astro-em-http-request 0.1.3.20090419
Sign up to get free protection for your applications and to get access to all the features.
- data/.autotest +1 -0
- data/LICENSE +58 -0
- data/README.rdoc +52 -0
- data/Rakefile +80 -0
- data/ext/buffer/em_buffer.c +630 -0
- data/ext/buffer/extconf.rb +53 -0
- data/ext/http11_client/ext_help.h +14 -0
- data/ext/http11_client/extconf.rb +6 -0
- data/ext/http11_client/http11_client.c +302 -0
- data/ext/http11_client/http11_parser.c +403 -0
- data/ext/http11_client/http11_parser.h +48 -0
- data/ext/http11_client/http11_parser.rl +173 -0
- data/lib/em-http.rb +17 -0
- data/lib/em-http/client.rb +428 -0
- data/lib/em-http/decoders.rb +119 -0
- data/lib/em-http/multi.rb +51 -0
- data/lib/em-http/request.rb +77 -0
- data/test/helper.rb +5 -0
- data/test/stallion.rb +127 -0
- data/test/test_multi.rb +34 -0
- data/test/test_request.rb +264 -0
- metadata +83 -0
data/.autotest
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
require 'autotest/redgreen'
|
data/LICENSE
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
Ruby is copyrighted free software by Yukihiro Matsumoto <matz@netlab.co.jp>.
|
2
|
+
You can redistribute it and/or modify it under either the terms of the GPL
|
3
|
+
(see COPYING.txt file), or the conditions below:
|
4
|
+
|
5
|
+
1. You may make and give away verbatim copies of the source form of the
|
6
|
+
software without restriction, provided that you duplicate all of the
|
7
|
+
original copyright notices and associated disclaimers.
|
8
|
+
|
9
|
+
2. You may modify your copy of the software in any way, provided that
|
10
|
+
you do at least ONE of the following:
|
11
|
+
|
12
|
+
a) place your modifications in the Public Domain or otherwise
|
13
|
+
make them Freely Available, such as by posting said
|
14
|
+
modifications to Usenet or an equivalent medium, or by allowing
|
15
|
+
the author to include your modifications in the software.
|
16
|
+
|
17
|
+
b) use the modified software only within your corporation or
|
18
|
+
organization.
|
19
|
+
|
20
|
+
c) rename any non-standard executables so the names do not conflict
|
21
|
+
with standard executables, which must also be provided.
|
22
|
+
|
23
|
+
d) make other distribution arrangements with the author.
|
24
|
+
|
25
|
+
3. You may distribute the software in object code or executable
|
26
|
+
form, provided that you do at least ONE of the following:
|
27
|
+
|
28
|
+
a) distribute the executables and library files of the software,
|
29
|
+
together with instructions (in the manual page or equivalent)
|
30
|
+
on where to get the original distribution.
|
31
|
+
|
32
|
+
b) accompany the distribution with the machine-readable source of
|
33
|
+
the software.
|
34
|
+
|
35
|
+
c) give non-standard executables non-standard names, with
|
36
|
+
instructions on where to get the original software distribution.
|
37
|
+
|
38
|
+
d) make other distribution arrangements with the author.
|
39
|
+
|
40
|
+
4. You may modify and include the part of the software into any other
|
41
|
+
software (possibly commercial). But some files in the distribution
|
42
|
+
are not written by the author, so that they are not under this terms.
|
43
|
+
|
44
|
+
They are gc.c(partly), utils.c(partly), regex.[ch], st.[ch] and some
|
45
|
+
files under the ./missing directory. See each file for the copying
|
46
|
+
condition.
|
47
|
+
|
48
|
+
5. The scripts and library files supplied as input to or produced as
|
49
|
+
output from the software do not automatically fall under the
|
50
|
+
copyright of the software, but belong to whomever generated them,
|
51
|
+
and may be sold commercially, and may be aggregated with this
|
52
|
+
software.
|
53
|
+
|
54
|
+
6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR
|
55
|
+
IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
|
56
|
+
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
57
|
+
PURPOSE.
|
58
|
+
|
data/README.rdoc
ADDED
@@ -0,0 +1,52 @@
|
|
1
|
+
= EM-HTTP-Client
|
2
|
+
|
3
|
+
EventMachine based HTTP Request interface. Supports streaming response processing, uses Ragel HTTP parser.
|
4
|
+
- Simple interface for single & parallel requests via deferred callbacks
|
5
|
+
- Automatic gzip & deflate decoding
|
6
|
+
- Basic-Auth support
|
7
|
+
- Custom timeouts
|
8
|
+
|
9
|
+
== Simple client example
|
10
|
+
|
11
|
+
EventMachine.run {
|
12
|
+
http = EventMachine::HttpRequest.new('http://127.0.0.1/').get :query => {'keyname' => 'value'}, :timeout => 10
|
13
|
+
|
14
|
+
http.callback {
|
15
|
+
p http.response_header.status
|
16
|
+
p http.response_header
|
17
|
+
p http.response
|
18
|
+
|
19
|
+
EventMachine.stop
|
20
|
+
}
|
21
|
+
}
|
22
|
+
|
23
|
+
== Multi request example
|
24
|
+
|
25
|
+
EventMachine.run {
|
26
|
+
multi = EventMachine::MultiRequest.new
|
27
|
+
|
28
|
+
# add multiple requests to the multi-handler
|
29
|
+
multi.add(EventMachine::HttpRequest.new('http://www.google.com/').get)
|
30
|
+
multi.add(EventMachine::HttpRequest.new('http://www.yahoo.com/').get)
|
31
|
+
|
32
|
+
multi.callback {
|
33
|
+
p multi.responses[:succeeded]
|
34
|
+
p multi.responses[:failed]
|
35
|
+
|
36
|
+
EventMachine.stop
|
37
|
+
}
|
38
|
+
}
|
39
|
+
|
40
|
+
== Basic-Auth example
|
41
|
+
|
42
|
+
EventMachine.run {
|
43
|
+
http = EventMachine::HttpRequest.new('http://www.website.com/').get :head => {'authorization' => ['user', 'pass']}
|
44
|
+
|
45
|
+
http.errback { failed }
|
46
|
+
http.callback {
|
47
|
+
p http.response_header
|
48
|
+
|
49
|
+
EventMachine.stop
|
50
|
+
}
|
51
|
+
}
|
52
|
+
|
data/Rakefile
ADDED
@@ -0,0 +1,80 @@
|
|
1
|
+
require 'rake'
|
2
|
+
require 'rake/clean'
|
3
|
+
require 'rake/rdoctask'
|
4
|
+
require 'rake/gempackagetask'
|
5
|
+
require 'fileutils'
|
6
|
+
include FileUtils
|
7
|
+
|
8
|
+
# copied from EventMachine.
|
9
|
+
MAKE = ENV['MAKE'] || if RUBY_PLATFORM =~ /mswin/ # mingw uses make.
|
10
|
+
'nmake'
|
11
|
+
else
|
12
|
+
'make'
|
13
|
+
end
|
14
|
+
|
15
|
+
# Default Rake task is compile
|
16
|
+
task :default => :compile
|
17
|
+
|
18
|
+
# RDoc
|
19
|
+
Rake::RDocTask.new(:rdoc) do |task|
|
20
|
+
task.rdoc_dir = 'doc'
|
21
|
+
task.title = 'EventMachine::HttpRequest'
|
22
|
+
task.options = %w(--title HttpRequest --main README --line-numbers)
|
23
|
+
task.rdoc_files.include(['lib/**/*.rb'])
|
24
|
+
task.rdoc_files.include(['README', 'LICENSE'])
|
25
|
+
end
|
26
|
+
|
27
|
+
# Rebuild parser Ragel
|
28
|
+
task :ragel do
|
29
|
+
Dir.chdir "ext/http11_client" do
|
30
|
+
target = "http11_parser.c"
|
31
|
+
File.unlink target if File.exist? target
|
32
|
+
sh "ragel http11_parser.rl | rlgen-cd -G2 -o #{target}"
|
33
|
+
raise "Failed to build C source" unless File.exist? target
|
34
|
+
end
|
35
|
+
end
|
36
|
+
|
37
|
+
def make(makedir)
|
38
|
+
Dir.chdir(makedir) { sh MAKE }
|
39
|
+
end
|
40
|
+
|
41
|
+
def extconf(dir)
|
42
|
+
Dir.chdir(dir) { ruby "extconf.rb" }
|
43
|
+
end
|
44
|
+
|
45
|
+
def setup_extension(dir, extension)
|
46
|
+
ext = "ext/#{dir}"
|
47
|
+
ext_so = "#{ext}/#{extension}.#{Config::CONFIG['DLEXT']}"
|
48
|
+
ext_files = FileList[
|
49
|
+
"#{ext}/*.c",
|
50
|
+
"#{ext}/*.h",
|
51
|
+
"#{ext}/extconf.rb",
|
52
|
+
"#{ext}/Makefile",
|
53
|
+
"lib"
|
54
|
+
]
|
55
|
+
|
56
|
+
task "lib" do
|
57
|
+
directory "lib"
|
58
|
+
end
|
59
|
+
|
60
|
+
desc "Builds just the #{extension} extension"
|
61
|
+
|
62
|
+
mf = (extension + '_makefile').to_sym
|
63
|
+
|
64
|
+
task mf do |t|
|
65
|
+
extconf "#{ext}"
|
66
|
+
end
|
67
|
+
|
68
|
+
task extension.to_sym => [mf] do
|
69
|
+
make "#{ext}"
|
70
|
+
cp ext_so, "lib"
|
71
|
+
end
|
72
|
+
end
|
73
|
+
|
74
|
+
setup_extension("buffer", "em_buffer")
|
75
|
+
setup_extension("http11_client", "http11_client")
|
76
|
+
|
77
|
+
task :compile => [:em_buffer, :http11_client]
|
78
|
+
|
79
|
+
CLEAN.include ['build/*', '**/*.o', '**/*.so', '**/*.a', '**/*.log', 'pkg']
|
80
|
+
CLEAN.include ['ext/buffer/Makefile', 'lib/em_buffer.*', 'lib/http11_client.*']
|
@@ -0,0 +1,630 @@
|
|
1
|
+
/*
|
2
|
+
* Copyright (C) 2007 Tony Arcieri
|
3
|
+
* You may redistribute this under the terms of the Ruby license.
|
4
|
+
* See LICENSE for details
|
5
|
+
*/
|
6
|
+
|
7
|
+
#include "ruby.h"
|
8
|
+
#include "rubyio.h"
|
9
|
+
|
10
|
+
#include <assert.h>
|
11
|
+
|
12
|
+
#include <string.h>
|
13
|
+
#include <time.h>
|
14
|
+
#include <errno.h>
|
15
|
+
|
16
|
+
#ifndef GetReadFile
|
17
|
+
#define FPTR_TO_FD(fptr) (fptr->fd)
|
18
|
+
#else
|
19
|
+
#define FPTR_TO_FD(fptr) (fileno(GetReadFile(fptr)))
|
20
|
+
#endif
|
21
|
+
|
22
|
+
/* Default number of bytes in each node's buffer */
|
23
|
+
#define DEFAULT_NODE_SIZE 16384
|
24
|
+
|
25
|
+
/* Maximum age of a buffer node in a memory pool, in seconds */
|
26
|
+
#define MAX_AGE 60
|
27
|
+
|
28
|
+
/* How often to scan the pool for old nodes */
|
29
|
+
#define PURGE_INTERVAL 10
|
30
|
+
|
31
|
+
struct buffer {
|
32
|
+
time_t last_purged_at;
|
33
|
+
unsigned size, node_size;
|
34
|
+
struct buffer_node *head, *tail;
|
35
|
+
struct buffer_node *pool_head, *pool_tail;
|
36
|
+
|
37
|
+
};
|
38
|
+
|
39
|
+
struct buffer_node {
|
40
|
+
time_t last_used_at;
|
41
|
+
unsigned start, end;
|
42
|
+
struct buffer_node *next;
|
43
|
+
unsigned char data[0];
|
44
|
+
};
|
45
|
+
|
46
|
+
static VALUE mEm = Qnil;
|
47
|
+
static VALUE cEm_Buffer = Qnil;
|
48
|
+
|
49
|
+
static VALUE Em_Buffer_allocate(VALUE klass);
|
50
|
+
static void Em_Buffer_mark(struct buffer *);
|
51
|
+
static void Em_Buffer_free(struct buffer *);
|
52
|
+
|
53
|
+
static VALUE Em_Buffer_initialize(int argc, VALUE *argv, VALUE self);
|
54
|
+
static VALUE Em_Buffer_clear(VALUE self);
|
55
|
+
static VALUE Em_Buffer_size(VALUE self);
|
56
|
+
static VALUE Em_Buffer_empty(VALUE self);
|
57
|
+
static VALUE Em_Buffer_append(VALUE self, VALUE data);
|
58
|
+
static VALUE Em_Buffer_prepend(VALUE self, VALUE data);
|
59
|
+
static VALUE Em_Buffer_read(int argc, VALUE *argv, VALUE self);
|
60
|
+
static VALUE Em_Buffer_to_str(VALUE self);
|
61
|
+
static VALUE Em_Buffer_read_from(VALUE self, VALUE io);
|
62
|
+
static VALUE Em_Buffer_write_to(VALUE self, VALUE io);
|
63
|
+
|
64
|
+
static struct buffer *buffer_new(void);
|
65
|
+
static void buffer_clear(struct buffer *buf);
|
66
|
+
static void buffer_free(struct buffer *buf);
|
67
|
+
static void buffer_gc(struct buffer *buf);
|
68
|
+
static void buffer_prepend(struct buffer *buf, char *str, unsigned len);
|
69
|
+
static void buffer_append(struct buffer *buf, char *str, unsigned len);
|
70
|
+
static void buffer_read(struct buffer *buf, char *str, unsigned len);
|
71
|
+
static void buffer_copy(struct buffer *buf, char *str, unsigned len);
|
72
|
+
static int buffer_read_from(struct buffer *buf, int fd);
|
73
|
+
static int buffer_write_to(struct buffer *buf, int fd);
|
74
|
+
|
75
|
+
/*
|
76
|
+
* High speed buffering geared towards non-blocking I/O.
|
77
|
+
*
|
78
|
+
* Data is stored in a byte queue implemented as a linked list of equal size
|
79
|
+
* chunks. Since every node in the list is the same size they are easily
|
80
|
+
* memory pooled. Routines are provided for high speed non-blocking reads
|
81
|
+
* and writes from Ruby IO objects.
|
82
|
+
*/
|
83
|
+
void Init_em_buffer()
|
84
|
+
{
|
85
|
+
mEm = rb_define_module("EventMachine");
|
86
|
+
cEm_Buffer = rb_define_class_under(mEm, "Buffer", rb_cObject);
|
87
|
+
rb_define_alloc_func(cEm_Buffer, Em_Buffer_allocate);
|
88
|
+
|
89
|
+
rb_define_method(cEm_Buffer, "initialize", Em_Buffer_initialize, -1);
|
90
|
+
rb_define_method(cEm_Buffer, "clear", Em_Buffer_clear, 0);
|
91
|
+
rb_define_method(cEm_Buffer, "size", Em_Buffer_size, 0);
|
92
|
+
rb_define_method(cEm_Buffer, "empty?", Em_Buffer_empty, 0);
|
93
|
+
rb_define_method(cEm_Buffer, "<<", Em_Buffer_append, 1);
|
94
|
+
rb_define_method(cEm_Buffer, "append", Em_Buffer_append, 1);
|
95
|
+
rb_define_method(cEm_Buffer, "prepend", Em_Buffer_prepend, 1);
|
96
|
+
rb_define_method(cEm_Buffer, "read", Em_Buffer_read, -1);
|
97
|
+
rb_define_method(cEm_Buffer, "to_str", Em_Buffer_to_str, 0);
|
98
|
+
rb_define_method(cEm_Buffer, "read_from", Em_Buffer_read_from, 1);
|
99
|
+
rb_define_method(cEm_Buffer, "write_to", Em_Buffer_write_to, 1);
|
100
|
+
}
|
101
|
+
|
102
|
+
static VALUE Em_Buffer_allocate(VALUE klass)
|
103
|
+
{
|
104
|
+
return Data_Wrap_Struct(klass, Em_Buffer_mark, Em_Buffer_free, buffer_new());
|
105
|
+
}
|
106
|
+
|
107
|
+
static void Em_Buffer_mark(struct buffer *buf)
|
108
|
+
{
|
109
|
+
/* Walks the pool of unused chunks and frees any that are beyond a certain age */
|
110
|
+
buffer_gc(buf);
|
111
|
+
}
|
112
|
+
|
113
|
+
static void Em_Buffer_free(struct buffer *buf)
|
114
|
+
{
|
115
|
+
buffer_free(buf);
|
116
|
+
}
|
117
|
+
|
118
|
+
/**
|
119
|
+
* call-seq:
|
120
|
+
* EventMachine::Buffer.new(size = DEFAULT_NODE_SIZE) -> EventMachine::Buffer
|
121
|
+
*
|
122
|
+
* Create a new EventMachine::Buffer with linked segments of the given size
|
123
|
+
*/
|
124
|
+
static VALUE Em_Buffer_initialize(int argc, VALUE *argv, VALUE self)
|
125
|
+
{
|
126
|
+
VALUE node_size_obj;
|
127
|
+
int node_size;
|
128
|
+
struct buffer *buf;
|
129
|
+
|
130
|
+
if(rb_scan_args(argc, argv, "01", &node_size_obj) == 1) {
|
131
|
+
node_size = NUM2INT(node_size_obj);
|
132
|
+
|
133
|
+
if(node_size < 1) rb_raise(rb_eArgError, "invalid buffer size");
|
134
|
+
|
135
|
+
Data_Get_Struct(self, struct buffer, buf);
|
136
|
+
|
137
|
+
/* Make sure we're not changing the buffer size after data has been allocated */
|
138
|
+
assert(!buf->head);
|
139
|
+
assert(!buf->pool_head);
|
140
|
+
|
141
|
+
buf->node_size = node_size;
|
142
|
+
}
|
143
|
+
|
144
|
+
return Qnil;
|
145
|
+
}
|
146
|
+
|
147
|
+
/**
|
148
|
+
* call-seq:
|
149
|
+
* EventMachine::Buffer#clear -> nil
|
150
|
+
*
|
151
|
+
* Clear all data from the EventMachine::Buffer
|
152
|
+
*/
|
153
|
+
static VALUE Em_Buffer_clear(VALUE self)
|
154
|
+
{
|
155
|
+
struct buffer *buf;
|
156
|
+
Data_Get_Struct(self, struct buffer, buf);
|
157
|
+
|
158
|
+
buffer_clear(buf);
|
159
|
+
|
160
|
+
return Qnil;
|
161
|
+
}
|
162
|
+
|
163
|
+
/**
|
164
|
+
* call-seq:
|
165
|
+
* EventMachine::Buffer#size -> Integer
|
166
|
+
*
|
167
|
+
* Return the size of the buffer in bytes
|
168
|
+
*/
|
169
|
+
static VALUE Em_Buffer_size(VALUE self)
|
170
|
+
{
|
171
|
+
struct buffer *buf;
|
172
|
+
Data_Get_Struct(self, struct buffer, buf);
|
173
|
+
|
174
|
+
return INT2NUM(buf->size);
|
175
|
+
}
|
176
|
+
|
177
|
+
/**
|
178
|
+
* call-seq:
|
179
|
+
* EventMachine::Buffer#empty? -> Boolean
|
180
|
+
*
|
181
|
+
* Is the buffer empty?
|
182
|
+
*/
|
183
|
+
static VALUE Em_Buffer_empty(VALUE self)
|
184
|
+
{
|
185
|
+
struct buffer *buf;
|
186
|
+
Data_Get_Struct(self, struct buffer, buf);
|
187
|
+
|
188
|
+
return buf->size > 0 ? Qfalse : Qtrue;
|
189
|
+
}
|
190
|
+
|
191
|
+
/**
|
192
|
+
* call-seq:
|
193
|
+
* EventMachine::Buffer#append(data) -> String
|
194
|
+
*
|
195
|
+
* Append the given data to the end of the buffer
|
196
|
+
*/
|
197
|
+
static VALUE Em_Buffer_append(VALUE self, VALUE data)
|
198
|
+
{
|
199
|
+
struct buffer *buf;
|
200
|
+
Data_Get_Struct(self, struct buffer, buf);
|
201
|
+
|
202
|
+
/* Is this needed? Never seen anyone else do it... */
|
203
|
+
data = rb_convert_type(data, T_STRING, "String", "to_str");
|
204
|
+
buffer_append(buf, RSTRING_PTR(data), RSTRING_LEN(data));
|
205
|
+
|
206
|
+
return data;
|
207
|
+
}
|
208
|
+
|
209
|
+
/**
|
210
|
+
* call-seq:
|
211
|
+
* EventMachine::Buffer#prepend(data) -> String
|
212
|
+
*
|
213
|
+
* Prepend the given data to the beginning of the buffer
|
214
|
+
*/
|
215
|
+
static VALUE Em_Buffer_prepend(VALUE self, VALUE data)
|
216
|
+
{
|
217
|
+
struct buffer *buf;
|
218
|
+
Data_Get_Struct(self, struct buffer, buf);
|
219
|
+
|
220
|
+
data = rb_convert_type(data, T_STRING, "String", "to_str");
|
221
|
+
buffer_prepend(buf, RSTRING_PTR(data), RSTRING_LEN(data));
|
222
|
+
|
223
|
+
return data;
|
224
|
+
}
|
225
|
+
|
226
|
+
/**
|
227
|
+
* call-seq:
|
228
|
+
* EventMachine::Buffer#read(length = nil) -> String
|
229
|
+
*
|
230
|
+
* Read the specified abount of data from the buffer. If no value
|
231
|
+
* is given the entire contents of the buffer are returned. Any data
|
232
|
+
* read from the buffer is cleared.
|
233
|
+
*/
|
234
|
+
static VALUE Em_Buffer_read(int argc, VALUE *argv, VALUE self)
|
235
|
+
{
|
236
|
+
VALUE length_obj, str;
|
237
|
+
int length;
|
238
|
+
struct buffer *buf;
|
239
|
+
|
240
|
+
Data_Get_Struct(self, struct buffer, buf);
|
241
|
+
|
242
|
+
if(rb_scan_args(argc, argv, "01", &length_obj) == 1) {
|
243
|
+
length = NUM2INT(length_obj);
|
244
|
+
} else {
|
245
|
+
if(buf->size == 0)
|
246
|
+
return rb_str_new2("");
|
247
|
+
|
248
|
+
length = buf->size;
|
249
|
+
}
|
250
|
+
|
251
|
+
if(length > buf->size)
|
252
|
+
length = buf->size;
|
253
|
+
|
254
|
+
if(length < 1)
|
255
|
+
rb_raise(rb_eArgError, "length must be greater than zero");
|
256
|
+
|
257
|
+
str = rb_str_new(0, length);
|
258
|
+
buffer_read(buf, RSTRING_PTR(str), length);
|
259
|
+
|
260
|
+
return str;
|
261
|
+
}
|
262
|
+
|
263
|
+
/**
|
264
|
+
* call-seq:
|
265
|
+
* EventMachine::Buffer#to_str -> String
|
266
|
+
*
|
267
|
+
* Convert the Buffer to a String. The original buffer is unmodified.
|
268
|
+
*/
|
269
|
+
static VALUE Em_Buffer_to_str(VALUE self) {
|
270
|
+
VALUE str;
|
271
|
+
struct buffer *buf;
|
272
|
+
|
273
|
+
Data_Get_Struct(self, struct buffer, buf);
|
274
|
+
|
275
|
+
str = rb_str_new(0, buf->size);
|
276
|
+
buffer_copy(buf, RSTRING_PTR(str), buf->size);
|
277
|
+
|
278
|
+
return str;
|
279
|
+
}
|
280
|
+
|
281
|
+
/**
|
282
|
+
* call-seq:
|
283
|
+
* EventMachine::Buffer#read_from(io) -> Integer
|
284
|
+
*
|
285
|
+
* Perform a nonblocking read of the the given IO object and fill
|
286
|
+
* the buffer with any data received. The call will read as much
|
287
|
+
* data as it can until the read would block.
|
288
|
+
*/
|
289
|
+
static VALUE Em_Buffer_read_from(VALUE self, VALUE io) {
|
290
|
+
struct buffer *buf;
|
291
|
+
#if HAVE_RB_IO_T
|
292
|
+
rb_io_t *fptr;
|
293
|
+
#else
|
294
|
+
OpenFile *fptr;
|
295
|
+
#endif
|
296
|
+
|
297
|
+
Data_Get_Struct(self, struct buffer, buf);
|
298
|
+
GetOpenFile(rb_convert_type(io, T_FILE, "IO", "to_io"), fptr);
|
299
|
+
rb_io_set_nonblock(fptr);
|
300
|
+
|
301
|
+
return INT2NUM(buffer_read_from(buf, FPTR_TO_FD(fptr)));
|
302
|
+
}
|
303
|
+
|
304
|
+
/**
|
305
|
+
* call-seq:
|
306
|
+
* EventMachine::Buffer#write_to(io) -> Integer
|
307
|
+
*
|
308
|
+
* Perform a nonblocking write of the buffer to the given IO object.
|
309
|
+
* As much data as possible is written until the call would block.
|
310
|
+
* Any data which is written is removed from the buffer.
|
311
|
+
*/
|
312
|
+
static VALUE Em_Buffer_write_to(VALUE self, VALUE io) {
|
313
|
+
struct buffer *buf;
|
314
|
+
#if HAVE_RB_IO_T
|
315
|
+
rb_io_t *fptr;
|
316
|
+
#else
|
317
|
+
OpenFile *fptr;
|
318
|
+
#endif
|
319
|
+
|
320
|
+
Data_Get_Struct(self, struct buffer, buf);
|
321
|
+
GetOpenFile(rb_convert_type(io, T_FILE, "IO", "to_io"), fptr);
|
322
|
+
rb_io_set_nonblock(fptr);
|
323
|
+
|
324
|
+
return INT2NUM(buffer_write_to(buf, FPTR_TO_FD(fptr)));
|
325
|
+
}
|
326
|
+
|
327
|
+
/*
|
328
|
+
* Ruby bindings end here. Below is the actual implementation of
|
329
|
+
* the underlying data structures.
|
330
|
+
*/
|
331
|
+
|
332
|
+
/* Create a new buffer */
|
333
|
+
static struct buffer *buffer_new(void)
|
334
|
+
{
|
335
|
+
struct buffer *buf;
|
336
|
+
|
337
|
+
buf = (struct buffer *)xmalloc(sizeof(struct buffer));
|
338
|
+
buf->head = buf->tail = buf->pool_head = buf->pool_tail = 0;
|
339
|
+
buf->size = 0;
|
340
|
+
buf->node_size = DEFAULT_NODE_SIZE;
|
341
|
+
time(&buf->last_purged_at);
|
342
|
+
|
343
|
+
return buf;
|
344
|
+
}
|
345
|
+
|
346
|
+
/* Clear all data from a buffer */
|
347
|
+
static void buffer_clear(struct buffer *buf)
|
348
|
+
{
|
349
|
+
struct buffer_node *tmp;
|
350
|
+
|
351
|
+
/* Move everything into the buffer pool */
|
352
|
+
if(!buf->pool_tail)
|
353
|
+
buf->pool_head = buf->pool_tail = buf->head;
|
354
|
+
else
|
355
|
+
buf->pool_tail->next = buf->head;
|
356
|
+
|
357
|
+
buf->head = buf->tail = 0;
|
358
|
+
buf->size = 0;
|
359
|
+
}
|
360
|
+
|
361
|
+
/* Free a buffer */
|
362
|
+
static void buffer_free(struct buffer *buf)
|
363
|
+
{
|
364
|
+
struct buffer_node *tmp;
|
365
|
+
|
366
|
+
buffer_clear(buf);
|
367
|
+
|
368
|
+
while(buf->pool_head) {
|
369
|
+
tmp = buf->pool_head;
|
370
|
+
buf->pool_head = tmp->next;
|
371
|
+
free(tmp);
|
372
|
+
}
|
373
|
+
|
374
|
+
free(buf);
|
375
|
+
}
|
376
|
+
|
377
|
+
/* Run through the pool and find elements that haven't been used for awhile */
|
378
|
+
static void buffer_gc(struct buffer *buf)
|
379
|
+
{
|
380
|
+
struct buffer_node *cur, *tmp;
|
381
|
+
time_t now;
|
382
|
+
time(&now);
|
383
|
+
|
384
|
+
/* Only purge if we've passed the purge interval */
|
385
|
+
if(now - buf->last_purged_at < PURGE_INTERVAL)
|
386
|
+
return;
|
387
|
+
|
388
|
+
buf->last_purged_at = now;
|
389
|
+
|
390
|
+
while(buf->pool_head && now - buf->pool_head->last_used_at >= MAX_AGE) {
|
391
|
+
tmp = buf->pool_head;
|
392
|
+
buf->pool_head = buf->pool_head->next;
|
393
|
+
free(tmp);
|
394
|
+
}
|
395
|
+
|
396
|
+
if(!buf->pool_head)
|
397
|
+
buf->pool_tail = 0;
|
398
|
+
}
|
399
|
+
|
400
|
+
/* Create a new buffer_node (or pull one from the memory pool) */
|
401
|
+
static struct buffer_node *buffer_node_new(struct buffer *buf)
|
402
|
+
{
|
403
|
+
struct buffer_node *node;
|
404
|
+
|
405
|
+
/* Pull from the memory pool if available */
|
406
|
+
if(buf->pool_head) {
|
407
|
+
node = buf->pool_head;
|
408
|
+
buf->pool_head = node->next;
|
409
|
+
|
410
|
+
if(node->next)
|
411
|
+
node->next = 0;
|
412
|
+
else
|
413
|
+
buf->pool_tail = 0;
|
414
|
+
} else {
|
415
|
+
node = (struct buffer_node *)xmalloc(sizeof(struct buffer_node) + buf->node_size);
|
416
|
+
node->next = 0;
|
417
|
+
}
|
418
|
+
|
419
|
+
node->start = node->end = 0;
|
420
|
+
return node;
|
421
|
+
}
|
422
|
+
|
423
|
+
/* Free a buffer node (i.e. return it to the memory pool) */
|
424
|
+
static void buffer_node_free(struct buffer *buf, struct buffer_node *node)
|
425
|
+
{
|
426
|
+
/* Store when the node was freed */
|
427
|
+
time(&node->last_used_at);
|
428
|
+
|
429
|
+
node->next = buf->pool_head;
|
430
|
+
buf->pool_head = node;
|
431
|
+
|
432
|
+
if(!buf->pool_tail)
|
433
|
+
buf->pool_tail = node;
|
434
|
+
}
|
435
|
+
|
436
|
+
/* Prepend data to the front of the buffer */
|
437
|
+
static void buffer_prepend(struct buffer *buf, char *str, unsigned len)
|
438
|
+
{
|
439
|
+
struct buffer_node *node, *tmp;
|
440
|
+
buf->size += len;
|
441
|
+
|
442
|
+
/* If it fits in the beginning of the head */
|
443
|
+
if(buf->head && buf->head->start >= len) {
|
444
|
+
buf->head->start -= len;
|
445
|
+
memcpy(buf->head->data + buf->head->start, str, len);
|
446
|
+
} else {
|
447
|
+
node = buffer_node_new(buf);
|
448
|
+
node->next = buf->head;
|
449
|
+
buf->head = node;
|
450
|
+
if(!buf->tail) buf->tail = node;
|
451
|
+
|
452
|
+
while(len > buf->node_size) {
|
453
|
+
memcpy(node->data, str, buf->node_size);
|
454
|
+
node->end = buf->node_size;
|
455
|
+
|
456
|
+
tmp = buffer_node_new(buf);
|
457
|
+
tmp->next = node->next;
|
458
|
+
node->next = tmp;
|
459
|
+
|
460
|
+
if(buf->tail == node) buf->tail = tmp;
|
461
|
+
node = tmp;
|
462
|
+
|
463
|
+
str += buf->node_size;
|
464
|
+
len -= buf->node_size;
|
465
|
+
}
|
466
|
+
|
467
|
+
if(len > 0) {
|
468
|
+
memcpy(node->data, str, len);
|
469
|
+
node->end = len;
|
470
|
+
}
|
471
|
+
}
|
472
|
+
}
|
473
|
+
|
474
|
+
/* Append data to the front of the buffer */
|
475
|
+
static void buffer_append(struct buffer *buf, char *str, unsigned len)
|
476
|
+
{
|
477
|
+
unsigned nbytes;
|
478
|
+
buf->size += len;
|
479
|
+
|
480
|
+
/* If it fits in the remaining space in the tail */
|
481
|
+
if(buf->tail && len <= buf->node_size - buf->tail->end) {
|
482
|
+
memcpy(buf->tail->data + buf->tail->end, str, len);
|
483
|
+
buf->tail->end += len;
|
484
|
+
return;
|
485
|
+
}
|
486
|
+
|
487
|
+
/* Empty list needs initialized */
|
488
|
+
if(!buf->head) {
|
489
|
+
buf->head = buffer_node_new(buf);
|
490
|
+
buf->tail = buf->head;
|
491
|
+
}
|
492
|
+
|
493
|
+
/* Build links out of the data */
|
494
|
+
while(len > 0) {
|
495
|
+
nbytes = buf->node_size - buf->tail->end;
|
496
|
+
if(len < nbytes) nbytes = len;
|
497
|
+
|
498
|
+
memcpy(buf->tail->data + buf->tail->end, str, nbytes);
|
499
|
+
str += nbytes;
|
500
|
+
len -= nbytes;
|
501
|
+
|
502
|
+
buf->tail->end += nbytes;
|
503
|
+
|
504
|
+
if(len > 0) {
|
505
|
+
buf->tail->next = buffer_node_new(buf);
|
506
|
+
buf->tail = buf->tail->next;
|
507
|
+
}
|
508
|
+
}
|
509
|
+
}
|
510
|
+
|
511
|
+
/* Read data from the buffer (and clear what we've read) */
|
512
|
+
static void buffer_read(struct buffer *buf, char *str, unsigned len)
|
513
|
+
{
|
514
|
+
unsigned nbytes;
|
515
|
+
struct buffer_node *tmp;
|
516
|
+
|
517
|
+
while(buf->size > 0 && len > 0) {
|
518
|
+
nbytes = buf->head->end - buf->head->start;
|
519
|
+
if(len < nbytes) nbytes = len;
|
520
|
+
|
521
|
+
memcpy(str, buf->head->data + buf->head->start, nbytes);
|
522
|
+
str += nbytes;
|
523
|
+
len -= nbytes;
|
524
|
+
|
525
|
+
buf->head->start += nbytes;
|
526
|
+
buf->size -= nbytes;
|
527
|
+
|
528
|
+
if(buf->head->start == buf->head->end) {
|
529
|
+
tmp = buf->head;
|
530
|
+
buf->head = tmp->next;
|
531
|
+
buffer_node_free(buf, tmp);
|
532
|
+
|
533
|
+
if(!buf->head) buf->tail = 0;
|
534
|
+
}
|
535
|
+
}
|
536
|
+
}
|
537
|
+
|
538
|
+
/* Copy data from the buffer without clearing it */
|
539
|
+
static void buffer_copy(struct buffer *buf, char *str, unsigned len)
|
540
|
+
{
|
541
|
+
unsigned nbytes;
|
542
|
+
struct buffer_node *node;
|
543
|
+
|
544
|
+
node = buf->head;
|
545
|
+
while(node && len > 0) {
|
546
|
+
nbytes = node->end - node->start;
|
547
|
+
if(len < nbytes) nbytes = len;
|
548
|
+
|
549
|
+
memcpy(str, node->data + node->start, nbytes);
|
550
|
+
str += nbytes;
|
551
|
+
len -= nbytes;
|
552
|
+
|
553
|
+
if(node->start + nbytes == node->end)
|
554
|
+
node = node->next;
|
555
|
+
}
|
556
|
+
}
|
557
|
+
|
558
|
+
/* Write data from the buffer to a file descriptor */
|
559
|
+
static int buffer_write_to(struct buffer *buf, int fd)
|
560
|
+
{
|
561
|
+
int bytes_written, total_bytes_written = 0;
|
562
|
+
struct buffer_node *tmp;
|
563
|
+
|
564
|
+
while(buf->head) {
|
565
|
+
bytes_written = write(fd, buf->head->data + buf->head->start, buf->head->end - buf->head->start);
|
566
|
+
|
567
|
+
/* If the write failed... */
|
568
|
+
if(bytes_written < 0) {
|
569
|
+
if(errno != EAGAIN)
|
570
|
+
rb_sys_fail("write");
|
571
|
+
|
572
|
+
return total_bytes_written;
|
573
|
+
}
|
574
|
+
|
575
|
+
total_bytes_written += bytes_written;
|
576
|
+
buf->size -= bytes_written;
|
577
|
+
|
578
|
+
/* If the write blocked... */
|
579
|
+
if(bytes_written < buf->head->end - buf->head->start) {
|
580
|
+
buf->head->start += bytes_written;
|
581
|
+
return total_bytes_written;
|
582
|
+
}
|
583
|
+
|
584
|
+
/* Otherwise we wrote the whole buffer */
|
585
|
+
tmp = buf->head;
|
586
|
+
buf->head = tmp->next;
|
587
|
+
buffer_node_free(buf, tmp);
|
588
|
+
|
589
|
+
if(!buf->head) buf->tail = 0;
|
590
|
+
}
|
591
|
+
|
592
|
+
return total_bytes_written;
|
593
|
+
}
|
594
|
+
|
595
|
+
/* Read data from a file descriptor to a buffer */
|
596
|
+
/* Append data to the front of the buffer */
|
597
|
+
static int buffer_read_from(struct buffer *buf, int fd)
|
598
|
+
{
|
599
|
+
int bytes_read, total_bytes_read = 0;
|
600
|
+
unsigned nbytes;
|
601
|
+
|
602
|
+
/* Empty list needs initialized */
|
603
|
+
if(!buf->head) {
|
604
|
+
buf->head = buffer_node_new(buf);
|
605
|
+
buf->tail = buf->head;
|
606
|
+
}
|
607
|
+
|
608
|
+
do {
|
609
|
+
nbytes = buf->node_size - buf->tail->end;
|
610
|
+
bytes_read = read(fd, buf->tail->data + buf->tail->end, nbytes);
|
611
|
+
|
612
|
+
if(bytes_read < 1) {
|
613
|
+
if(errno != EAGAIN)
|
614
|
+
rb_sys_fail("read");
|
615
|
+
|
616
|
+
return total_bytes_read;
|
617
|
+
}
|
618
|
+
|
619
|
+
total_bytes_read += bytes_read;
|
620
|
+
buf->tail->end += nbytes;
|
621
|
+
buf->size += nbytes;
|
622
|
+
|
623
|
+
if(buf->tail->end == buf->node_size) {
|
624
|
+
buf->tail->next = buffer_node_new(buf);
|
625
|
+
buf->tail = buf->tail->next;
|
626
|
+
}
|
627
|
+
} while(bytes_read == nbytes);
|
628
|
+
|
629
|
+
return total_bytes_read;
|
630
|
+
}
|