moonbridge
view reference.txt @ 97:0561abcc68ee
Bugfix regarding TCP PSH
| author | jbe |
|---|---|
| date | Wed Apr 08 01:35:06 2015 +0200 (2015-04-08) |
| parents | de3982f17d05 |
| children | acaa85256c4b |
line source
2 Moonbridge reference
3 ====================
7 Global function listen{...}
8 ---------------------------
10 This function initializes the Moonbridge Network Server. It may be called
11 multiple times. However, it is not allowed to register additional listeners by
12 calling listen(...) from a "prepare", "connect", or "finish" handler.
14 See file "example.lua" for parametrization of the listen(...) function.
16 Warning: Moonbridge will fork the Lua environment to handle parallel requests.
17 Functions provided as "prepare", "connect", and "finish" handlers may access
18 global variables, but for every child process these global variables will not
19 be shared! If you require a global state, a DBMS, cache server, or similar is
20 necessary.
24 Global function timeout(...)
25 ----------------------------
27 Calling this function with a positive number (time in seconds) sets a timer
28 that kills the current process after the selected time runs out. The remaining
29 time can be queried by calling this function without arguments.
31 Calling this function with a single argument that is the number zero will
32 disable the timeout.
34 Another mode of operation is selected by passing two arguments: a time (in
35 seconds) as first argument and a function as second argument. In this case, a
36 sub-timer will be used to limit the execution time of the function. In case of
37 timeout, the process will be killed (and the timeout function does not return).
38 If the time for the sub-timer is longer than a previously set timeout (using
39 the timeout(...) function with one argument), the shorter timeout (of the
40 previous call of timeout(...)) will have precedence.
42 Timers are also automatically reset (disabled) when a handler (prepare handler
43 or connect handler) returns. To shutdown processes after a certain time waiting
44 for a new request, use the idle_time parameter of the listen function.
48 Socket object passed to "connect" handler
49 -----------------------------------------
51 For every incoming connection, the registered "connect" handler is called with
52 a single socket object as argument, which is described below:
55 ### socket:close()
57 Closes the socket connection (input and output stream) by flushing all data and
58 sending a TCP FIN packet.
60 Returns true on success, or nil plus error message in case of an I/O error.
61 Using this method on sockets that have already been closed (or reset) will
62 throw an error.
64 Warning: Pending data on the input stream may cause connection aborts (TCP RST)
65 when network connections are used. All pending input data should have been read
66 (or drained) before calling socket:close(). Use socket:finish() to send a
67 TCP FIN packet to the peer before waiting for EOF from the peer.
70 ### socket:drain(maxlen, terminator)
72 Same as socket:read(maxlen, terminator), but discards the input and returns the
73 number of discarded bytes. If no bytes could be read but EOF was encountered,
74 then true is returned.
76 In case of an I/O error, nil (as first return value) plus an error message (as
77 second result value) are returned.
80 ### socket:drain_nb(maxlen, terminator)
82 Same as socket:read_nb(maxlen, terminator), but discards the input and returns
83 the number of discarded bytes. If no bytes could be read but EOF was
84 encountered, then true is returned.
86 In case of an I/O error, nil (as first return value) plus an error message (as
87 second result value) are returned.
90 ### socket:finish()
92 Sends a TCP FIN packet to indicate EOF on write stream. Subsequent reads are
93 still possible. When there is no more input data to be read, the connection
94 should finally be closed with socket:close().
96 In case of local sockets (Unix Domain Sockets), socket:finish() simply closes
97 the underlying socket and emulates EOF on subsequent reads. Also in this case,
98 the connection should be finally closed with socket:close().
101 ### socket:flush(...)
103 Same as socket:write(...) but additionally flushes the socket (i.e. all pending
104 data is passed to the operating system).
106 In case of an I/O error, nil (as first return value) plus an error message (as
107 second result value) are returned. On success, the socket userdata object is
108 returned.
111 ### socket:flush_nb(...)
113 Same as socket:write_nb(...) but additionally flushes the socket (i.e. all
114 pending data is passed to the operating system). The total number of bytes that
115 could not be passed yet to the operating system is returned. Zero is returned
116 if all data could be flushed out.
118 In case of an I/O error, nil (as first return value) plus an error message (as
119 second result value) are returned.
122 ### socket.interval
124 Set to the name of an interval timer if the "connect" handler was called due to
125 an elapsed interval timer. Otherwise nil.
128 ### socket.local_ip4
130 Local IPv4 address used for the connection. Encoded as 4 raw bytes in form of a
131 string.
134 ### socket.local_ip6
136 Local IPv6 address used for the connection. Encoded as 16 raw bytes in form of
137 a string.
140 ### socket.local_tcpport
142 Local TCP port used for the connection.
145 ### socket:read(maxlen, terminator)
147 Read up to maxlen bytes or until an optional termination character is
148 encountered (which is included in the result). The maxlen value may be nil, in
149 which case there is no limit on the number of bytes read.
151 If EOF is encountered before any data could be read, then false (as first
152 return value) plus a notice string (as second return value) are returned.
154 In case of an I/O error, nil (as first return value) plus an error message (as
155 second result value) are returned.
158 ### socket:read_nb(maxlen, terminator)
160 Read up to maxlen bytes, until an optional termination character is encountered
161 (which is included in the result), or until no more data is available for
162 reading. The maxlen value may be nil, in which case there is no limit on the
163 number of bytes read.
165 If EOF is encountered before any data could be read, then false (as first
166 return value) plus a notice string (as second return value) are returned.
168 If no data was available for reading, but no EOF was encountered, then an empty
169 string is returned.
171 In case of an I/O error, nil (as first return value) plus an error message (as
172 second result value) are returned.
175 ### socket.remote_ip4
177 Remote IPv4 address used for the connection. Encoded as 4 raw bytes in form of
178 a string.
181 ### socket.remote_ip6
183 Remote IPv6 address used for the connection. Encoded as 16 raw bytes in form of
184 a string.
187 ### socket.remote_tcpport
189 Remote TCP port used for the connection.
192 ### socket:reset()
194 Alias for socket:close(0). Closes the socket connection by sending a TCP RST
195 packet if possible to indicate error condition.
197 Returns true on success, or nil (as first result value) plus error message (as
198 second result value) in case of an I/O error. Using this method on sockets that
199 have already been closed (or reset) will throw an error.
201 Warning: Previously sent (and flushed) data may be lost during transmission.
204 ### socket:write(...)
206 Takes a variable number of strings and sends them to the peer. The operation is
207 buffered, so to actually send out the data, it is necessary to eventually call
208 socket:flush(), socket:finish(), or socket:close().
210 In case of an I/O error, nil (as first return value) plus an error message (as
211 second result value) are returned. On success, the socket userdata object is
212 returned.
215 ### socket:write_nb(...)
217 Takes a variable number of strings and sends them to the peer. The operation is
218 buffered, so to actually send out the data, it is necessary to eventually call
219 socket:flush_nb(), socket:flush(), socket:finish(), or socket:close().
221 This function always returns immediately (i.e. it does not block). If all data
222 (but a small buffered portion) could be sent out, then zero is returned.
223 Otherwise, all arguments that could not be sent are stored in a buffer of
224 unlimited size (up to memory capabilities) and an integer is returned that
225 indicates the number of bytes currently in the buffer.
227 In case of an I/O error, nil (as first return value) plus an error message (as
228 second result value) are returned.
232 HTTP module
233 -----------
235 The http module exports the function http.generate_handler(callback) that
236 converts an HTTP handler to a "connect" handler. See file "example.lua" for an
237 example of invocation. A table with options may be passed either as a second
238 argument, or as a first argument preceeding the callback function (whichever is
239 more convenient).
241 The following options are supported:
243 - request_body_size_limit: maximum size of payload of HTTP request body
244 (transfer encoding is allowed to add a limited amount of extra data)
245 - chunk_size: optional default value for maximum_input_chunk_size and
246 minimum_output_chunk_size
247 - request_header_size_limit: maximum size of HTTP request headers
248 - maximum_input_chunk_size: maximum chunk size when streaming a request body or
249 certain POST fields (bigger chunks will be fragmented automatically)
250 - minimum_output_chunk_size: minimum size for a chunk when sending a response
251 body (smaller chunks will be buffered and concatenated with future data;
252 ignored when request:flush() is called)
253 - static_headers: a set of headers to be included in every HTTP response
254 (may be a string, a table or strings, or a table of key-value pairs)
256 The callback function receives a single request object as argument, which is
257 described below.
260 ### request.body
262 The request body (without headers) as a string. Accessing this value makes
263 further access to request.post_params and request.post_params_list, or
264 invocation of request:stream_request_body(...) impossible.
267 ### request:close_after_finish()
269 Closes the connection after answering the request.
271 This method can only be called before the HTTP response header section has been
272 finished (i.e. before request:finish_headers(), request:send_data(...), or
273 request:finish() were called), but it may be called before a status code has
274 been sent using request:send_status(...).
276 A corresponding "Connection: close" header is automatically sent.
279 ### request.cookies
281 A table with all cookies sent by the client.
284 ### request.defer_reading()
286 Disables automatic request body processing on write. Can be called before
287 sending a HTTP status code to send a response before the request has been fully
288 received.
290 CAUTION: Responding to a request before the request body has been processed may
291 lead to a deadlock if the browser does not process the response while trying to
292 send the request. Therefore, this function should only be used if:
294 - the TCP stack has enough buffer space for the response (i.e. if the response
295 is small enough), and if
296 - a timer is used to cancel the response in case of a deadlock.
298 It is recommended to not use this function unless certain performance tweaks
299 are desired.
302 ### request.faulty
304 Normally set to false. In case of a read or write error on the client
305 connection, this value is set to true before a Lua error is raised.
307 A faulty request handle must not be used, or another Lua error will be raised.
310 ### request:finish()
312 Finishes and flushes a HTTP response. May be called multiple times. An
313 HTTP status, all headers, and the response body (if applicable) must have been
314 previously sent. After calling this method, no further data may be written.
317 ### request:finish_headers()
319 Finishes and flushes the HTTP response header section. May be called multiple
320 times, as long as the request is not finished completely. This method is
321 automatically invoked if the application is beginning to send a response body.
322 After calling this method, no further headers may be sent.
325 ### request:flush()
327 Flushes any pending output data. Note: In order to mark the end of a response
328 body, it is required to call request:finish().
331 ### request.get_params
333 A table that maps field names to their corresponding GET value. If there are
334 several GET values with the given field name, then the first value is used.
336 Note: May be implemented through metamethods, but does support iteration
337 through pairs(...).
340 ### request.get_params_list
342 A table that maps field names to a sequence of their corresponding GET values.
344 Note: May be implemented through metamethods, but does support iteration
345 through pairs(...).
348 ### request.headers
350 A table that maps (case-insensitively) a HTTP header field name to a sequence
351 of values. One entry is created for every occurrence of a header line with the
352 given field name).
355 ### request.headers_csv_string
357 A table that maps (case-insensitively) a HTTP header field name to a comma
358 separated string. Multiple occurrences of the header with the given field name
359 are automatically merged into the comma separated string.
362 ### request.headers_csv_table
364 A table that maps (case-insensitively) a HTTP header field name to a sequence
365 of values. One entry is created for every comma separated value of each header
366 with the given field name.
369 ### request.headers_flags
371 A table that maps (case-insensitively) a HTTP header field name to another
372 table which (again case-insensitively) maps a string to a boolean, depending on
373 whether this string occurred in the list of comma separated values of one
374 header line with the given field name that was the key in the first table.
377 ### request.headers_value
379 A table that maps (case-insensitively) a HTTP header field name to a value. If
380 multiple header lines with the given field name have been received, false is
381 used as value.
384 ### request.method
386 The HTTP request method, e.g. "HEAD", "GET", or "POST".
389 ### request.path
391 The requested path without a leading slash and without the query part (e.g.
392 "index.html" if "/index.html?a=b&c=d" has been requested). For the query part,
393 see request.query.
395 This value will be nil if (and only if) the request method is "OPTIONS" with a
396 request target equal to "*" (see also asterisk-form of request-target in
397 section 5.3.4 in RFC 7230).
400 ### request.post_metadata
402 Only set for multipart/form-data POST requests. A table that maps field names
403 to their corresponding POST metadata table which contains two entries:
404 "file_name" and "content_type". If there are several POST values with the given
405 field name, then the first value/file is used.
408 ### request.post_metadata_list
410 Only set for multipart/form-data POST requests. A table that maps field names
411 to a sequence with their corresponding POST metadata tables. Needed if multiple
412 files are uploaded with the same field name.
415 ### request.post_params
417 A table that maps field names to their corresponding POST value. If there are
418 several POST values with the given field name, then the first value is used.
420 Note: May be implemented through metamethods, but does support iteration
421 through pairs(...).
424 ### request.post_params_list
426 A table that maps field names to a sequence of their corresponding POST values.
428 Note: May be implemented through metamethods, but does support iteration
429 through pairs(...).
432 ### request.query
434 Query part of the request target including the leading question mark, e.g.
435 "?a=b&c=d" if the requested target is "/index.html?a=b&c=d". The data is
436 automatically parsed and made available through request.get_params and
437 request.get_params_list.
439 If there is no query part given in the request target, then this string is
440 the empty string. This value will be nil if (and only if) the request method
441 is "OPTIONS" with a request target equal to "*" (see also asterisk-form of
442 request-target in section 5.3.4 in RFC 7230).
445 ### request:process_request_body()
447 Starts processing the request body (if existent) to set the values
448 request.post_params, request.post_params_list, request.post_metadata, and
449 and request.post_metadata_list and/or to call POST field stream handlers that
450 have been previously registered with request:stream_post_param(...) or
451 request:stream_post_params(...).
453 This method gets invoked automatically when the POST param tables
454 (request.post_params, etc.) are accessed, or if a response is sent (to avoid
455 deadlocks with the webbrowser). (Note: Automatic request body processing on
456 write may be disabled by calling request:defer_reading().)
458 After this method returned, all registered POST field stream handlers have
459 received all data. Registration of other POST field stream handlers is not
460 possible after this method has been called (or after request.post_params_list
461 or request.post_params have been accessed).
464 ### request:send_data(...)
466 Sends data as response body. All arguments are converted via tostring(...) and
467 concatenated. May be called multiple times until the request has been finished
468 by calling request:finish().
470 If the request method (see request.method) is "HEAD", then calls to
471 request:send_data(...) are automatically ignored.
474 ### request:send_header(key, value)
476 Sends a HTTP response header that consists of the given key and the given
477 value. Note: Key and value must be provided as separate arguments. Before any
478 headers can be sent, a HTTP status must have been set with
479 request:send_status(status_string).
482 ### request:send_status(status_string)
484 Sends a HTTP response status that is given as a string consisting of a 3-digit
485 number and an explanatory string, e.g. "200 OK" or "404 Not Found". This
486 function must be called once before any headers or response body data may be
487 sent.
490 ### request.socket
492 The underlaying socket. Can be used to force a TCP RST, etc.
495 ### request:stream_post_param(field_name, callback)
497 Registers a stream handler for the given POST parameter. The callback function
498 will be called in the following manner:
500 - For the initial chunk, the first chunk gets passed as first argument while a
501 table with metadata ("field_name" and "content_type") gets passed as second
502 argument. In case of an immediate EOF (i.e. an empty file), the passed
503 chunk is the empty string. In all other cases the chunk has a length greater
504 than zero.
505 - For any remaining chunks, the respective chunk gets passed as first and only
506 argument (no metadata). Here, the chunk has always a length greater than
507 zero.
508 - To indicate the end of the stream, the callback function is called without
509 arguments. This also happens in case of an immediate EOF (see above).
511 In case of an immediate EOF (i.e. an empty file), the callback function is thus
512 called as follows:
514 - The first time with an empty string as first argument, and with the metadata
515 as second argument.
516 - The second time without any arguments.
519 ### request:stream_post_params(pattern, callback)
521 Same as request:stream_post_param(...) but providing a string pattern to match
522 multiple field names (e.g. "^file_[0-9]+$").
525 ### request:stream_request_body(callback)
527 Start streaming of request body. For each chunk of the request body, the
528 callback function is called with the corresponding chunk. End of data is
529 indicated through return of request:stream_request_body(...) (not by calling
530 the callback without arguments).
532 The function may be called with nil instead of a callback function. In this
533 case, the request body is read and discarded. Only if nil is passed instead of
534 a callback, then the function may also be invoked when the request body has
535 already been read and/or processed. In the latter case, the function performs
536 no operation.