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
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
|
// $Id$
#ifndef _CLIENT_HANDLER_H
#define _CLIENT_HANDLER_H
/*
A connection handler will also be derived from ACE_Event_Handler so that we
can register with a reactor.
*/
#include "ace/Event_Handler.h"
#if !defined (ACE_LACKS_PRAGMA_ONCE)
# pragma once
#endif /* ACE_LACKS_PRAGMA_ONCE */
#include "ace/INET_Addr.h"
/*
Since we're doing TCP/IP, we'll need a SOCK_Stream for the connection.
*/
#include "ace/SOCK_Stream.h"
class Logging_Handler : public ACE_Event_Handler
{
public:
/*
Like the acceptor, we're simple enough to avoid constructor and destructor.
*/
/*
To open the client handler, we have to register ourselves with the reactor.
Notice that we don't have to "open" our ACE_SOCK_Stream member variable.
Why? Because the call to the acceptor's accept() method took care of those
details for us.
*/
int open ( ACE_Reactor * _reactor )
{
/*
Remember our reactor...
*/
reactor_ = _reactor;
/*
In this case we're using the READ_MASK. Like the acceptor, handle_input()
will be called due to this mask but it's a nice piece of bookkeeping to
have separate masks for the separate types of activity.
*/
if (_reactor-> register_handler (this, ACE_Event_Handler::READ_MASK) == -1)
ACE_ERROR_RETURN ((LM_ERROR, "(%P|%t) can't register with reactor\n"), -1);
return 0;
}
/*
If we're explicitly closed we'll close our "file handle". The net result
is to close the connection to the client and remove ourselves from the
reactor if we're registered
*/
int close (void)
{
return this->handle_close (ACE_INVALID_HANDLE, ACE_Event_Handler::RWE_MASK);
}
/*
This is a bit of magic... When we call the accept() method of the acceptor
object, it wants to do work on an ACE_SOCK_Stream. We have one of those as
our connection to the client but it would be gross to provide a method to
access that object. It's much cooler if the acceptor can just treat the
Logging_Handler as an ACE_SOCK_Stream. Providing this cast operator lets
that happen cleanly.
*/
operator ACE_SOCK_Stream &()
{
return this->cli_stream_;
}
protected:
/*
Again, like the acceptor, we need to provide the connection handle to the reactor.
*/
ACE_HANDLE get_handle (void) const
{
return this->cli_stream_.get_handle ();
}
/*
And here's the handle_input(). This is really the workhorse of the application.
*/
virtual int handle_input (ACE_HANDLE)
{
/*
Create and initialize a small receive buffer. The extra byte is
there to allow us to have a null-terminated string when it's over.
*/
char buf[BUFSIZ+1];
/*
Invoke the recv() method of the ACE_SOCK_Stream to get some data. It will
return -1 if there is an error. Otherwise, it will return the number of bytes
read. Of course, if it read zero bytes then the connection must be gone.
How do I know that? Because handle_input() would not be called by the reactor
if there wasn't *some* kind of activity and a closed connection looks like a
read request to the reactor. But when you read from a closed connection you'll
read zero bytes.
Notice that in the error case or closed case we return -1. That tells the reactor
to call our handle_close() where we'll take care of shutting down cleanly.
Although we don't make use of them, there are additional parameters you can
use with the recv() call. One of these is an ACE_Time_Value that allows you to
limit the amount of time blocking on the recv(). You would use that if you
weren't sure if data was available. Since we only get to handle_input() when
data is ready, that would be redundant. On the other hand, if you use recv_n()
to read *exactly* a number of bytes then limiting the time you wait for those
bytes might be good.
The other paramter that may come in handy is an integer <i>flags</i>. This is
passed directly to the underlying OS recv() call. See the man page recv(2)
and the header sys/socket.h for the gory details.
*/
ssize_t retval;
switch( retval = this->cli_stream_.recv(buf,BUFSIZ) )
{
case -1:
ACE_ERROR_RETURN ((LM_ERROR, "(%P|%t) %p bad read\n", "client logger"), -1);
case 0:
ACE_ERROR_RETURN ((LM_ERROR, "(%P|%t) closing log daemon (fd = %d)\n",
this->get_handle ()), -1);
default:
buf[retval] = '\0';
ACE_DEBUG ((LM_DEBUG, "(%P|%t) from client: %s",buf));
}
return 0;
}
/*
When handle_input() returns -1, we'll end up here. There are a few housekeeping
chores to handle.
*/
int handle_close (ACE_HANDLE, ACE_Reactor_Mask _mask)
{
/*
Remove ourselves from the reactor. We have to include the DONT_CALL in the
mask so that it won't call handle_close() on us again!
*/
reactor_->remove_handler(this,_mask|ACE_Event_Handler::DONT_CALL);
/*
Close the socket that we're connected to the client with.
*/
cli_stream_.close();
/*
Since we know we were dynamically allocated by the acceptor, now is a good
time to get rid of ourselves.
*/
delete this;
return 0;
}
protected:
/*
Our peer connection.
*/
ACE_SOCK_Stream cli_stream_;
/*
Our reactor (and our acceptor's reactor).
*/
ACE_Reactor * reactor_;
};
#endif /* _CLIENT_HANDLER_H */
|
