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
|
=pod
=head1 NAME
B<XPATemplate: Access Point Names and Templates>
=head1 SYNOPSIS
XPA access points are composed of two parts: a general class and a
specific name. Both parts accept template characters so that you
can send/retrieve data to/from multiple servers at one time.
=head1 DESCRIPTION
When XPA servers call
XPANew(),
or
XPACmdNew()
to define XPA access points, they specify a string identifier composed of a
class and a name. When clients communicate with XPA access points,
they specify which access points to communicate with using
an identifier of the form:
class:name
All registered XPA access points that match the specified identifier
will be available for communication (subject to access control rules,
etc.)
As of XPA 2.1.5, the length of both the class and name designations are
limited to 1024 characters.
The XPA class:name identifier actually is a template: it accepts wild
cards in its syntax, so a single specifier can match more than one XPA
access point. (Note that the class is optional and defaults to "*".)
The allowed syntax for clients to specify the class:name template is
of the form shown below. (Note that "*" is used to denote a generic
wild card, but other wild cards characters are supported, as described
below).
template explanation
-------- -----------
class:name exact match of class and name
name match any class with this name
*:name match any class with this name
class:* match any name of this class
*:* match any access point
In general, the following wild-cards can be applied to class and name:
wildcard explanation
-------- -----------
? match any character, but there must be one
* match anything, or nothing
[...] match an inclusive set
Although the class:name template normally is used to refer to XPA
access points, these also can be specified using their individual
socket identifiers. For inet sockets, the socket identifier is
B<ip:port>, where ip can be the DNS-registered name,
the ASCII IP number (e.g. 123.45.67.890) or the hex IP number
(e.g. 838f3a60). For unix sockets, the identifier is the socket file
name. These socket identifiers are displayed as the fourth argument
in the xpans display of registered access points. For example,
consider the ds9 program started using inet sockets. The xpans name
server will register something like this:
csh> xpaget xpans
DS9 ds9 gs saord.harvard.edu:3236 eric
You can access ds9 using ip:3236 in any of the three forms:
csh> xpaget saord:3236 file
/home/eric/data/snr.ev
csh> xpaget 123.45.67.890:3236 file
/home/eric/data/snr.ev
csh> xpaget 838f3a60:3236 file
/home/eric/data/snr.ev
In the case of unix sockets, the socket identifier is a file:
csh> xpaget xpans
DS9 ds9 gs /tmp/.xpa/DS9_ds9.2631 eric
csh> xpaget /tmp/.xpa/DS9_ds9.2631 file
/home/eric/data/snr.ev
This feature can be useful in distinguishing between multiple
instances of a program that all have the same class:name designation.
=head1 SEE ALSO
See xpa(n) for a list of XPA help pages
=cut
|
