enclosing the regular expression in
202
.L ?
203
reverses the search direction.
204
.TP
205
9.
206
If an address begins with
207
.L +
208
or
209
.L -
210
the addition or subtraction is taken with respect to the current line;
211
e.g.\&
212
.L -5
213
is understood to mean
214
.LR .-5 .
215
.TP
216
10.
217
If an address ends with
218
.L +
219
or
220
.LR - ,
221
then 1 is added (resp. subtracted).
222
As a consequence of this rule and rule 9,
223
the address
224
.L -
225
refers to the line before the current line.
226
Moreover,
227
trailing
228
.L +
229
and
230
.L -
231
characters
232
have cumulative effect, so
233
.L --
234
refers to the current
235
line less 2.
236
.TP
237
11.
238
To maintain compatibility with earlier versions of the editor,
239
the character
240
.L ^
241
in addresses is
242
equivalent to
243
.LR - .
244
.PP
245
Commands may require zero, one, or two addresses.
246
Commands which require no addresses regard the presence
247
of an address as an error.
248
Commands which accept one or two addresses
249
assume default addresses when insufficient are given.
250
If more addresses are given than a command requires,
251
the last one or two (depending on what is accepted) are used.
252
.PP
253
Addresses are separated from each other typically by a comma
254
.LR , .
255
They may also be separated by a semicolon
256
.LR ; .
257
In this case the current line
258
is set to
259
the previous address before the next address is interpreted.
260
If no address precedes a comma or semicolon, line 1 is assumed;
261
if no address follows, the last line of the buffer is assumed.
262
The second address of any two-address sequence
263
must correspond to a line following the line corresponding to the first address.
264
.PP
265
In the following list of
266
.I ed
267
commands, the default addresses
268
are shown in parentheses.
269
The parentheses are not part of
270
the address, but are used to show that the given addresses are
271
the default.
272
`Dot' means the current line.
273
.TP
274
.RB (\|\fL.\fP\|) \|a
275
.br
276
.ns
277
.TP
278
<text>
279
.br
280
.ns
281
.TP
282
.B .
283
Read the given text
284
and append it after the addressed line.
285
Dot is left
286
on the last line input, if there
287
were any, otherwise at the addressed line.
288
Address
289
.L 0
290
is legal for this command; text is placed
291
at the beginning of the buffer.
292
.TP
293
.RB (\|\fL.,.\fP\|) \|b [ +- ][\fIpagesize\fP][ pln\fR]
294
Browse.
295
Print a `page', normally 20 lines.
296
The optional
297
.L +
298
(default) or
299
.L -
300
specifies whether the next or previous
301
page is to be printed.
302
The optional
303
.I pagesize
304
is the number of lines in a page.
305
The optional
306
.LR p ,
307
.LR n ,
308
or
309
.L l
310
causes printing in the specified format, initially
311
.LR p .
312
Pagesize and format are remembered between
313
.L b
314
commands.
315
Dot is left at the last line displayed.
316
.TP
317
.RB (\|\fL.,.\fP\|) \|c
318
.br
319
.ns
320
.TP
321
<text>
322
.br
323
.ns
324
.TP
325
.B .
326
Change.
327
Delete the addressed lines, then accept input
328
text to replace these lines.
329
Dot is left at the last line input; if there were none,
330
it is left at the line preceding the deleted lines.
331
.TP
332
.RB (\|\fL.,.\fP\|) \|d
333
Delete the addressed lines from the buffer.
334
Dot is set to the line following the last line deleted, or to
335
the last line of the buffer if the deleted lines had no successor.
336
.TP
337
.BI e " filename"
338
Edit.
339
Delete the entire contents of the buffer;
340
then read the named file into the buffer.
341
Dot is set to the last line of the buffer.
342
The number of characters read is typed.
343
The file name is remembered for possible use in later
344
.LR e ,
345
.LR r ,
346
or
347
.L w
348
commands.
349
If
350
.I filename
351
is missing, the remembered name is used.
352
.TP
353
.BI E " filename"
354
Unconditional
355
.LR e ;
356
see
357
.RL ` q '
358
below.
359
.TP
360
.BI f " filename"
361
Print the currently remembered file name.
362
If
363
.I filename
364
is given,
365
the currently remembered file name is first changed to
366
.IR filename .
367
.TP
368
.RB (\|\fL1,$\fP\|) \|g/\fIregular\ expression\fP/\fIcommand\ list\fP
369
.PD 0
370
.TP
371
.RB (\|\fL1,$\fP\|) \|g/\fIregular\ expression\fP/
372
.TP
373
.RB (\|\fL1,$\fP\|) \|g/\fIregular\ expression\fP
374
.PD
375
Global.
376
First mark every line which matches
377
the given
378
.IR regular expression .
379
Then for every such line, execute the
380
.I command list
381
with dot initially set to that line.
382
A single command or the first of multiple commands
383
appears on the same line with the global command.
384
All lines of a multi-line list except the last line must end with
385
.LR \e .
386
The
387
.RB \&` \&. \&'
388
terminating input mode for an
389
.LR a ,
390
.LR i ,
391
.L c
392
command may be omitted if it would be on the
393
last line of the command list.
394
The commands
395
.L g
396
and
397
.L v
398
are not permitted in the command list.
399
Any character other than space or newline may
400
be used instead of
401
.L /
402
to delimit the regular expression.
403
The second and third forms mean
404
.BI g/ regular\ expression /p \f1.
405
.TP
406
.RB (\| .\| ) \|i
407
.PD 0
408
.TP
409
<text>
410
.TP
411
.B .
412
Insert the given text before the addressed line.
413
Dot is left at the last line input, or, if there were none,
414
at the line before the addressed line.
415
This command differs from the
416
.I a
417
command only in the placement of the
418
text.
419
.PD
420
.TP
421
.RB (\| .,.+1 \|) \|j
422
Join the addressed lines into a single line;
423
intermediate newlines are deleted.
424
Dot is left at the resulting line.
425
.TP
426
.RB (\|\fL.\fP\|) \|k\fIx\fP
427
Mark the addressed line with name
428
.IR x ,
429
which must be a lower-case letter.
430
The address form
431
.BI \' x
432
then addresses this line.
433
.ne 2.5
434
.TP
435
.RB (\|\fL.,.\fP\|) \|l
436
List.
437
Print the addressed lines in an unambiguous way:
438
a tab is printed as
439
.LR \et ,
440
a backspace as
441
.LR \eb ,
442
backslashes as
443
.LR \e\e ,
444
and non-printing characters as
445
a backslash, an
446
.LR x ,
447
and four hexadecimal digits.
448
Long lines are folded,
449
with the second and subsequent sub-lines indented one tab stop.
450
If the last character in the line is a blank,
451
it is followed by
452
.LR \en .
453
An
454
.L l
455
may be appended, like
456
.LR p ,
457
to any non-I/O command.
458
.TP
459
.RB (\|\fL.,.\fP\|) \|m\fIa
460
Move.
461
Reposition the addressed lines after the line
462
addressed by
463
.IR a .
464
Dot is left at the last moved line.
465
.TP
466
.RB (\|\fL.,.\fP\|) \|n
467
Number.
468
Perform
469
.LR p ,
470
prefixing each line with its line number and a tab.
471
An
472
.L n
473
may be appended, like
474
.LR p ,
475
to any non-I/O command.
476
.TP
477
.RB (\|\fL.,.\fP\|) \|p
478
Print the addressed lines.
479
Dot is left at the last line printed.
480
A
481
.L p
482
appended to any non-I/O command causes the then current line
483
to be printed after the command is executed.
484
.TP
485
.RB (\|\fL.,.\fP\|) \|P
486
This command is a synonym for
487
.LR p .
488
.TP
489
.B q
490
Quit the editor.
491
No automatic write
492
of a file is done.
493
A
494
.L q
495
or
496
.L e
497
command is considered to be in error if the buffer has
498
been modified since the last
499
.LR w ,
500
.LR q ,
501
or
502
.L e
503
command.
504
.TP
505
.B Q
506
Quit unconditionally.
507
.TP
508
.RB ( $ )\|r\ \fIfilename\fP
509
Read in the given file after the addressed line.
510
If no
511
.I filename
512
is given, the remembered file name is used.
513
The file name is remembered if there were no
514
remembered file name already.
515
If the read is successful, the number of characters
516
read is printed.
517
Dot is left at the last line read from the file.
518
.TP
519
.RB (\|\fL.,.\fP\|) \|s\fIn\fP/\fIregular\ expression\fP/\fIreplacement\fP/
520
.PD 0
521
.TP
522
.RB (\|\fL.,.\fP\|) \|s\fIn\fP/\fIregular\ expression\fP/\fIreplacement\fP/g
523
.TP
524
.RB (\|\fL.,.\fP\|) \|s\fIn\fP/\fIregular\ expression\fP/\fIreplacement\fP
525
.PD
526
Substitute.
527
Search each addressed
528
line for an occurrence of the specified regular expression.
529
On each line in which
530
.I n
531
matches are found
532
.RI ( n
533
defaults to 1 if missing),
534
the
535
.IR n th
536
matched string is replaced by the replacement specified.
537
If the global replacement indicator
538
.L g
539
appears after the command,
540
all subsequent matches on the line are also replaced.
541
It is an error for the substitution to fail on all addressed lines.
542
Any character other than space or newline
543
may be used instead of
544
.L /
545
to delimit the regular expression
546
and the replacement.
547
Dot is left at the last line substituted.
548
The third form means
549
.BI s n / regular\ expression / replacement\fP/p\f1.
550
The second
551
.L /
552
may be omitted if the replacement is
553
empty.
554
.IP
555
An ampersand
556
.L &
557
appearing in the replacement
558
is replaced by the string matching the regular expression.
559
The characters
560
.BI \e n\f1,
561
where
562
.I n
563
is a digit,
564
are replaced by the text matched by the
565
.IR n -th
566
regular subexpression
567
enclosed between
568
.L (
569
and
570
.LR ) .
571
When
572
nested parenthesized subexpressions
573
are present,
574
.I n
575
is determined by counting occurrences of
576
.L (
577
starting from the left.
578
.IP
579
A literal
580
.LR & ,
581
.LR / ,
582
.L \e
583
or newline may be included in a replacement
584
by prefixing it with
585
.LR \e .
586
.TP
587
.RB (\|\fL.,.\fP\|) \|t\|\fIa
588
Transfer.
589
Copy the addressed lines
590
after the line addressed by
591
.IR a .
592
Dot is left at the last line of the copy.
593
.TP
594
.RB (\|\fL.,.\fP\|) \|u
595
Undo.
596
Restore the preceding contents
597
of the first addressed line (sic), which must be the last line
598
in which a substitution was made (double sic).
599
.TP
600
.RB (\|\fL1,$\fP\|) \|v/\fIregular\ expression\fP/\fIcommand\ list\fP
601
This command is the same as the global command
602
.L g
603
except that the command list is executed with
604
dot initially set to every line
605
.I except
606
those
607
matching the regular expression.
608
.TP
609
.RB (\|\fL1,$\fP\|) \|w " \fIfilename\fP"
610
Write the addressed lines to
611
the given file.
612
If the file does not exist,
613
it is created with mode 666 (readable and writable by everyone).
614
If no
615
.I filename
616
is given, the remembered file name, if any, is used.
617
The file name is remembered if there were no
618
remembered file name already.
619
Dot is unchanged.
620
If the write is successful, the number of characters written is
621
printed.
622
.TP
623
.RB (\|\fL1,$\fP\|) \|W " \fIfilename\fP"
624
Perform
625
.LR w ,
626
but append to, instead of overwriting, any existing file contents.
627
.TP
628
.RB ( $ ) \|=
629
Print the line number of the addressed line.
630
Dot is unchanged.
631
.TP
632
.BI ! shell\ command
633
Send the remainder of the line after the
634
.L !
635
to
636
.IR rc (1)
637
to be interpreted as a command.
638
Dot is unchanged.
639
.TP
640
.RB (\| .+1 )\|<newline>
641
An address without a command is taken as a
642
.L p
643
command.
644
A terminal
645
.L /
646
may be omitted from the address.
647
A blank line alone is equivalent to
648
.LR .+1p ;
649
it is useful
650
for stepping through text.
651
.PP
652
If an interrupt signal
653
.SM (DEL)
654
is sent,
655
.I ed
656
prints a
657
.L ?
658
and returns to its command level.
659
.PP
660
When reading a file,
661
.I ed
662
discards
663
.SM NUL
664
characters
665
and all characters after the last newline.
666
.SH FILES
667
.B /tmp/e*
668
.br
669
.B ed.hup
670
\ \ work is saved here if terminal hangs up
671
.SH SOURCE
672
.B \*9/src/cmd/ed.c
673
.SH "SEE ALSO"
674
.IR sam (1),
675
.IR sed (1),
676
.IR regexp (7)
677
.SH DIAGNOSTICS
678
.BI ? name
679
for inaccessible file;
680
.L ?TMP
681
for temporary file overflow;
682
.L ?
683
for errors in commands or other overflows.
Loggerhead is a web-based interface for Breezy
Version: 2.0.1