173 lines
9.6 KiB
HTML
173 lines
9.6 KiB
HTML
<!DOCTYPE HTML PUBLIC "-//W3O//DTD W3 HTML 2.0//EN">
|
|
<!Converted with LaTeX2HTML 0.6.5 (Tue Nov 15 1994) by Nikos Drakos (nikos@cbl.leeds.ac.uk), CBLU, University of Leeds >
|
|
<HEAD>
|
|
<TITLE>A.2.4. Conditional and Other Complex Transducers</TITLE>
|
|
</HEAD>
|
|
<BODY>
|
|
<meta name="description" value=" Conditional and Other Complex Transducers">
|
|
<meta name="keywords" value="clm">
|
|
<meta name="resource-type" value="document">
|
|
<meta name="distribution" value="global">
|
|
<P>
|
|
<b>Common Lisp the Language, 2nd Edition</b>
|
|
<BR> <HR><A NAME=tex2html6116 HREF="node354.html"><IMG ALIGN=BOTTOM ALT="next" SRC="icons/next_motif.gif"></A> <A NAME=tex2html6114 HREF="node349.html"><IMG ALIGN=BOTTOM ALT="up" SRC="icons/up_motif.gif"></A> <A NAME=tex2html6108 HREF="node352.html"><IMG ALIGN=BOTTOM ALT="previous" SRC="icons/previous_motif.gif"></A> <A NAME=tex2html6118 HREF="node1.html"><IMG ALIGN=BOTTOM ALT="contents" SRC="icons/contents_motif.gif"></A> <A NAME=tex2html6119 HREF="index.html"><IMG ALIGN=BOTTOM ALT="index" SRC="icons/index_motif.gif"></A> <BR>
|
|
<B> Next:</B> <A NAME=tex2html6117 HREF="node354.html"> Collectors</A>
|
|
<B>Up:</B> <A NAME=tex2html6115 HREF="node349.html"> Series Functions</A>
|
|
<B> Previous:</B> <A NAME=tex2html6109 HREF="node352.html"> Truncation and Other </A>
|
|
<HR> <P>
|
|
<H2><A NAME=SECTION003424000000000000000>A.2.4. Conditional and Other Complex Transducers</A></H2>
|
|
<P>
|
|
<img align=bottom alt="change_begin" src="gif/change_begin.gif"><br>
|
|
<A NAME=SERIESOLSECTION>This</A>
|
|
section presents a number of complex transducers, including ones that
|
|
support conditional computation.
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>choose <i>bools</i> &optional (<i>items</i> bools) <BR></tt><tt>choose-if <i>pred</i> <i>items</i></tt><P>Each of these functions takes in a series of elements (<i>items</i>) and
|
|
returns a series containing the same elements in the same order, but with
|
|
some elements removed. <tt>choose</tt> removes <i>items</i>j if <i>bools</i>j is <tt>nil</tt> or <b><i>j</i></b> is beyond the end of <i>bools</i>. If <i>items</i> is omitted, <tt>choose</tt> returns the non-null elements of <i>bools</i>. <tt>choose-if</tt> removes <i>items</i>j if
|
|
<tt>(<i>pred</i> <i>items</i>j)</tt> is <tt>nil</tt>.
|
|
<P><pre>
|
|
(choose #Z(t nil t nil) #Z(a b c d)) => #Z(a c)
|
|
(collect-sum (choose-if #'plusp #Z(-1 2 -3 4))) => 6
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>expand <i>bools</i> <i>items</i> &optional (<i>default</i> nil)</tt><P><tt>expand</tt> is a quasi-inverse of <tt>choose</tt>. The output contains the
|
|
elements of the input series <i>items</i> spread out into the positions
|
|
specified by the non-null elements
|
|
in <i>bools</i>-that is, <i>items</i>j
|
|
is in the position occupied by the <i>j</i>th non-null element in <i>bools</i>.
|
|
The other positions in the output are occupied by <i>default</i>. The
|
|
output stops as soon as <i>bools</i> runs out of elements or a non-null
|
|
element in <i>bools</i> is encountered for which there is no corresponding
|
|
element in <i>items</i>.
|
|
<P><pre>
|
|
(expand #Z(nil t nil t t) #Z(a b c)) => #Z(nil a nil b c)
|
|
(expand #Z(nil t nil t t) #Z(a)) => #Z(nil a nil)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>split <i>items</i> &rest <i>test-series-inputs</i> <BR></tt><tt>split-if <i>items</i> &rest <i>test-predicates</i></tt><P>These functions are like <tt>choose</tt> and <tt>choose-if</tt> except that
|
|
instead of producing one restricted output, they partition the input series
|
|
<i>items</i> between several outputs. If there are <b><i>n</i></b> test inputs
|
|
following <i>items</i>, then there are <i>n</i>+1 outputs. Each input element is
|
|
placed in exactly one output series, depending on the outcome of a sequence
|
|
of tests. If the element <i>items</i>j fails the first <b><i>k</i>-1</b> tests and
|
|
passes the <i>k</i>h test, it is put in the <i>k</i>th output.
|
|
If <i>items</i>j
|
|
fails every test, it is placed in the last output. In addition, all output
|
|
stops as soon as any series input runs out of elements. The test inputs to
|
|
<tt>split</tt> are series of values; <i>items</i>j passes the <i>k</i>th test
|
|
if the <i>j</i>th element of the <i>k</i>th test series is not <tt>nil</tt>. The test
|
|
inputs to <tt>split-if</tt> are predicates; <i>items</i>j passes the <i>k</i>th
|
|
test if the <i>k</i>th test predicate returns non-null when applied to <i>items</i>j.
|
|
<P><pre>
|
|
(split #Z(-1 2 3 -4) #Z(t nil nil t))
|
|
=> #Z(-1 -4) and #Z(2 3)
|
|
(multiple-value-bind (+x -x) (split-if #Z(-1 2 3 -4) #'plusp)
|
|
(values (collect-sum +x) (collect-sum -x)))
|
|
=> 5 and -5
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>catenate &rest <i>series-inputs</i></tt><P><tt>catenate</tt> combines two or more series into one long series by appending
|
|
them end to end. The length of the output is the sum of the lengths of the
|
|
inputs.
|
|
<P><pre>
|
|
(catenate #Z(b c) #Z() #Z(d)) => #Z(b c d)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>subseries <i>items</i> <i>start</i> &optional <i>below</i></tt><P><tt>subseries</tt> returns a series containing the elements of the input
|
|
series <i>items</i> indexed by the non-negative integers from <i>start</i> up
|
|
to, but not including, <i>below</i>. If <i>below</i> is omitted or greater
|
|
than the length of <i>items</i>, the output goes all the way to the end
|
|
of <i>items</i>.
|
|
<P><pre>
|
|
(subseries #Z(a b c d) 1) => #Z(b c d)
|
|
(subseries #Z(a b c d) 1 3) => #Z(b c)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>positions <i>bools</i></tt><P><tt>positions</tt> returns a series of the indices of the non-null elements in
|
|
the series input <i>bools</i>.
|
|
<P><pre>
|
|
(positions #Z(t nil t 44)) => #Z(0 2 3)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>mask <i>monotonic-indices</i></tt><P><tt>mask</tt> is a quasi-inverse of <tt>positions</tt>. The series input <i>monotonic-indices</i> must be a strictly increasing series of non-negative
|
|
integers. The output, which is always unbounded, contains <tt>t</tt> in the
|
|
positions specified by <i>monotonic-indices</i> and <tt>nil</tt> everywhere else.
|
|
<P><pre>
|
|
(mask #Z(0 2 3)) => #Z(t nil t t nil nil ...)
|
|
(mask #Z()) => #Z(nil nil ...)
|
|
(mask (positions #Z(nil a nil b nil)))
|
|
=> #Z(nil t nil t nil ...)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>mingle <i>items1</i> <i>items2</i> <i>comparator</i></tt><P>The series returned by <tt>mingle</tt> contains all and only the elements of
|
|
the two input series. The length of the output is the sum of the lengths
|
|
of the inputs and is unbounded if either input is unbounded. The order of
|
|
the elements remains unchanged; however, the elements from the two inputs
|
|
are stably intermixed under the control of the <i>comparator</i>.
|
|
<P>
|
|
The <i>comparator</i> must accept two arguments and return non-null if and only
|
|
if its first argument is strictly less than its second argument (in some
|
|
appropriate sense). At each step, the <i>comparator</i> is used to compare
|
|
the current elements in the two series. If the current element from <i>items2</i> is strictly less than the current element from <i>items1</i>, the
|
|
current element is removed from <i>items2</i> and transferred to the output.
|
|
Otherwise, the next output element comes from <i>items1</i>.
|
|
<P><pre>
|
|
(mingle #Z(1 3 7 9) #Z(4 5 8) #'<) => #Z(1 3 4 5 7 8 9)
|
|
(mingle #Z(1 7 3 9) #Z(4 5 8) #'<) => #Z(1 4 5 7 3 8 9)
|
|
</pre><P>
|
|
<P>
|
|
<BR><b>[Function]</b><BR>
|
|
<tt>chunk <i>m</i> <i>n</i> <i>items</i></tt><P>This function has the effect of breaking up the input series <i>items</i> into
|
|
(possibly overlapping) chunks of length <i>m</i>. The starting positions of successive chunks differ
|
|
by <i>n</i>. The inputs <i>m</i> and <i>n</i> must both be positive integers.
|
|
<P>
|
|
<tt>chunk</tt> produces <i>m</i> output series. The <i>i</i>th chunk provides
|
|
the <i>i</i>th element for
|
|
each of the <i>m</i> outputs. Suppose that the length of <i>items</i> is <i>l</i>.
|
|
The length of
|
|
each output is <IMG ALIGN=BOTTOM ALT="" SRC="_24769_tex2html_wrap44201.gif">.
|
|
The <i>i</i>th element of the <i>k</i>th output is the
|
|
(i*n+k)th element of <i>items</i> (<i>i</i> and <b><i>k</i></b> counting from zero).
|
|
<P>
|
|
Note that if <b><i>l</i><<i>m</i></b>, there will be no
|
|
output elements, and if <b><i>l</i>-<i>m</i></b> is not a multiple of <i>n</i>,
|
|
the last few input elements will
|
|
not appear in the output. If <b><i>m</i><i>n</i></b>,
|
|
one can guarantee that the last chunk will contain the last
|
|
element of <i>items</i> by catenating <br><i>n</i>-1
|
|
copies of an appropriate padding value to the end of <i>items</i>.
|
|
<P>
|
|
The first example below shows <tt>chunk</tt>
|
|
being used to compute a moving average. The second example shows
|
|
<tt>chunk</tt> being used to convert a property list into an association list.
|
|
<P><pre>
|
|
(mapping (((xi xi+1 xi+2) (chunk 3 1 #Z(1 5 3 4 5 6))))
|
|
(/ (+ xi xi+1 xi+2) 3))
|
|
=> #Z(3 4 4 5)
|
|
|
|
(collect
|
|
(mapping (((prop val) (chunk 2 2 (scan '(a 2 b 5 c 8)))))
|
|
(cons prop val)))
|
|
=> ((a . 2) (b . 5) (c . 8))
|
|
</pre><P>
|
|
<br><img align=bottom alt="change_end" src="gif/change_end.gif">
|
|
<P>
|
|
<BR> <HR><A NAME=tex2html6116 HREF="node354.html"><IMG ALIGN=BOTTOM ALT="next" SRC="icons/next_motif.gif"></A> <A NAME=tex2html6114 HREF="node349.html"><IMG ALIGN=BOTTOM ALT="up" SRC="icons/up_motif.gif"></A> <A NAME=tex2html6108 HREF="node352.html"><IMG ALIGN=BOTTOM ALT="previous" SRC="icons/previous_motif.gif"></A> <A NAME=tex2html6118 HREF="node1.html"><IMG ALIGN=BOTTOM ALT="contents" SRC="icons/contents_motif.gif"></A> <A NAME=tex2html6119 HREF="index.html"><IMG ALIGN=BOTTOM ALT="index" SRC="icons/index_motif.gif"></A> <BR>
|
|
<B> Next:</B> <A NAME=tex2html6117 HREF="node354.html"> Collectors</A>
|
|
<B>Up:</B> <A NAME=tex2html6115 HREF="node349.html"> Series Functions</A>
|
|
<B> Previous:</B> <A NAME=tex2html6109 HREF="node352.html"> Truncation and Other </A>
|
|
<HR> <P>
|
|
<HR>
|
|
<P><ADDRESS>
|
|
AI.Repository@cs.cmu.edu
|
|
</ADDRESS>
|
|
</BODY>
|