Files
pvaClient/documentation/htmldoxygen/pvaClientChannel.html
2017-07-12 06:11:29 -04:00

124 lines
3.9 KiB
HTML

<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="content-type" content="text/html; charset=iso-8859-1" />
<title>PvaClientChannel</title>
<link rel="stylesheet" type="text/css"
href="http://epics-pvdata.sourceforge.net/base.css" />
<link rel="stylesheet" type="text/css"
href="http://epics-pvdata.sourceforge.net/epicsv4.css" />
<style type="text/css">
/*<![CDATA[*/
.about { margin-left: 3em; margin-right: 3em; font-size: .83em}
table { margin-left: auto; margin-right: auto }
.diagram { text-align: center; margin: 2.5em 0 }
span.opt { color: grey }
span.nterm { font-style:italic }
span.term { font-family:courier }
span.user { font-family:courier }
span.user:before { content:"<" }
span.user:after { content:">" }
.nonnorm { font-style:italic }
p.ed { color: #AA0000 }
span.ed { color: #AA0000 }
p.ed.priv { display: inline; }
span.ed.priv { display: inline; }
/*]]>*/</style>
<!-- Script that generates the Table of Contents -->
<script type="text/javascript"
src="http://epics-pvdata.sourceforge.net/script/tocgen.js">
</script>
</head>
<body>
<h1>PvaClientChannel</h1>
<h2>Overview</h2>
<p>
pvaClientChannel is a synchronous wrapper for the pvAccess::Channel API, which is a callback based API.
Thus it is easier to use than pvAccess::Channel itself.
</p>
<p>An instance of <b>PvaClientChannel</b> connects to a single channel.
An instance can only be created via class <b>PvaClient</b> which has both synchronous methods, which block, and non blocking methods.
The synchrouous methods block until a connection is made to the channel and throw an exception if a
timeout occurs while trying to make a connection.
The non blocking methods leave connection to the caller.
</p>
<p><b>PvaClientChannel</b> has methods:</p>
<dl>
<dt>Connect to channel</dt>
<dd>These can be called indirectly by a blocking request to <b>PvaClient</b>
or by the client if a non blocking request is made to <b>PvaClient</b>.
</dd>
<dt>Channel state change requester</dt>
<dd>The client can provide a callback that is called each time the connection state
of the channel changes.
</dd>
<dt>Creating all of the following</dt>
<dd>
<pre>
PvaClientField NOT IMPLEMENTED
PvaClientProcess
PvaClientGet
PvaClientPut
PvaClientPutGet
PvaClientMonitor
PvaClientArray NOT IMPLEMENTED
PvaClientRPC
</pre>
</dd>
</dl>
<h2>Connect: Blocking vs Non-Blocking </h2>
<p><b>PvaClientChannel</b> has methods:</p>
<dl>
<dt>connect</dt>
<dd>
This calls issueConnect and then waitConnect.
If waitConnect fails an exception is thrown.
Since waitConnect is a blocking method so is this.
</dd>
<dt>issueConnect</dt>
<dd>
This is a request to connect to the channel. This is a non blocking call.
</dd>
<dt>waitConnect</dt>
<dd>
This waits for the server to respond to issueConnect.
It blocks until the server responds or a timeout occurs.
</dd>
</dl>
<h2>Get and Put Caching</h2>
<p>
<b>PvaClientChannel</b> has methods:
</p>
<pre>
PvaClientGetPtr get(std::string const &amp; request);
PvaClientPutPtr put(std::string const &amp; request);
</pre>
<p>
Each of these caches.
For example all calls to <b>get</b> with the same <b>request</b> will share the same
<b>PvaChannelGet</b>
</p>
<p>
For example consider a client that makes multiple calls like:
</p>
<pre>
double value;
value = pva->channel(channelName)->get()->getData()->getDouble();
...
value = pva->channel(channelName)->get()->getData()->getDouble();
</pre>
<p>
Only the first call creates a new PvaClientChannel and a new PvaClientGet.
The second call reuses the cached PvaClientChannel and PvaClientGet.
</p>
</body>
</html>