1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
|
\par
\section{Data Structure}
\label{section:InpMtx:dataStructure}
\par
\par
The {\tt InpMtx} structure has the following fields.
\begin{itemize}
\item
{\tt int coordType} : coordinate type. The following types are
supported.
\begin{itemize}
\item {\tt INPMTX\_BY\_ROWS} --- row triples,
the coordinates for $a_{i,j}$ is $(i,j)$.
\item {\tt INPMTX\_BY\_COLUMNS} --- column triples,
the coordinates for $a_{i,j}$ is $(j,i)$.
\item {\tt INPMTX\_BY\_CHEVRONS} --- chevron triples,
the coordinates for $a_{i,j}$ is $(\min(i,j), j-i)$.
(Chevron $j$ contains $a_{j,j}$, $a_{j,k} \ne 0$ and $a_{k,j} \ne 0$
for $k > j$.)
\item {\tt INPMTX\_CUSTOM} --- custom coordinates.
\end{itemize}
\item
{\tt int storageMode} : mode of storage
\begin{itemize}
\item {\tt INPMTX\_RAW\_DATA} ---
data is raw pairs or triples, two coordinates and (optionally)
one or two double precision values.
\item {\tt INPMTX\_SORTED} --- data is sorted and distinct triples,
the primary key is the first coordinate,
the secondary key is the second coordinate.
\item {\tt INPMTX\_BY\_VECTORS} --- data is sorted and distinct vectors.
All entries in a vector share something in common.
For example, when {\tt coordType} is {\tt INPMTX\_BY\_ROWS},
{\tt INPMTX\_BY\_COLUMNS} or {\tt INPMTX\_BY\_CHEVRONS},
row vectors, column vectors, or chevron vectors are stored,
respectively.
When {\tt coordType} is {\tt INPMTX\_CUSTOM},
a custom type, entries in the same
vector have something in common but it need not be a common row,
column or chevron coordinate.
\end{itemize}
\item
{\tt int inputMode} : mode of data input
\begin{itemize}
\item {\tt INPMTX\_INDICES\_ONLY}
--- only indices are stored, not entries.
\item {\tt SPOOLES\_REAL}
--- indices and real entries are stored.
\item {\tt SPOOLES\_COMPLEX}
--- indices and complex entries are stored.
\end{itemize}
\item
{\tt int maxnent} -- present maximum number of entries in the object.
This quantity is initialized by the {\tt InpMtx\_init()} method,
but will be changed as the object resizes itself as necessary.
\item
{\tt int nent} -- present number of entries in the object.
This quantity changes as data is input or when the raw triples are
sorted and compressed.
\item
{\tt double resizeMultiple} -- governs how the workspace grows as
necessary. The default value is 1.25.
\item
{\tt IV ivec1IV} -- an {\tt IV} vector object
of size {\tt mxnent} that holds first coordinates.
\item
{\tt IV ivec2IV} -- an {\tt IV} vector object
of size {\tt mxnent} that holds second coordinates.
\item
{\tt DV dvecDV} -- a {\tt DV} vector object of size {\tt mxnent}
that holds double precision entries. Used only when {\tt inputMode}
is {\tt SPOOLES\_REAL} or {\tt SPOOLES\_COMPLEX}.
\item
{\tt int maxnvector} -- present maximum number of vectors.
This quantity is initialized by the {\tt InpMtx\_init()} method,
but will be changed as the object resizes itself as necessary.
Used only when {\tt storageMode} is {\tt INPMTX\_BY\_VECTORS}.
\item
{\tt int nvector} -- present number of vectors. Used only when {\tt
storageMode} is {\tt INPMTX\_BY\_VECTORS}.
\item
{\tt IV vecidsIV} -- an {\tt IV} vector object of size {\tt nvector}
to hold the id of each vector.
Used only when {\tt storageMode} is {\tt INPMTX\_BY\_VECTORS}.
\item
{\tt IV sizesIV} -- an {\tt IV} vector object of size {\tt nvector}
to hold the size of each vector.
Used only when {\tt storageMode} is {\tt INPMTX\_BY\_VECTORS}.
\item
{\tt IV offsetsIV} -- an {\tt IV} vector object of size {\tt
nvector} to hold the offset of each vector into the {\tt ivec1IV},
{\tt ivec2IV} and {\tt dvecDV} vector objects.
Used only when {\tt storageMode} is {\tt INPMTX\_BY\_VECTORS}.
\end{itemize}
\par
One can query the attributes of the object with the following
macros.
\begin{itemize}
\item
{\tt INPMTX\_IS\_BY\_ROWS(mtx)} returns {\tt 1} if the entries are
stored by rows, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_BY\_COLUMNS(mtx)} returns {\tt 1} if the entries are
stored by columns, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_BY\_CHEVRONS(mtx)} returns {\tt 1} if the entries are
stored by chevrons, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_BY\_CUSTOM(mtx)} returns {\tt 1} if the entries are
stored by some custom coordinate, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_RAW\_DATA(mtx)} returns {\tt 1} if the entries are
stored as unsorted pairs or triples, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_SORTED(mtx)} returns {\tt 1} if the entries are
stored as sorted pairs or triples, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_BY\_VECTORS(mtx)} returns {\tt 1} if the entries are
stored as vectors, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_INDICES\_ONLY(mtx)} returns {\tt 1} if the entries are
not stored, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_REAL\_ENTRIES(mtx)} returns {\tt 1}
if the entries are real, and {\tt 0} otherwise.
\item
{\tt INPMTX\_IS\_COMPLEX\_ENTRIES(mtx)} returns {\tt 1}
if the entries are complex, and {\tt 0} otherwise.
\end{itemize}
|
