summaryrefslogtreecommitdiffstats
path: root/runtime/doc/usr_30.txt
blob: a23df9660697b663f583c9b9ace4d28dd6de38fd (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
*usr_30.txt*	For Vim version 9.0.  Last change: 2007 Nov 10

		     VIM USER MANUAL - by Bram Moolenaar

			      Editing programs


Vim has various commands that aid in writing computer programs.  Compile a
program and directly jump to reported errors.  Automatically set the indent
for many languages and format comments.

|30.1|	Compiling
|30.2|	Indenting C files
|30.3|	Automatic indenting
|30.4|	Other indenting
|30.5|	Tabs and spaces
|30.6|	Formatting comments

     Next chapter: |usr_31.txt|  Exploiting the GUI
 Previous chapter: |usr_29.txt|  Moving through programs
Table of contents: |usr_toc.txt|

==============================================================================
*30.1*	Compiling

Vim has a set of so called "quickfix" commands.  They enable you to compile a
program from within Vim and then go through the errors generated and fix them
(hopefully).  You can then recompile and fix any new errors that are found
until finally your program compiles without any error.

The following command runs the program "make" (supplying it with any argument
you give) and captures the results: >

	:make {arguments}

If errors were generated, they are captured and the editor positions you where
the first error occurred.
   Take a look at an example ":make" session.  (Typical :make sessions generate
far more errors and fewer stupid ones.)  After typing ":make" the screen looks
like this:

	:!make | &tee /tmp/vim215953.err ~
	gcc -g -Wall -o prog main.c sub.c ~
	main.c: In function 'main': ~
	main.c:6: too many arguments to function 'do_sub' ~
	main.c: At top level: ~
	main.c:10: parse error before '}' ~
	make: *** [prog] Error 1 ~

	2 returned ~
	"main.c" 11L, 111C ~
	(3 of 6): too many arguments to function 'do_sub' ~
	Press ENTER or type command to continue ~

From this you can see that you have errors in the file "main.c".  When you
press <Enter>, Vim displays the file "main.c", with the cursor positioned on
line 6, the first line with an error.  You did not need to specify the file or
the line number, Vim knew where to go by looking in the error messages.

		+---------------------------------------------------+
		|int main()					    |
		|{						    |
		|	int i=3;				    |
      cursor -> |	do_sub("foo");				    |
		|	++i;					    |
		|	return (0);				    |
		|}						    |
		|}						    |
		| ~						    |
		|(3 of 12): too many arguments to function 'do_sub' |
		+---------------------------------------------------+

The following command goes to where the next error occurs: >

	:cnext

Vim jumps to line 10, the last line in the file, where there is an extra '}'.
   When there is not enough room, Vim will shorten the error message.  To see
the whole message use: >

	:cc

You can get an overview of all the error messages with the ":clist" command.
The output looks like this: >

	:clist
<	3 main.c: 6:too many arguments to function 'do_sub' ~
	5 main.c: 10:parse error before '}' ~

Only the lines where Vim recognized a file name and line number are listed
here.  It assumes those are the interesting lines and the rest is just boring
messages.  However, sometimes unrecognized lines do contain something you want
to see.  Output from the linker, for example, about an undefined function.
To see all the messages add a "!" to the command: >

	:clist!
<	1 gcc -g -Wall -o prog main.c sub.c ~
	2 main.c: In function 'main': ~
	3 main.c:6: too many arguments to function 'do_sub' ~
	4 main.c: At top level: ~
	5 main.c:10: parse error before '}' ~
	6 make: *** [prog] Error 1 ~

Vim will highlight the current error.  To go back to the previous error, use:
>
	:cprevious

Other commands to move around in the error list:

	:cfirst		to first error
	:clast		to last error
	:cc 3		to error nr 3


USING ANOTHER COMPILER

The name of the program to run when the ":make" command is executed is defined
by the 'makeprg' option.  Usually this is set to "make", but Visual C++ users
should set this to "nmake" by executing the following command: >

	:set makeprg=nmake

You can also include arguments in this option.  Special characters need to
be escaped with a backslash.  Example: >

	:set makeprg=nmake\ -f\ project.mak

You can include special Vim keywords in the command specification.  The %
character expands to the name of the current file.  So if you execute the
command: >
	:set makeprg=make\ %:S

When you are editing main.c, then ":make" executes the following command: >

	make main.c

This is not too useful, so you will refine the command a little and use the :r
(root) modifier: >

	:set makeprg=make\ %:r:S.o

Now the command executed is as follows: >

	make main.o

More about these modifiers here: |filename-modifiers|.


OLD ERROR LISTS

Suppose you ":make" a program.  There is a warning message in one file and an
error message in another.  You fix the error and use ":make" again to check if
it was really fixed.  Now you want to look at the warning message.  It doesn't
show up in the last error list, since the file with the warning wasn't
compiled again.  You can go back to the previous error list with: >

	:colder

Then use ":clist" and ":cc {nr}" to jump to the place with the warning.
   To go forward to the next error list: >

	:cnewer

Vim remembers ten error lists.


SWITCHING COMPILERS

You have to tell Vim what format the error messages are that your compiler
produces.  This is done with the 'errorformat' option.  The syntax of this
option is quite complicated and it can be made to fit almost any compiler.
You can find the explanation here: |errorformat|.

You might be using various different compilers.  Setting the 'makeprg' option,
and especially the 'errorformat' each time is not easy.  Vim offers a simple
method for this.  For example, to switch to using the Microsoft Visual C++
compiler: >

	:compiler msvc

This will find the Vim script for the "msvc" compiler and set the appropriate
options.
   You can write your own compiler files.  See |write-compiler-plugin|.


OUTPUT REDIRECTION

The ":make" command redirects the output of the executed program to an error
file.  How this works depends on various things, such as the 'shell'.  If your
":make" command doesn't capture the output, check the 'makeef' and
'shellpipe' options.  The 'shellquote' and 'shellxquote' options might also
matter.

In case you can't get ":make" to redirect the file for you, an alternative is
to compile the program in another window and redirect the output into a file.
Then have Vim read this file with: >

	:cfile {filename}

Jumping to errors will work like with the ":make" command.

==============================================================================
*30.2*	Indenting C style text

A program is much easier to understand when the lines have been properly
indented.  Vim offers various ways to make this less work.  For C or C style
programs like Java or C++, set the 'cindent' option.  Vim knows a lot about C
programs and will try very hard to automatically set the indent for you.  Set
the 'shiftwidth' option to the amount of spaces you want for a deeper level.
Four spaces will work fine.  One ":set" command will do it: >

	:set cindent shiftwidth=4

With this option enabled, when you type something such as "if (x)", the next
line will automatically be indented an additional level.

				    if (flag)
	Automatic indent   --->		do_the_work();
	Automatic unindent <--	    if (other_flag) {
	Automatic indent   --->		do_file();
	keep indent			do_some_more();
	Automatic unindent <--	    }

When you type something in curly braces ({}), the text will be indented at the
start and unindented at the end.  The unindenting will happen after typing the
'}', since Vim can't guess what you are going to type.

One side effect of automatic indentation is that it helps you catch errors in
your code early.  When you type a } to finish a function, only to find that
the automatic indentation gives it more indent than what you expected, there
is probably a } missing.  Use the "%" command to find out which { matches the
} you typed.
   A missing ) and ; also cause extra indent.  Thus if you get more white
space than you would expect, check the preceding lines.

When you have code that is badly formatted, or you inserted and deleted lines,
you need to re-indent the lines.  The "=" operator does this.  The simplest
form is: >

	==

This indents the current line.  Like with all operators, there are three ways
to use it.  In Visual mode "=" indents the selected lines.  A useful text
object is "a{".  This selects the current {} block.  Thus, to re-indent the
code block the cursor is in: >

	=a{

I you have really badly indented code, you can re-indent the whole file with:
>
	gg=G

However, don't do this in files that have been carefully indented manually.
The automatic indenting does a good job, but in some situations you might want
to overrule it.


SETTING INDENT STYLE

Different people have different styles of indentation.  By default Vim does a
pretty good job of indenting in a way that 90% of programmers do.  There are
different styles, however; so if you want to, you can customize the
indentation style with the 'cinoptions' option.
   By default 'cinoptions' is empty and Vim uses the default style.  You can
add various items where you want something different.  For example, to make
curly braces be placed like this:

	if (flag) ~
	  { ~
	    i = 8; ~
	    j = 0; ~
	  } ~

Use this command: >

	:set cinoptions+={2

There are many of these items.  See |cinoptions-values|.

==============================================================================
*30.3*	Automatic indenting

You don't want to switch on the 'cindent' option manually every time you edit
a C file.  This is how you make it work automatically: >

	:filetype indent on

Actually, this does a lot more than switching on 'cindent' for C files.  First
of all, it enables detecting the type of a file.  That's the same as what is
used for syntax highlighting.
   When the filetype is known, Vim will search for an indent file for this
type of file.  The Vim distribution includes a number of these for various
programming languages.  This indent file will then prepare for automatic
indenting specifically for this file.

If you don't like the automatic indenting, you can switch it off again: >

	:filetype indent off

If you don't like the indenting for one specific type of file, this is how you
avoid it.  Create a file with just this one line: >

	:let b:did_indent = 1

Now you need to write this in a file with a specific name:

	{directory}/indent/{filetype}.vim

The {filetype} is the name of the file type, such as "cpp" or "java".  You can
see the exact name that Vim detected with this command: >

	:set filetype

In this file the output is:

	filetype=help ~

Thus you would use "help" for {filetype}.
   For the {directory} part you need to use your runtime directory.  Look at
the output of this command: >

	set runtimepath

Now use the first item, the name before the first comma.  Thus if the output
looks like this:

	runtimepath=~/.vim,/usr/local/share/vim/vim60/runtime,~/.vim/after ~

You use "~/.vim" for {directory}.  Then the resulting file name is:

	~/.vim/indent/help.vim ~

Instead of switching the indenting off, you could write your own indent file.
How to do that is explained here: |indent-expression|.

==============================================================================
*30.4*	Other indenting

The simplest form of automatic indenting is with the 'autoindent' option.
It uses the indent from the previous line.  A bit smarter is the 'smartindent'
option.  This is useful for languages where no indent file is available.
'smartindent' is not as smart as 'cindent', but smarter than 'autoindent'.
   With 'smartindent' set, an extra level of indentation is added for each {
and removed for each }.  An extra level of indentation will also be added for
any of the words in the 'cinwords' option.  Lines that begin with # are
treated specially: all indentation is removed.  This is done so that
preprocessor directives will all start in column 1.  The indentation is
restored for the next line.


CORRECTING INDENTS

When you are using 'autoindent' or 'smartindent' to get the indent of the
previous line, there will be many times when you need to add or remove one
'shiftwidth' worth of indent.  A quick way to do this is using the CTRL-D and
CTRL-T commands in Insert mode.
   For example, you are typing a shell script that is supposed to look like
this:

	if test -n a; then ~
	   echo a ~
	   echo "-------" ~
	fi ~

Start off by setting these options: >

	:set autoindent shiftwidth=3

You start by typing the first line, <Enter> and the start of the second line:

	if test -n a; then ~
	echo ~

Now you see that you need an extra indent.  Type CTRL-T.  The result:

	if test -n a; then ~
	   echo ~

The CTRL-T command, in Insert mode, adds one 'shiftwidth' to the indent, no
matter where in the line you are.
   You continue typing the second line, <Enter> and the third line.  This time
the indent is OK.  Then <Enter> and the last line.  Now you have this:

	if test -n a; then ~
	   echo a ~
	   echo "-------" ~
	   fi ~

To remove the superfluous indent in the last line press CTRL-D.  This deletes
one 'shiftwidth' worth of indent, no matter where you are in the line.
   When you are in Normal mode, you can use the ">>" and "<<" commands to
shift lines.  ">" and "<" are operators, thus you have the usual three ways to
specify the lines you want to indent.  A useful combination is: >

	>i{

This adds one indent to the current block of lines, inside {}.  The { and }
lines themselves are left unmodified.  ">a{" includes them.  In this example
the cursor is on "printf":

	original text		after ">i{"		after ">a{"

	if (flag)		if (flag)		if (flag) ~
	{			{			    { ~
	printf("yes");		    printf("yes");	    printf("yes"); ~
	flag = 0;		    flag = 0;		    flag = 0;  ~
	}			}			    } ~

==============================================================================
*30.5*	Tabs and spaces

'tabstop' is set to eight by default.  Although you can change it, you quickly
run into trouble later.  Other programs won't know what tabstop value you
used.  They probably use the default value of eight, and your text suddenly
looks very different.  Also, most printers use a fixed tabstop value of eight.
Thus it's best to keep 'tabstop' alone.  (If you edit a file which was written
with a different tabstop setting, see |25.3| for how to fix that.)
   For indenting lines in a program, using a multiple of eight spaces makes
you quickly run into the right border of the window.  Using a single space
doesn't provide enough visual difference.  Many people prefer to use four
spaces, a good compromise.
   Since a <Tab> is eight spaces and you want to use an indent of four spaces,
you can't use a <Tab> character to make your indent.  There are two ways to
handle this:

1.  Use a mix of <Tab> and space characters.  Since a <Tab> takes the place of
    eight spaces, you have fewer characters in your file.  Inserting a <Tab>
    is quicker than eight spaces.  Backspacing works faster as well.

2.  Use spaces only.  This avoids the trouble with programs that use a
    different tabstop value.

Fortunately, Vim supports both methods quite well.


SPACES AND TABS

If you are using a combination of tabs and spaces, you just edit normally.
The Vim defaults do a fine job of handling things.
   You can make life a little easier by setting the 'softtabstop' option.
This option tells Vim to make the <Tab> key look and feel as if tabs were set
at the value of 'softtabstop', but actually use a combination of tabs and
spaces.
   After you execute the following command, every time you press the <Tab> key
the cursor moves to the next 4-column boundary: >

	:set softtabstop=4

When you start in the first column and press <Tab>, you get 4 spaces inserted
in your text.  The second time, Vim takes out the 4 spaces and puts in a <Tab>
(thus taking you to column 8).  Thus Vim uses as many <Tab>s as possible, and
then fills up with spaces.
   When backspacing it works the other way around.  A <BS> will always delete
the amount specified with 'softtabstop'.  Then <Tab>s are used as many as
possible and spaces to fill the gap.
   The following shows what happens pressing <Tab> a few times, and then using
<BS>.  A "." stands for a space and "------->" for a <Tab>.

	type			  result ~
	<Tab>			  ....
	<Tab><Tab>		  ------->
	<Tab><Tab><Tab>		  ------->....
	<Tab><Tab><Tab><BS>	  ------->
	<Tab><Tab><Tab><BS><BS>   ....

An alternative is to use the 'smarttab' option.  When it's set, Vim uses
'shiftwidth' for a <Tab> typed in the indent of a line, and a real <Tab> when
typed after the first non-blank character.  However, <BS> doesn't work like
with 'softtabstop'.


JUST SPACES

If you want absolutely no tabs in your file, you can set the 'expandtab'
option: >

	:set expandtab

When this option is set, the <Tab> key inserts a series of spaces.  Thus you
get the same amount of white space as if a <Tab> character was inserted, but
there isn't a real <Tab> character in your file.
   The backspace key will delete each space by itself.  Thus after typing one
<Tab> you have to press the <BS> key up to eight times to undo it.  If you are
in the indent, pressing CTRL-D will be a lot quicker.


CHANGING TABS IN SPACES (AND BACK)

Setting 'expandtab' does not affect any existing tabs.  In other words, any
tabs in the document remain tabs.  If you want to convert tabs to spaces, use
the ":retab" command.  Use these commands: >

	:set expandtab
	:%retab

Now Vim will have changed all indents to use spaces instead of tabs.  However,
all tabs that come after a non-blank character are kept.  If you want these to
be converted as well, add a !: >

	:%retab!

This is a little bit dangerous, because it can also change tabs inside a
string.  To check if these exist, you could use this: >

	/"[^"\t]*\t[^"]*"

It's recommended not to use hard tabs inside a string.  Replace them with
"\t" to avoid trouble.

The other way around works just as well: >

	:set noexpandtab
	:%retab!

==============================================================================
*30.6*	Formatting comments

One of the great things about Vim is that it understands comments.  You can
ask Vim to format a comment and it will do the right thing.
   Suppose, for example, that you have the following comment:

	/* ~
	 * This is a test ~
	 * of the text formatting. ~
	 */ ~

You then ask Vim to format it by positioning the cursor at the start of the
comment and type: >

	gq]/

"gq" is the operator to format text.  "]/" is the motion that takes you to the
end of a comment.  The result is:

	/* ~
	 * This is a test of the text formatting. ~
	 */ ~

Notice that Vim properly handled the beginning of each line.
  An alternative is to select the text that is to be formatted in Visual mode
and type "gq".

To add a new line to the comment, position the cursor on the middle line and
press "o".  The result looks like this:

	/* ~
	 * This is a test of the text formatting. ~
	 * ~
	 */ ~

Vim has automatically inserted a star and a space for you.  Now you can type
the comment text.  When it gets longer than 'textwidth', Vim will break the
line.  Again, the star is inserted automatically:

	/* ~
	 * This is a test of the text formatting. ~
	 * Typing a lot of text here will make Vim ~
	 * break ~
	 */ ~

For this to work some flags must be present in 'formatoptions':

	r	insert the star when typing <Enter> in Insert mode
	o	insert the star when using "o" or "O" in Normal mode
	c	break comment text according to 'textwidth'

See |fo-table| for more flags.


DEFINING A COMMENT

The 'comments' option defines what a comment looks like.  Vim distinguishes
between a single-line comment and a comment that has a different start, end
and middle part.
   Many single-line comments start with a specific character.  In C++ // is
used, in Makefiles #, in Vim scripts ".  For example, to make Vim understand
C++ comments: >

	:set comments=://

The colon separates the flags of an item from the text by which the comment is
recognized.  The general form of an item in 'comments' is:

	{flags}:{text}

The {flags} part can be empty, as in this case.
   Several of these items can be concatenated, separated by commas.  This
allows recognizing different types of comments at the same time.  For example,
let's edit an e-mail message.  When replying, the text that others wrote is
preceded with ">" and "!" characters.  This command would work: >

	:set comments=n:>,n:!

There are two items, one for comments starting with ">" and one for comments
that start with "!".  Both use the flag "n".  This means that these comments
nest.  Thus a line starting with ">" may have another comment after the ">".
This allows formatting a message like this:

	> ! Did you see that site? ~
	> ! It looks really great. ~
	> I don't like it.  The ~
	> colors are terrible. ~
	What is the URL of that ~
	site? ~

Try setting 'textwidth' to a different value, e.g., 80, and format the text by
Visually selecting it and typing "gq".  The result is:

	> ! Did you see that site?  It looks really great. ~
	> I don't like it.  The colors are terrible. ~
	What is the URL of that site? ~

You will notice that Vim did not move text from one type of comment to
another.  The "I" in the second line would have fit at the end of the first
line, but since that line starts with "> !" and the second line with ">", Vim
knows that this is a different kind of comment.


A THREE PART COMMENT

A C comment starts with "/*", has "*" in the middle and "*/" at the end.  The
entry in 'comments' for this looks like this: >

	:set comments=s1:/*,mb:*,ex:*/

The start is defined with "s1:/*".  The "s" indicates the start of a
three-piece comment.  The colon separates the flags from the text by which the
comment is recognized: "/*".  There is one flag: "1".  This tells Vim that the
middle part has an offset of one space.
   The middle part "mb:*" starts with "m", which indicates it is a middle
part.  The "b" flag means that a blank must follow the text.  Otherwise Vim
would consider text like "*pointer" also to be the middle of a comment.
   The end part "ex:*/" has the "e" for identification.  The "x" flag has a
special meaning.  It means that after Vim automatically inserted a star,
typing / will remove the extra space.

For more details see |format-comments|.

==============================================================================

Next chapter: |usr_31.txt|  Exploiting the GUI

Copyright: see |manual-copyright|  vim:tw=78:ts=8:noet:ft=help:norl: