3
@node Syntax, Evaluation and Compilation, Introduction (Introduction), Top
9
* Interpretation of Tokens::
10
* Standard Macro Characters::
13
@node Character Syntax, Reader Algorithm, Syntax, Syntax
14
@section Character Syntax
16
@c including concept-syntax
18
The @i{Lisp reader} takes @i{characters} from a @i{stream},
19
interprets them as a printed representation of an @i{object},
20
constructs that @i{object}, and returns it.
22
The syntax described by this chapter is called the @i{standard syntax}
23
@IGindex{standard syntax}
25
Operations are provided by @r{Common Lisp} so that
26
various aspects of the syntax information represented by a @i{readtable}
27
can be modified under program control; see @ref{Reader}.
28
Except as explicitly stated otherwise,
29
the syntax used throughout this document is @i{standard syntax}.
33
* Variables that affect the Lisp Reader::
34
* Standard Characters::
35
* Character Syntax Types::
38
@node Readtables, Variables that affect the Lisp Reader, Character Syntax, Character Syntax
39
@subsection Readtables
41
Syntax information for use by the @i{Lisp reader} is embodied in an
42
@i{object} called a @i{readtable}
45
the @i{readtable} contains the association between @i{characters}
48
Figure 2--1 lists some @i{defined names} that are applicable to
54
@w{ *readtable* readtable-case }
55
@w{ copy-readtable readtablep }
56
@w{ get-dispatch-macro-character set-dispatch-macro-character }
57
@w{ get-macro-character set-macro-character }
58
@w{ make-dispatch-macro-character set-syntax-from-char }
61
@w{ Figure 2--1: Readtable defined names }
67
* The Current Readtable::
68
* The Standard Readtable::
69
* The Initial Readtable::
72
@node The Current Readtable, The Standard Readtable, Readtables, Readtables
73
@subsubsection The Current Readtable
75
Several @i{readtables} describing different syntaxes can exist,
76
but at any given time only one, called the @i{current readtable}
77
@IGindex{current readtable}
79
affects the way in which @i{expressions}_2 are parsed
80
into @i{objects} by the @i{Lisp reader}.
81
The @i{current readtable} in a given @i{dynamic environment}
82
is the @i{value} of @b{*readtable*} in that @i{environment}.
83
To make a different @i{readtable} become the @i{current readtable},
84
@b{*readtable*} can be @i{assigned} or @i{bound}.
86
@node The Standard Readtable, The Initial Readtable, The Current Readtable, Readtables
87
@subsubsection The Standard Readtable
89
The @i{standard readtable}
90
@IGindex{standard readtable}
91
conforms to @i{standard syntax}.
92
The consequences are undefined if an attempt is made
93
to modify the @i{standard readtable}.
94
To achieve the effect of altering or extending @i{standard syntax},
95
a copy of the @i{standard readtable} can be created; see the @i{function} @b{copy-readtable}.
97
The @i{readtable case} of the @i{standard readtable} is @t{:upcase}.
99
@node The Initial Readtable, , The Standard Readtable, Readtables
100
@subsubsection The Initial Readtable
102
The @i{initial readtable}
103
@IGindex{initial readtable}
105
the @i{readtable} that is the @i{current readtable}
106
at the time when the @i{Lisp image} starts.
107
At that time, it conforms to @i{standard syntax}.
108
The @i{initial readtable} is @i{distinct}
109
from the @i{standard readtable}.
110
It is permissible for a @i{conforming program}
111
to modify the @i{initial readtable}.
113
@node Variables that affect the Lisp Reader, Standard Characters, Readtables, Character Syntax
114
@subsection Variables that affect the Lisp Reader
116
The @i{Lisp reader} is influenced not only by the @i{current readtable},
117
but also by various @i{dynamic variables}. Figure 2--2 lists
118
the @i{variables} that influence the behavior of the @i{Lisp reader}.
123
@w{ *package* *read-default-float-format* *readtable* }
124
@w{ *read-base* *read-suppress* }
127
@w{ Figure 2--2: Variables that influence the Lisp reader. }
132
@node Standard Characters, Character Syntax Types, Variables that affect the Lisp Reader, Character Syntax
133
@subsection Standard Characters
135
All @i{implementations} must support a @i{character} @i{repertoire}
136
called @b{standard-char}; @i{characters} that are members of that
137
@i{repertoire} are called @i{standard characters}
138
@IGindex{standard character}
141
The @b{standard-char} @i{repertoire} consists of
142
the @i{non-graphic} @i{character} @i{newline},
143
the @i{graphic} @i{character} @i{space},
144
and the following additional
145
ninety-four @i{graphic} @i{characters} or their equivalents:
150
@w{ Graphic ID Glyph Description Graphic ID Glyph Description }
151
@w{ LA01 @t{a} small a LN01 @t{n} small n }
152
@w{ LA02 @t{A} capital A LN02 @t{N} capital N }
153
@w{ LB01 @t{b} small b LO01 @t{o} small o }
154
@w{ LB02 @t{B} capital B LO02 @t{O} capital O }
155
@w{ LC01 @t{c} small c LP01 @t{p} small p }
156
@w{ LC02 @t{C} capital C LP02 @t{P} capital P }
157
@w{ LD01 @t{d} small d LQ01 @t{q} small q }
158
@w{ LD02 @t{D} capital D LQ02 @t{Q} capital Q }
159
@w{ LE01 @t{e} small e LR01 @t{r} small r }
160
@w{ LE02 @t{E} capital E LR02 @t{R} capital R }
161
@w{ LF01 @t{f} small f LS01 @t{s} small s }
162
@w{ LF02 @t{F} capital F LS02 @t{S} capital S }
163
@w{ LG01 @t{g} small g LT01 @t{t} small t }
164
@w{ LG02 @t{G} capital G LT02 @t{T} capital T }
165
@w{ LH01 @t{h} small h LU01 @t{u} small u }
166
@w{ LH02 @t{H} capital H LU02 @t{U} capital U }
167
@w{ LI01 @t{i} small i LV01 @t{v} small v }
168
@w{ LI02 @t{I} capital I LV02 @t{V} capital V }
169
@w{ LJ01 @t{j} small j LW01 @t{w} small w }
170
@w{ LJ02 @t{J} capital J LW02 @t{W} capital W }
171
@w{ LK01 @t{k} small k LX01 @t{x} small x }
172
@w{ LK02 @t{K} capital K LX02 @t{X} capital X }
173
@w{ LL01 @t{l} small l LY01 @t{y} small y }
174
@w{ LL02 @t{L} capital L LY02 @t{Y} capital Y }
175
@w{ LM01 @t{m} small m LZ01 @t{z} small z }
176
@w{ LM02 @t{M} capital M LZ02 @t{Z} capital Z }
179
@w{ Figure 2--3: Standard Character Subrepertoire (Part 1 of 3: Latin Characters)}
187
@w{ Graphic ID Glyph Description Graphic ID Glyph Description }
188
@w{ ND01 @t{1} digit 1 ND06 @t{6} digit 6 }
189
@w{ ND02 @t{2} digit 2 ND07 @t{7} digit 7 }
190
@w{ ND03 @t{3} digit 3 ND08 @t{8} digit 8 }
191
@w{ ND04 @t{4} digit 4 ND09 @t{9} digit 9 }
192
@w{ ND05 @t{5} digit 5 ND10 @t{0} digit 0 }
195
@w{ Figure 2--4: Standard Character Subrepertoire (Part 2 of 3: Numeric Characters)}
203
@w{ Graphic ID Glyph Description }
204
@w{ SP02 @t{!} exclamation mark }
205
@w{ SC03 @t{$} dollar sign }
206
@w{ SP04 @t{"} quotation mark, or double quote }
207
@w{ SP05 @t{'} apostrophe, or @r{[}single@r{]} quote }
208
@w{ SP06 @t{(} left parenthesis, or open parenthesis }
209
@w{ SP07 @t{)} right parenthesis, or close parenthesis }
210
@w{ SP08 @t{,} comma }
211
@w{ SP09 @t{_} low line, or underscore }
212
@w{ SP10 @t{-} hyphen, or minus @r{[}sign@r{]} }
213
@w{ SP11 @t{.} full stop, period, or dot }
214
@w{ SP12 @t{/} solidus, or slash }
215
@w{ SP13 @t{:} colon }
216
@w{ SP14 @t{;} semicolon }
217
@w{ SP15 @t{?} question mark }
218
@w{ SA01 @t{+} plus @r{[}sign@r{]} }
219
@w{ SA03 @t{<} less-than @r{[}sign@r{]} }
220
@w{ SA04 @t{=} equals @r{[}sign@r{]} }
221
@w{ SA05 @t{>} greater-than @r{[}sign@r{]} }
222
@w{ SM01 @t{#} number sign, or sharp@r{[}sign@r{]} }
223
@w{ SM02 @t{%} percent @r{[}sign@r{]} }
224
@w{ SM03 @t{&} ampersand }
225
@w{ SM04 @t{*} asterisk, or star }
226
@w{ SM05 @t{@@} commercial at, or at-sign }
227
@w{ SM06 @t{[} left @r{[}square@r{]} bracket }
228
@w{ SM07 @t{\} reverse solidus, or backslash }
229
@w{ SM08 @t{]} right @r{[}square@r{]} bracket }
230
@w{ SM11 @t{@{} left curly bracket, or left brace }
231
@w{ SM13 @t{|} vertical bar }
232
@w{ SM14 @t{@}} right curly bracket, or right brace }
233
@w{ SD13 @t{`} grave accent, or backquote }
234
@w{ SD15 @t{@t{^}} circumflex accent }
235
@w{ SD19 @t{~} tilde }
238
@w{ Figure 2--5: Standard Character Subrepertoire (Part 3 of 3: Special Characters)}
243
The graphic IDs are not used within @r{Common Lisp},
244
but are provided for cross reference purposes with @r{ISO 6937/2}.
245
Note that the first letter of the graphic ID
246
categorizes the character as follows:
247
L---Latin, N---Numeric, S---Special.
249
@node Character Syntax Types, , Standard Characters, Character Syntax
250
@subsection Character Syntax Types
252
The @i{Lisp reader} constructs an @i{object}
253
from the input text by interpreting each @i{character}
254
according to its @i{syntax type}.
255
The @i{Lisp reader} cannot accept as input
256
everything that the @i{Lisp printer} produces,
257
and the @i{Lisp reader} has features that are not used by the @i{Lisp printer}.
258
The @i{Lisp reader} can be used as a lexical analyzer
259
for a more general user-written parser.
261
When the @i{Lisp reader} is invoked, it reads a single character from
262
the @i{input} @i{stream} and dispatches according to the
264
@IGindex{syntax type}
265
of that @i{character}.
266
Every @i{character} that can appear in the @i{input} @i{stream}
267
is of one of the @i{syntax types} shown in @i{Figure~2--6}.
272
@w{ @i{constituent} @i{macro character} @i{single escape} }
273
@w{ @i{invalid} @i{multiple escape} @i{whitespace}_2 }
276
@w{ Figure 2--6: Possible Character Syntax Types }
281
The @i{syntax type} of a @i{character} in a @i{readtable}
282
determines how that character is interpreted by the @i{Lisp reader}
283
while that @i{readtable} is the @i{current readtable}.
284
At any given time, every character has exactly one @i{syntax type}.
287
lists the @i{syntax type} of each @i{character} in @i{standard syntax}.
294
@w{ character syntax type character syntax type }
295
@w{ Backspace @i{constituent} 0--9 @i{constituent} }
296
@w{ Tab @i{whitespace}_2 : @i{constituent} }
297
@w{ Newline @i{whitespace}_2 ; @i{terminating} @i{macro char} }
298
@w{ Linefeed @i{whitespace}_2 @t{<} @i{constituent} }
299
@w{ Page @i{whitespace}_2 = @i{constituent} }
300
@w{ Return @i{whitespace}_2 @t{>} @i{constituent} }
301
@w{ Space @i{whitespace}_2 ? @i{constituent}* }
302
@w{ ! @i{constituent}* @t{@@} @i{constituent} }
303
@w{ @t{"} @i{terminating} @i{macro char} A--Z @i{constituent} }
304
@w{ # @i{non-terminating} @i{macro char} @t{[} @i{constituent}* }
305
@w{ $ @i{constituent} @t{\} @i{single escape} }
306
@w{ % @i{constituent} @t{]} @i{constituent}* }
307
@w{ & @i{constituent} @t{^} @i{constituent} }
308
@w{ ' @i{terminating} @i{macro char} @t{_} @i{constituent} }
309
@w{ ( @i{terminating} @i{macro char} ` @i{terminating} @i{macro char} }
310
@w{ ) @i{terminating} @i{macro char} a--z @i{constituent} }
311
@w{ @t{*} @i{constituent} @t{@{} @i{constituent}* }
312
@w{ + @i{constituent} @t{|} @i{multiple escape} }
313
@w{ , @i{terminating} @i{macro char} @t{@}} @i{constituent}* }
314
@w{ - @i{constituent} @t{~} @i{constituent} }
315
@w{ . @i{constituent} Rubout @i{constituent} }
316
@w{ / @i{constituent} }
319
@w{ Figure 2--7: Character Syntax Types in Standard Syntax }
325
The characters marked with an asterisk (*) are initially @i{constituents},
326
but they are not used in any standard @r{Common Lisp} notations.
327
These characters are explicitly reserved to the @i{programmer}.
328
@t{~} is not used in @r{Common Lisp}, and reserved to implementors.
329
@t{$} and @t{%} are @i{alphabetic}_2 @i{characters},
330
but are not used in the names of any standard @r{Common Lisp} @i{defined names}.
332
@i{Whitespace}_2 characters serve as separators but are otherwise
333
ignored. @i{Constituent} and @i{escape} @i{characters} are accumulated
334
to make a @i{token}, which is then interpreted as a @i{number} or @i{symbol}.
335
@i{Macro characters} trigger the invocation of @i{functions} (possibly
336
user-supplied) that can perform arbitrary parsing actions.
337
@i{Macro characters} are divided into two kinds,
338
@i{terminating} and @i{non-terminating},
339
depending on whether or not they terminate a @i{token}.
340
The following are descriptions of each kind of @i{syntax type}.
343
* Constituent Characters::
344
* Constituent Traits::
345
* Invalid Characters::
347
* Multiple Escape Characters::
348
* Examples of Multiple Escape Characters::
349
* Single Escape Character::
350
* Examples of Single Escape Characters::
351
* Whitespace Characters::
352
* Examples of Whitespace Characters::
355
@node Constituent Characters, Constituent Traits, Character Syntax Types, Character Syntax Types
356
@subsubsection Constituent Characters
358
@i{Constituent} @i{characters} are used in @i{tokens}.
361
is a representation of a @i{number} or a @i{symbol}.
362
Examples of @i{constituent} @i{characters} are letters and digits.
364
Letters in symbol names are sometimes converted to
365
letters in the opposite @i{case} when the name is read;
366
see @ref{Effect of Readtable Case on the Lisp Reader}.
367
@i{Case} conversion can be suppressed by the use
368
of @i{single escape} or @i{multiple escape} characters.
370
@node Constituent Traits, Invalid Characters, Constituent Characters, Character Syntax Types
371
@subsubsection Constituent Traits
373
Every @i{character} has one or more @i{constituent traits}
374
that define how the @i{character} is to be interpreted by the @i{Lisp reader}
375
when the @i{character} is a @i{constituent} @i{character}.
376
These @i{constituent traits} are
387
@i{Figure~2--8} shows the @i{constituent traits}
388
of the @i{standard characters}
389
and of certain @i{semi-standard} @i{characters};
390
no mechanism is provided for changing the @i{constituent trait} of a @i{character}.
391
Any @i{character} with the alphadigit @i{constituent trait}
392
in that figure is a digit if the @i{current input base} is greater
393
than that character's digit value,
394
otherwise the @i{character} is @i{alphabetic}_2.
395
Any @i{character} quoted by a @i{single escape}
396
is treated as an @i{alphabetic}_2 constituent, regardless of its normal syntax.
401
@w{ constituent traits constituent traits }
402
@w{ character character }
403
@w{ ________________________________________________________________________________}
404
@w{ Backspace @i{invalid} @t{@{} @i{alphabetic}_2 }
405
@w{ Tab @i{invalid}* @t{@}} @i{alphabetic}_2 }
406
@w{ Newline @i{invalid}* + @i{alphabetic}_2, plus sign }
407
@w{ Linefeed @i{invalid}* - @i{alphabetic}_2, minus sign }
408
@w{ Page @i{invalid}* . @i{alphabetic}_2, dot, decimal point }
409
@w{ Return @i{invalid}* / @i{alphabetic}_2, @i{ratio marker} }
410
@w{ Space @i{invalid}* A, a alphadigit }
411
@w{ ! @i{alphabetic}_2 B, b alphadigit }
412
@w{ @t{"} @i{alphabetic}_2* C, c alphadigit }
413
@w{ # @i{alphabetic}_2* D, d alphadigit, double-float @i{exponent marker} }
414
@w{ $ @i{alphabetic}_2 E, e alphadigit, float @i{exponent marker} }
415
@w{ % @i{alphabetic}_2 F, f alphadigit, single-float @i{exponent marker} }
416
@w{ & @i{alphabetic}_2 G, g alphadigit }
417
@w{ ' @i{alphabetic}_2* H, h alphadigit }
418
@w{ ( @i{alphabetic}_2* I, i alphadigit }
419
@w{ ) @i{alphabetic}_2* J, j alphadigit }
420
@w{ @t{*} @i{alphabetic}_2 K, k alphadigit }
421
@w{ , @i{alphabetic}_2* L, l alphadigit, long-float @i{exponent marker} }
422
@w{ 0-9 alphadigit M, m alphadigit }
423
@w{ : @i{package marker} N, n alphadigit }
424
@w{ ; @i{alphabetic}_2* O, o alphadigit }
425
@w{ @t{<} @i{alphabetic}_2 P, p alphadigit }
426
@w{ = @i{alphabetic}_2 Q, q alphadigit }
427
@w{ @t{>} @i{alphabetic}_2 R, r alphadigit }
428
@w{ ? @i{alphabetic}_2 S, s alphadigit, short-float @i{exponent marker} }
429
@w{ @t{@@} @i{alphabetic}_2 T, t alphadigit }
430
@w{ @t{[} @i{alphabetic}_2 U, u alphadigit }
431
@w{ @t{\} @i{alphabetic}_2* V, v alphadigit }
432
@w{ @t{]} @i{alphabetic}_2 W, w alphadigit }
433
@w{ @t{^} @i{alphabetic}_2 X, x alphadigit }
434
@w{ @t{_} @i{alphabetic}_2 Y, y alphadigit }
435
@w{ ` @i{alphabetic}_2* Z, z alphadigit }
436
@w{ @t{|} @i{alphabetic}_2* Rubout @i{invalid} }
437
@w{ @t{~} @i{alphabetic}_2 }
441
@w{ Figure 2--8: Constituent Traits of Standard Characters and Semi-Standard Characters}
443
The interpretations in this table apply only to @i{characters}
444
whose @i{syntax type} is @i{constituent}.
445
Entries marked with an asterisk (*) are normally @i{shadowed}_2
446
because the indicated @i{characters} are of @i{syntax type}
450
or @i{multiple escape};
451
these @i{constituent traits} apply to them only if their @i{syntax types}
452
are changed to @i{constituent}.
454
@node Invalid Characters, Macro Characters, Constituent Traits, Character Syntax Types
455
@subsubsection Invalid Characters
457
@i{Characters} with the @i{constituent trait} @i{invalid}
458
cannot ever appear in a @i{token}
459
except under the control of a @i{single escape} @i{character}.
460
If an @i{invalid} @i{character} is encountered while an @i{object} is
461
being read, an error of @i{type} @b{reader-error} is signaled.
462
If an @i{invalid} @i{character} is preceded by a @i{single escape} @i{character},
463
it is treated as an @i{alphabetic}_2 @i{constituent} instead.
465
@node Macro Characters, Multiple Escape Characters, Invalid Characters, Character Syntax Types
466
@subsubsection Macro Characters
468
When the @i{Lisp reader} encounters a @i{macro character}
469
on an @i{input} @i{stream},
470
special parsing of subsequent @i{characters}
471
on the @i{input} @i{stream}
474
A @i{macro character} has an associated @i{function}
475
called a @i{reader macro function}
476
@IGindex{reader macro function}
477
that implements its specialized parsing behavior.
478
An association of this kind can be established or modified under control of
479
a @i{conforming program} by using
480
the @i{functions} @b{set-macro-character} and @b{set-dispatch-macro-character}.
482
Upon encountering a @i{macro character}, the @i{Lisp reader} calls its
483
@i{reader macro function}, which parses one specially formatted object
484
from the @i{input} @i{stream}.
485
The @i{function} either returns the parsed @i{object},
486
or else it returns no @i{values}
487
to indicate that the characters scanned by the @i{function}
488
are being ignored (@i{e.g.}, in the case of a comment).
489
Examples of @i{macro characters}
490
are @i{backquote}, @i{single-quote}, @i{left-parenthesis}, and
491
@i{right-parenthesis}.
493
A @i{macro character} is either @i{terminating} or @i{non-terminating}.
494
The difference between @i{terminating} and @i{non-terminating} @i{macro characters}
495
lies in what happens when such characters occur in the middle of a @i{token}.
496
If a @i{non-terminating}
497
@IGindex{non-terminating}
498
@i{macro character} occurs in the middle of a @i{token},
499
the @i{function} associated
500
with the @i{non-terminating} @i{macro character} is not called,
502
@i{non-terminating} @i{macro character} does not terminate the @i{token}'s name; it
503
becomes part of the name as if the @i{macro character} were really a constituent
504
character. A @i{terminating}
505
@IGindex{terminating}
506
@i{macro character} terminates any @i{token},
507
and its associated @i{reader macro function}
508
is called no matter where the @i{character} appears.
509
The only @i{non-terminating} @i{macro character} in @i{standard syntax}
512
If a @i{character} is a @i{dispatching macro character} C_1,
513
its @i{reader macro function} is a @i{function} supplied by the @i{implementation}.
514
This @i{function} reads decimal @i{digit} @i{characters} until a non-@i{digit}
516
If any @i{digits} were read,
517
they are converted into a corresponding @i{integer} infix parameter P;
518
otherwise, the infix parameter P is @b{nil}.
519
The terminating non-@i{digit} C_2 is a @i{character}
520
(sometimes called a ``sub-character'' to emphasize its subordinate role in the dispatching)
521
that is looked up in the dispatch table associated with
522
the @i{dispatching macro character} C_1.
523
The @i{reader macro function} associated with the sub-character C_2
524
is invoked with three arguments:
526
the sub-character C_2,
527
and the infix parameter P.
528
For more information about dispatch characters,
529
see the @i{function} @b{set-dispatch-macro-character}.
531
For information about the @i{macro characters}
532
that are available in @i{standard syntax},
533
see @ref{Standard Macro Characters}.
535
@node Multiple Escape Characters, Examples of Multiple Escape Characters, Macro Characters, Character Syntax Types
536
@subsubsection Multiple Escape Characters
538
A pair of @i{multiple escape}
539
@IGindex{multiple escape}
541
is used to indicate that an enclosed sequence of characters,
542
including possible @i{macro characters} and @i{whitespace}_2 @i{characters},
543
are to be treated as @i{alphabetic}_2 @i{characters}
544
with @i{case} preserved.
545
Any @i{single escape} and @i{multiple escape} @i{characters}
546
that are to appear in the sequence must be preceded by a @i{single escape}
549
@i{Vertical-bar} is a @i{multiple escape} @i{character}
550
in @i{standard syntax}.
552
@node Examples of Multiple Escape Characters, Single Escape Character, Multiple Escape Characters, Character Syntax Types
553
@subsubsection Examples of Multiple Escape Characters
556
;; The following examples assume the readtable case of *readtable*
557
;; and *print-case* are both :upcase.
558
(eq 'abc 'ABC) @result{} @i{true}
559
(eq 'abc '|ABC|) @result{} @i{true}
560
(eq 'abc 'a|B|c) @result{} @i{true}
561
(eq 'abc '|abc|) @result{} @i{false}
564
@node Single Escape Character, Examples of Single Escape Characters, Examples of Multiple Escape Characters, Character Syntax Types
565
@subsubsection Single Escape Character
568
@IGindex{single escape}
569
is used to indicate that
570
the next @i{character} is to be treated as
571
an @i{alphabetic}_2 @i{character}
572
with its @i{case} preserved,
573
no matter what the @i{character} is
574
or which @i{constituent traits} it has.
576
@i{Slash} is a @i{single escape} @i{character}
577
in @i{standard syntax}.
579
@node Examples of Single Escape Characters, Whitespace Characters, Single Escape Character, Character Syntax Types
580
@subsubsection Examples of Single Escape Characters
583
;; The following examples assume the readtable case of *readtable*
584
;; and *print-case* are both :upcase.
585
(eq 'abc '\A\B\C) @result{} @i{true}
586
(eq 'abc 'a\Bc) @result{} @i{true}
587
(eq 'abc '\ABC) @result{} @i{true}
588
(eq 'abc '\abc) @result{} @i{false}
591
@node Whitespace Characters, Examples of Whitespace Characters, Examples of Single Escape Characters, Character Syntax Types
592
@subsubsection Whitespace Characters
594
@i{Whitespace}_2 @i{characters} are used to separate @i{tokens}.
596
@i{Space} and @i{newline} are @i{whitespace}_2 @i{characters}
597
in @i{standard syntax}.
599
@node Examples of Whitespace Characters, , Whitespace Characters, Character Syntax Types
600
@subsubsection Examples of Whitespace Characters
603
(length '(this-that)) @result{} 1
604
(length '(this - that)) @result{} 3
611
@c end of including concept-syntax
613
@node Reader Algorithm, Interpretation of Tokens, Character Syntax, Syntax
614
@section Reader Algorithm
616
@c including concept-reader-algorithm
618
This section describes the algorithm used by the @i{Lisp reader}
619
to parse @i{objects} from an @i{input} @i{character} @i{stream},
620
including how the @i{Lisp reader} processes @i{macro characters}.
622
When dealing with @i{tokens}, the reader's basic function is to distinguish
623
representations of @i{symbols} from those of @i{numbers}.
624
When a @i{token} is accumulated, it is assumed to represent a @i{number} if it
625
satisfies the syntax for numbers listed in @i{Figure~2--9}.
626
If it does not represent a @i{number},
627
it is then assumed to be a @i{potential number}
628
if it satisfies the rules governing the syntax for a @i{potential number}.
629
If a valid @i{token} is neither a representation of a @i{number}
630
nor a @i{potential number},
631
it represents a @i{symbol}.
633
The algorithm performed by the @i{Lisp reader} is as follows:
638
If at end of file, end-of-file processing is performed as specified
641
one @i{character}, @i{x}, is read from the @i{input} @i{stream}, and
642
dispatched according to the @i{syntax type} of @i{x} to one
646
If @i{x} is an @i{invalid} @i{character},
647
an error of @i{type} @b{reader-error} is signaled.
650
If @i{x} is a @i{whitespace}_2 @i{character},
651
then it is discarded and step 1 is re-entered.
654
If @i{x} is a @i{terminating} or @i{non-terminating} @i{macro character}
655
then its associated @i{reader macro function} is called with two @i{arguments},
656
the @i{input} @i{stream} and @i{x}.
658
The @i{reader macro function} may read @i{characters}
659
from the @i{input} @i{stream};
660
if it does, it will see those @i{characters} following the @i{macro character}.
661
The @i{Lisp reader} may be invoked recursively from the @i{reader macro function}.
663
The @i{reader macro function} must not have any side effects other than on the
664
@i{input} @i{stream};
665
because of backtracking and restarting of the @b{read} operation,
666
front ends to the @i{Lisp reader} (@i{e.g.}, ``editors'' and ``rubout handlers'')
667
may cause the @i{reader macro function} to be called repeatedly during the
668
reading of a single @i{expression} in which @i{x} only appears once.
670
The @i{reader macro function} may return zero values or one value.
671
If one value is returned,
672
then that value is returned as the result of the read operation;
673
the algorithm is done.
674
If zero values are returned, then step 1 is re-entered.
677
If @i{x} is a @i{single escape} @i{character}
678
then the next @i{character}, @i{y}, is read, or an error of @i{type} @b{end-of-file}
679
is signaled if at the end of file.
680
@i{y} is treated as if it is a @i{constituent}
681
whose only @i{constituent trait} is @i{alphabetic}_2.
682
@i{y} is used to begin a @i{token}, and step 8 is entered.
685
If @i{x} is a @i{multiple escape} @i{character}
686
then a @i{token} (initially
687
containing no @i{characters}) is begun and step 9 is entered.
690
If @i{x} is a @i{constituent} @i{character}, then it begins a @i{token}.
691
After the @i{token} is read in, it will be interpreted
692
either as a @r{Lisp} @i{object} or as being of invalid syntax.
693
If the @i{token} represents an @i{object},
694
that @i{object} is returned as the result of the read operation.
695
If the @i{token} is of invalid syntax, an error is signaled.
696
If @i{x} is a @i{character} with @i{case},
697
it might be replaced with the corresponding @i{character} of the opposite @i{case},
698
depending on the @i{readtable case} of the @i{current readtable},
699
as outlined in @ref{Effect of Readtable Case on the Lisp Reader}.
700
@i{X} is used to begin a @i{token}, and step 8 is entered.
703
At this point a @i{token} is being accumulated, and an even number
704
of @i{multiple escape} @i{characters} have been encountered.
705
If at end of file, step 10 is entered.
706
Otherwise, a @i{character}, @i{y}, is read, and
707
one of the following actions is performed according to its @i{syntax type}:
712
If @i{y} is a @i{constituent} or @i{non-terminating} @i{macro character}:
716
If @i{y} is a @i{character} with @i{case},
717
it might be replaced with the corresponding @i{character} of the opposite @i{case},
718
depending on the @i{readtable case} of the @i{current readtable},
719
as outlined in @ref{Effect of Readtable Case on the Lisp Reader}.
721
@i{Y} is appended to the @i{token} being built.
727
If @i{y} is a @i{single escape} @i{character}, then the next @i{character},
728
@i{z}, is read, or an error of @i{type} @b{end-of-file} is signaled if at end of file.
729
@i{Z} is treated as if it is a @i{constituent}
730
whose only @i{constituent trait} is @i{alphabetic}_2.
731
@i{Z} is appended to the @i{token} being built,
732
and step 8 is repeated.
735
If @i{y} is a @i{multiple escape} @i{character},
736
then step 9 is entered.
739
If @i{y} is an @i{invalid} @i{character},
740
an error of @i{type} @b{reader-error} is signaled.
743
If @i{y} is a @i{terminating} @i{macro character},
744
then it terminates the @i{token}.
745
First the @i{character} @i{y} is unread (see @b{unread-char}),
746
and then step 10 is entered.
749
If @i{y} is a @i{whitespace}_2 @i{character}, then it terminates
750
the @i{token}. First the @i{character} @i{y} is unread
751
if appropriate (see @b{read-preserving-whitespace}),
752
and then step 10 is entered.
756
At this point a @i{token} is being accumulated, and an odd number
757
of @i{multiple escape} @i{characters} have been encountered.
758
If at end of file, an error of @i{type} @b{end-of-file} is signaled.
759
Otherwise, a @i{character}, @i{y}, is read, and
760
one of the following actions is performed according to its @i{syntax type}:
765
If @i{y} is a @i{constituent}, macro, or @i{whitespace}_2 @i{character},
766
@i{y} is treated as a @i{constituent}
767
whose only @i{constituent trait} is @i{alphabetic}_2.
768
@i{Y} is appended to the @i{token} being built, and step 9 is repeated.
771
If @i{y} is a @i{single escape} @i{character}, then the next @i{character},
772
@i{z}, is read, or an error of @i{type} @b{end-of-file} is signaled if at end of file.
773
@i{Z} is treated as a @i{constituent}
774
whose only @i{constituent trait} is @i{alphabetic}_2.
775
@i{Z} is appended to the @i{token} being built,
776
and step 9 is repeated.
779
If @i{y} is a @i{multiple escape} @i{character},
780
then step 8 is entered.
783
If @i{y} is an @i{invalid} @i{character},
784
an error of @i{type} @b{reader-error} is signaled.
788
An entire @i{token} has been accumulated.
789
The @i{object} represented by the @i{token} is returned
790
as the result of the read operation,
791
or an error of @i{type} @b{reader-error} is signaled if the @i{token} is not of valid syntax.
794
@c end of including concept-reader-algorithm
796
@node Interpretation of Tokens, Standard Macro Characters, Reader Algorithm, Syntax
797
@section Interpretation of Tokens
799
@c including concept-tokens
802
* Numbers as Tokens::
803
* Constructing Numbers from Tokens::
805
* Symbols as Tokens::
806
* Valid Patterns for Tokens::
807
* Package System Consistency Rules::
810
@node Numbers as Tokens, Constructing Numbers from Tokens, Interpretation of Tokens, Interpretation of Tokens
811
@subsection Numbers as Tokens
813
When a @i{token} is read,
814
it is interpreted as a @i{number} or @i{symbol}.
815
The @i{token} is interpreted as a @i{number} if it satisfies
816
the syntax for numbers specified in Figure 2--9.
821
@w{ @i{numeric-token} ::= !@i{integer} | !@i{ratio} | !@i{float} }
822
@w{ @i{integer} ::= @t{[}@i{sign}@t{]} @{@i{decimal-digit}@}^+ @i{decimal-point} | @t{[}@i{sign}@t{]} @{@i{digit}@}^+ }
823
@w{ @i{ratio} ::= @t{[}@i{sign}@t{]} @{@i{digit}@}^+ @i{slash} @{@i{digit}@}^+ }
824
@w{ @i{float} ::= @t{[}@i{sign}@t{]} @{@i{decimal-digit}@}* @i{decimal-point} @{@i{decimal-digit}@}^+ @t{[}!@i{exponent}@t{]} }
825
@w{ | @t{[}@i{sign}@t{]} @{@i{decimal-digit}@}^+ @t{[}@i{decimal-point} @{@i{decimal-digit}@}*@t{]} !@i{exponent} }
826
@w{ @i{exponent} ::= @i{exponent-marker} @t{[}@i{sign}@t{]} @{@i{digit}@}^+ }
827
@w{ @i{sign}---a @i{sign}.}
828
@w{ @i{slash}---a @i{slash}}
829
@w{ @i{decimal-point}---a @i{dot}.}
830
@w{ @i{exponent-marker}---an @i{exponent marker}.}
831
@w{ @i{decimal-digit}---a @i{digit} in @i{radix} @t{10}.}
832
@w{ @i{digit}---a @i{digit} in the @i{current input radix}.}
836
@w{ Figure 2--9: Syntax for Numeric Tokens}
839
* Potential Numbers as Tokens::
840
* Escape Characters and Potential Numbers::
841
* Examples of Potential Numbers::
844
@node Potential Numbers as Tokens, Escape Characters and Potential Numbers, Numbers as Tokens, Numbers as Tokens
845
@subsubsection Potential Numbers as Tokens
847
To allow implementors and future @r{Common Lisp} standards
848
to extend the syntax of numbers, a
849
syntax for @i{potential numbers} is defined that is
850
more general than the syntax for numbers.
851
A @i{token} is a @i{potential number} if it satisfies all of the following
857
The @i{token} consists entirely of
861
decimal points (@t{.}),
862
extension characters (@t{^} or @t{_}),
864
A number marker is a letter.
865
Whether a letter may be treated as a number marker depends on context,
866
but no letter that is adjacent to another letter may ever be treated as a number marker.
867
@i{Exponent markers} are number markers.
870
The @i{token} contains at least one digit. Letters may be considered to be
871
digits, depending on the @i{current input base}, but only
872
in @i{tokens} containing no decimal points.
875
The @i{token} begins with a @i{digit}, @i{sign}, decimal point, or extension character,
877
[Reviewer Note by Barmar: This section is unnecessary because the first bullet already
878
omits discussion of a colon (@i{package marker}).]
881
The syntax involving a leading
882
@i{package marker} followed by a @i{potential number} is
883
not well-defined. The consequences of the use
884
of notation such as @t{:1}, @t{:1/2}, and @t{:2^3} in a
885
position where an expression appropriate for @b{read}
886
is expected are unspecified.
889
The @i{token} does not end with a sign.
892
If a @i{potential number} has number syntax,
893
a @i{number} of the appropriate type is constructed and returned,
894
if the @i{number} is representable in an implementation.
895
A @i{number} will not be representable in an implementation
896
if it is outside the boundaries set by the @i{implementation-dependent}
897
constants for @i{numbers}.
898
For example, specifying too large or too small an exponent for a @i{float}
899
may make the @i{number} impossible to represent in the implementation.
900
A @i{ratio} with denominator zero (such as @t{-35/000})
901
is not represented in any implementation.
902
When a @i{token} with the syntax of a number cannot be converted to an internal
903
@i{number}, an error of @i{type} @b{reader-error} is signaled. An error
904
must not be signaled for specifying too many significant digits
905
for a @i{float}; a truncated or rounded value should be produced.
907
If there is an ambiguity as to whether
908
a letter should be treated as a digit or as a number marker,
909
the letter is treated as a digit.
911
@node Escape Characters and Potential Numbers, Examples of Potential Numbers, Potential Numbers as Tokens, Numbers as Tokens
912
@subsubsection Escape Characters and Potential Numbers
914
A @i{potential number} cannot contain any @i{escape}
915
@i{characters}. An @i{escape} @i{character} robs the following
916
@i{character} of all syntactic qualities, forcing it to be strictly
917
@i{alphabetic}_2 and therefore unsuitable for use in a
918
@i{potential number}. For example, all of the following
919
representations are interpreted as @i{symbols}, not @i{numbers}:
922
\256 25\64 1.0\E6 |100| 3\.14159 |3/4| 3\/4 5||
925
In each case, removing the @i{escape} @i{character} (or @i{characters})
927
cause the token to be a @i{potential number}.
929
@node Examples of Potential Numbers, , Escape Characters and Potential Numbers, Numbers as Tokens
930
@subsubsection Examples of Potential Numbers
932
As examples, the @i{tokens} in Figure 2--10 are @i{potential numbers},
933
but they are not actually numbers, and so are reserved @i{tokens};
934
a @i{conforming implementation} is permitted, but not required,
935
to define their meaning.
940
@w{ @t{1b5000} @t{777777q} @t{1.7J} @t{-3/4+6.7J} @t{12/25/83} }
941
@w{ @t{27^19} @t{3^4/5} @t{6//7} @t{3.1.2.6} @t{@t{^}-43@t{^}} }
942
@w{ @t{3.141_592_653_589_793_238_4} @t{-3.7+2.6i-6.17j+19.6k} }
945
@w{ Figure 2--10: Examples of reserved tokens }
950
The @i{tokens} in Figure 2--11 are not @i{potential numbers};
951
they are always treated as @i{symbols}:
956
@w{ @t{/} @t{/5} @t{+} @t{1+} @t{1-} }
957
@w{ @t{foo+} @t{ab.cd} @t{_} @t{@t{^}} @t{@t{^}/-} }
960
@w{ Figure 2--11: Examples of symbols}
965
The @i{tokens} in Figure 2--12 are @i{potential numbers}
966
if the @i{current input base} is @t{16},
967
but they are always treated as @i{symbols} if the @i{current input base} is @t{10}.
972
@w{ @t{bad-face} @t{25-dec-83} @t{a/b} @t{fad_cafe} @t{f@t{^}} }
975
@w{ Figure 2--12: Examples of symbols or potential numbers}
980
@node Constructing Numbers from Tokens, The Consing Dot, Numbers as Tokens, Interpretation of Tokens
981
@subsection Constructing Numbers from Tokens
983
A @i{real} is constructed directly from a corresponding numeric @i{token};
986
A @i{complex} is notated as a @t{#C} (or @t{#c}) followed by a @i{list}
987
of two @i{reals}; see @ref{Sharpsign C}.
989
The @i{reader macros} @t{#B}, @t{#O}, @t{#X}, and @t{#R} may also be useful
990
in controlling the input @i{radix} in which @i{rationals} are parsed;
991
see @ref{Sharpsign B},
994
and @ref{Sharpsign R}.
996
This section summarizes the full syntax for @i{numbers}.
999
* Syntax of a Rational::
1000
* Syntax of an Integer::
1001
* Syntax of a Ratio::
1002
* Syntax of a Float::
1003
* Syntax of a Complex::
1006
@node Syntax of a Rational, Syntax of an Integer, Constructing Numbers from Tokens, Constructing Numbers from Tokens
1007
@subsubsection Syntax of a Rational
1009
@node Syntax of an Integer, Syntax of a Ratio, Syntax of a Rational, Constructing Numbers from Tokens
1010
@subsubsection Syntax of an Integer
1012
@i{Integers} can be written as a sequence of @i{digits},
1013
optionally preceded by a @i{sign} and optionally followed by a decimal point;
1014
see @i{Figure~2--9}.
1015
When a decimal point is used,
1016
the @i{digits} are taken to be in @i{radix} @t{10};
1017
when no decimal point is used,
1018
the @i{digits} are taken to be in radix given by the @i{current input base}.
1020
For information on how @i{integers} are printed, see @ref{Printing Integers}.
1022
@node Syntax of a Ratio, Syntax of a Float, Syntax of an Integer, Constructing Numbers from Tokens
1023
@subsubsection Syntax of a Ratio
1025
@i{Ratios} can be written as an optional @i{sign} followed by two
1026
non-empty sequences of @i{digits} separated by a @i{slash};
1027
see @i{Figure~2--9}.
1028
The second sequence may not consist
1030
Examples of @i{ratios} are in Figure 2--13.
1035
@w{ @t{2/3} ;This is in canonical form }
1036
@w{ @t{4/6} ;A non-canonical form for 2/3 }
1037
@w{ @t{-17/23} ;A ratio preceded by a sign }
1038
@w{ @t{-30517578125/32768} ;This is (-5/2)^15 }
1039
@w{ @t{10/5} ;The canonical form for this is @t{2} }
1040
@w{ @t{#o-101/75} ;Octal notation for -65/61 }
1041
@w{ @t{#3r120/21} ;Ternary notation for 15/7 }
1042
@w{ @t{#Xbc/ad} ;Hexadecimal notation for 188/173 }
1043
@w{ @t{#xFADED/FACADE} ;Hexadecimal notation for 1027565/16435934 }
1046
@w{ Figure 2--13: Examples of Ratios }
1051
[Reviewer Note by Barmar: #o, #3r, #X, and #x mentioned above
1052
are not in the syntax rules defined just above that.]
1054
For information on how @i{ratios} are printed,
1055
see @ref{Printing Ratios}.
1057
@node Syntax of a Float, Syntax of a Complex, Syntax of a Ratio, Constructing Numbers from Tokens
1058
@subsubsection Syntax of a Float
1060
@i{Floats} can be written in either decimal fraction or computerized
1061
scientific notation: an optional sign, then a non-empty sequence of digits
1062
with an embedded decimal point,
1063
then an optional decimal exponent specification.
1064
If there is no exponent specifier, then
1065
the decimal point is required, and there must be digits
1067
The exponent specifier consists of an @i{exponent marker},
1068
an optional sign, and a non-empty sequence of digits.
1069
If no exponent specifier is present, or if the @i{exponent marker} @t{e}
1070
(or @t{E}) is used, then
1071
the format specified
1072
by @b{*read-default-float-format*} is used.
1073
See @i{Figure~2--9}.
1075
An implementation may provide one or more kinds of @i{float}
1076
that collectively make up the @i{type} @b{float}.
1077
The letters @t{s}, @t{f}, @t{d}, and @t{l} (or their
1078
respective uppercase equivalents) explicitly specify the
1079
use of the @i{types} @b{short-float}, @b{single-float},
1080
@b{double-float}, and @b{long-float}, respectively.
1082
The internal format used for an external representation depends only
1083
on the @i{exponent marker}, and not on the number of decimal digits
1084
in the external representation.
1086
Figure 2--14 contains examples of notations for @i{floats}:
1091
@w{ @t{0.0} ;Floating-point zero in default format }
1092
@w{ @t{0E0} ;As input, this is also floating-point zero in default format. }
1093
@w{ ;As output, this would appear as @t{0.0}. }
1094
@w{ @t{0e0} ;As input, this is also floating-point zero in default format. }
1095
@w{ ;As output, this would appear as @t{0.0}. }
1096
@w{ @t{-.0} ;As input, this might be a zero or a minus zero, }
1097
@w{ ; depending on whether the implementation supports }
1098
@w{ ; a distinct minus zero. }
1099
@w{ ;As output, @t{0.0} is zero and @t{-0.0} is minus zero. }
1100
@w{ @t{0.} ;On input, the integer zero---@i{not} a floating-point number! }
1101
@w{ ;Whether this appears as @t{0} or @t{0.} on output depends }
1102
@w{ ;on the @i{value} of @b{*print-radix*}. }
1103
@w{ @t{0.0s0} ;A floating-point zero in short format }
1104
@w{ @t{0s0} ;As input, this is a floating-point zero in short format. }
1105
@w{ ;As output, such a zero would appear as @t{0.0s0} }
1106
@w{ ; (or as @t{0.0} if @b{short-float} was the default format). }
1107
@w{ @t{6.02E+23} ;Avogadro's number, in default format }
1108
@w{ @t{602E+21} ;Also Avogadro's number, in default format }
1111
@w{ Figure 2--14: Examples of Floating-point numbers }
1116
For information on how @i{floats} are printed,
1117
see @ref{Printing Floats}.
1119
@node Syntax of a Complex, , Syntax of a Float, Constructing Numbers from Tokens
1120
@subsubsection Syntax of a Complex
1122
A @i{complex} has a Cartesian structure,
1123
with a real part and an imaginary part each of which is a
1127
The parts of a @i{complex} are not necessarily @i{floats}
1128
but both parts must be of the same @i{type}:
1130
[Editorial Note by KMP: This is not the same as saying they must be the same type.
1131
Maybe we mean they are of the same `precision' or `format'?
1132
GLS had suggestions which are not yet merged.]
1133
either both are @i{rationals}, or both are of the same @i{float} @i{subtype}.
1134
When constructing a @i{complex}, if the specified parts are not the
1135
same @i{type}, the parts are converted to be the same @i{type}
1136
internally (@i{i.e.}, the @i{rational} part is converted to a @i{float}).
1137
An @i{object} of type @t{(complex rational)} is converted internally
1138
and represented thereafter as a @i{rational} if its imaginary part is an
1139
@i{integer} whose value is 0.
1141
For further information, see @ref{Sharpsign C} and @ref{Printing Complexes}.
1143
@node The Consing Dot, Symbols as Tokens, Constructing Numbers from Tokens, Interpretation of Tokens
1144
@subsection The Consing Dot
1146
If a @i{token} consists solely of dots (with no escape characters),
1147
then an error of @i{type} @b{reader-error} is signaled,
1148
except in one circumstance:
1149
if the @i{token} is a single @i{dot}
1150
and appears in a situation where @i{dotted pair} notation permits a @i{dot},
1151
then it is accepted as part of such syntax and no error is signaled.
1152
See @ref{Left-Parenthesis}.
1154
@node Symbols as Tokens, Valid Patterns for Tokens, The Consing Dot, Interpretation of Tokens
1155
@subsection Symbols as Tokens
1157
Any @i{token} that is not a @i{potential number},
1158
does not contain a @i{package marker},
1159
and does not consist entirely of dots
1160
will always be interpreted as a @i{symbol}.
1161
Any @i{token} that is a @i{potential number} but does not fit the
1162
number syntax is a reserved @i{token} and
1163
has an @i{implementation-dependent} interpretation.
1164
In all other cases, the @i{token} is construed to be the name of a @i{symbol}.
1166
Examples of the printed representation of @i{symbols} are in Figure 2--15.
1167
For presentational simplicity,
1168
these examples assume that
1169
the @i{readtable case} of the @i{current readtable} is @t{:upcase}.
1174
@w{ @t{FROBBOZ} The @i{symbol} whose @i{name} is @t{FROBBOZ}. }
1175
@w{ @t{frobboz} Another way to notate the same @i{symbol}. }
1176
@w{ @t{fRObBoz} Yet another way to notate it. }
1177
@w{ @t{unwind-protect} A @i{symbol} with a hyphen in its @i{name}. }
1178
@w{ @t{+$} The @i{symbol} named @t{+$}. }
1179
@w{ @t{1+} The @i{symbol} named @t{1+}. }
1180
@w{ @t{+1} This is the @i{integer} @t{1}, not a @i{symbol}. }
1181
@w{ @t{pascal_style} This @i{symbol} has an underscore in its @i{name}. }
1182
@w{ @t{file.rel.43} This @i{symbol} has periods in its @i{name}. }
1183
@w{ @t{\(} The @i{symbol} whose @i{name} is @t{(}. }
1184
@w{ @t{\+1} The @i{symbol} whose @i{name} is @t{+1}. }
1185
@w{ @t{+\1} Also the @i{symbol} whose @i{name} is @t{+1}. }
1186
@w{ @t{\frobboz} The @i{symbol} whose @i{name} is @t{fROBBOZ}. }
1187
@w{ @t{3.14159265\s0} The @i{symbol} whose @i{name} is @t{3.14159265s0}. }
1188
@w{ @t{3.14159265\S0} A different @i{symbol}, whose @i{name} is @t{3.14159265S0}. }
1189
@w{ @t{3.14159265s0} A possible @i{short float} approximation to \pi. }
1192
@w{ Figure 2--15: Examples of the printed representation of symbols (Part 1 of 2)}
1200
@w{ @t{APL\\360} The @i{symbol} whose @i{name} is @t{APL\360}. }
1201
@w{ @t{apl\\360} Also the @i{symbol} whose @i{name} is @t{APL\360}. }
1202
@w{ @t{\(b@t{^}2\)\ -\ 4*a@t{*c}} The @i{name} is @t{(B@t{^}2) - 4*A*C}. }
1203
@w{ Parentheses and two spaces in it. }
1204
@w{ @t{\(\b@t{^}2\)\ -\4*\a*\c} The @i{name} is @t{(b@t{^}2) - 4*a*c}. }
1205
@w{ Letters explicitly lowercase. }
1206
@w{ @t{|"|} The same as writing @t{\"}. }
1207
@w{ @t{|(b@t{^}2) - 4*a*c|} The @i{name} is @t{(b@t{^}2) - 4*a*c}. }
1208
@w{ @t{|frobboz|} The @i{name} is @t{frobboz}, not @t{FROBBOZ}. }
1209
@w{ @t{|APL\360|} The @i{name} is @t{APL360}. }
1210
@w{ @t{|APL\\360|} The @i{name} is @t{APL\360}. }
1211
@w{ @t{|apl\\360|} The @i{name} is @t{apl\360}. }
1212
@w{ @t{|\|\||} Same as @t{\|\|} ---the @i{name} is @t{||}. }
1213
@w{ @t{|(B@t{^}2) - 4*A*C|} The @i{name} is @t{(B@t{^}2) - 4*A*C}. }
1214
@w{ Parentheses and two spaces in it. }
1215
@w{ @t{|(b@t{^}2) - 4*a*c|} The @i{name} is @t{(b@t{^}2) - 4*a*c}. }
1218
@w{ Figure 2--16: Examples of the printed representation of symbols (Part 2 of 2)}
1223
In the process of parsing a @i{symbol},
1224
it is @i{implementation-dependent} which
1225
@i{implementation-defined} @i{attributes} are removed
1226
from the @i{characters} forming a @i{token} that represents a @i{symbol}.
1228
When parsing the syntax for a @i{symbol},
1229
the @i{Lisp reader} looks up the @i{name} of that @i{symbol}
1230
in the @i{current package}.
1231
This lookup may involve looking in other
1232
@i{packages} whose @i{external symbols}
1233
are inherited by the @i{current package}. If the name is found,
1234
the corresponding @i{symbol} is returned. If the name is not found
1235
(that is, there is no @i{symbol}
1236
of that name @i{accessible} in the @i{current package}),
1237
a new @i{symbol} is created and is placed in the @i{current package}
1238
as an @i{internal symbol}. The @i{current package} becomes the owner
1239
(@i{home package}) of the @i{symbol},
1240
and the @i{symbol} becomes interned in the @i{current package}.
1241
If the name is later read again while this same @i{package} is
1242
current, the same @i{symbol} will be found and returned.
1244
@node Valid Patterns for Tokens, Package System Consistency Rules, Symbols as Tokens, Interpretation of Tokens
1245
@subsection Valid Patterns for Tokens
1247
The valid patterns for @i{tokens} are summarized in Figure 2--17.
1252
@w{ @t{@i{nnnnn}} a @i{number} }
1253
@w{ @t{@i{xxxxx}} a @i{symbol} in the @i{current package} }
1254
@w{ @t{:@i{xxxxx}} a @i{symbol} in the the @t{KEYWORD} @i{package} }
1255
@w{ @t{@i{ppppp}:@i{xxxxx}} an @i{external symbol} in the @i{ppppp} @i{package} }
1256
@w{ @t{@i{ppppp}::@i{xxxxx}} a (possibly internal) @i{symbol} in the @i{ppppp} @i{package} }
1257
@w{ @t{:@i{nnnnn}} undefined }
1258
@w{ @t{@i{ppppp}:@i{nnnnn}} undefined }
1259
@w{ @t{@i{ppppp}::@i{nnnnn}} undefined }
1260
@w{ @t{::@i{aaaaa}} undefined }
1261
@w{ @t{@i{aaaaa}:} undefined }
1262
@w{ @t{@i{aaaaa}:@i{aaaaa}:@i{aaaaa}} undefined }
1265
@w{ Figure 2--17: Valid patterns for tokens }
1270
Note that @i{nnnnn} has number syntax,
1271
neither @i{xxxxx} nor @i{ppppp} has number syntax,
1272
and @i{aaaaa} has any syntax.
1274
A summary of rules concerning @i{package markers} follows.
1275
In each case, examples are offered to illustrate the case;
1276
for presentational simplicity, the examples assume that
1277
the @i{readtable case} of the @i{current readtable} is @t{:upcase}.
1282
If there is a single @i{package marker}, and it occurs at the beginning of the
1283
@i{token}, then the @i{token} is interpreted as a @i{symbol} in the @t{KEYWORD} @i{package}.
1284
It also sets the @b{symbol-value} of the newly-created @i{symbol} to that
1285
same @i{symbol} so that the @i{symbol} will self-evaluate.
1288
@t{:bar}, when read, interns @t{BAR} as an @i{external symbol} in the @t{KEYWORD} @i{package}.
1291
If there is a single @i{package marker} not at the beginning or end of the
1292
@i{token}, then it divides the @i{token} into two parts. The first part
1293
specifies a @i{package};
1294
the second part is the name of an @i{external symbol}
1295
available in that package.
1298
@t{foo:bar}, when read, looks up @t{BAR} among the @i{external symbols} of
1299
the @i{package} named @t{FOO}.
1302
If there are two adjacent @i{package markers} not at the beginning or end of the
1303
@i{token}, then they divide the @i{token} into two parts. The first part
1304
specifies a @i{package};
1305
the second part is the name of a @i{symbol} within
1306
that @i{package} (possibly an @i{internal symbol}).
1309
@t{foo::bar}, when read, interns @t{BAR} in the @i{package} named @t{FOO}.
1312
If the @i{token} contains no @i{package markers},
1313
and does not have @i{potential number} syntax,
1314
then the entire @i{token} is the name of the @i{symbol}.
1315
The @i{symbol} is looked up in the @i{current package}.
1318
@t{bar}, when read, interns @t{BAR} in the @i{current package}.
1321
The consequences are unspecified if any other pattern of @i{package markers}
1322
in a @i{token} is used.
1323
All other uses of @i{package markers} within names of @i{symbols}
1324
are not defined by this standard
1325
but are reserved for @i{implementation-dependent} use.
1329
assuming the @i{readtable case} of the @i{current readtable} is @t{:upcase},
1330
@t{editor:buffer} refers to the @i{external symbol}
1331
named @t{BUFFER} present in the @i{package} named @t{editor},
1332
regardless of whether there is a @i{symbol} named @t{BUFFER} in
1333
the @i{current package}. If there is no @i{package} named
1334
@t{editor}, or if no @i{symbol} named @t{BUFFER}
1335
is present in @t{editor}, or if @t{BUFFER} is not exported by
1336
@t{editor}, the reader signals
1337
a correctable error.
1338
If @t{editor::buffer} is seen, the effect is exactly the same as
1339
reading @t{buffer} with the @t{EDITOR} @i{package} being the @i{current package}.
1341
@node Package System Consistency Rules, , Valid Patterns for Tokens, Interpretation of Tokens
1342
@subsection Package System Consistency Rules
1344
The following rules apply to the package system as long as
1345
the @i{value} of @b{*package*} is not changed:
1349
@item @b{Read-read consistency}
1350
Reading the same @i{symbol} @i{name}
1351
always results in the @i{same} @i{symbol}.
1353
@item @b{Print-read consistency}
1354
An @i{interned symbol} always prints as a sequence of characters that,
1355
when read back in, yields the @i{same} @i{symbol}.
1357
For information about how the @i{Lisp printer} treats @i{symbols},
1358
see @ref{Printing Symbols}.
1360
@item @b{Print-print consistency}
1361
If two interned @i{symbols} are not the @i{same},
1362
then their printed representations will be different sequences of characters.
1365
These rules are true regardless of any implicit interning.
1366
As long as the @i{current package} is not changed,
1367
results are reproducible regardless of the order of @i{loading} files
1368
or the exact history of what @i{symbols} were typed in when.
1369
If the @i{value} of @b{*package*} is changed and then changed back to the previous value,
1370
consistency is maintained.
1371
The rules can be violated by
1372
changing the @i{value} of @b{*package*},
1373
forcing a change to @i{symbols}
1376
by continuing from an error,
1377
or calling one of the following @i{functions}:
1381
@b{shadowing-import},
1382
or @b{unuse-package}.
1384
An inconsistency only applies if one of the restrictions is violated
1385
between two of the named @i{symbols}.
1386
@b{shadow}, @b{unexport}, @b{unintern},
1387
and @b{shadowing-import} can only affect the consistency of
1388
@i{symbols} with the same @i{names} (under @b{string=})
1389
as the ones supplied as arguments.
1391
@c end of including concept-tokens
1393
@node Standard Macro Characters, , Interpretation of Tokens, Syntax
1394
@section Standard Macro Characters
1396
@c including concept-macro-chars
1398
If the reader encounters a @i{macro character},
1399
then its associated @i{reader macro function}
1400
is invoked and may produce an @i{object} to be returned.
1401
This @i{function} may read the @i{characters}
1402
following the @i{macro character} in the @i{stream}
1403
in any syntax and return the @i{object} represented by that syntax.
1405
Any @i{character} can be made to be a @i{macro character}.
1406
The @i{macro characters} defined initially in a @i{conforming implementation}
1411
* Left-Parenthesis::
1412
* Right-Parenthesis::
1419
* Re-Reading Abbreviated Expressions::
1422
@node Left-Parenthesis, Right-Parenthesis, Standard Macro Characters, Standard Macro Characters
1423
@subsection Left-Parenthesis
1425
The @i{left-parenthesis} initiates reading of a @i{list}.
1426
@b{read} is called recursively to read successive @i{objects}
1427
until a right parenthesis is found in the input @i{stream}.
1428
A @i{list} of the @i{objects} read is returned. Thus
1434
is read as a @i{list} of three @i{objects}
1435
(the @i{symbols} @t{a}, @t{b}, and @t{c}).
1436
The right parenthesis need not immediately follow the printed representation of
1437
the last @i{object}; @i{whitespace}_2
1438
characters and comments may precede it.
1440
If no @i{objects} precede the right parenthesis,
1441
it reads as a @i{list} of zero @i{objects}
1442
(the @i{empty list}).
1444
If a @i{token} that is just a dot
1445
not immediately preceded by an escape character
1446
is read after some @i{object}
1447
then exactly one more @i{object} must follow the dot,
1448
possibly preceded or followed by @i{whitespace}_2 or a comment,
1449
followed by the right parenthesis:
1455
This means that the @i{cdr} of the last @i{cons} in the
1456
@i{list} is not @b{nil},
1457
but rather the @i{object} whose representation followed the dot.
1458
The above example might have been the result of evaluating
1461
(cons 'a (cons 'b (cons 'c 'd)))
1467
(cons 'this-one 'that-one) @result{} (this-one . that-one)
1470
It is permissible for the @i{object}
1471
following the dot to be a @i{list}:
1474
(a b c d . (e f . (g))) @equiv{} (a b c d e f g)
1477
For information on how the @i{Lisp printer} prints @i{lists} and @i{conses},
1478
see @ref{Printing Lists and Conses}.
1480
@node Right-Parenthesis, Single-Quote, Left-Parenthesis, Standard Macro Characters
1481
@subsection Right-Parenthesis
1483
The @i{right-parenthesis} is invalid
1484
except when used in conjunction with the left parenthesis character.
1485
For more information, see @ref{Reader Algorithm}.
1487
@node Single-Quote, Semicolon, Right-Parenthesis, Standard Macro Characters
1488
@subsection Single-Quote
1490
@b{Syntax:} @t{'<<@i{exp}>>}
1492
A @i{single-quote} introduces an @i{expression} to be ``quoted.''
1493
@i{Single-quote} followed by an @i{expression} @i{exp}
1494
is treated by the @i{Lisp reader} as an abbreviation for
1495
and is parsed identically to the @i{expression} @t{(quote @i{exp})}.
1496
See the @i{special operator} @b{quote}.
1499
* Examples of Single-Quote::
1502
@node Examples of Single-Quote, , Single-Quote, Single-Quote
1503
@subsubsection Examples of Single-Quote
1507
''foo @result{} (QUOTE FOO)
1508
(car ''foo) @result{} QUOTE
1511
@node Semicolon, Double-Quote, Single-Quote, Standard Macro Characters
1512
@subsection Semicolon
1514
@b{Syntax:} @t{;<<@i{text}>>}
1516
A @i{semicolon} introduces @i{characters} that are to be ignored,
1517
such as comments. The @i{semicolon} and all @i{characters} up to
1518
and including the next @i{newline} or end of file are ignored.
1521
* Examples of Semicolon::
1522
* Notes about Style for Semicolon::
1523
* Use of Single Semicolon::
1524
* Use of Double Semicolon::
1525
* Use of Triple Semicolon::
1526
* Use of Quadruple Semicolon::
1527
* Examples of Style for Semicolon::
1530
@node Examples of Semicolon, Notes about Style for Semicolon, Semicolon, Semicolon
1531
@subsubsection Examples of Semicolon
1539
@node Notes about Style for Semicolon, Use of Single Semicolon, Examples of Semicolon, Semicolon
1540
@subsubsection Notes about Style for Semicolon
1542
Some text editors make assumptions about desired indentation based on
1543
the number of @i{semicolons} that begin a comment. The following style
1544
conventions are common, although not by any means universal.
1546
@node Use of Single Semicolon, Use of Double Semicolon, Notes about Style for Semicolon, Semicolon
1547
@subsubsection Use of Single Semicolon
1549
Comments that begin with a single @i{semicolon} are all aligned to
1550
the same column at the right (sometimes called the ``comment column'').
1551
The text of such a comment generally applies only to the line on which it appears.
1552
Occasionally two or three contain a single sentence together;
1553
this is sometimes indicated by indenting all but the first with an additional
1554
space (after the @i{semicolon}).
1556
@node Use of Double Semicolon, Use of Triple Semicolon, Use of Single Semicolon, Semicolon
1557
@subsubsection Use of Double Semicolon
1559
Comments that begin with a double @i{semicolon} are all aligned to
1560
the same level of indentation as a @i{form} would be at that same
1561
position in the @i{code}.
1562
The text of such a comment usually describes
1563
the state of the @i{program} at the point where the comment occurs,
1564
the @i{code} which follows the comment,
1567
@node Use of Triple Semicolon, Use of Quadruple Semicolon, Use of Double Semicolon, Semicolon
1568
@subsubsection Use of Triple Semicolon
1570
Comments that begin with a triple @i{semicolon} are all aligned to
1571
the left margin. Usually they are used prior to a definition or set
1572
of definitions, rather than within a definition.
1574
@node Use of Quadruple Semicolon, Examples of Style for Semicolon, Use of Triple Semicolon, Semicolon
1575
@subsubsection Use of Quadruple Semicolon
1577
Comments that begin with a quadruple @i{semicolon} are all aligned to
1578
the left margin, and generally contain only a short piece of text that
1579
serve as a title for the code which follows, and might be used in the
1580
header or footer of a program that prepares code for presentation as
1581
a hardcopy document.
1583
@node Examples of Style for Semicolon, , Use of Quadruple Semicolon, Semicolon
1584
@subsubsection Examples of Style for Semicolon
1589
;;; FIB computes the the Fibonacci function in the traditional
1593
(check-type n integer)
1594
;; At this point we're sure we have an integer argument.
1595
;; Now we can get down to some serious computation.
1597
;; Hey, this is just supposed to be a simple example.
1598
;; Did you really expect me to handle the general case?
1599
(error "FIB got ~D as an argument." n))
1600
((< n 2) n) ;fib[0]=0 and fib[1]=1
1601
;; The cheap cases didn't work.
1602
;; Nothing more to do but recurse.
1603
(t (+ (fib (- n 1)) ;The traditional formula
1604
(fib (- n 2)))))) ; is fib[n-1]+fib[n-2].
1607
@node Double-Quote, Backquote, Semicolon, Standard Macro Characters
1608
@subsection Double-Quote
1610
@b{Syntax:} @t{"<<@i{text}>>"}
1612
The @i{double-quote} is used to begin and end a @i{string}.
1613
When a @i{double-quote} is encountered,
1614
@i{characters} are read from the @i{input} @i{stream}
1615
and accumulated until another @i{double-quote} is encountered.
1616
If a @i{single escape} @i{character} is seen,
1617
the @i{single escape} @i{character} is discarded,
1618
the next @i{character} is accumulated, and accumulation continues.
1619
The accumulated @i{characters}
1620
up to but not including the matching @i{double-quote}
1621
are made into a @i{simple string} and returned.
1623
It is @i{implementation-dependent}
1624
which @i{attributes} of the accumulated characters are removed in this process.
1626
Examples of the use of the @i{double-quote} character are in Figure 2--18.
1631
@w{ @t{"Foo"} ;A string with three characters in it }
1632
@w{ @t{""} ;An empty string }
1633
@w{ @t{"\"APL\\360?\" he cried."} ;A string with twenty characters }
1634
@w{ @t{"|x| = |-x|"} ;A ten-character string }
1637
@w{ Figure 2--18: Examples of the use of double-quote }
1642
Note that to place a single escape character or a @i{double-quote} into a string,
1643
such a character must be preceded by a single escape character.
1644
Note, too, that a multiple escape character need not be quoted by a
1645
single escape character within a string.
1647
For information on how the @i{Lisp printer} prints @i{strings},
1648
see @ref{Printing Strings}.
1650
@node Backquote, Comma, Double-Quote, Standard Macro Characters
1651
@subsection Backquote
1653
The @i{backquote} introduces a template of a data structure to be built.
1654
For example, writing
1657
`(cond ((numberp ,x) ,@@y) (t (print ,x) ,@@y))
1660
is roughly equivalent to writing
1664
(cons (list 'numberp x) y)
1665
(list* 't (list 'print x) y))
1669
occurs in the template,
1671
following the comma is to be evaluated to produce an @i{object} to
1672
be inserted at that point. Assume @t{b} has the value 3, for example, then
1673
evaluating the @i{form} denoted by @t{`(a b ,b ,(+ b 1) b)} produces
1674
the result @t{(a b 3 4 b)}.
1676
If a comma is immediately followed by an @i{at-sign},
1677
then the @i{form} following the @i{at-sign}
1678
is evaluated to produce a @i{list} of @i{objects}.
1679
These @i{objects} are then ``spliced'' into place in the template. For
1680
example, if @t{x} has the value @t{(a b c)}, then
1683
`(x ,x ,@@x foo ,(cadr x) bar ,(cdr x) baz ,@@(cdr x))
1684
@result{} (x (a b c) a b c foo b bar (b c) baz b c)
1687
The backquote syntax can be summarized formally as follows.
1692
@t{`@i{basic}} is the same as @t{'@i{basic}},
1693
that is, @t{(quote @i{basic})}, for any @i{expression}
1694
@i{basic} that is not a @i{list} or a general @i{vector}.
1697
@t{`,@i{form}} is the same as @i{form}, for any @i{form}, provided
1698
that the representation of @i{form} does not begin with @i{at-sign}
1699
or @i{dot}. (A similar caveat holds for all occurrences of a form after a @i{comma}.)
1702
@t{`,@@@i{form}} has undefined consequences.
1705
@t{`(x1 x2 x3 ... xn . atom)}
1706
may be interpreted to mean
1709
(append [ x1 ] [ x2 ] [ x3 ] ... [ xn ] (quote atom))
1712
where the brackets are used to indicate
1713
a transformation of an @i{xj} as follows:
1718
@t{[@i{form}]} is interpreted as @t{(list `@i{form})},
1719
which contains a backquoted form that must then be further interpreted.
1722
@t{[,@i{form}]} is interpreted as @t{(list @i{form})}.
1725
@t{[,@@@i{form}]} is interpreted as @i{form}.
1729
@t{`(x1 x2 x3 ... xn)} may be interpreted to mean
1730
the same as the backquoted form
1731
@t{`(x1 x2 x3 ... xn . @b{nil})},
1732
thereby reducing it to the previous case.
1735
@t{`(x1 x2 x3 ... xn . ,form)} may be interpreted to mean
1738
(append [ x1 ] [ x2 ] [ x3 ] ... [ xn ] form)
1741
where the brackets indicate a transformation of an @t{xj} as described above.
1744
@t{`(x1 x2 x3 ... xn . ,@@form)} has undefined consequences.
1747
@t{`#(x1 x2 x3 ... xn)} may be interpreted to mean
1748
@t{(apply #'vector `(x1 x2 x3 ... xn))}.
1751
Anywhere ``@t{,@@}'' may be used, the syntax ``@t{,.}'' may be used instead
1752
to indicate that it is permissible to operate @i{destructively} on
1753
the @i{list structure} produced by the form following the ``@t{,.}''
1754
(in effect, to use @b{nconc} instead of @b{append}).
1756
If the backquote syntax is nested, the innermost backquoted form
1757
should be expanded first. This means that if several commas occur
1758
in a row, the leftmost one belongs to the innermost @i{backquote}.
1760
An @i{implementation} is free to interpret a backquoted @i{form} F_1
1761
as any @i{form} F_2 that, when evaluated, will produce a result that is
1762
the @i{same} under @b{equal} as the result implied by the above definition,
1763
provided that the side-effect behavior of the substitute @i{form} F_2
1764
is also consistent with the description given above.
1766
copy of the template might or might not share @i{list} structure with the
1767
template itself. As an example, the above definition implies that
1773
will be interpreted as if it were
1776
(append (list (append (list a) (list 'b) '@b{nil})) (list c) d '@b{nil})
1779
but it could also be legitimately interpreted to mean any of the following:
1782
(append (list (append (list a) (list 'b))) (list c) d)
1783
(append (list (append (list a) '(b))) (list c) d)
1784
(list* (cons a '(b)) c d)
1785
(list* (cons a (list 'b)) c d)
1786
(append (list (cons a '(b))) (list c) d)
1787
(list* (cons a '(b)) c (copy-list d))
1791
* Notes about Backquote::
1794
@node Notes about Backquote, , Backquote, Backquote
1795
@subsubsection Notes about Backquote
1797
Since the exact manner in which the @i{Lisp reader} will parse
1798
an @i{expression} involving the @i{backquote} @i{reader macro}
1799
is not specified, an @i{implementation} is free to choose any
1800
representation that preserves the semantics described.
1802
Often an @i{implementation} will choose a representation that facilitates
1803
pretty printing of the expression, so that @t{(pprint `(a ,b))} will display
1804
@t{`(a ,b)} and not, for example, @t{(list 'a b)}. However, this is not a
1807
Implementors who have no particular reason to make one choice or another
1808
might wish to refer to @b{IEEE Standard for the Scheme Programming Language}, which identifies a popular choice of
1809
representation for such expressions that might provide useful to be useful
1810
compatibility for some user communities. There is no requirement, however,
1811
that any @i{conforming implementation} use this particular representation.
1812
This information is provided merely for cross-reference purposes.
1814
@node Comma, Sharpsign, Backquote, Standard Macro Characters
1817
The @i{comma} is part of the backquote syntax; see @ref{Backquote}.
1818
@i{Comma} is invalid if used other than inside the body of a
1819
backquote @i{expression} as described above.
1821
@node Sharpsign, Re-Reading Abbreviated Expressions, Comma, Standard Macro Characters
1822
@subsection Sharpsign
1824
@i{Sharpsign} is a @i{non-terminating} @i{dispatching macro character}.
1825
It reads an optional
1826
sequence of digits and then one more character,
1827
and uses that character to select a @i{function} to run as a
1828
@i{reader macro function}.
1830
The @i{standard syntax} includes constructs introduced by the @t{#} character.
1831
The syntax of these constructs is as follows:
1832
a character that identifies the type of construct is
1833
followed by arguments in some form.
1834
If the character is a letter, its @i{case} is not important;
1835
@t{#O} and @t{#o} are considered to be equivalent, for example.
1837
Certain @t{#} constructs allow an unsigned decimal number to appear
1838
between the @t{#} and the character.
1840
The @i{reader macros} associated with the @i{dispatching macro character} @t{#}
1841
are described later in this section and summarized in Figure 2--19.
1848
@w{ dispatch char purpose dispatch char purpose }
1849
@w{ Backspace signals error @t{@{} undefined* }
1850
@w{ Tab signals error @t{@}} undefined* }
1851
@w{ Newline signals error + read-time conditional }
1852
@w{ Linefeed signals error - read-time conditional }
1853
@w{ Page signals error . read-time evaluation }
1854
@w{ Return signals error / undefined }
1855
@w{ Space signals error A, a array }
1856
@w{ ! undefined* B, b binary rational }
1857
@w{ @t{"} undefined C, c complex number }
1858
@w{ # reference to = label D, d undefined }
1859
@w{ $ undefined E, e undefined }
1860
@w{ % undefined F, f undefined }
1861
@w{ & undefined G, g undefined }
1862
@w{ ' function abbreviation H, h undefined }
1863
@w{ ( simple vector I, i undefined }
1864
@w{ ) signals error J, j undefined }
1865
@w{ @t{*} bit vector K, k undefined }
1866
@w{ , undefined L, l undefined }
1867
@w{ : uninterned symbol M, m undefined }
1868
@w{ ; undefined N, n undefined }
1869
@w{ @t{<} signals error O, o octal rational }
1870
@w{ @t{=} labels following object P, p pathname }
1871
@w{ @t{>} undefined Q, q undefined }
1872
@w{ ? undefined* R, r radix-n rational }
1873
@w{ @@ undefined S, s structure }
1874
@w{ [ undefined* T, t undefined }
1875
@w{ @t{\} character object U, u undefined }
1876
@w{ ] undefined* V, v undefined }
1877
@w{ @t{^} undefined W, w undefined }
1878
@w{ @t{_} undefined X, x hexadecimal rational }
1879
@w{ ` undefined Y, y undefined }
1880
@w{ @t{|} balanced comment Z, z undefined }
1881
@w{ @t{~} undefined Rubout undefined }
1884
@w{ Figure 2--19: Standard # Dispatching Macro Character Syntax }
1890
The combinations marked by an asterisk (*) are explicitly reserved to the
1891
user. No @i{conforming implementation} defines them.
1893
Note also that @i{digits} do not appear in the preceding table. This is
1894
because the notations @t{#0}, @t{#1}, ..., @t{#9} are
1895
reserved for another purpose which occupies the same syntactic space.
1896
When a @i{digit} follows a @i{sharpsign},
1897
it is not treated as a dispatch character.
1898
Instead, an unsigned integer argument is accumulated
1899
and passed as an @i{argument} to the @i{reader macro}
1900
for the @i{character} that follows the digits.
1902
@t{#2A((1 2) (3 4))} is a use of @t{#A} with an argument of @t{2}.
1905
* Sharpsign Backslash::
1906
* Sharpsign Single-Quote::
1907
* Sharpsign Left-Parenthesis::
1908
* Sharpsign Asterisk::
1909
* Examples of Sharpsign Asterisk::
1920
* Sharpsign Equal-Sign::
1921
* Sharpsign Sharpsign::
1924
* Sharpsign Vertical-Bar::
1925
* Examples of Sharpsign Vertical-Bar::
1926
* Notes about Style for Sharpsign Vertical-Bar::
1927
* Sharpsign Less-Than-Sign::
1928
* Sharpsign Whitespace::
1929
* Sharpsign Right-Parenthesis::
1932
@node Sharpsign Backslash, Sharpsign Single-Quote, Sharpsign, Sharpsign
1933
@subsubsection Sharpsign Backslash
1935
@b{Syntax:} @t{#\<<@i{x}>>}
1937
When the @i{token} @i{x} is a single @i{character} long,
1938
this parses as the literal @i{character} @i{char}.
1939
@i{Uppercase} and @i{lowercase} letters are distinguished after @t{#\};
1940
@t{#\A} and @t{#\a} denote different @i{character} @i{objects}.
1941
Any single @i{character} works after @t{#\},
1942
even those that are normally special to @b{read},
1943
such as @i{left-parenthesis} and @i{right-parenthesis}.
1945
In the single @i{character} case,
1946
the @i{x} must be followed by a non-constituent @i{character}.
1947
After @t{#\} is read,
1948
the reader backs up over the @i{slash} and then reads a @i{token},
1949
treating the initial @i{slash} as a @i{single escape} @i{character}
1950
(whether it really is or not in the @i{current readtable}).
1952
When the @i{token} @i{x} is more than one @i{character} long,
1953
the @i{x} must have the syntax of a @i{symbol}
1954
with no embedded @i{package markers}.
1955
In this case, the @i{sharpsign} @i{backslash} notation
1956
parses as the @i{character} whose @i{name} is @t{(string-upcase @i{x})};
1957
see @ref{Character Names}.
1959
For information about how the @i{Lisp printer} prints @i{character} @i{objects},
1960
see @ref{Printing Characters}.
1962
@node Sharpsign Single-Quote, Sharpsign Left-Parenthesis, Sharpsign Backslash, Sharpsign
1963
@subsubsection Sharpsign Single-Quote
1965
Any @i{expression} preceded by @t{#'}
1966
(@i{sharpsign} followed by @i{single-quote}),
1967
as in @t{#'@i{expression}},
1968
is treated by the @i{Lisp reader} as an abbreviation for and parsed identically
1969
to the @i{expression} @t{(function @i{expression})}.
1970
See @b{function}. For example,
1973
(apply #'+ l) @equiv{} (apply (function +) l)
1976
@node Sharpsign Left-Parenthesis, Sharpsign Asterisk, Sharpsign Single-Quote, Sharpsign
1977
@subsubsection Sharpsign Left-Parenthesis
1979
@t{#(} and @t{)} are used to notate a @i{simple vector}.
1981
If an unsigned decimal integer
1982
appears between the @t{#} and @t{(},
1983
it specifies explicitly the length of the @i{vector}.
1984
The consequences are undefined if the number of @i{objects}
1985
specified before the closing @t{)}
1986
exceeds the unsigned decimal integer.
1987
If the number of @i{objects} supplied before the closing @t{)}
1988
is less than the unsigned decimal integer but greater than zero,
1991
remaining elements of the @i{vector}.
1993
[Editorial Note by Barmar: This should say "signals...".]
1994
The consequences are undefined if the unsigned decimal integer is non-zero and
1995
number of @i{objects} supplied before the closing @t{)}
2006
all mean the same thing: a @i{vector} of length @t{6}
2007
with @i{elements} @t{a}, @t{b}, and four occurrences of @t{c}.
2008
Other examples follow:
2011
#(a b c) ;A vector of length 3
2012
#(2 3 5 7 11 13 17 19 23 29 31 37 41 43 47)
2013
;A vector containing the primes below 50
2014
#() ;An empty vector
2017
The notation @t{#()} denotes an empty @i{vector}, as does @t{#0()}.
2019
For information on how the @i{Lisp printer} prints @i{vectors},
2020
see @ref{Printing Strings},
2021
@ref{Printing Bit Vectors},
2022
or @ref{Printing Other Vectors}.
2024
@node Sharpsign Asterisk, Examples of Sharpsign Asterisk, Sharpsign Left-Parenthesis, Sharpsign
2025
@subsubsection Sharpsign Asterisk
2027
@b{Syntax:} @t{#*<<@i{bits}>>}
2029
A @i{simple bit vector} is constructed containing the indicated @i{bits}
2030
(@t{0}'s and @t{1}'s), where the leftmost @i{bit} has index zero
2031
and the subsequent @i{bits} have increasing indices.
2033
@b{Syntax:} @t{#<<@i{n}>>*<<@i{bits}>>}
2035
With an argument @i{n},
2036
the @i{vector} to be created is of @i{length} @i{n}.
2037
If the number of @i{bits} is less than @i{n} but greater than zero,
2038
the last bit is used to fill all remaining bits of the @i{bit vector}.
2040
The notations @t{#*} and @t{#0*} each denote an empty @i{bit vector}.
2042
Regardless of whether the optional numeric argument @i{n} is provided,
2043
the @i{token} that follows the @i{asterisk} is delimited by
2044
a normal @i{token} delimiter.
2045
However, (unless the @i{value} of @b{*read-suppress*} is @i{true})
2046
an error of @i{type} @b{reader-error} is signaled
2047
if that @i{token} is not composed entirely of @t{0}'s and @t{1}'s,
2048
or if @i{n} was supplied
2049
and the @i{token} is composed of more than @i{n} @i{bits},
2050
or if @i{n} is greater than one, but no @i{bits} were specified.
2051
Neither a @i{single escape} nor a @i{multiple escape} is permitted in this @i{token}.
2053
For information on how the @i{Lisp printer} prints @i{bit vectors},
2054
see @ref{Printing Bit Vectors}.
2056
@node Examples of Sharpsign Asterisk, Sharpsign Colon, Sharpsign Asterisk, Sharpsign
2057
@subsubsection Examples of Sharpsign Asterisk
2067
all mean the same thing: a @i{vector} of length @t{6}
2068
with @i{elements} @t{1}, @t{0}, @t{1}, @t{1}, @t{1}, and @t{1}.
2073
#* ;An empty bit-vector
2076
@node Sharpsign Colon, Sharpsign Dot, Examples of Sharpsign Asterisk, Sharpsign
2077
@subsubsection Sharpsign Colon
2079
@b{Syntax:} @t{#:<<@i{symbol-name}>>}
2081
@t{#:} introduces an @i{uninterned} @i{symbol} whose @i{name}
2082
is @i{symbol-name}. Every time this syntax is encountered,
2083
a @i{distinct} @i{uninterned} @i{symbol} is created.
2084
The @i{symbol-name} must have the syntax of a @i{symbol}
2085
with no @i{package prefix}.
2087
For information on how the @i{Lisp reader}
2088
prints @i{uninterned} @i{symbols},
2089
see @ref{Printing Symbols}.
2091
@node Sharpsign Dot, Sharpsign B, Sharpsign Colon, Sharpsign
2092
@subsubsection Sharpsign Dot
2094
@t{#.@i{foo}} is read as the @i{object} resulting from the evaluation
2095
of the @i{object} represented by @i{foo}.
2096
The evaluation is done during the @b{read} process,
2097
when the @t{#.} notation is encountered.
2098
The @t{#.} syntax therefore performs a read-time evaluation of @i{foo}.
2100
The normal effect of @t{#.} is inhibited when the @i{value} of @b{*read-eval*} is @i{false}.
2102
In that situation, an error of @i{type} @b{reader-error} is signaled.
2105
that does not have a convenient printed
2106
representation, a @i{form} that computes the @i{object} can be given using
2107
the @t{#.} notation.
2109
@node Sharpsign B, Sharpsign O, Sharpsign Dot, Sharpsign
2110
@subsubsection Sharpsign B
2112
@t{#B}@i{rational} reads @i{rational} in binary (radix 2).
2116
#B1101 @equiv{} 13 ;1101_2
2117
#b101/11 @equiv{} 5/3
2120
The consequences are undefined if the token immediately following
2121
the @t{#B} does not have the syntax of a binary (@i{i.e.}, radix 2) @i{rational}.
2123
@node Sharpsign O, Sharpsign X, Sharpsign B, Sharpsign
2124
@subsubsection Sharpsign O
2126
@t{#O}@i{rational} reads @i{rational} in octal (radix 8).
2130
#o37/15 @equiv{} 31/13
2132
#o105 @equiv{} 69 ;105_8
2135
The consequences are undefined if the token immediately following
2136
the @t{#O} does not have the syntax of an octal (@i{i.e.}, radix 8) @i{rational}.
2138
@node Sharpsign X, Sharpsign R, Sharpsign O, Sharpsign
2139
@subsubsection Sharpsign X
2141
@t{#X}@i{rational} reads @i{rational} in hexadecimal (radix 16).
2142
The digits above @t{9} are the letters @t{A} through @t{F} (the lowercase
2143
letters @t{a} through @t{f} are also acceptable). For example,
2147
#x105 @equiv{} 261 ;105_@t{16}
2150
The consequences are undefined if the token immediately following
2151
the @t{#X} does not have the syntax of a hexadecimal (@i{i.e.}, radix 16) @i{rational}.
2153
@node Sharpsign R, Sharpsign C, Sharpsign X, Sharpsign
2154
@subsubsection Sharpsign R
2158
@t{#@i{radix}R@i{rational}} reads @i{rational} in radix @i{radix}.
2159
@i{radix} must consist of only digits
2160
that are interpreted as an @i{integer}
2161
in decimal radix; its value must be between 2 and 36 (inclusive).
2163
for the specified radix may be used.
2165
For example, @t{#3r102} is another way of writing @t{11} (decimal),
2167
is another way of writing @t{35} (decimal).
2168
For radices larger than 10, letters of
2169
the alphabet are used in order for the digits after @t{9}.
2170
No alternate @t{#} notation exists for the decimal radix since a
2171
decimal point suffices.
2173
Figure 2--20 contains examples of the use of @t{#B},
2174
@t{#O}, @t{#X}, and @t{#R}.
2179
@w{ @t{#2r11010101} ;Another way of writing @t{213} decimal }
2180
@w{ @t{#b11010101} ;Ditto }
2181
@w{ @t{#b+11010101} ;Ditto }
2182
@w{ @t{#o325} ;Ditto, in octal radix }
2183
@w{ @t{#xD5} ;Ditto, in hexadecimal radix }
2184
@w{ @t{#16r+D5} ;Ditto }
2185
@w{ @t{#o-300} ;Decimal @t{-192}, written in base 8 }
2186
@w{ @t{#3r-21010} ;Same thing in base 3 }
2187
@w{ @t{#25R-7H} ;Same thing in base 25 }
2188
@w{ @t{#xACCEDED} ;@t{181202413}, in hexadecimal radix }
2191
@w{ Figure 2--20: Radix Indicator Example }
2196
The consequences are undefined if the token immediately following
2197
the @t{#@i{n}R} does not have the syntax of a @i{rational} in radix @i{n}.
2199
@node Sharpsign C, Sharpsign A, Sharpsign R, Sharpsign
2200
@subsubsection Sharpsign C
2202
@t{#C} reads a following @i{object}, which must be a @i{list} of
2203
length two whose @i{elements} are both @i{reals}.
2204
These @i{reals} denote, respectively,
2205
the real and imaginary parts of a @i{complex} number.
2207
If the two parts as notated are not of the same data type,
2208
then they are converted
2209
according to the rules of floating-point @i{contagion}
2210
described in @ref{Contagion in Numeric Operations}.
2212
@t{#C(@i{real} @i{imag})} is equivalent to
2213
@t{#.(complex (quote @i{real}) (quote @i{imag}))},
2214
except that @t{#C} is not affected by @b{*read-eval*}.
2215
See the @i{function} @b{complex}.
2217
Figure 2--21 contains examples of the use of @t{#C}.
2222
@w{ @t{#C(3.0s1 2.0s-1)} ;A @i{complex} with @i{small float} parts. }
2223
@w{ @t{#C(5 -3) } ;A ``Gaussian integer'' }
2224
@w{ @t{#C(5/3 7.0) } ;Will be converted internally to @t{#C(1.66666 7.0)} }
2225
@w{ @t{#C(0 1)} ;The imaginary unit; that is, i. }
2228
@w{ Figure 2--21: Complex Number Example }
2233
For further information,
2234
see @ref{Printing Complexes} and @ref{Syntax of a Complex}.
2236
@node Sharpsign A, Sharpsign S, Sharpsign C, Sharpsign
2237
@subsubsection Sharpsign A
2241
@t{#@i{n}@t{A}@i{object}} constructs an @i{n}-dimensional @i{array},
2242
using @i{object} as the value of the @t{:initial-contents} argument
2245
For example, @t{#2A((0 1 5) (foo 2 (hot dog)))} represents a 2-by-3 matrix:
2252
In contrast, @t{#1A((0 1 5) (foo 2 (hot dog)))}
2253
represents a @i{vector} of @i{length} @t{2}
2254
whose @i{elements} are @i{lists}:
2257
(0 1 5) (foo 2 (hot dog))
2260
@t{#0A((0 1 5) (foo 2 (hot dog)))} represents a zero-dimensional
2261
@i{array} whose sole element is a @i{list}:
2264
((0 1 5) (foo 2 (hot dog)))
2267
@t{#0A foo} represents
2268
a zero-dimensional @i{array} whose sole element is the
2270
The notation @t{#1A foo} is not valid because @t{foo} is
2273
If some @i{dimension} of the @i{array}
2274
whose representation is being parsed is found to be @t{0},
2275
all @i{dimensions} to the right
2276
(@i{i.e.}, the higher numbered @i{dimensions})
2277
are also considered to be @t{0}.
2279
For information on how the @i{Lisp printer} prints @i{arrays},
2280
see @ref{Printing Strings},
2281
@ref{Printing Bit Vectors},
2282
@ref{Printing Other Vectors},
2283
or @ref{Printing Other Arrays}.
2285
@node Sharpsign S, Sharpsign P, Sharpsign A, Sharpsign
2286
@subsubsection Sharpsign S
2288
@t{#s(name slot1 value1 slot2 value2 ...)}
2289
denotes a @i{structure}. This is valid only if @i{name} is the name
2290
of a @i{structure} @i{type} already defined by @b{defstruct}
2291
and if the @i{structure} @i{type} has a standard constructor function.
2292
Let @i{cm} stand for the name of this constructor function;
2293
then this syntax is equivalent to
2296
#.(cm keyword1 'value1 keyword2 'value2 ...)
2299
where each @i{keywordj} is the result of computing
2302
(intern (string slotj) (find-package 'keyword))
2305
The net effect is that the constructor function is called with the specified
2306
slots having the specified values.
2308
(This coercion feature is deprecated; in the future, keyword names will
2309
be taken in the package they are read in, so @i{symbols} that are
2310
actually in the @t{KEYWORD} @i{package} should be used if that is what is desired.)
2312
Whatever @i{object} the constructor function returns
2313
is returned by the @t{#S} syntax.
2315
For information on how the @i{Lisp printer} prints @i{structures},
2316
see @ref{Printing Structures}.
2318
@node Sharpsign P, Sharpsign Equal-Sign, Sharpsign S, Sharpsign
2319
@subsubsection Sharpsign P
2321
@t{#P} reads a following @i{object}, which must be a @i{string}.
2323
@t{#P<<@i{expression}>>} is equivalent to
2324
@t{#.(parse-namestring '<<@i{expression}>>)},
2325
except that @t{#P} is not affected by @b{*read-eval*}.
2327
For information on how the @i{Lisp printer} prints @i{pathnames},
2328
see @ref{Printing Pathnames}.
2330
@node Sharpsign Equal-Sign, Sharpsign Sharpsign, Sharpsign P, Sharpsign
2331
@subsubsection Sharpsign Equal-Sign
2335
@t{#@i{n}=@i{object}} reads as whatever @i{object}
2336
has @i{object} as its printed representation. However, that @i{object}
2337
is labeled by @i{n}, a required unsigned decimal integer, for
2338
possible reference by the syntax @t{#@i{n}#}.
2339
The scope of the label is the @i{expression} being read by the outermost
2340
call to @b{read}; within this @i{expression},
2341
the same label may not appear twice.
2343
@node Sharpsign Sharpsign, Sharpsign Plus, Sharpsign Equal-Sign, Sharpsign
2344
@subsubsection Sharpsign Sharpsign
2348
@t{#@i{n}#}, where @i{n} is a required unsigned decimal
2350
provides a reference to some @i{object} labeled by @t{#@i{n}=};
2351
that is, @t{#@i{n}#} represents a pointer to the same
2352
(@b{eq}) @i{object} labeled by @t{#@i{n}=}.
2353
For example, a structure created in the variable @t{y} by this code:
2356
(setq x (list 'p 'q))
2357
(setq y (list (list 'a 'b) x 'foo x))
2358
(rplacd (last y) (cdr y))
2361
could be represented in this way:
2364
((a b) . #1=(#2=(p q) foo #2# . #1#))
2367
Without this notation, but with @b{*print-length*} set to @t{10}
2368
and @b{*print-circle*} set to @b{nil},
2369
the structure would print in this way:
2372
((a b) (p q) foo (p q) (p q) foo (p q) (p q) foo (p q) ...)
2375
A reference @t{#@i{n}#} may only occur after a label @t{#@i{n}=};
2376
forward references are not permitted. The reference
2377
may not appear as the labeled object itself (that is,
2378
@t{#@i{n}=#@i{n}#}) may not be written
2379
because the @i{object}
2380
labeled by @t{#@i{n}=} is not well defined in this case.
2382
@node Sharpsign Plus, Sharpsign Minus, Sharpsign Sharpsign, Sharpsign
2383
@subsubsection Sharpsign Plus
2385
@t{#+} provides a read-time conditionalization facility;
2386
the syntax is @t{#+@i{test} @i{expression}}.
2387
If the @i{feature expression} @i{test} succeeds,
2388
then this textual notation represents an @i{object}
2389
whose printed representation is @i{expression}.
2390
If the @i{feature expression} @i{test} fails,
2391
then this textual notation is treated as @i{whitespace}_2;
2392
that is, it is as if the ``@t{#+} @i{test} @i{expression}''
2393
did not appear and only a @i{space} appeared in its place.
2395
For a detailed description of success and failure in @i{feature expressions},
2396
see @ref{Feature Expressions}.
2398
@t{#+} operates by first reading the @i{feature expression}
2399
and then skipping over the @i{form} if the @i{feature expression} fails.
2401
While reading the @i{test}, the @i{current package} is the @t{KEYWORD} @i{package}.
2403
Skipping over the @i{form} is accomplished by @i{binding}
2404
@b{*read-suppress*} to @i{true} and then calling @b{read}.
2406
For examples, see @ref{Examples of Feature Expressions}.
2408
@node Sharpsign Minus, Sharpsign Vertical-Bar, Sharpsign Plus, Sharpsign
2409
@subsubsection Sharpsign Minus
2411
@t{#-} is like @t{#+}
2412
except that it skips the @i{expression} if the @i{test} succeeds;
2416
#-@i{test} @i{expression} @equiv{} #+(not @i{test}) @i{expression}
2419
For examples, see @ref{Examples of Feature Expressions}.
2421
@node Sharpsign Vertical-Bar, Examples of Sharpsign Vertical-Bar, Sharpsign Minus, Sharpsign
2422
@subsubsection Sharpsign Vertical-Bar
2424
@t{#|...|#} is treated as a comment by the reader.
2425
It must be balanced with respect to other occurrences of @t{#|} and @t{|#},
2426
but otherwise may contain any characters whatsoever.
2428
@node Examples of Sharpsign Vertical-Bar, Notes about Style for Sharpsign Vertical-Bar, Sharpsign Vertical-Bar, Sharpsign
2429
@subsubsection Examples of Sharpsign Vertical-Bar
2431
The following are some examples that exploit the @t{#|...|#} notation:
2434
;;; In this example, some debugging code is commented out with #|...|#
2435
;;; Note that this kind of comment can occur in the middle of a line
2436
;;; (because a delimiter marks where the end of the comment occurs)
2437
;;; where a semicolon comment can only occur at the end of a line
2438
;;; (because it comments out the rest of the line).
2439
(defun add3 (n) #|(format t "~&Adding 3 to ~D." n)|# (+ n 3))
2441
;;; The examples that follow show issues related to #| ... |# nesting.
2443
;;; In this first example, #| and |# always occur properly paired,
2444
;;; so nesting works naturally.
2445
(defun mention-fun-fact-1a ()
2446
(format t "CL uses ; and #|...|# in comments."))
2447
@result{} MENTION-FUN-FACT-1A
2448
(mention-fun-fact-1a)
2449
@t{ |> } CL uses ; and #|...|# in comments.
2451
#| (defun mention-fun-fact-1b ()
2452
(format t "CL uses ; and #|...|# in comments.")) |#
2453
(fboundp 'mention-fun-fact-1b) @result{} NIL
2455
;;; In this example, vertical-bar followed by sharpsign needed to appear
2456
;;; in a string without any matching sharpsign followed by vertical-bar
2457
;;; having preceded this. To compensate, the programmer has included a
2458
;;; slash separating the two characters. In case 2a, the slash is
2459
;;; unnecessary but harmless, but in case 2b, the slash is critical to
2460
;;; allowing the outer #| ... |# pair match. If the slash were not present,
2461
;;; the outer comment would terminate prematurely.
2462
(defun mention-fun-fact-2a ()
2463
(format t "Don't use |\# unmatched or you'll get in trouble!"))
2464
@result{} MENTION-FUN-FACT-2A
2465
(mention-fun-fact-2a)
2466
@t{ |> } Don't use |# unmatched or you'll get in trouble!
2468
#| (defun mention-fun-fact-2b ()
2469
(format t "Don't use |\# unmatched or you'll get in trouble!") |#
2470
(fboundp 'mention-fun-fact-2b) @result{} NIL
2472
;;; In this example, the programmer attacks the mismatch problem in a
2473
;;; different way. The sharpsign vertical bar in the comment is not needed
2474
;;; for the correct parsing of the program normally (as in case 3a), but
2475
;;; becomes important to avoid premature termination of a comment when such
2476
;;; a program is commented out (as in case 3b).
2477
(defun mention-fun-fact-3a () ; #|
2478
(format t "Don't use |# unmatched or you'll get in trouble!"))
2479
@result{} MENTION-FUN-FACT-3A
2480
(mention-fun-fact-3a)
2481
@t{ |> } Don't use |# unmatched or you'll get in trouble!
2484
(defun mention-fun-fact-3b () ; #|
2485
(format t "Don't use |# unmatched or you'll get in trouble!"))
2487
(fboundp 'mention-fun-fact-3b) @result{} NIL
2490
@node Notes about Style for Sharpsign Vertical-Bar, Sharpsign Less-Than-Sign, Examples of Sharpsign Vertical-Bar, Sharpsign
2491
@subsubsection Notes about Style for Sharpsign Vertical-Bar
2493
Some text editors that purport to understand Lisp syntax treat any @t{|...|}
2494
as balanced pairs that cannot nest (as if they were just balanced pairs of
2495
the multiple escapes used in notating certain symbols). To compensate for
2496
this deficiency, some programmers use the notation @t{#||...#||...||#...||#}
2497
instead of @t{#|...#|...|#...|#}. Note that this alternate usage is not
2498
a different @i{reader macro}; it merely exploits the fact that the additional
2499
vertical-bars occur within the comment in a way that tricks certain text editor
2500
into better supporting nested comments. As such, one might sometimes see code
2504
#|| (+ #|| 3 ||# 4 5) ||#
2507
Such code is equivalent to:
2510
#| (+ #| 3 |# 4 5) |#
2513
@node Sharpsign Less-Than-Sign, Sharpsign Whitespace, Notes about Style for Sharpsign Vertical-Bar, Sharpsign
2514
@subsubsection Sharpsign Less-Than-Sign
2516
@t{#<} is not valid reader syntax.
2517
The @i{Lisp reader} will signal an error
2519
of @i{type} @b{reader-error}
2521
on encountering @t{#<}.
2522
This syntax is typically used in the printed representation
2523
of @i{objects} that cannot be read back in.
2525
@node Sharpsign Whitespace, Sharpsign Right-Parenthesis, Sharpsign Less-Than-Sign, Sharpsign
2526
@subsubsection Sharpsign Whitespace
2528
@t{#} followed immediately by @i{whitespace}_1 is not valid reader syntax.
2529
The @i{Lisp reader} will signal an error of @i{type} @b{reader-error} if it
2530
encounters the reader macro notation @t{#<@i{Newline}>} or @t{#<@i{Space}>}.
2532
@node Sharpsign Right-Parenthesis, , Sharpsign Whitespace, Sharpsign
2533
@subsubsection Sharpsign Right-Parenthesis
2535
This is not valid reader syntax.
2537
The @i{Lisp reader} will signal an error
2539
of @i{type} @b{reader-error}
2541
upon encountering @t{#)}.
2543
@node Re-Reading Abbreviated Expressions, , Sharpsign, Standard Macro Characters
2544
@subsection Re-Reading Abbreviated Expressions
2546
Note that the @i{Lisp reader} will
2548
signal an error of @i{type} @b{reader-error}
2549
when reading an @i{expression}_2 that has been
2550
abbreviated because of length or level limits
2551
(see @b{*print-level*},
2553
and @b{*print-lines*})
2554
due to restrictions on ``@t{..}'', ``@t{...}'', ``@t{#}'' followed by @i{whitespace}_1,
2557
@c end of including concept-macro-chars
2559
@c %**end of chapter
added
removed
info/chap-2.texi
