HyperLib 1997 Winter - Disc 1/HYPERLIB-1997-Winter-CD1.ISO.7z/HYPERLIB-1997-Winter-CD1.ISO/オンラインウェア/PRG/PL 2.0 SupplementDoc Folder.sit/PL 2.0 SupplementDoc Folder/Documentation/Chapter 18. Strings

Text File | 1995-03-27 | 17KB | 418 lines

Common Lisp the Language, 2nd Edition
-------------------------------------------------------------------------------

18. Strings

A string is a specialized vector (one-dimensional array) whose elements are
characters.

[old_change_begin]
Specifically, the type string is identical to the type (vector string-char),
which in turn is the same as (array string-char (*)).
[old_change_end]

[change_begin]
X3J13 voted in March 1989 (CHARACTER-PROPOSAL)   to eliminate the type
string-char and to redefine the type string to be the union of one or more
specialized vector types, the types of whose elements are subtypes of the type
character.
[change_end]

Any string-specific function defined in this chapter whose name begins with the
prefix string will accept a symbol instead of a string as an argument provided
that the operation never modifies that argument; the print name of the symbol
is used.   In this respect the string-specific sequence operations are not
simply specializations of generic versions; the generic sequence operations
described in chapter 14 never accept symbols as sequences. This slight
inelegance is permitted in Common Lisp in the name of pragmatic utility. One
may get the effect of having a generic sequence function operate on either
symbols or strings by applying the coercion function string to any argument
whose data type is in doubt.

[change_begin]
Note that this remark, predating the design of the Common Lisp Object System,
uses the term ``generic'' in a generic sense and not necessarily in the
technical sense used by CLOS (see chapter 2).
[change_end]

Also, there is a slight non-parallelism in the names of string functions. Where
the suffixes equalp and eql would be more appropriate, for historical
compatibility the suffixes equal and = are used instead to indicate
case-insensitive and case-sensitive character comparison, respectively.

Any Lisp object may be tested for being a string by the predicate stringp.

Note that strings, like all vectors, may have fill pointers (though such
strings are not necessarily simple). String operations generally operate only
on the active portion of the string (below the fill pointer). See fill-pointer
and related functions.

-------------------------------------------------------------------------------

   *  String Access
   *  String Comparison
   *  String Construction and Manipulation

-------------------------------------------------------------------------------

18.1. String Access

The following functions access a single character element of a string.

[Function]
char string index
schar simple-string index

The given index must be a non-negative integer less than the length of string,
which must be a string. The character at position index of the string is
returned as a character object. (This character will necessarily satisfy the
predicate string-char-p.)

[change_begin]
X3J13 voted in March 1989 (CHARACTER-PROPOSAL)   to eliminate string-char-p.
[change_end]

As with all sequences in Common Lisp, indexing is zero-origin. For example:

(char "Floob-Boober-Bab-Boober-Bubs" 0) => #¥F
(char "Floob-Boober-Bab-Boober-Bubs" 1) => #¥l

See aref and elt. In effect,

(char s j) == (aref (the string s) j)

setf may be used with char to destructively replace a character within a
string.

For char, the string may be any string; for schar, it must be a simple string.
In some implementations of Common Lisp, the function schar may be faster than
char when it is applicable.

-------------------------------------------------------------------------------

18.2. String Comparison

The naming conventions for these functions and for their keyword arguments
generally follow the conventions for the generic sequence functions (see
chapter 14).

[change_begin]
Note that this remark, predating the design of the Common Lisp Object System,
uses the term ``generic'' in a generic sense and not necessarily in the
technical sense used by CLOS (see chapter 2).
[change_end]

[Function]
string= string1 string2 &key :start1 :end1 :start2 :end2

string= compares two strings and is true if they are the same (corresponding
characters are identical) but is false if they are not. The function equal
calls string= if applied to two strings.

The keyword arguments :start1 and :start2 are the places in the strings to
start the comparison. The arguments :end1 and :end2 are the places in the
strings to stop comparing; comparison stops just before the position specified
by a limit. The ``start'' arguments default to zero (beginning of string), and
the ``end'' arguments (if either omitted or nil) default to the lengths of the
strings (end of string), so that by default the entirety of each string is
examined. These arguments are provided so that substrings can be compared
efficiently.

string= is necessarily false if the (sub)strings being compared are of unequal
length; that is, if

(not (= (- end1 start1) (- end2 start2)))

is true, then string= is false.

(string= "foo" "foo") is true
(string= "foo" "Foo") is false
(string= "foo" "bar") is false
(string= "together" "frog" :start1 1 :end1 3 :start2 2)
   is true

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

-------------------------------------------------------------------------------
Compatibility note: string= is called strequal in Interlisp.
-------------------------------------------------------------------------------

[Function]
string-equal string1 string2 &key :start1 :end1 :start2 :end2

string-equal is just like string= except that differences in case are ignored;
two characters are considered to be the same if char-equal is true of them. For
example:

(string-equal "foo" "Foo") is true

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

[Function]

string< string1 string2 &key :start1 :end1 :start2 :end2
string> string1 string2 &key :start1 :end1 :start2 :end2
string<= string1 string2 &key :start1 :end1 :start2 :end2
string>= string1 string2 &key :start1 :end1 :start2 :end2
string/= string1 string2 &key :start1 :end1 :start2 :end2

These functions compare the two string arguments lexicographically, and the
result is nil unless string1 is respectively less than, greater than, less than
or equal to, greater than or equal to, or not equal to string2. If the
condition is satisfied, however, then the result is the index within the
strings of the first character position at which the strings fail to match; put
another way, the result is the length of the longest common prefix of the
strings.

A string a is less than a string b if in the first position in which they
differ the character of a is less than the corresponding character of b
according to the function char<, or if string a is a proper prefix of string b
(of shorter length and matching in all the characters of a).

The keyword arguments :start1 and :start2 are the places in the strings to
start the comparison. The keyword arguments :end1 and :end2 are the places in
the strings to stop comparing; comparison stops just before the position
specified by a limit. The ``start'' arguments default to zero (beginning of
string), and the ``end'' arguments (if either omitted or nil) default to the
lengths of the strings (end of string), so that by default the entirety of each
string is examined. These arguments are provided so that substrings can be
compared efficiently. The index returned in case of a mismatch is an index into
string1.

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

[Function]

string-lessp string1 string2 &key :start1 :end1 :start2 :end2
string-greaterp string1 string2 &key :start1 :end1 :start2 :end2
string-not-greaterp string1 string2 &key :start1 :end1 :start2 :end2
string-not-lessp string1 string2 &key :start1 :end1 :start2 :end2
string-not-equal string1 string2 &key :start1 :end1 :start2 :end2

These are exactly like string<, string>, string<=, string>=, and string/=,
respectively, except that distinctions between uppercase and lowercase letters
are ignored. It is as if char-lessp were used instead of char< for comparing
characters.

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

-------------------------------------------------------------------------------

18.3. String Construction and Manipulation

Most of the interesting operations on strings may be performed with the generic
sequence functions described in chapter 14. The following functions perform
additional operations that are specific to strings.

[change_begin]
Note that this remark, predating the design of the Common Lisp Object System,
uses the term ``generic'' in a generic sense and not necessarily in the
technical sense used by CLOS (see chapter 2).
[change_end]

[old_change_begin]

[Function]
make-string size &key :initial-element

This returns a string (in fact a simple string) of length size, each of whose
characters has been initialized to the :initial-element argument. If an
:initial-element argument is not specified, then the string will be initialized
in an implementation-dependent way.

-------------------------------------------------------------------------------
Implementation note: It may be convenient to initialize the string to null
characters, or to spaces, or to garbage (``whatever was there'').
-------------------------------------------------------------------------------

A string is really just a one-dimensional array of ``string characters'' (that
is, those characters that are members of type string-char). More complex
character arrays may be constructed using the function make-array.
[old_change_end]

[change_begin]
X3J13 voted in March 1989 (CHARACTER-PROPOSAL)   to eliminate the type
string-char and to add a keyword argument :element-type to make-string. The new
function description is as follows.

[Function]
make-string size &key :initial-element :element-type

This returns a simple string of length size, each of whose characters has been
initialized to the :initial-element argument. If an :initial-element argument
is not specified, then the string will be initialized in an
implementation-dependent way.

The :element-type argument names the type of the elements of the string; a
string is constructed of the most specialized type that can accommodate
elements of the given type. If :element-type is omitted, the type character is
the default.

X3J13 voted in January 1989 (ARGUMENTS-UNDERSPECIFIED)   to clarify that the
size argument must be a non-negative integer less than the value of
array-dimension-limit.
[change_end]

[Function]
string-trim character-bag string
string-left-trim character-bag string
string-right-trim character-bag string

string-trim returns a substring of string, with all characters in character-bag
stripped off the beginning and end. The function string-left-trim is similar
but strips characters off only the beginning; string-right-trim strips off only
the end. The argument character-bag may be any sequence containing characters.
For example:

(string-trim '(#¥Space #¥Tab #¥Newline) " garbanzo beans
        ") => "garbanzo beans"
(string-trim " (*)" " ( *three (silly) words* ) ")
   => "three (silly) words"
(string-left-trim " (*)" " ( *three (silly) words* ) ")
   => "three (silly) words* ) "
(string-right-trim " (*)" " ( *three (silly) words* ) ")
   => " ( *three (silly) words"

If no characters need to be trimmed from the string, then either the argument
string itself or a copy of it may be returned, at the discretion of the
implementation.

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

[Function]
string-upcase string &key :start :end
string-downcase string &key :start :end
string-capitalize string &key :start :end

string-upcase returns a string just like string with all lowercase characters
replaced by the corresponding uppercase characters. More precisely, each
character of the result string is produced by applying the function char-upcase
to the corresponding character of string.

string-downcase is similar, except that uppercase characters are converted to
lowercase characters (using char-downcase).

The keyword arguments :start and :end delimit the portion of the string to be
affected. The result is always of the same length as string, however.

The argument is not destroyed. However, if no characters in the argument
require conversion, the result may be either the argument or a copy of it, at
the implementation's discretion. For example:

(string-upcase "Dr. Livingstone, I presume?")
   => "DR. LIVINGSTONE, I PRESUME?"
(string-downcase "Dr. Livingstone, I presume?")
   => "dr. livingstone, i presume?"
(string-upcase "Dr. Livingstone, I presume?" :start 6 :end 10)
   => "Dr. LiVINGstone, I presume?"

string-capitalize produces a copy of string such that, for every word in the
copy, the first character of the word, if case-modifiable, is uppercase and any
other case-modifiable characters in the word are lowercase. For the purposes of
string-capitalize, a word is defined to be a consecutive subsequence consisting
of alphanumeric characters or digits, delimited at each end either by a
non-alphanumeric character or by an end of the string. For example:

(string-capitalize " hello ") => " Hello "
(string-capitalize
    ｯ"occlUDeD cASEmenTs FOreSTAll iNADVertent DEFenestraTION")
=>¥>"Occluded Casements Forestall Inadvertent Defenestration"
(string-capitalize 'kludgy-hash-search) => "Kludgy-Hash-Search"
(string-capitalize "DON'T!") => "Don'T!"     ;not "Don't!"
(string-capitalize "pipe 13a, foo16c") => "Pipe 13a, Foo16c"

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to clarify string coercion (see
string).
[change_end]

-------------------------------------------------------------------------------
Compatibility note: Some very approximate Interlisp equivalents to
string-upcase, string-downcase, and string-capitalize are u-case, l-case with
second argument nil, and l-case with second argument t.
-------------------------------------------------------------------------------

[Function]
nstring-upcase string &key :start :end
nstring-downcase string &key :start :end
nstring-capitalize string &key :start :end

These three functions are just like string-upcase, string-downcase, and
string-capitalize but destructively modify the argument string by altering
case-modifiable characters as necessary.

The keyword arguments :start and :end delimit the portion of the string to be
affected. The argument string is returned as the result.

[Function]
string x

Most of the string functions effectively apply string to such of their
arguments as are supposed to be strings. If x is a string, it is returned. If x
is a symbol, its print name is returned.

[old_change_begin]
If x is a string character (a character of type string-char), then a string
containing that one character is returned.
[old_change_end]

[change_begin]
X3J13 voted in March 1989 (CHARACTER-PROPOSAL)   to eliminate the type
string-char and to redefine the type string to be the union of one or more
specialized vector types, the types of whose elements are subtypes of the type
character. Presumably converting a character to a string always works according
to this vote.
[change_end]

In any other situation, an error is signaled.

To convert a sequence of characters to a string, use coerce. (Note that (coerce
x 'string) will not succeed if x is a symbol. Conversely, string will not
convert a list or other sequence to be a string.)

To get the string representation of a number or any other Lisp object, use
prin1-to-string, princ-to-string, or format.

[change_begin]
X3J13 voted in June 1989 (STRING-COERCION)   to specify that the following
functions perform coercion on their string arguments identical to that
performed by the function string.

string=       string-equal           string-trim
string<       string-lessp           string-left-trim
string>       string-greaterp        string-right-trim
string<=      string-not-greaterp    string-upcase
string>=      string-not-lessp       string-downcase
string/=      string-not-equal       string-capitalize

Note that nstring-upcase, nstring-downcase, and nstring-capitalize are absent
from this list; because they modify destructively, the argument must be a
string.

As part of the same vote X3J13 specified that string may perform additional
implementation-dependent coercions but the returned value must be of type
string. Only when no coercion is defined, whether standard or
implementation-dependent, is string required to signal an error, in which case
the error condition must be of type type-error.
[change_end]

-------------------------------------------------------------------------------