See Section <Ref Sect="cvecs"/> for a documentation of the data format
of <C>cvec</C>s and its restrictions.<P/>
<Section>
<Heading>Creation</Heading>
Note that many functions described later in this chapter implicitly
create new <C>cvec</C>s, such that it is usually only in the beginning
of a calculation necessary to create <C>cvec</C>s explicitly.
<ManSection>
<Oper Name="CVec" Arg="arg"/>
<Returns>a new mutable <C>cvec</C></Returns>
<Description>
Creates a new <C>cvec</C>. See the method descriptions for details.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVec" Arg="cvecclass" Label="cvecclass"/>
<Returns>a new mutable <C>cvec</C></Returns>
<Description>
<A>cvecclass</A> must be a <C>cvecclass</C>. Creates a new
<C>cvec</C> containing only zeroes. For an explanation of the term
<C>cvecclass</C> see Section <Ref Sect="cvecs"/> and
<Ref Oper="CVecClass"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVec" Arg="coeffs, cvecclass" Label="coefflcvecclass"/>
<Returns>a new mutable <C>cvec</C></Returns>
<Description>
This method takes a coefficient list and a <C>cvecclass</C>
as arguments. Assume the vector will be over <M>GF(p,d)</M> with
<M>q=p^d</M>. For the coefficient list <A>coeffs</A> there exist
several possibilities, partially depending on the base field size.
The easiest way is to use &GAP; finite
field elements, which will be put into the new vector in the same
order. If <M>d=1</M>, one can always use &GAP; immediate integers
between <M>0</M> and <M>p-1</M>, the vector will then contain the
corresponding cosets in <M>GF(p)=Z/pZ</M>. If <M>q</M> is small enough
to be a &GAP; immediate integer, then one can use &GAP; immediate
integers that are equal to the <M>p</M>-adic expansion using the
coefficients of the representing polynomial as coefficients. That is,
if an element in <M>GF(p,d)</M> is represented by the polynomial
<M>\sum_{{i=0}}^{{d-1}} a_i x^i</M> with <M>a_i \in \{0,\ldots,p-1\}</M>,
then one has to put <M>\sum_{{i=0}}^{{d-1}} a_i p^i</M> as integer into
the coefficient list <A>coeffs</A>. If <M>q</M> is larger, then
<A>coeffs</A> must be a list of lists of length <M>d</M>
and contains for each field element of <M>GF(p,d)</M>
in the vector a list of its <M>d</M> coefficients of the representing
polynomial.
For an explanation of the term <C>cvecclass</C> see Section <Ref
Sect="cvecs"/> and <Ref Oper="CVecClass"/>. Of course, the length of
the list <A>coeffs</A> must match the length of the vector given in the
<C>cvecclass</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVec" Arg="coeffs, p, d" Label="coefflpd"/>
<Returns>a new mutable <C>cvec</C></Returns>
<Description>
This method takes a coefficient list and two positive integers
<A>p</A> and <A>d</A> as arguments. A new <C>cvec</C> over <M>GF(p,d)</M>
will be created. Let <M>q := p^d</M>.<P/>
For a description of the possible values of the coefficient list
<A>coeffs</A> see the description of the method <Ref Meth="CVec"
Label="coefflcvecclass"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVec" Arg="v" Label="compvec"/>
<Returns>a new <C>cvec</C></Returns>
<Description>
<A>v</A> must be a &GAP; compressed vector either over <M>GF(2)</M> or
<M>GF(p,d)</M> with <M>p^d \leq 256</M>. Creates a new
<C>cvec</C> containing the same numbers as <A>v</A> over the field
which the <C>Field</C> operation returns for <A>v</A>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVec" Arg="coeffs, f" Label="coefflff"/>
<Returns>a new mutable <C>cvec</C></Returns>
<Description>
This method takes a coefficient list and a finite field
<A>f</A> as arguments. A new <C>cvec</C> over <A>f</A>
will be created. Let <M>q := </M><C>Size(</C><A>f</A><C>)</C>.<P/>
For a description of the possible values of the coefficient list
<A>coeffs</A> see the description of the method <Ref Meth="CVec"
Label="coefflcvecclass"/>.
</Description>
</ManSection>
After having encountered the concept of a <C>cvecclass</C> already
a few times it is time to learn how to create one. The following
operation is used first to create new <C>cvecclass</C>es and second
to ask a <C>cvec</C> for its class. In addition, it is used for
<C>csca</C>s.
<ManSection>
<Oper Name="CVecClass" Arg="arg"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
Creates new <C>cvecclass</C>es and asks <C>cvec</C>s
for their class. See the following method descriptions for details.
For an explanation of the term <C>cvecclass</C> see Section <Ref
Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVecClass" Arg="p, d, l" Label="for three integers"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
All three arguments must be integers. The arguments <A>p</A> and <A>d</A>
must be positive and describe the base field <M>GF(p,d)</M>. The third
argument must be non-negative and the method returns
the <C>cvecclass</C> of vectors over <M>GF(p,d)</M> of length <A>l</A>.
<P/>
For an explanation of the term <C>cvecclass</C> and its data structure
see Section <Ref Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVecClass" Arg="v" Label="for a cvec"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
The argument <A>v</A> must be a <C>cvec</C>.
The method returns the corresponding <C>cvecclass</C>.
For an explanation of the term <C>cvecclass</C> and its data structure
see Section <Ref Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVecClass" Arg="v, l" Label="for a cvec and an integer"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
The first argument <A>v</A> must be a <C>cvec</C>.
The method returns the corresponding <C>cvecclass</C> for vectors
over the same field as <A>v</A> but with length <A>l</A>. This is
much faster than producing the same object by giving <M>p</M> and
<M>d</M>.
For an explanation of the term <C>cvecclass</C> and its data structure
see Section <Ref Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVecClass" Arg="c, l" Label="for a cvecclass and an integer"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
The argument <A>c</A> must be a <C>cvecclass</C>.
The method returns the corresponding <C>cvecclass</C> for vectors
over the same field as those in <A>c</A> but with length <A>l</A>.
This is much faster than producing the same object by giving <M>p</M> and
<M>d</M>.
For an explanation of the term <C>cvecclass</C> and its data structure
see Section <Ref Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVecClass" Arg="f, l" Label="for a finite field and an integer"/>
<Returns>a <C>cvecclass</C></Returns>
<Description>
The argument <A>f</A> must be a <C>finite field</C>.
The method returns the corresponding <C>cvecclass</C> for vectors
over the field <A>f</A> with length <A>l</A>.
For an explanation of the term <C>cvecclass</C> and its data structure
see Section <Ref Sect="cvecs"/>.
</Description>
</ManSection>
<ManSection>
<Meth Name="ZeroSameMutability" Arg="v" Label="cvec"/>
<Returns>the zero <C>cvec</C> in the same <C>cvecclass</C> as <A>v</A>
</Returns>
<Description>
<A>v</A> must be a <C>cvec</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="ShallowCopy" Arg="v" Label="cvec"/>
<Returns>a mutable copy of <A>v</A>
</Returns>
<Description>
<A>v</A> must be a <C>cvec</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="Randomize" Arg="v" Label="cvec"/>
<Meth Name="Randomize" Arg="v, rs" Label="cvecrandomsource"/>
<Returns>the vector <A>v</A></Returns>
<Description>
<A>v</A> must be a <C>cvec</C> and <A>rs</A> must be a random source
object if given. This method changes the vector <A>v</A> in place by
(pseudo) random values in the field over which the vector lives. If
a random source is given, the pseudo random numbers used are taken from
this source, otherwise the global random source in the &GAP; library
is taken.
</Description>
</ManSection>
</Section>
<Section>
<Heading>Arithmetic</Heading>
Of course, the standard arithmetic infix operations <M>+</M>, <M>-</M> and
<M>*</M> (for vectors and scalars) work as expected by using the
methods below. We start this section with a subsection on the usage of
scalars in arithmetic operations involving vectors.
<Subsection Label="scahandl">
<Heading>Handling of scalars in arithmetic operations</Heading>
In all places (like in <C>\*</C>) where vectors and scalars occur, the
following conventions apply to scalars:<P/>
&GAP; finite field elements can be used as scalars.<P/>
Integers between <M>0</M> and <M>p-1</M> (inclusively) can always be used as
scalars representing prime field elements via the isomorphism
<M>GF(p)=Z/pZ</M>, also for extension fields. Larger integers than
<M>p-1</M>, as long as they are &GAP; immediate integers, are interpreted
as the <M>p</M>-adic expansion of the coefficient list of the polynomial
representing the scalar. Note that this usage of immediate integers
differs from the standard list arithmetic in &GAP; because multiplication
with the integer <M>n</M> not necessarily means the same than <M>n</M> times
addition! Larger integers than immediate integers are not supported.
</Subsection>
<ManSection>
<Meth Name="\+" Arg="v, w" Label="cveccvec"/>
<Returns>the sum <M><A>v</A>+<A>w</A></M> as a new <C>cvec</C></Returns>
<Description>
For two <C>cvec</C>s <A>v</A> and <A>w</A>. Works as expected.
</Description>
</ManSection>
<ManSection>
<Meth Name="\-" Arg="v, w" Label="cveccvec"/>
<Returns>the difference <M><A>v</A>-<A>w</A></M> as a new <C>cvec</C></Returns>
<Description>
For two <C>cvec</C>s <A>v</A> and <A>w</A>. Works as expected.
</Description>
</ManSection>
<ManSection>
<Meth Name="AdditiveInverseSameMutability" Arg="v" Label="cvec"/>
<Meth Name="\-" Arg="v" Label="cvec"/>
<Returns>the additive inverse of <A>v</A> as a new <C>cvec</C></Returns>
<Description>
For a <C>cvec</C> <A>v</A>. Works as expected.
</Description>
</ManSection>
<ManSection>
<Meth Name="AdditiveInverseMutable" Arg="v" Label="cvec"/>
<Returns>the additive inverse of <A>v</A> as a new mutable <C>cvec</C>
</Returns>
<Description>
For a <C>cvec</C> <A>v</A>. Works as expected.
</Description>
</ManSection>
<ManSection>
<Meth Name="\*" Arg="v, s" Label="cvecsca"/>
<Meth Name="\*" Arg="s, v" Label="scacvec"/>
<Returns>the scalar multiple <A>s</A><M>\cdot</M><A>v</A></Returns>
<Description>
For a <C>cvec</C> <A>v</A> and a scalar <A>s</A>. For the format
of the scalar see <Ref Subsect="scahandl"/>. Works as expected.
</Description>
</ManSection>
<ManSection>
<Meth Name="AddRowVector" Arg="v, w, s" Label="cveccvecsca"/>
<Meth Name="AddRowVector" Arg="v, w, s, fr, to" Label="cveccvecscapospos"/>
<Returns>nothing</Returns>
<Description>
<A>v</A> and <A>w</A> must be <C>cvec</C>s over the same field with
equal length, <A>s</A> a scalar (see Subsection <Ref Subsect="scahandl"/>)
and <A>v</A> must be mutable.
Adds <A>s</A><M>\cdot</M><A>w</A> to <A>v</A> modifying <A>v</A>. If
<A>fr</A> and <A>to</A> are given, they give a hint, that <A>w</A> is
zero outside positions <A>fr</A> to <A>to</A> (inclusively). The method
can, if it chooses so, save time by not computing outside that range.
In fact, the implemented method restricts the operation to the
<C>Word</C>s involved.<P/>
If either <A>fr</A> or <A>to</A> is <M>0</M> it defaults to <M>1</M> and
<C>Length(v)</C> respectively.
</Description>
</ManSection>
<ManSection>
<Meth Name="MultVector" Arg="v, s" Label="cvecsca"/>
<Meth Name="MultVector" Arg="v, s, fr, to" Label="cvecscapospos"/>
<Returns>nothing</Returns>
<Description>
<A>v</A> must be a mutable <C>cvec</C> and <A>s</A> a scalar (see
Subsection <Ref Subsect="scahandl"/>).
Multiplies <A>v</A> by <A>s</A> modifying <A>v</A>. If
<A>fr</A> and <A>to</A> are given, they give a hint, that <A>v</A> is
zero outside positions <A>fr</A> to <A>to</A> (inclusively). The method
can, if it chooses so, save time by not computing outside that range.
In fact, the implemented method restricts the operation to the
<C>Word</C>s involved.<P/>
If either <A>fr</A> or <A>to</A> is <M>0</M> it defaults to <M>1</M> and
<C>Length(v)</C> respectively.
</Description>
</ManSection>
<ManSection>
<Meth Name="ScalarProduct" Arg="v, w" Label="cveccvec"/>
<Returns>the scalar product</Returns>
<Description>
Both <A>v</A> and <A>w</A> must be <C>cvec</C>s over the same field
with equal length. The function returns the scalar product of the
two vectors. Note that there is a very efficient method for prime
fields with <M>p < 65536</M>. In the other cases the method taken
is not extremely fast.
</Description>
</ManSection>
<ManSection>
<Meth Name="ZeroMutable" Arg="v" Label="cvec"/>
<Returns>a mutable copy of the zero <C>cvec</C> in the same <C>cvecclass</C>
as <A>v</A>
</Returns>
<Description>
<A>v</A> must be a <C>cvec</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="ZeroVector" Arg="l, v" Label="cvec"/>
<Returns>a mutable copy of the zero <C>cvec</C> over the same
field than <A>v</A> but with length <A>l</A></Returns>
<Description>
<A>v</A> must be a <C>cvec</C> and <A>l</A> a &GAP; integer.
</Description>
</ManSection>
</Section>
<Section>
<Heading>Element access and slicing</Heading>
<C>cvec</C>s behave to some extend like lists with respect to element access
and slicing. However, they allow only actions that can be implemented
efficiently and that do not change their length. In addition there is a
highly optimised function to copy contiguous sections of <C>cvec</C>s
into another <C>cvec</C>.
<ManSection>
<Meth Name="ELM_LIST" Arg="v, pos" Label="for vectors"/>
<Returns>a finite field element</Returns>
<Description>
This is a method for (reading) element access of vectors. <A>v</A> must
be a <A>cvec</A> and <A>pos</A> must be a positive integer not greater
than the length of <A>v</A>. The finite field element at position
<A>pos</A> in <A>v</A> is returned.<P/>
Note that the list access syntax <Q><A>v</A>[<A>pos</A>]</Q> triggers a call
to the <Ref Meth="ELM_LIST" Label="for vectors"/> operation.
</Description>
</ManSection>
<ManSection>
<Meth Name="ASS_LIST" Arg="v, pos, s" Label="for vectors"/>
<Returns>nothing</Returns>
<Description>
This is a method for (writing) element access of vectors. <A>v</A> must
be a <A>cvec</A> and <A>pos</A> must be a positive integer not greater
than the length of <A>v</A>. <A>s</A> must be a finite field element
or an integer. The finite field element at position
<A>pos</A> in <A>v</A> is set to <A>s</A>.<P/>
The scalar <A>s</A> is treated exactly as described in Subsection
<Ref Subsect="scahandl"/>.<P/>
Note that the list access syntax
<Q><A>v</A><C>[</C><A>pos</A><C>] := </C><A>s</A></Q>
triggers a call to the <Ref Meth="ASS_LIST" Label="for vectors"/> operation.
</Description>
</ManSection>
<ManSection>
<Meth Name="ELMS_LIST" Arg="v, l" Label="for vectors"/>
<Returns>a <C>cvec</C></Returns>
<Description>
This is a method for (reading) slice access of vectors. <A>v</A> must
be a <A>cvec</A> and <A>l</A> must be a range object or a list of positive
integers not greater than the length of <A>v</A>. In both cases the list
of numbers must be contiguous list of valid positions in the vector.
A new <C>cvec</C> over the same field as <A>v</A> and with the same
length as <A>l</A> is created and returned. The finite field element at i
positions <A>l</A> in <A>v</A> are copied into the new vector.<P/>
Note that the list slice access syntax <Q><A>v</A>{<A>l</A>}</Q> triggers a
call to the <Ref Meth="ELMS_LIST" Label="for vectors"/> operation.
</Description>
</ManSection>
Note that there intentionally is no write slice access to <C>cvec</C>s,
because in most cases this would lead to code that unnecessarily
copies data around in an expensive way. Please use the following function
instead:
<ManSection>
<Func Name="CVEC_Slice" Arg="src, dst, srcpos, len, dstpos"/>
<Returns>nothing</Returns>
<Description>
<A>src</A> and <A>dst</A> must be <C>cvec</C>s over the same field.
The field elements from positions <A>srcpos</A> to
<M><A>srcpos</A>+<A>len</A>-1</M> in <A>src</A> are copied to positions
from <A>dstpos</A> in <A>dst</A>. Of course all positions must be within
the vectors. <P/>
Note that this functions is quite efficient, however, the ranges are
checked. If you want to avoid this, use <C>CVEC&uscore;SLICE</C> instead (with
same calling convention), but do not complain later if crashes occur in
case of illegal positions used.
</Description>
</ManSection>
<ManSection>
<Meth Name="CopySubVector" Arg="src, dst, srcposs, dstposs"/>
<Returns>nothing</Returns>
<Description>
Implements the operation <Ref Oper="CopySubVector" BookName="Ref"/>
for <C>cvec</C>s <A>src</A> and <A>dst</A>, that is, <A>srcposs</A>
and <A>dstposs</A> must be ranges or plain lists of integers of equal
length such that all numbers contained lie between
<M>1</M> and the length of <A>src</A> and <A>dst</A> respectively.
The result is undefined if <A>src</A> and <A>dst</A> are the same
objects. The method is particularly efficient if both <A>srcposs</A>
and <A>dstposs</A> are ranges with increment <M>1</M>.
</Description>
</ManSection>
<ManSection>
<Meth Name="CVEC_Concatenation" Arg="arg"/>
<Returns>a new <C>cvec</C> by concatenating all arguments</Returns>
<Description>
This function provides concatenation of <C>cvec</C>s over the same
base field. The result is a new <C>cvec</C>. A variable number of
<C>cvec</C>s over the same field can be given.
</Description>
</ManSection>
</Section>
<Section>
<Heading>Comparison of Vectors and other information</Heading>
<ManSection>
<Meth Name="=" Arg="v, w" Label="for vectors"/>
<Returns><C>true</C> or <C>false</C></Returns>
<Description>
Returns <C>true</C> if the <C>cvec</C>s <A>v</A> and <A>w</A> are
equal. The vectors must be over the same field and must have equal
length.
</Description>
</ManSection>
<ManSection>
<Meth Name="LT" Arg="v, w" Label="for vectors"/>
<Returns><C>true</C> or <C>false</C></Returns>
<Description>
Returns <C>true</C> if the <C>cvec</C> <A>v</A> is smaller than <A>w</A>.
The vectors must be over the same field and must have equal
length. The order implemented is very efficient but is <E>not</E>
compatible with lexicographic ordering of lists of
finite field elements! This method should therefore only be used
for binary search purposes.
Note that the operation <C>LT</C> is the same as <C>&bslash;<</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="IsZero" Arg="v" Label="for a vector"/>
<Returns><C>true</C> or <C>false</C></Returns>
<Description>
Returns <C>true</C> if the <C>cvec</C> <A>v</A> is equal to zero, meaning
that all entries are equal to zero.
</Description>
</ManSection>
<ManSection>
<Meth Name="PositionNonZero" Arg="v"/>
<Returns>a positive integer</Returns>
<Description>
Returns the index of the first entry in the <C>cvec</C> <A>v</A> that is
not equal to zero. If the vector is equal to zero, then its <C>Length</C>
plus one is returned.
</Description>
</ManSection>
<ManSection>
<Meth Name="PositionLastNonZero" Arg="v"/>
<Returns>a non-negative integer</Returns>
<Description>
Returns the index of the last entry in the <C>cvec</C> <A>v</A> that is
not equal to zero. If the vector is equal to zero, then <M>0</M>
is returned.
</Description>
</ManSection>
<ManSection>
<Meth Name="Length" Arg="v"/>
<Returns>a non-negative integer</Returns>
<Description>
Returns the length of the <C>cvec</C> <A>v</A>.
</Description>
</ManSection>
<ManSection>
<Meth Name="Unpack" Arg="v" Label="cvec"/>
<Returns>a list of finite field elements</Returns>
<Description>
This operation unpacks a <C>cvec</C> <A>v</A>. A new plain list containing
the corresponding numbers as &GAP; finite field elements. Note that
the vector <A>v</A> remains unchanged.
</Description>
</ManSection>
<ManSection>
<Meth Name="IntegerRep" Arg="v" Label="cvec"/>
<Returns>a list of integers or of lists of integers</Returns>
<Description>
This operation unpacks a <C>cvec</C> <A>v</A> into its integer
representation. A new plain list containing
the corresponding numbers of the vector is returned. The integer
representation of a finite field element is either one integer
(containing the p-adic expansion of the polynomial representative of
the residue class modulo the Conway polynomial, if the field has
less or equal to <M>65536</M> elements. For larger finite fields each
field element is represented as a list of <M>d</M> integers between
<M>0</M> and <M>p-1</M>, where <M>d</M> is the degree of the finite
field over its prime field. Note that
the vector <A>v</A> remains unchanged.
</Description>
</ManSection>
<ManSection>
<Meth Name="NumberFFVector" Arg="v, sz" Label="cvec"/>
<Returns>an integer</Returns>
<Description>
This implements the library operation <Ref Oper="NumberFFVector"
BookName="Ref"/>. Note that only the case that <A>sz</A> is the
number of elements in the base field of <A>v</A> is implemented.
There is an inverse operation called <Ref Oper="CVecNumber" Label="for an integer and a cvecclass"/>.
</Description>
</ManSection>
<ManSection>
<Oper Name="CVecNumber" Arg="nr, cl" Label="for an integer and a cvecclass"/>
<Oper Name="CVecNumber" Arg="nr, p, d, l" Label="for four integers"/>
<Returns>a <C>cvec</C></Returns>
<Description>
This is the inverse operation to <Ref Oper="NumberFFVector"
BookName="Ref"/>. Of course, the base field of the vector and its
length has to be specified, either by giving a <C>cvecclass</C> <A>cl</A>
or the parameters <A>p</A>, <A>d</A>, and <A>l</A>. For both cases
corresponding methods are available.
</Description>
</ManSection>
</Section>
<Section>
<Heading>Access to the base field</Heading>
<ManSection>
<Meth Name="BaseDomain" Arg="v" Label="cvec"/>
<Returns>the base field of <A>v</A>
</Returns>
<Description>
For a <C>cvec</C> <A>v</A> this returns
the &GAP; object <C>GF(p,d)</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="BaseField" Arg="v" Label="cvec"/>
<Returns>the base field of <A>v</A>
</Returns>
<Description>
For a <C>cvec</C> <A>v</A> this returns
the &GAP; object <C>GF(p,d)</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="Characteristic" Arg="v" Label="cvec"/>
<Returns>the characteristic of the base field of <A>v</A>
</Returns>
<Description>
Returns the characteristic of the base field of <A>v</A> (see
<Ref Meth="BaseField" Label="cvec"/>).
</Description>
</ManSection>
<ManSection>
<Meth Name="DegreeFFE" Arg="v" Label="cvec"/>
<Returns>the degree of the base field of <A>v</A> over its prime field
</Returns>
<Description>
Returns the degree of the base field of <A>v</A> over its prime field (see
<Ref Meth="BaseField" Label="cvec"/>).
</Description>
</ManSection>
</Section>
<Section>
<Heading>Hashing techniques for <C>cvec</C>s</Heading>
<ManSection>
<Func Name="CVEC_HashFunctionForCVecs" Arg="v, data"/>
<Returns>an integer hash value</Returns>
<Description>
This is a hash function usable for the <Ref Oper="ChooseHashFunction"
BookName="orb"/> framework. It takes as arguments a <C>cvec</C>
<A>v</A> and a list <A>data</A> of length <M>2</M>. The first entry of
<A>data</A> is the length of the hash table used and the second entry
is the number of bytes looked at in the <C>cvec</C>.
</Description>
</ManSection>
<ManSection>
<Meth Name="ChooseHashFunction" Arg="v, hashlen" Label="for vectors"/>
<Returns>a record describing a hash function</Returns>
<Description>
Chooses a hash function to store <C>cvec</C>s like <A>v</A> in a hash table
of length <A>hashlen</A>. The return value is a record with components
<C>func</C> and <C>data</C> bound to <Ref
Func="CVEC_HashFunctionForCVecs"/> and a list of length <M>2</M> to be
given as a second argument to <Ref Func="CVEC_HashFunctionForCVecs"/>.
This allows to use <C>cvec</C>s in the <Ref Oper="ChooseHashFunction"
BookName="orb"/> framework.
</Description>
</ManSection>
<!-- This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
-->
¤ Dauer der Verarbeitung: 0.3 Sekunden
(vorverarbeitet)
¤
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung ist noch experimentell.
