2 @c This is part of the GNU Guile Reference Manual.
3 @c Copyright (C) 1996, 1997, 2000, 2001, 2002, 2003, 2004, 2007, 2009,
4 @c 2010, 2011, 2013 Free Software Foundation, Inc.
5 @c See the file guile.texi for copying conditions.
8 @section Input and Output
11 * Ports:: The idea of the port abstraction.
12 * Reading:: Procedures for reading from a port.
13 * Writing:: Procedures for writing to a port.
14 * Closing:: Procedures to close a port.
15 * Random Access:: Moving around a random access port.
16 * Line/Delimited:: Read and write lines or delimited text.
17 * Block Reading and Writing:: Reading and writing blocks of text.
18 * Default Ports:: Defaults for input, output and errors.
19 * Port Types:: Types of port and how to make them.
20 * R6RS I/O Ports:: The R6RS port API.
21 * I/O Extensions:: Using and extending ports in C.
22 * BOM Handling:: Handling of Unicode byte order marks.
30 Sequential input/output in Scheme is represented by operations on a
31 @dfn{port}. This chapter explains the operations that Guile provides
32 for working with ports.
34 Ports are created by opening, for instance @code{open-file} for a file
35 (@pxref{File Ports}). Characters can be read from an input port and
36 written to an output port, or both on an input/output port. A port
37 can be closed (@pxref{Closing}) when no longer required, after which
38 any attempt to read or write is an error.
40 The formal definition of a port is very generic: an input port is
41 simply ``an object which can deliver characters on demand,'' and an
42 output port is ``an object which can accept characters.'' Because
43 this definition is so loose, it is easy to write functions that
44 simulate ports in software. @dfn{Soft ports} and @dfn{string ports}
45 are two interesting and powerful examples of this technique.
46 (@pxref{Soft Ports}, and @ref{String Ports}.)
48 Ports are garbage collected in the usual way (@pxref{Memory
49 Management}), and will be closed at that time if not already closed.
50 In this case any errors occurring in the close will not be reported.
51 Usually a program will want to explicitly close so as to be sure all
52 its operations have been successful. Of course if a program has
53 abandoned something due to an error or other condition then closing
54 problems are probably not of interest.
56 It is strongly recommended that file ports be closed explicitly when
57 no longer required. Most systems have limits on how many files can be
58 open, both on a per-process and a system-wide basis. A program that
59 uses many files should take care not to hit those limits. The same
60 applies to similar system resources such as pipes and sockets.
62 Note that automatic garbage collection is triggered only by memory
63 consumption, not by file or other resource usage, so a program cannot
64 rely on that to keep it away from system limits. An explicit call to
65 @code{gc} can of course be relied on to pick up unreferenced ports.
66 If program flow makes it hard to be certain when to close then this
67 may be an acceptable way to control resource usage.
69 All file access uses the ``LFS'' large file support functions when
70 available, so files bigger than 2 Gbytes (@math{2^31} bytes) can be
71 read and written on a 32-bit system.
73 Each port has an associated character encoding that controls how bytes
74 read from the port are converted to characters and string and controls
75 how characters and strings written to the port are converted to bytes.
76 When ports are created, they inherit their character encoding from the
77 current locale, but, that can be modified after the port is created.
79 Currently, the ports only work with @emph{non-modal} encodings. Most
80 encodings are non-modal, meaning that the conversion of bytes to a
81 string doesn't depend on its context: the same byte sequence will always
82 return the same string. A couple of modal encodings are in common use,
83 like ISO-2022-JP and ISO-2022-KR, and they are not yet supported.
85 Each port also has an associated conversion strategy: what to do when
86 a Guile character can't be converted to the port's encoded character
87 representation for output. There are three possible strategies: to
88 raise an error, to replace the character with a hex escape, or to
89 replace the character with a substitute character.
92 @deffn {Scheme Procedure} input-port? x
93 @deffnx {C Function} scm_input_port_p (x)
94 Return @code{#t} if @var{x} is an input port, otherwise return
95 @code{#f}. Any object satisfying this predicate also satisfies
100 @deffn {Scheme Procedure} output-port? x
101 @deffnx {C Function} scm_output_port_p (x)
102 Return @code{#t} if @var{x} is an output port, otherwise return
103 @code{#f}. Any object satisfying this predicate also satisfies
107 @deffn {Scheme Procedure} port? x
108 @deffnx {C Function} scm_port_p (x)
109 Return a boolean indicating whether @var{x} is a port.
110 Equivalent to @code{(or (input-port? @var{x}) (output-port?
114 @deffn {Scheme Procedure} set-port-encoding! port enc
115 @deffnx {C Function} scm_set_port_encoding_x (port, enc)
116 Sets the character encoding that will be used to interpret all port I/O.
117 @var{enc} is a string containing the name of an encoding. Valid
118 encoding names are those
119 @url{http://www.iana.org/assignments/character-sets, defined by IANA}.
122 @defvr {Scheme Variable} %default-port-encoding
123 A fluid containing @code{#f} or the name of the encoding to
124 be used by default for newly created ports (@pxref{Fluids and Dynamic
125 States}). The value @code{#f} is equivalent to @code{"ISO-8859-1"}.
127 New ports are created with the encoding appropriate for the current
128 locale if @code{setlocale} has been called or the value specified by
129 this fluid otherwise.
132 @deffn {Scheme Procedure} port-encoding port
133 @deffnx {C Function} scm_port_encoding (port)
134 Returns, as a string, the character encoding that @var{port} uses to interpret
135 its input and output. The value @code{#f} is equivalent to @code{"ISO-8859-1"}.
138 @deffn {Scheme Procedure} set-port-conversion-strategy! port sym
139 @deffnx {C Function} scm_set_port_conversion_strategy_x (port, sym)
140 Sets the behavior of the interpreter when outputting a character that
141 is not representable in the port's current encoding. @var{sym} can be
142 either @code{'error}, @code{'substitute}, or @code{'escape}. If it is
143 @code{'error}, an error will be thrown when an nonconvertible character
144 is encountered. If it is @code{'substitute}, then nonconvertible
145 characters will be replaced with approximate characters, or with
146 question marks if no approximately correct character is available. If
147 it is @code{'escape}, it will appear as a hex escape when output.
149 If @var{port} is an open port, the conversion error behavior
150 is set for that port. If it is @code{#f}, it is set as the
151 default behavior for any future ports that get created in
155 @deffn {Scheme Procedure} port-conversion-strategy port
156 @deffnx {C Function} scm_port_conversion_strategy (port)
157 Returns the behavior of the port when outputting a character that is
158 not representable in the port's current encoding. It returns the
159 symbol @code{error} if unrepresentable characters should cause
160 exceptions, @code{substitute} if the port should try to replace
161 unrepresentable characters with question marks or approximate
162 characters, or @code{escape} if unrepresentable characters should be
163 converted to string escapes.
165 If @var{port} is @code{#f}, then the current default behavior will be
166 returned. New ports will have this default behavior when they are
170 @deffn {Scheme Variable} %default-port-conversion-strategy
171 The fluid that defines the conversion strategy for newly created ports,
172 and for other conversion routines such as @code{scm_to_stringn},
173 @code{scm_from_stringn}, @code{string->pointer}, and
174 @code{pointer->string}.
176 Its value must be one of the symbols described above, with the same
177 semantics: @code{'error}, @code{'substitute}, or @code{'escape}.
179 When Guile starts, its value is @code{'substitute}.
181 Note that @code{(set-port-conversion-strategy! #f @var{sym})} is
182 equivalent to @code{(fluid-set! %default-port-conversion-strategy
191 [Generic procedures for reading from ports.]
193 These procedures pertain to reading characters and strings from
194 ports. To read general S-expressions from ports, @xref{Scheme Read}.
197 @cindex End of file object
198 @deffn {Scheme Procedure} eof-object? x
199 @deffnx {C Function} scm_eof_object_p (x)
200 Return @code{#t} if @var{x} is an end-of-file object; otherwise
205 @deffn {Scheme Procedure} char-ready? [port]
206 @deffnx {C Function} scm_char_ready_p (port)
207 Return @code{#t} if a character is ready on input @var{port}
208 and return @code{#f} otherwise. If @code{char-ready?} returns
209 @code{#t} then the next @code{read-char} operation on
210 @var{port} is guaranteed not to hang. If @var{port} is a file
211 port at end of file then @code{char-ready?} returns @code{#t}.
213 @code{char-ready?} exists to make it possible for a
214 program to accept characters from interactive ports without
215 getting stuck waiting for input. Any input editors associated
216 with such ports must make sure that characters whose existence
217 has been asserted by @code{char-ready?} cannot be rubbed out.
218 If @code{char-ready?} were to return @code{#f} at end of file,
219 a port at end of file would be indistinguishable from an
220 interactive port that has no ready characters.
224 @deffn {Scheme Procedure} read-char [port]
225 @deffnx {C Function} scm_read_char (port)
226 Return the next character available from @var{port}, updating
227 @var{port} to point to the following character. If no more
228 characters are available, the end-of-file object is returned.
230 When @var{port}'s data cannot be decoded according to its
231 character encoding, a @code{decoding-error} is raised and
232 @var{port} points past the erroneous byte sequence.
235 @deftypefn {C Function} size_t scm_c_read (SCM port, void *buffer, size_t size)
236 Read up to @var{size} bytes from @var{port} and store them in
237 @var{buffer}. The return value is the number of bytes actually read,
238 which can be less than @var{size} if end-of-file has been reached.
240 Note that this function does not update @code{port-line} and
241 @code{port-column} below.
245 @deffn {Scheme Procedure} peek-char [port]
246 @deffnx {C Function} scm_peek_char (port)
247 Return the next character available from @var{port},
248 @emph{without} updating @var{port} to point to the following
249 character. If no more characters are available, the
250 end-of-file object is returned.
252 The value returned by
253 a call to @code{peek-char} is the same as the value that would
254 have been returned by a call to @code{read-char} on the same
255 port. The only difference is that the very next call to
256 @code{read-char} or @code{peek-char} on that @var{port} will
257 return the value returned by the preceding call to
258 @code{peek-char}. In particular, a call to @code{peek-char} on
259 an interactive port will hang waiting for input whenever a call
260 to @code{read-char} would have hung.
262 As for @code{read-char}, a @code{decoding-error} may be raised
263 if such a situation occurs. However, unlike with @code{read-char},
264 @var{port} still points at the beginning of the erroneous byte
265 sequence when the error is raised.
268 @deffn {Scheme Procedure} unread-char cobj [port]
269 @deffnx {C Function} scm_unread_char (cobj, port)
270 Place character @var{cobj} in @var{port} so that it will be read by the
271 next read operation. If called multiple times, the unread characters
272 will be read again in last-in first-out order. If @var{port} is
273 not supplied, the current input port is used.
276 @deffn {Scheme Procedure} unread-string str port
277 @deffnx {C Function} scm_unread_string (str, port)
278 Place the string @var{str} in @var{port} so that its characters will
279 be read from left-to-right as the next characters from @var{port}
280 during subsequent read operations. If called multiple times, the
281 unread characters will be read again in last-in first-out order. If
282 @var{port} is not supplied, the @code{current-input-port} is used.
285 @deffn {Scheme Procedure} drain-input port
286 @deffnx {C Function} scm_drain_input (port)
287 This procedure clears a port's input buffers, similar
288 to the way that force-output clears the output buffer. The
289 contents of the buffers are returned as a single string, e.g.,
292 (define p (open-input-file ...))
293 (drain-input p) => empty string, nothing buffered yet.
294 (unread-char (read-char p) p)
295 (drain-input p) => initial chars from p, up to the buffer size.
298 Draining the buffers may be useful for cleanly finishing
299 buffered I/O so that the file descriptor can be used directly
303 @deffn {Scheme Procedure} port-column port
304 @deffnx {Scheme Procedure} port-line port
305 @deffnx {C Function} scm_port_column (port)
306 @deffnx {C Function} scm_port_line (port)
307 Return the current column number or line number of @var{port}.
309 unknown, the result is #f. Otherwise, the result is a 0-origin integer
310 - i.e.@: the first character of the first line is line 0, column 0.
311 (However, when you display a file position, for example in an error
312 message, we recommend you add 1 to get 1-origin integers. This is
313 because lines and column numbers traditionally start with 1, and that is
314 what non-programmers will find most natural.)
317 @deffn {Scheme Procedure} set-port-column! port column
318 @deffnx {Scheme Procedure} set-port-line! port line
319 @deffnx {C Function} scm_set_port_column_x (port, column)
320 @deffnx {C Function} scm_set_port_line_x (port, line)
321 Set the current column or line number of @var{port}.
328 [Generic procedures for writing to ports.]
330 These procedures are for writing characters and strings to
331 ports. For more information on writing arbitrary Scheme objects to
332 ports, @xref{Scheme Write}.
334 @deffn {Scheme Procedure} get-print-state port
335 @deffnx {C Function} scm_get_print_state (port)
336 Return the print state of the port @var{port}. If @var{port}
337 has no associated print state, @code{#f} is returned.
341 @deffn {Scheme Procedure} newline [port]
342 @deffnx {C Function} scm_newline (port)
343 Send a newline to @var{port}.
344 If @var{port} is omitted, send to the current output port.
347 @deffn {Scheme Procedure} port-with-print-state port [pstate]
348 @deffnx {C Function} scm_port_with_print_state (port, pstate)
349 Create a new port which behaves like @var{port}, but with an
350 included print state @var{pstate}. @var{pstate} is optional.
351 If @var{pstate} isn't supplied and @var{port} already has
352 a print state, the old print state is reused.
355 @deffn {Scheme Procedure} simple-format destination message . args
356 @deffnx {C Function} scm_simple_format (destination, message, args)
357 Write @var{message} to @var{destination}, defaulting to
358 the current output port.
359 @var{message} can contain @code{~A} (was @code{%s}) and
360 @code{~S} (was @code{%S}) escapes. When printed,
361 the escapes are replaced with corresponding members of
363 @code{~A} formats using @code{display} and @code{~S} formats
365 If @var{destination} is @code{#t}, then use the current output
366 port, if @var{destination} is @code{#f}, then return a string
367 containing the formatted text. Does not add a trailing newline.
371 @deffn {Scheme Procedure} write-char chr [port]
372 @deffnx {C Function} scm_write_char (chr, port)
373 Send character @var{chr} to @var{port}.
376 @deftypefn {C Function} void scm_c_write (SCM port, const void *buffer, size_t size)
377 Write @var{size} bytes at @var{buffer} to @var{port}.
379 Note that this function does not update @code{port-line} and
380 @code{port-column} (@pxref{Reading}).
384 @deffn {Scheme Procedure} force-output [port]
385 @deffnx {C Function} scm_force_output (port)
386 Flush the specified output port, or the current output port if @var{port}
387 is omitted. The current output buffer contents are passed to the
388 underlying port implementation (e.g., in the case of fports, the
389 data will be written to the file and the output buffer will be cleared.)
390 It has no effect on an unbuffered port.
392 The return value is unspecified.
395 @deffn {Scheme Procedure} flush-all-ports
396 @deffnx {C Function} scm_flush_all_ports ()
397 Equivalent to calling @code{force-output} on
398 all open output ports. The return value is unspecified.
404 @cindex Closing ports
407 @deffn {Scheme Procedure} close-port port
408 @deffnx {C Function} scm_close_port (port)
409 Close the specified port object. Return @code{#t} if it
410 successfully closes a port or @code{#f} if it was already
411 closed. An exception may be raised if an error occurs, for
412 example when flushing buffered output. See also @ref{Ports and
413 File Descriptors, close}, for a procedure which can close file
417 @deffn {Scheme Procedure} close-input-port port
418 @deffnx {Scheme Procedure} close-output-port port
419 @deffnx {C Function} scm_close_input_port (port)
420 @deffnx {C Function} scm_close_output_port (port)
421 @rnindex close-input-port
422 @rnindex close-output-port
423 Close the specified input or output @var{port}. An exception may be
424 raised if an error occurs while closing. If @var{port} is already
425 closed, nothing is done. The return value is unspecified.
427 See also @ref{Ports and File Descriptors, close}, for a procedure
428 which can close file descriptors.
431 @deffn {Scheme Procedure} port-closed? port
432 @deffnx {C Function} scm_port_closed_p (port)
433 Return @code{#t} if @var{port} is closed or @code{#f} if it is
439 @subsection Random Access
440 @cindex Random access, ports
441 @cindex Port, random access
443 @deffn {Scheme Procedure} seek fd_port offset whence
444 @deffnx {C Function} scm_seek (fd_port, offset, whence)
445 Sets the current position of @var{fd_port} to the integer
446 @var{offset}, which is interpreted according to the value of
449 One of the following variables should be supplied for
452 Seek from the beginning of the file.
455 Seek from the current position.
458 Seek from the end of the file.
460 If @var{fd_port} is a file descriptor, the underlying system
461 call is @code{lseek}. @var{port} may be a string port.
463 The value returned is the new position in the file. This means
464 that the current position of a port can be obtained using:
466 (seek port 0 SEEK_CUR)
470 @deffn {Scheme Procedure} ftell fd_port
471 @deffnx {C Function} scm_ftell (fd_port)
472 Return an integer representing the current position of
473 @var{fd_port}, measured from the beginning. Equivalent to:
476 (seek port 0 SEEK_CUR)
482 @deffn {Scheme Procedure} truncate-file file [length]
483 @deffnx {C Function} scm_truncate_file (file, length)
484 Truncate @var{file} to @var{length} bytes. @var{file} can be a
485 filename string, a port object, or an integer file descriptor. The
486 return value is unspecified.
488 For a port or file descriptor @var{length} can be omitted, in which
489 case the file is truncated at the current position (per @code{ftell}
492 On most systems a file can be extended by giving a length greater than
493 the current size, but this is not mandatory in the POSIX standard.
497 @subsection Line Oriented and Delimited Text
498 @cindex Line input/output
499 @cindex Port, line input/output
501 The delimited-I/O module can be accessed with:
504 (use-modules (ice-9 rdelim))
507 It can be used to read or write lines of text, or read text delimited by
508 a specified set of characters. It's similar to the @code{(scsh rdelim)}
509 module from guile-scsh, but does not use multiple values or character
510 sets and has an extra procedure @code{write-line}.
512 @c begin (scm-doc-string "rdelim.scm" "read-line")
513 @deffn {Scheme Procedure} read-line [port] [handle-delim]
514 Return a line of text from @var{port} if specified, otherwise from the
515 value returned by @code{(current-input-port)}. Under Unix, a line of text
516 is terminated by the first end-of-line character or by end-of-file.
518 If @var{handle-delim} is specified, it should be one of the following
522 Discard the terminating delimiter. This is the default, but it will
523 be impossible to tell whether the read terminated with a delimiter or
526 Append the terminating delimiter (if any) to the returned string.
528 Push the terminating delimiter (if any) back on to the port.
530 Return a pair containing the string read from the port and the
531 terminating delimiter or end-of-file object.
534 Like @code{read-char}, this procedure can throw to @code{decoding-error}
535 (@pxref{Reading, @code{read-char}}).
538 @c begin (scm-doc-string "rdelim.scm" "read-line!")
539 @deffn {Scheme Procedure} read-line! buf [port]
540 Read a line of text into the supplied string @var{buf} and return the
541 number of characters added to @var{buf}. If @var{buf} is filled, then
542 @code{#f} is returned.
543 Read from @var{port} if
544 specified, otherwise from the value returned by @code{(current-input-port)}.
547 @c begin (scm-doc-string "rdelim.scm" "read-delimited")
548 @deffn {Scheme Procedure} read-delimited delims [port] [handle-delim]
549 Read text until one of the characters in the string @var{delims} is found
550 or end-of-file is reached. Read from @var{port} if supplied, otherwise
551 from the value returned by @code{(current-input-port)}.
552 @var{handle-delim} takes the same values as described for @code{read-line}.
555 @c begin (scm-doc-string "rdelim.scm" "read-delimited!")
556 @deffn {Scheme Procedure} read-delimited! delims buf [port] [handle-delim] [start] [end]
557 Read text into the supplied string @var{buf}.
559 If a delimiter was found, return the number of characters written,
560 except if @var{handle-delim} is @code{split}, in which case the return
561 value is a pair, as noted above.
563 As a special case, if @var{port} was already at end-of-stream, the EOF
564 object is returned. Also, if no characters were written because the
565 buffer was full, @code{#f} is returned.
567 It's something of a wacky interface, to be honest.
570 @deffn {Scheme Procedure} write-line obj [port]
571 @deffnx {C Function} scm_write_line (obj, port)
572 Display @var{obj} and a newline character to @var{port}. If
573 @var{port} is not specified, @code{(current-output-port)} is
574 used. This function is equivalent to:
581 In the past, Guile did not have a procedure that would just read out all
582 of the characters from a port. As a workaround, many people just called
583 @code{read-delimited} with no delimiters, knowing that would produce the
584 behavior they wanted. This prompted Guile developers to add some
585 routines that would read all characters from a port. So it is that
586 @code{(ice-9 rdelim)} is also the home for procedures that can reading
589 @deffn {Scheme Procedure} read-string [port] [count]
590 Read all of the characters out of @var{port} and return them as a
591 string. If the @var{count} is present, treat it as a limit to the
592 number of characters to read.
594 By default, read from the current input port, with no size limit on the
595 result. This procedure always returns a string, even if no characters
599 @deffn {Scheme Procedure} read-string! buf [port] [start] [end]
600 Fill @var{buf} with characters read from @var{port}, defaulting to the
601 current input port. Return the number of characters read.
603 If @var{start} or @var{end} are specified, store data only into the
604 substring of @var{str} bounded by @var{start} and @var{end} (which
605 default to the beginning and end of the string, respectively).
608 Some of the aforementioned I/O functions rely on the following C
609 primitives. These will mainly be of interest to people hacking Guile
612 @deffn {Scheme Procedure} %read-delimited! delims str gobble [port [start [end]]]
613 @deffnx {C Function} scm_read_delimited_x (delims, str, gobble, port, start, end)
614 Read characters from @var{port} into @var{str} until one of the
615 characters in the @var{delims} string is encountered. If
616 @var{gobble} is true, discard the delimiter character;
617 otherwise, leave it in the input stream for the next read. If
618 @var{port} is not specified, use the value of
619 @code{(current-input-port)}. If @var{start} or @var{end} are
620 specified, store data only into the substring of @var{str}
621 bounded by @var{start} and @var{end} (which default to the
622 beginning and end of the string, respectively).
624 Return a pair consisting of the delimiter that terminated the
625 string and the number of characters read. If reading stopped
626 at the end of file, the delimiter returned is the
627 @var{eof-object}; if the string was filled without encountering
628 a delimiter, this value is @code{#f}.
631 @deffn {Scheme Procedure} %read-line [port]
632 @deffnx {C Function} scm_read_line (port)
633 Read a newline-terminated line from @var{port}, allocating storage as
634 necessary. The newline terminator (if any) is removed from the string,
635 and a pair consisting of the line and its delimiter is returned. The
636 delimiter may be either a newline or the @var{eof-object}; if
637 @code{%read-line} is called at the end of file, it returns the pair
638 @code{(#<eof> . #<eof>)}.
641 @node Block Reading and Writing
642 @subsection Block reading and writing
643 @cindex Block read/write
644 @cindex Port, block read/write
646 The Block-string-I/O module can be accessed with:
649 (use-modules (ice-9 rw))
652 It currently contains procedures that help to implement the
653 @code{(scsh rw)} module in guile-scsh.
655 @deffn {Scheme Procedure} read-string!/partial str [port_or_fdes [start [end]]]
656 @deffnx {C Function} scm_read_string_x_partial (str, port_or_fdes, start, end)
657 Read characters from a port or file descriptor into a
658 string @var{str}. A port must have an underlying file
659 descriptor --- a so-called fport. This procedure is
660 scsh-compatible and can efficiently read large strings.
665 attempt to fill the entire string, unless the @var{start}
666 and/or @var{end} arguments are supplied. i.e., @var{start}
667 defaults to 0 and @var{end} defaults to
668 @code{(string-length str)}
670 use the current input port if @var{port_or_fdes} is not
673 return fewer than the requested number of characters in some
674 cases, e.g., on end of file, if interrupted by a signal, or if
675 not all the characters are immediately available.
677 wait indefinitely for some input if no characters are
679 unless the port is in non-blocking mode.
681 read characters from the port's input buffers if available,
682 instead from the underlying file descriptor.
684 return @code{#f} if end-of-file is encountered before reading
685 any characters, otherwise return the number of characters
688 return 0 if the port is in non-blocking mode and no characters
689 are immediately available.
691 return 0 if the request is for 0 bytes, with no
696 @deffn {Scheme Procedure} write-string/partial str [port_or_fdes [start [end]]]
697 @deffnx {C Function} scm_write_string_partial (str, port_or_fdes, start, end)
698 Write characters from a string @var{str} to a port or file
699 descriptor. A port must have an underlying file descriptor
700 --- a so-called fport. This procedure is
701 scsh-compatible and can efficiently write large strings.
706 attempt to write the entire string, unless the @var{start}
707 and/or @var{end} arguments are supplied. i.e., @var{start}
708 defaults to 0 and @var{end} defaults to
709 @code{(string-length str)}
711 use the current output port if @var{port_of_fdes} is not
714 in the case of a buffered port, store the characters in the
715 port's output buffer, if all will fit. If they will not fit
716 then any existing buffered characters will be flushed
718 to write the new characters directly to the underlying file
719 descriptor. If the port is in non-blocking mode and
720 buffered characters can not be flushed immediately, then an
721 @code{EAGAIN} system-error exception will be raised (Note:
722 scsh does not support the use of non-blocking buffered ports.)
724 write fewer than the requested number of
725 characters in some cases, e.g., if interrupted by a signal or
726 if not all of the output can be accepted immediately.
728 wait indefinitely for at least one character
729 from @var{str} to be accepted by the port, unless the port is
730 in non-blocking mode.
732 return the number of characters accepted by the port.
734 return 0 if the port is in non-blocking mode and can not accept
735 at least one character from @var{str} immediately
737 return 0 immediately if the request size is 0 bytes.
742 @subsection Default Ports for Input, Output and Errors
743 @cindex Default ports
744 @cindex Port, default
746 @rnindex current-input-port
747 @deffn {Scheme Procedure} current-input-port
748 @deffnx {C Function} scm_current_input_port ()
749 @cindex standard input
750 Return the current input port. This is the default port used
751 by many input procedures.
753 Initially this is the @dfn{standard input} in Unix and C terminology.
754 When the standard input is a tty the port is unbuffered, otherwise
757 Unbuffered input is good if an application runs an interactive
758 subprocess, since any type-ahead input won't go into Guile's buffer
759 and be unavailable to the subprocess.
761 Note that Guile buffering is completely separate from the tty ``line
762 discipline''. In the usual cooked mode on a tty Guile only sees a
763 line of input once the user presses @key{Return}.
766 @rnindex current-output-port
767 @deffn {Scheme Procedure} current-output-port
768 @deffnx {C Function} scm_current_output_port ()
769 @cindex standard output
770 Return the current output port. This is the default port used
771 by many output procedures.
773 Initially this is the @dfn{standard output} in Unix and C terminology.
774 When the standard output is a tty this port is unbuffered, otherwise
777 Unbuffered output to a tty is good for ensuring progress output or a
778 prompt is seen. But an application which always prints whole lines
779 could change to line buffered, or an application with a lot of output
780 could go fully buffered and perhaps make explicit @code{force-output}
781 calls (@pxref{Writing}) at selected points.
784 @deffn {Scheme Procedure} current-error-port
785 @deffnx {C Function} scm_current_error_port ()
786 @cindex standard error output
787 Return the port to which errors and warnings should be sent.
789 Initially this is the @dfn{standard error} in Unix and C terminology.
790 When the standard error is a tty this port is unbuffered, otherwise
794 @deffn {Scheme Procedure} set-current-input-port port
795 @deffnx {Scheme Procedure} set-current-output-port port
796 @deffnx {Scheme Procedure} set-current-error-port port
797 @deffnx {C Function} scm_set_current_input_port (port)
798 @deffnx {C Function} scm_set_current_output_port (port)
799 @deffnx {C Function} scm_set_current_error_port (port)
800 Change the ports returned by @code{current-input-port},
801 @code{current-output-port} and @code{current-error-port}, respectively,
802 so that they use the supplied @var{port} for input or output.
805 @deftypefn {C Function} void scm_dynwind_current_input_port (SCM port)
806 @deftypefnx {C Function} void scm_dynwind_current_output_port (SCM port)
807 @deftypefnx {C Function} void scm_dynwind_current_error_port (SCM port)
808 These functions must be used inside a pair of calls to
809 @code{scm_dynwind_begin} and @code{scm_dynwind_end} (@pxref{Dynamic
810 Wind}). During the dynwind context, the indicated port is set to
813 More precisely, the current port is swapped with a `backup' value
814 whenever the dynwind context is entered or left. The backup value is
815 initialized with the @var{port} argument.
819 @subsection Types of Port
820 @cindex Types of ports
823 [Types of port; how to make them.]
826 * File Ports:: Ports on an operating system file.
827 * String Ports:: Ports on a Scheme string.
828 * Soft Ports:: Ports on arbitrary Scheme procedures.
829 * Void Ports:: Ports on nothing at all.
834 @subsubsection File Ports
838 The following procedures are used to open file ports.
839 See also @ref{Ports and File Descriptors, open}, for an interface
840 to the Unix @code{open} system call.
842 Most systems have limits on how many files can be open, so it's
843 strongly recommended that file ports be closed explicitly when no
844 longer required (@pxref{Ports}).
846 @deffn {Scheme Procedure} open-file filename mode @
847 [#:guess-encoding=#f] [#:encoding=#f]
848 @deffnx {C Function} scm_open_file_with_encoding @
849 (filename, mode, guess_encoding, encoding)
850 @deffnx {C Function} scm_open_file (filename, mode)
851 Open the file whose name is @var{filename}, and return a port
852 representing that file. The attributes of the port are
853 determined by the @var{mode} string. The way in which this is
854 interpreted is similar to C stdio. The first character must be
855 one of the following:
859 Open an existing file for input.
861 Open a file for output, creating it if it doesn't already exist
862 or removing its contents if it does.
864 Open a file for output, creating it if it doesn't already
865 exist. All writes to the port will go to the end of the file.
866 The "append mode" can be turned off while the port is in use
867 @pxref{Ports and File Descriptors, fcntl}
870 The following additional characters can be appended:
874 Open the port for both input and output. E.g., @code{r+}: open
875 an existing file for both input and output.
877 Create an "unbuffered" port. In this case input and output
878 operations are passed directly to the underlying port
879 implementation without additional buffering. This is likely to
880 slow down I/O operations. The buffering mode can be changed
881 while a port is in use @pxref{Ports and File Descriptors,
884 Add line-buffering to the port. The port output buffer will be
885 automatically flushed whenever a newline character is written.
887 Use binary mode, ensuring that each byte in the file will be read as one
890 To provide this property, the file will be opened with the 8-bit
891 character encoding "ISO-8859-1", ignoring the default port encoding.
892 @xref{Ports}, for more information on port encodings.
894 Note that while it is possible to read and write binary data as
895 characters or strings, it is usually better to treat bytes as octets,
896 and byte sequences as bytevectors. @xref{R6RS Binary Input}, and
897 @ref{R6RS Binary Output}, for more.
899 This option had another historical meaning, for DOS compatibility: in
900 the default (textual) mode, DOS reads a CR-LF sequence as one LF byte.
901 The @code{b} flag prevents this from happening, adding @code{O_BINARY}
902 to the underlying @code{open} call. Still, the flag is generally useful
903 because of its port encoding ramifications.
906 Unless binary mode is requested, the character encoding of the new port
907 is determined as follows: First, if @var{guess-encoding} is true, the
908 @code{file-encoding} procedure is used to guess the encoding of the file
909 (@pxref{Character Encoding of Source Files}). If @var{guess-encoding}
910 is false or if @code{file-encoding} fails, @var{encoding} is used unless
911 it is also false. As a last resort, the default port encoding is used.
912 @xref{Ports}, for more information on port encodings. It is an error to
913 pass a non-false @var{guess-encoding} or @var{encoding} if binary mode
916 If a file cannot be opened with the access requested, @code{open-file}
919 When the file is opened, its encoding is set to the current
920 @code{%default-port-encoding}, unless the @code{b} flag was supplied.
921 Sometimes it is desirable to honor Emacs-style coding declarations in
922 files@footnote{Guile 2.0.0 to 2.0.7 would do this by default. This
923 behavior was deemed inappropriate and disabled starting from Guile
924 2.0.8.}. When that is the case, the @code{file-encoding} procedure can
925 be used as follows (@pxref{Character Encoding of Source Files,
926 @code{file-encoding}}):
929 (let* ((port (open-input-file file))
930 (encoding (file-encoding port)))
931 (set-port-encoding! port (or encoding (port-encoding port))))
934 In theory we could create read/write ports which were buffered
935 in one direction only. However this isn't included in the
939 @rnindex open-input-file
940 @deffn {Scheme Procedure} open-input-file filename @
941 [#:guess-encoding=#f] [#:encoding=#f] [#:binary=#f]
943 Open @var{filename} for input. If @var{binary} is true, open the port
944 in binary mode, otherwise use text mode. @var{encoding} and
945 @var{guess-encoding} determine the character encoding as described above
946 for @code{open-file}. Equivalent to
948 (open-file @var{filename}
949 (if @var{binary} "rb" "r")
950 #:guess-encoding @var{guess-encoding}
951 #:encoding @var{encoding})
955 @rnindex open-output-file
956 @deffn {Scheme Procedure} open-output-file filename @
957 [#:encoding=#f] [#:binary=#f]
959 Open @var{filename} for output. If @var{binary} is true, open the port
960 in binary mode, otherwise use text mode. @var{encoding} specifies the
961 character encoding as described above for @code{open-file}. Equivalent
964 (open-file @var{filename}
965 (if @var{binary} "wb" "w")
966 #:encoding @var{encoding})
970 @deffn {Scheme Procedure} call-with-input-file filename proc @
971 [#:guess-encoding=#f] [#:encoding=#f] [#:binary=#f]
972 @deffnx {Scheme Procedure} call-with-output-file filename proc @
973 [#:encoding=#f] [#:binary=#f]
974 @rnindex call-with-input-file
975 @rnindex call-with-output-file
976 Open @var{filename} for input or output, and call @code{(@var{proc}
977 port)} with the resulting port. Return the value returned by
978 @var{proc}. @var{filename} is opened as per @code{open-input-file} or
979 @code{open-output-file} respectively, and an error is signaled if it
982 When @var{proc} returns, the port is closed. If @var{proc} does not
983 return (e.g.@: if it throws an error), then the port might not be
984 closed automatically, though it will be garbage collected in the usual
985 way if not otherwise referenced.
988 @deffn {Scheme Procedure} with-input-from-file filename thunk @
989 [#:guess-encoding=#f] [#:encoding=#f] [#:binary=#f]
990 @deffnx {Scheme Procedure} with-output-to-file filename thunk @
991 [#:encoding=#f] [#:binary=#f]
992 @deffnx {Scheme Procedure} with-error-to-file filename thunk @
993 [#:encoding=#f] [#:binary=#f]
994 @rnindex with-input-from-file
995 @rnindex with-output-to-file
996 Open @var{filename} and call @code{(@var{thunk})} with the new port
997 setup as respectively the @code{current-input-port},
998 @code{current-output-port}, or @code{current-error-port}. Return the
999 value returned by @var{thunk}. @var{filename} is opened as per
1000 @code{open-input-file} or @code{open-output-file} respectively, and an
1001 error is signaled if it cannot be opened.
1003 When @var{thunk} returns, the port is closed and the previous setting
1004 of the respective current port is restored.
1006 The current port setting is managed with @code{dynamic-wind}, so the
1007 previous value is restored no matter how @var{thunk} exits (eg.@: an
1008 exception), and if @var{thunk} is re-entered (via a captured
1009 continuation) then it's set again to the @var{filename} port.
1011 The port is closed when @var{thunk} returns normally, but not when
1012 exited via an exception or new continuation. This ensures it's still
1013 ready for use if @var{thunk} is re-entered by a captured continuation.
1014 Of course the port is always garbage collected and closed in the usual
1015 way when no longer referenced anywhere.
1018 @deffn {Scheme Procedure} port-mode port
1019 @deffnx {C Function} scm_port_mode (port)
1020 Return the port modes associated with the open port @var{port}.
1021 These will not necessarily be identical to the modes used when
1022 the port was opened, since modes such as "append" which are
1023 used only during port creation are not retained.
1026 @deffn {Scheme Procedure} port-filename port
1027 @deffnx {C Function} scm_port_filename (port)
1028 Return the filename associated with @var{port}, or @code{#f} if no
1029 filename is associated with the port.
1031 @var{port} must be open, @code{port-filename} cannot be used once the
1035 @deffn {Scheme Procedure} set-port-filename! port filename
1036 @deffnx {C Function} scm_set_port_filename_x (port, filename)
1037 Change the filename associated with @var{port}, using the current input
1038 port if none is specified. Note that this does not change the port's
1039 source of data, but only the value that is returned by
1040 @code{port-filename} and reported in diagnostic output.
1043 @deffn {Scheme Procedure} file-port? obj
1044 @deffnx {C Function} scm_file_port_p (obj)
1045 Determine whether @var{obj} is a port that is related to a file.
1050 @subsubsection String Ports
1052 @cindex Port, string
1054 The following allow string ports to be opened by analogy to R4RS
1055 file port facilities:
1057 With string ports, the port-encoding is treated differently than other
1058 types of ports. When string ports are created, they do not inherit a
1059 character encoding from the current locale. They are given a
1060 default locale that allows them to handle all valid string characters.
1061 Typically one should not modify a string port's character encoding
1062 away from its default.
1064 @deffn {Scheme Procedure} call-with-output-string proc
1065 @deffnx {C Function} scm_call_with_output_string (proc)
1066 Calls the one-argument procedure @var{proc} with a newly created output
1067 port. When the function returns, the string composed of the characters
1068 written into the port is returned. @var{proc} should not close the port.
1071 @deffn {Scheme Procedure} call-with-input-string string proc
1072 @deffnx {C Function} scm_call_with_input_string (string, proc)
1073 Calls the one-argument procedure @var{proc} with a newly
1074 created input port from which @var{string}'s contents may be
1075 read. The value yielded by the @var{proc} is returned.
1078 @deffn {Scheme Procedure} with-output-to-string thunk
1079 Calls the zero-argument procedure @var{thunk} with the current output
1080 port set temporarily to a new string port. It returns a string
1081 composed of the characters written to the current output.
1084 @deffn {Scheme Procedure} with-input-from-string string thunk
1085 Calls the zero-argument procedure @var{thunk} with the current input
1086 port set temporarily to a string port opened on the specified
1087 @var{string}. The value yielded by @var{thunk} is returned.
1090 @deffn {Scheme Procedure} open-input-string str
1091 @deffnx {C Function} scm_open_input_string (str)
1092 Take a string and return an input port that delivers characters
1093 from the string. The port can be closed by
1094 @code{close-input-port}, though its storage will be reclaimed
1095 by the garbage collector if it becomes inaccessible.
1098 @deffn {Scheme Procedure} open-output-string
1099 @deffnx {C Function} scm_open_output_string ()
1100 Return an output port that will accumulate characters for
1101 retrieval by @code{get-output-string}. The port can be closed
1102 by the procedure @code{close-output-port}, though its storage
1103 will be reclaimed by the garbage collector if it becomes
1107 @deffn {Scheme Procedure} get-output-string port
1108 @deffnx {C Function} scm_get_output_string (port)
1109 Given an output port created by @code{open-output-string},
1110 return a string consisting of the characters that have been
1111 output to the port so far.
1113 @code{get-output-string} must be used before closing @var{port}, once
1114 closed the string cannot be obtained.
1117 A string port can be used in many procedures which accept a port
1118 but which are not dependent on implementation details of fports.
1119 E.g., seeking and truncating will work on a string port,
1120 but trying to extract the file descriptor number will fail.
1124 @subsubsection Soft Ports
1128 A @dfn{soft-port} is a port based on a vector of procedures capable of
1129 accepting or delivering characters. It allows emulation of I/O ports.
1131 @deffn {Scheme Procedure} make-soft-port pv modes
1132 @deffnx {C Function} scm_make_soft_port (pv, modes)
1133 Return a port capable of receiving or delivering characters as
1134 specified by the @var{modes} string (@pxref{File Ports,
1135 open-file}). @var{pv} must be a vector of length 5 or 6. Its
1136 components are as follows:
1140 procedure accepting one character for output
1142 procedure accepting a string for output
1144 thunk for flushing output
1146 thunk for getting one character
1148 thunk for closing port (not by garbage collection)
1150 (if present and not @code{#f}) thunk for computing the number of
1151 characters that can be read from the port without blocking.
1154 For an output-only port only elements 0, 1, 2, and 4 need be
1155 procedures. For an input-only port only elements 3 and 4 need
1156 be procedures. Thunks 2 and 4 can instead be @code{#f} if
1157 there is no useful operation for them to perform.
1159 If thunk 3 returns @code{#f} or an @code{eof-object}
1160 (@pxref{Input, eof-object?, ,r5rs, The Revised^5 Report on
1161 Scheme}) it indicates that the port has reached end-of-file.
1165 (define stdout (current-output-port))
1166 (define p (make-soft-port
1168 (lambda (c) (write c stdout))
1169 (lambda (s) (display s stdout))
1170 (lambda () (display "." stdout))
1171 (lambda () (char-upcase (read-char)))
1172 (lambda () (display "@@" stdout)))
1175 (write p p) @result{} #<input-output: soft 8081e20>
1181 @subsubsection Void Ports
1185 This kind of port causes any data to be discarded when written to, and
1186 always returns the end-of-file object when read from.
1188 @deffn {Scheme Procedure} %make-void-port mode
1189 @deffnx {C Function} scm_sys_make_void_port (mode)
1190 Create and return a new void port. A void port acts like
1191 @file{/dev/null}. The @var{mode} argument
1192 specifies the input/output modes for this port: see the
1193 documentation for @code{open-file} in @ref{File Ports}.
1197 @node R6RS I/O Ports
1198 @subsection R6RS I/O Ports
1203 The I/O port API of the @uref{http://www.r6rs.org/, Revised Report^6 on
1204 the Algorithmic Language Scheme (R6RS)} is provided by the @code{(rnrs
1205 io ports)} module. It provides features, such as binary I/O and Unicode
1206 string I/O, that complement or refine Guile's historical port API
1207 presented above (@pxref{Input and Output}). Note that R6RS ports are not
1208 disjoint from Guile's native ports, so Guile-specific procedures will
1209 work on ports created using the R6RS API, and vice versa.
1211 The text in this section is taken from the R6RS standard libraries
1212 document, with only minor adaptions for inclusion in this manual. The
1213 Guile developers offer their thanks to the R6RS editors for having
1214 provided the report's text under permissive conditions making this
1217 @c FIXME: Update description when implemented.
1218 @emph{Note}: The implementation of this R6RS API is not complete yet.
1221 * R6RS File Names:: File names.
1222 * R6RS File Options:: Options for opening files.
1223 * R6RS Buffer Modes:: Influencing buffering behavior.
1224 * R6RS Transcoders:: Influencing port encoding.
1225 * R6RS End-of-File:: The end-of-file object.
1226 * R6RS Port Manipulation:: Manipulating R6RS ports.
1227 * R6RS Input Ports:: Input Ports.
1228 * R6RS Binary Input:: Binary input.
1229 * R6RS Textual Input:: Textual input.
1230 * R6RS Output Ports:: Output Ports.
1231 * R6RS Binary Output:: Binary output.
1232 * R6RS Textual Output:: Textual output.
1235 A subset of the @code{(rnrs io ports)} module, plus one non-standard
1236 procedure @code{unget-bytevector} (@pxref{R6RS Binary Input}), is
1237 provided by the @code{(ice-9 binary-ports)} module. It contains binary
1238 input/output procedures and does not rely on R6RS support.
1240 @node R6RS File Names
1241 @subsubsection File Names
1243 Some of the procedures described in this chapter accept a file name as an
1244 argument. Valid values for such a file name include strings that name a file
1245 using the native notation of file system paths on an implementation's
1246 underlying operating system, and may include implementation-dependent
1249 A @var{filename} parameter name means that the
1250 corresponding argument must be a file name.
1252 @node R6RS File Options
1253 @subsubsection File Options
1254 @cindex file options
1256 When opening a file, the various procedures in this library accept a
1257 @code{file-options} object that encapsulates flags to specify how the
1258 file is to be opened. A @code{file-options} object is an enum-set
1259 (@pxref{rnrs enums}) over the symbols constituting valid file options.
1261 A @var{file-options} parameter name means that the corresponding
1262 argument must be a file-options object.
1264 @deffn {Scheme Syntax} file-options @var{file-options-symbol} ...
1266 Each @var{file-options-symbol} must be a symbol.
1268 The @code{file-options} syntax returns a file-options object that
1269 encapsulates the specified options.
1271 When supplied to an operation that opens a file for output, the
1272 file-options object returned by @code{(file-options)} specifies that the
1273 file is created if it does not exist and an exception with condition
1274 type @code{&i/o-file-already-exists} is raised if it does exist. The
1275 following standard options can be included to modify the default
1280 If the file does not already exist, it is not created;
1281 instead, an exception with condition type @code{&i/o-file-does-not-exist}
1283 If the file already exists, the exception with condition type
1284 @code{&i/o-file-already-exists} is not raised
1285 and the file is truncated to zero length.
1287 If the file already exists, the exception with condition type
1288 @code{&i/o-file-already-exists} is not raised,
1289 even if @code{no-create} is not included,
1290 and the file is truncated to zero length.
1292 If the file already exists and the exception with condition type
1293 @code{&i/o-file-already-exists} has been inhibited by inclusion of
1294 @code{no-create} or @code{no-fail}, the file is not truncated, but
1295 the port's current position is still set to the beginning of the
1299 These options have no effect when a file is opened only for input.
1300 Symbols other than those listed above may be used as
1301 @var{file-options-symbol}s; they have implementation-specific meaning,
1305 Only the name of @var{file-options-symbol} is significant.
1309 @node R6RS Buffer Modes
1310 @subsubsection Buffer Modes
1312 Each port has an associated buffer mode. For an output port, the
1313 buffer mode defines when an output operation flushes the buffer
1314 associated with the output port. For an input port, the buffer mode
1315 defines how much data will be read to satisfy read operations. The
1316 possible buffer modes are the symbols @code{none} for no buffering,
1317 @code{line} for flushing upon line endings and reading up to line
1318 endings, or other implementation-dependent behavior,
1319 and @code{block} for arbitrary buffering. This section uses
1320 the parameter name @var{buffer-mode} for arguments that must be
1321 buffer-mode symbols.
1323 If two ports are connected to the same mutable source, both ports
1324 are unbuffered, and reading a byte or character from that shared
1325 source via one of the two ports would change the bytes or characters
1326 seen via the other port, a lookahead operation on one port will
1327 render the peeked byte or character inaccessible via the other port,
1328 while a subsequent read operation on the peeked port will see the
1329 peeked byte or character even though the port is otherwise unbuffered.
1331 In other words, the semantics of buffering is defined in terms of side
1332 effects on shared mutable sources, and a lookahead operation has the
1333 same side effect on the shared source as a read operation.
1335 @deffn {Scheme Syntax} buffer-mode @var{buffer-mode-symbol}
1337 @var{buffer-mode-symbol} must be a symbol whose name is one of
1338 @code{none}, @code{line}, and @code{block}. The result is the
1339 corresponding symbol, and specifies the associated buffer mode.
1342 Only the name of @var{buffer-mode-symbol} is significant.
1346 @deffn {Scheme Procedure} buffer-mode? obj
1347 Returns @code{#t} if the argument is a valid buffer-mode symbol, and
1348 returns @code{#f} otherwise.
1351 @node R6RS Transcoders
1352 @subsubsection Transcoders
1354 @cindex end-of-line style
1357 @cindex textual port
1359 Several different Unicode encoding schemes describe standard ways to
1360 encode characters and strings as byte sequences and to decode those
1361 sequences. Within this document, a @dfn{codec} is an immutable Scheme
1362 object that represents a Unicode or similar encoding scheme.
1364 An @dfn{end-of-line style} is a symbol that, if it is not @code{none},
1365 describes how a textual port transcodes representations of line endings.
1367 A @dfn{transcoder} is an immutable Scheme object that combines a codec
1368 with an end-of-line style and a method for handling decoding errors.
1369 Each transcoder represents some specific bidirectional (but not
1370 necessarily lossless), possibly stateful translation between byte
1371 sequences and Unicode characters and strings. Every transcoder can
1372 operate in the input direction (bytes to characters) or in the output
1373 direction (characters to bytes). A @var{transcoder} parameter name
1374 means that the corresponding argument must be a transcoder.
1376 A @dfn{binary port} is a port that supports binary I/O, does not have an
1377 associated transcoder and does not support textual I/O. A @dfn{textual
1378 port} is a port that supports textual I/O, and does not support binary
1379 I/O. A textual port may or may not have an associated transcoder.
1381 @deffn {Scheme Procedure} latin-1-codec
1382 @deffnx {Scheme Procedure} utf-8-codec
1383 @deffnx {Scheme Procedure} utf-16-codec
1385 These are predefined codecs for the ISO 8859-1, UTF-8, and UTF-16
1388 A call to any of these procedures returns a value that is equal in the
1389 sense of @code{eqv?} to the result of any other call to the same
1393 @deffn {Scheme Syntax} eol-style @var{eol-style-symbol}
1395 @var{eol-style-symbol} should be a symbol whose name is one of
1396 @code{lf}, @code{cr}, @code{crlf}, @code{nel}, @code{crnel}, @code{ls},
1399 The form evaluates to the corresponding symbol. If the name of
1400 @var{eol-style-symbol} is not one of these symbols, the effect and
1401 result are implementation-dependent; in particular, the result may be an
1402 eol-style symbol acceptable as an @var{eol-style} argument to
1403 @code{make-transcoder}. Otherwise, an exception is raised.
1405 All eol-style symbols except @code{none} describe a specific
1406 line-ending encoding:
1414 carriage return, linefeed
1418 carriage return, next line
1423 For a textual port with a transcoder, and whose transcoder has an
1424 eol-style symbol @code{none}, no conversion occurs. For a textual input
1425 port, any eol-style symbol other than @code{none} means that all of the
1426 above line-ending encodings are recognized and are translated into a
1427 single linefeed. For a textual output port, @code{none} and @code{lf}
1428 are equivalent. Linefeed characters are encoded according to the
1429 specified eol-style symbol, and all other characters that participate in
1430 possible line endings are encoded as is.
1433 Only the name of @var{eol-style-symbol} is significant.
1437 @deffn {Scheme Procedure} native-eol-style
1438 Returns the default end-of-line style of the underlying platform, e.g.,
1439 @code{lf} on Unix and @code{crlf} on Windows.
1442 @deffn {Condition Type} &i/o-decoding
1443 @deffnx {Scheme Procedure} make-i/o-decoding-error port
1444 @deffnx {Scheme Procedure} i/o-decoding-error? obj
1446 This condition type could be defined by
1449 (define-condition-type &i/o-decoding &i/o-port
1450 make-i/o-decoding-error i/o-decoding-error?)
1453 An exception with this type is raised when one of the operations for
1454 textual input from a port encounters a sequence of bytes that cannot be
1455 translated into a character or string by the input direction of the
1458 When such an exception is raised, the port's position is past the
1462 @deffn {Condition Type} &i/o-encoding
1463 @deffnx {Scheme Procedure} make-i/o-encoding-error port char
1464 @deffnx {Scheme Procedure} i/o-encoding-error? obj
1465 @deffnx {Scheme Procedure} i/o-encoding-error-char condition
1467 This condition type could be defined by
1470 (define-condition-type &i/o-encoding &i/o-port
1471 make-i/o-encoding-error i/o-encoding-error?
1472 (char i/o-encoding-error-char))
1475 An exception with this type is raised when one of the operations for
1476 textual output to a port encounters a character that cannot be
1477 translated into bytes by the output direction of the port's transcoder.
1478 @var{char} is the character that could not be encoded.
1481 @deffn {Scheme Syntax} error-handling-mode @var{error-handling-mode-symbol}
1483 @var{error-handling-mode-symbol} should be a symbol whose name is one of
1484 @code{ignore}, @code{raise}, and @code{replace}. The form evaluates to
1485 the corresponding symbol. If @var{error-handling-mode-symbol} is not
1486 one of these identifiers, effect and result are
1487 implementation-dependent: The result may be an error-handling-mode
1488 symbol acceptable as a @var{handling-mode} argument to
1489 @code{make-transcoder}. If it is not acceptable as a
1490 @var{handling-mode} argument to @code{make-transcoder}, an exception is
1494 Only the name of @var{error-handling-mode-symbol} is significant.
1497 The error-handling mode of a transcoder specifies the behavior
1498 of textual I/O operations in the presence of encoding or decoding
1501 If a textual input operation encounters an invalid or incomplete
1502 character encoding, and the error-handling mode is @code{ignore}, an
1503 appropriate number of bytes of the invalid encoding are ignored and
1504 decoding continues with the following bytes.
1506 If the error-handling mode is @code{replace}, the replacement
1507 character U+FFFD is injected into the data stream, an appropriate
1508 number of bytes are ignored, and decoding
1509 continues with the following bytes.
1511 If the error-handling mode is @code{raise}, an exception with condition
1512 type @code{&i/o-decoding} is raised.
1514 If a textual output operation encounters a character it cannot encode,
1515 and the error-handling mode is @code{ignore}, the character is ignored
1516 and encoding continues with the next character. If the error-handling
1517 mode is @code{replace}, a codec-specific replacement character is
1518 emitted by the transcoder, and encoding continues with the next
1519 character. The replacement character is U+FFFD for transcoders whose
1520 codec is one of the Unicode encodings, but is the @code{?} character
1521 for the Latin-1 encoding. If the error-handling mode is @code{raise},
1522 an exception with condition type @code{&i/o-encoding} is raised.
1525 @deffn {Scheme Procedure} make-transcoder codec
1526 @deffnx {Scheme Procedure} make-transcoder codec eol-style
1527 @deffnx {Scheme Procedure} make-transcoder codec eol-style handling-mode
1529 @var{codec} must be a codec; @var{eol-style}, if present, an eol-style
1530 symbol; and @var{handling-mode}, if present, an error-handling-mode
1533 @var{eol-style} may be omitted, in which case it defaults to the native
1534 end-of-line style of the underlying platform. @var{handling-mode} may
1535 be omitted, in which case it defaults to @code{replace}. The result is
1536 a transcoder with the behavior specified by its arguments.
1539 @deffn {Scheme procedure} native-transcoder
1540 Returns an implementation-dependent transcoder that represents a
1541 possibly locale-dependent ``native'' transcoding.
1544 @deffn {Scheme Procedure} transcoder-codec transcoder
1545 @deffnx {Scheme Procedure} transcoder-eol-style transcoder
1546 @deffnx {Scheme Procedure} transcoder-error-handling-mode transcoder
1548 These are accessors for transcoder objects; when applied to a
1549 transcoder returned by @code{make-transcoder}, they return the
1550 @var{codec}, @var{eol-style}, and @var{handling-mode} arguments,
1554 @deffn {Scheme Procedure} bytevector->string bytevector transcoder
1556 Returns the string that results from transcoding the
1557 @var{bytevector} according to the input direction of the transcoder.
1560 @deffn {Scheme Procedure} string->bytevector string transcoder
1562 Returns the bytevector that results from transcoding the
1563 @var{string} according to the output direction of the transcoder.
1566 @node R6RS End-of-File
1567 @subsubsection The End-of-File Object
1572 R5RS' @code{eof-object?} procedure is provided by the @code{(rnrs io
1575 @deffn {Scheme Procedure} eof-object? obj
1576 @deffnx {C Function} scm_eof_object_p (obj)
1577 Return true if @var{obj} is the end-of-file (EOF) object.
1580 In addition, the following procedure is provided:
1582 @deffn {Scheme Procedure} eof-object
1583 @deffnx {C Function} scm_eof_object ()
1584 Return the end-of-file (EOF) object.
1587 (eof-object? (eof-object))
1593 @node R6RS Port Manipulation
1594 @subsubsection Port Manipulation
1596 The procedures listed below operate on any kind of R6RS I/O port.
1598 @deffn {Scheme Procedure} port? obj
1599 Returns @code{#t} if the argument is a port, and returns @code{#f}
1603 @deffn {Scheme Procedure} port-transcoder port
1604 Returns the transcoder associated with @var{port} if @var{port} is
1605 textual and has an associated transcoder, and returns @code{#f} if
1606 @var{port} is binary or does not have an associated transcoder.
1609 @deffn {Scheme Procedure} binary-port? port
1610 Return @code{#t} if @var{port} is a @dfn{binary port}, suitable for
1611 binary data input/output.
1613 Note that internally Guile does not differentiate between binary and
1614 textual ports, unlike the R6RS. Thus, this procedure returns true when
1615 @var{port} does not have an associated encoding---i.e., when
1616 @code{(port-encoding @var{port})} is @code{#f} (@pxref{Ports,
1617 port-encoding}). This is the case for ports returned by R6RS procedures
1618 such as @code{open-bytevector-input-port} and
1619 @code{make-custom-binary-output-port}.
1621 However, Guile currently does not prevent use of textual I/O procedures
1622 such as @code{display} or @code{read-char} with binary ports. Doing so
1623 ``upgrades'' the port from binary to textual, under the ISO-8859-1
1624 encoding. Likewise, Guile does not prevent use of
1625 @code{set-port-encoding!} on a binary port, which also turns it into a
1629 @deffn {Scheme Procedure} textual-port? port
1630 Always return @code{#t}, as all ports can be used for textual I/O in
1634 @deffn {Scheme Procedure} transcoded-port binary-port transcoder
1635 The @code{transcoded-port} procedure
1636 returns a new textual port with the specified @var{transcoder}.
1637 Otherwise the new textual port's state is largely the same as
1638 that of @var{binary-port}.
1639 If @var{binary-port} is an input port, the new textual
1640 port will be an input port and
1641 will transcode the bytes that have not yet been read from
1643 If @var{binary-port} is an output port, the new textual
1644 port will be an output port and
1645 will transcode output characters into bytes that are
1646 written to the byte sink represented by @var{binary-port}.
1648 As a side effect, however, @code{transcoded-port}
1649 closes @var{binary-port} in
1650 a special way that allows the new textual port to continue to
1651 use the byte source or sink represented by @var{binary-port},
1652 even though @var{binary-port} itself is closed and cannot
1653 be used by the input and output operations described in this
1657 @deffn {Scheme Procedure} port-position port
1658 If @var{port} supports it (see below), return the offset (an integer)
1659 indicating where the next octet will be read from/written to in
1660 @var{port}. If @var{port} does not support this operation, an error
1661 condition is raised.
1663 This is similar to Guile's @code{seek} procedure with the
1664 @code{SEEK_CUR} argument (@pxref{Random Access}).
1667 @deffn {Scheme Procedure} port-has-port-position? port
1668 Return @code{#t} is @var{port} supports @code{port-position}.
1671 @deffn {Scheme Procedure} set-port-position! port offset
1672 If @var{port} supports it (see below), set the position where the next
1673 octet will be read from/written to @var{port} to @var{offset} (an
1674 integer). If @var{port} does not support this operation, an error
1675 condition is raised.
1677 This is similar to Guile's @code{seek} procedure with the
1678 @code{SEEK_SET} argument (@pxref{Random Access}).
1681 @deffn {Scheme Procedure} port-has-set-port-position!? port
1682 Return @code{#t} is @var{port} supports @code{set-port-position!}.
1685 @deffn {Scheme Procedure} call-with-port port proc
1686 Call @var{proc}, passing it @var{port} and closing @var{port} upon exit
1687 of @var{proc}. Return the return values of @var{proc}.
1690 @node R6RS Input Ports
1691 @subsubsection Input Ports
1693 @deffn {Scheme Procedure} input-port? obj
1694 Returns @code{#t} if the argument is an input port (or a combined input
1695 and output port), and returns @code{#f} otherwise.
1698 @deffn {Scheme Procedure} port-eof? input-port
1700 if the @code{lookahead-u8} procedure (if @var{input-port} is a binary port)
1701 or the @code{lookahead-char} procedure (if @var{input-port} is a textual port)
1703 the end-of-file object, and @code{#f} otherwise.
1704 The operation may block indefinitely if no data is available
1705 but the port cannot be determined to be at end of file.
1708 @deffn {Scheme Procedure} open-file-input-port filename
1709 @deffnx {Scheme Procedure} open-file-input-port filename file-options
1710 @deffnx {Scheme Procedure} open-file-input-port filename file-options buffer-mode
1711 @deffnx {Scheme Procedure} open-file-input-port filename file-options buffer-mode maybe-transcoder
1712 @var{maybe-transcoder} must be either a transcoder or @code{#f}.
1714 The @code{open-file-input-port} procedure returns an
1715 input port for the named file. The @var{file-options} and
1716 @var{maybe-transcoder} arguments are optional.
1718 The @var{file-options} argument, which may determine
1719 various aspects of the returned port (@pxref{R6RS File Options}),
1720 defaults to the value of @code{(file-options)}.
1722 The @var{buffer-mode} argument, if supplied,
1723 must be one of the symbols that name a buffer mode.
1724 The @var{buffer-mode} argument defaults to @code{block}.
1726 If @var{maybe-transcoder} is a transcoder, it becomes the transcoder associated
1727 with the returned port.
1729 If @var{maybe-transcoder} is @code{#f} or absent,
1730 the port will be a binary port and will support the
1731 @code{port-position} and @code{set-port-position!} operations.
1732 Otherwise the port will be a textual port, and whether it supports
1733 the @code{port-position} and @code{set-port-position!} operations
1734 is implementation-dependent (and possibly transcoder-dependent).
1737 @deffn {Scheme Procedure} standard-input-port
1738 Returns a fresh binary input port connected to standard input. Whether
1739 the port supports the @code{port-position} and @code{set-port-position!}
1740 operations is implementation-dependent.
1743 @deffn {Scheme Procedure} current-input-port
1744 This returns a default textual port for input. Normally, this default
1745 port is associated with standard input, but can be dynamically
1746 re-assigned using the @code{with-input-from-file} procedure from the
1747 @code{io simple (6)} library (@pxref{rnrs io simple}). The port may or
1748 may not have an associated transcoder; if it does, the transcoder is
1749 implementation-dependent.
1752 @node R6RS Binary Input
1753 @subsubsection Binary Input
1755 @cindex binary input
1757 R6RS binary input ports can be created with the procedures described
1760 @deffn {Scheme Procedure} open-bytevector-input-port bv [transcoder]
1761 @deffnx {C Function} scm_open_bytevector_input_port (bv, transcoder)
1762 Return an input port whose contents are drawn from bytevector @var{bv}
1763 (@pxref{Bytevectors}).
1765 @c FIXME: Update description when implemented.
1766 The @var{transcoder} argument is currently not supported.
1769 @cindex custom binary input ports
1771 @deffn {Scheme Procedure} make-custom-binary-input-port id read! get-position set-position! close
1772 @deffnx {C Function} scm_make_custom_binary_input_port (id, read!, get-position, set-position!, close)
1773 Return a new custom binary input port@footnote{This is similar in spirit
1774 to Guile's @dfn{soft ports} (@pxref{Soft Ports}).} named @var{id} (a
1775 string) whose input is drained by invoking @var{read!} and passing it a
1776 bytevector, an index where bytes should be written, and the number of
1777 bytes to read. The @code{read!} procedure must return an integer
1778 indicating the number of bytes read, or @code{0} to indicate the
1781 Optionally, if @var{get-position} is not @code{#f}, it must be a thunk
1782 that will be called when @code{port-position} is invoked on the custom
1783 binary port and should return an integer indicating the position within
1784 the underlying data stream; if @var{get-position} was not supplied, the
1785 returned port does not support @code{port-position}.
1787 Likewise, if @var{set-position!} is not @code{#f}, it should be a
1788 one-argument procedure. When @code{set-port-position!} is invoked on the
1789 custom binary input port, @var{set-position!} is passed an integer
1790 indicating the position of the next byte is to read.
1792 Finally, if @var{close} is not @code{#f}, it must be a thunk. It is
1793 invoked when the custom binary input port is closed.
1795 The returned port is fully buffered by default, but its buffering mode
1796 can be changed using @code{setvbuf} (@pxref{Ports and File Descriptors,
1799 Using a custom binary input port, the @code{open-bytevector-input-port}
1800 procedure could be implemented as follows:
1803 (define (open-bytevector-input-port source)
1805 (define length (bytevector-length source))
1807 (define (read! bv start count)
1808 (let ((count (min count (- length position))))
1809 (bytevector-copy! source position
1811 (set! position (+ position count))
1814 (define (get-position) position)
1816 (define (set-position! new-position)
1817 (set! position new-position))
1819 (make-custom-binary-input-port "the port" read!
1823 (read (open-bytevector-input-port (string->utf8 "hello")))
1828 @cindex binary input
1829 Binary input is achieved using the procedures below:
1831 @deffn {Scheme Procedure} get-u8 port
1832 @deffnx {C Function} scm_get_u8 (port)
1833 Return an octet read from @var{port}, a binary input port, blocking as
1834 necessary, or the end-of-file object.
1837 @deffn {Scheme Procedure} lookahead-u8 port
1838 @deffnx {C Function} scm_lookahead_u8 (port)
1839 Like @code{get-u8} but does not update @var{port}'s position to point
1843 @deffn {Scheme Procedure} get-bytevector-n port count
1844 @deffnx {C Function} scm_get_bytevector_n (port, count)
1845 Read @var{count} octets from @var{port}, blocking as necessary and
1846 return a bytevector containing the octets read. If fewer bytes are
1847 available, a bytevector smaller than @var{count} is returned.
1850 @deffn {Scheme Procedure} get-bytevector-n! port bv start count
1851 @deffnx {C Function} scm_get_bytevector_n_x (port, bv, start, count)
1852 Read @var{count} bytes from @var{port} and store them in @var{bv}
1853 starting at index @var{start}. Return either the number of bytes
1854 actually read or the end-of-file object.
1857 @deffn {Scheme Procedure} get-bytevector-some port
1858 @deffnx {C Function} scm_get_bytevector_some (port)
1859 Read from @var{port}, blocking as necessary, until bytes are available
1860 or an end-of-file is reached. Return either the end-of-file object or a
1861 new bytevector containing some of the available bytes (at least one),
1862 and update the port position to point just past these bytes.
1865 @deffn {Scheme Procedure} get-bytevector-all port
1866 @deffnx {C Function} scm_get_bytevector_all (port)
1867 Read from @var{port}, blocking as necessary, until the end-of-file is
1868 reached. Return either a new bytevector containing the data read or the
1869 end-of-file object (if no data were available).
1872 The @code{(ice-9 binary-ports)} module provides the following procedure
1873 as an extension to @code{(rnrs io ports)}:
1875 @deffn {Scheme Procedure} unget-bytevector port bv [start [count]]
1876 @deffnx {C Function} scm_unget_bytevector (port, bv, start, count)
1877 Place the contents of @var{bv} in @var{port}, optionally starting at
1878 index @var{start} and limiting to @var{count} octets, so that its bytes
1879 will be read from left-to-right as the next bytes from @var{port} during
1880 subsequent read operations. If called multiple times, the unread bytes
1881 will be read again in last-in first-out order.
1884 @node R6RS Textual Input
1885 @subsubsection Textual Input
1887 @deffn {Scheme Procedure} get-char textual-input-port
1888 Reads from @var{textual-input-port}, blocking as necessary, until a
1889 complete character is available from @var{textual-input-port},
1890 or until an end of file is reached.
1892 If a complete character is available before the next end of file,
1893 @code{get-char} returns that character and updates the input port to
1894 point past the character. If an end of file is reached before any
1895 character is read, @code{get-char} returns the end-of-file object.
1898 @deffn {Scheme Procedure} lookahead-char textual-input-port
1899 The @code{lookahead-char} procedure is like @code{get-char}, but it does
1900 not update @var{textual-input-port} to point past the character.
1903 @deffn {Scheme Procedure} get-string-n textual-input-port count
1905 @var{count} must be an exact, non-negative integer object, representing
1906 the number of characters to be read.
1908 The @code{get-string-n} procedure reads from @var{textual-input-port},
1909 blocking as necessary, until @var{count} characters are available, or
1910 until an end of file is reached.
1912 If @var{count} characters are available before end of file,
1913 @code{get-string-n} returns a string consisting of those @var{count}
1914 characters. If fewer characters are available before an end of file, but
1915 one or more characters can be read, @code{get-string-n} returns a string
1916 containing those characters. In either case, the input port is updated
1917 to point just past the characters read. If no characters can be read
1918 before an end of file, the end-of-file object is returned.
1921 @deffn {Scheme Procedure} get-string-n! textual-input-port string start count
1923 @var{start} and @var{count} must be exact, non-negative integer objects,
1924 with @var{count} representing the number of characters to be read.
1925 @var{string} must be a string with at least $@var{start} + @var{count}$
1928 The @code{get-string-n!} procedure reads from @var{textual-input-port}
1929 in the same manner as @code{get-string-n}. If @var{count} characters
1930 are available before an end of file, they are written into @var{string}
1931 starting at index @var{start}, and @var{count} is returned. If fewer
1932 characters are available before an end of file, but one or more can be
1933 read, those characters are written into @var{string} starting at index
1934 @var{start} and the number of characters actually read is returned as an
1935 exact integer object. If no characters can be read before an end of
1936 file, the end-of-file object is returned.
1939 @deffn {Scheme Procedure} get-string-all textual-input-port
1940 Reads from @var{textual-input-port} until an end of file, decoding
1941 characters in the same manner as @code{get-string-n} and
1942 @code{get-string-n!}.
1944 If characters are available before the end of file, a string containing
1945 all the characters decoded from that data are returned. If no character
1946 precedes the end of file, the end-of-file object is returned.
1949 @deffn {Scheme Procedure} get-line textual-input-port
1950 Reads from @var{textual-input-port} up to and including the linefeed
1951 character or end of file, decoding characters in the same manner as
1952 @code{get-string-n} and @code{get-string-n!}.
1954 If a linefeed character is read, a string containing all of the text up
1955 to (but not including) the linefeed character is returned, and the port
1956 is updated to point just past the linefeed character. If an end of file
1957 is encountered before any linefeed character is read, but some
1958 characters have been read and decoded as characters, a string containing
1959 those characters is returned. If an end of file is encountered before
1960 any characters are read, the end-of-file object is returned.
1963 The end-of-line style, if not @code{none}, will cause all line endings
1964 to be read as linefeed characters. @xref{R6RS Transcoders}.
1968 @deffn {Scheme Procedure} get-datum textual-input-port count
1969 Reads an external representation from @var{textual-input-port} and returns the
1970 datum it represents. The @code{get-datum} procedure returns the next
1971 datum that can be parsed from the given @var{textual-input-port}, updating
1972 @var{textual-input-port} to point exactly past the end of the external
1973 representation of the object.
1975 Any @emph{interlexeme space} (comment or whitespace, @pxref{Scheme
1976 Syntax}) in the input is first skipped. If an end of file occurs after
1977 the interlexeme space, the end-of-file object (@pxref{R6RS End-of-File})
1980 If a character inconsistent with an external representation is
1981 encountered in the input, an exception with condition types
1982 @code{&lexical} and @code{&i/o-read} is raised. Also, if the end of
1983 file is encountered after the beginning of an external representation,
1984 but the external representation is incomplete and therefore cannot be
1985 parsed, an exception with condition types @code{&lexical} and
1986 @code{&i/o-read} is raised.
1989 @node R6RS Output Ports
1990 @subsubsection Output Ports
1992 @deffn {Scheme Procedure} output-port? obj
1993 Returns @code{#t} if the argument is an output port (or a
1994 combined input and output port), @code{#f} otherwise.
1997 @deffn {Scheme Procedure} flush-output-port port
1998 Flushes any buffered output from the buffer of @var{output-port} to the
1999 underlying file, device, or object. The @code{flush-output-port}
2000 procedure returns an unspecified values.
2003 @deffn {Scheme Procedure} open-file-output-port filename
2004 @deffnx {Scheme Procedure} open-file-output-port filename file-options
2005 @deffnx {Scheme Procedure} open-file-output-port filename file-options buffer-mode
2006 @deffnx {Scheme Procedure} open-file-output-port filename file-options buffer-mode maybe-transcoder
2008 @var{maybe-transcoder} must be either a transcoder or @code{#f}.
2010 The @code{open-file-output-port} procedure returns an output port for the named file.
2012 The @var{file-options} argument, which may determine various aspects of
2013 the returned port (@pxref{R6RS File Options}), defaults to the value of
2014 @code{(file-options)}.
2016 The @var{buffer-mode} argument, if supplied,
2017 must be one of the symbols that name a buffer mode.
2018 The @var{buffer-mode} argument defaults to @code{block}.
2020 If @var{maybe-transcoder} is a transcoder, it becomes the transcoder
2021 associated with the port.
2023 If @var{maybe-transcoder} is @code{#f} or absent,
2024 the port will be a binary port and will support the
2025 @code{port-position} and @code{set-port-position!} operations.
2026 Otherwise the port will be a textual port, and whether it supports
2027 the @code{port-position} and @code{set-port-position!} operations
2028 is implementation-dependent (and possibly transcoder-dependent).
2031 @deffn {Scheme Procedure} standard-output-port
2032 @deffnx {Scheme Procedure} standard-error-port
2033 Returns a fresh binary output port connected to the standard output or
2034 standard error respectively. Whether the port supports the
2035 @code{port-position} and @code{set-port-position!} operations is
2036 implementation-dependent.
2039 @deffn {Scheme Procedure} current-output-port
2040 @deffnx {Scheme Procedure} current-error-port
2041 These return default textual ports for regular output and error output.
2042 Normally, these default ports are associated with standard output, and
2043 standard error, respectively. The return value of
2044 @code{current-output-port} can be dynamically re-assigned using the
2045 @code{with-output-to-file} procedure from the @code{io simple (6)}
2046 library (@pxref{rnrs io simple}). A port returned by one of these
2047 procedures may or may not have an associated transcoder; if it does, the
2048 transcoder is implementation-dependent.
2051 @node R6RS Binary Output
2052 @subsubsection Binary Output
2054 Binary output ports can be created with the procedures below.
2056 @deffn {Scheme Procedure} open-bytevector-output-port [transcoder]
2057 @deffnx {C Function} scm_open_bytevector_output_port (transcoder)
2058 Return two values: a binary output port and a procedure. The latter
2059 should be called with zero arguments to obtain a bytevector containing
2060 the data accumulated by the port, as illustrated below.
2065 (open-bytevector-output-port))
2066 (lambda (port get-bytevector)
2067 (display "hello" port)
2070 @result{} #vu8(104 101 108 108 111)
2073 @c FIXME: Update description when implemented.
2074 The @var{transcoder} argument is currently not supported.
2077 @cindex custom binary output ports
2079 @deffn {Scheme Procedure} make-custom-binary-output-port id write! get-position set-position! close
2080 @deffnx {C Function} scm_make_custom_binary_output_port (id, write!, get-position, set-position!, close)
2081 Return a new custom binary output port named @var{id} (a string) whose
2082 output is sunk by invoking @var{write!} and passing it a bytevector, an
2083 index where bytes should be read from this bytevector, and the number of
2084 bytes to be ``written''. The @code{write!} procedure must return an
2085 integer indicating the number of bytes actually written; when it is
2086 passed @code{0} as the number of bytes to write, it should behave as
2087 though an end-of-file was sent to the byte sink.
2089 The other arguments are as for @code{make-custom-binary-input-port}
2090 (@pxref{R6RS Binary Input, @code{make-custom-binary-input-port}}).
2093 @cindex binary output
2094 Writing to a binary output port can be done using the following
2097 @deffn {Scheme Procedure} put-u8 port octet
2098 @deffnx {C Function} scm_put_u8 (port, octet)
2099 Write @var{octet}, an integer in the 0--255 range, to @var{port}, a
2103 @deffn {Scheme Procedure} put-bytevector port bv [start [count]]
2104 @deffnx {C Function} scm_put_bytevector (port, bv, start, count)
2105 Write the contents of @var{bv} to @var{port}, optionally starting at
2106 index @var{start} and limiting to @var{count} octets.
2109 @node R6RS Textual Output
2110 @subsubsection Textual Output
2112 @deffn {Scheme Procedure} put-char port char
2113 Writes @var{char} to the port. The @code{put-char} procedure returns
2114 an unspecified value.
2117 @deffn {Scheme Procedure} put-string port string
2118 @deffnx {Scheme Procedure} put-string port string start
2119 @deffnx {Scheme Procedure} put-string port string start count
2121 @var{start} and @var{count} must be non-negative exact integer objects.
2122 @var{string} must have a length of at least @math{@var{start} +
2123 @var{count}}. @var{start} defaults to 0. @var{count} defaults to
2124 @math{@code{(string-length @var{string})} - @var{start}}$. The
2125 @code{put-string} procedure writes the @var{count} characters of
2126 @var{string} starting at index @var{start} to the port. The
2127 @code{put-string} procedure returns an unspecified value.
2130 @deffn {Scheme Procedure} put-datum textual-output-port datum
2131 @var{datum} should be a datum value. The @code{put-datum} procedure
2132 writes an external representation of @var{datum} to
2133 @var{textual-output-port}. The specific external representation is
2134 implementation-dependent. However, whenever possible, an implementation
2135 should produce a representation for which @code{get-datum}, when reading
2136 the representation, will return an object equal (in the sense of
2137 @code{equal?}) to @var{datum}.
2140 Not all datums may allow producing an external representation for which
2141 @code{get-datum} will produce an object that is equal to the
2142 original. Specifically, NaNs contained in @var{datum} may make
2147 The @code{put-datum} procedure merely writes the external
2148 representation, but no trailing delimiter. If @code{put-datum} is
2149 used to write several subsequent external representations to an
2150 output port, care should be taken to delimit them properly so they can
2151 be read back in by subsequent calls to @code{get-datum}.
2155 @node I/O Extensions
2156 @subsection Using and Extending Ports in C
2159 * C Port Interface:: Using ports from C.
2160 * Port Implementation:: How to implement a new port type in C.
2164 @node C Port Interface
2165 @subsubsection C Port Interface
2166 @cindex C port interface
2167 @cindex Port, C interface
2169 This section describes how to use Scheme ports from C.
2171 @subsubheading Port basics
2174 @tindex scm_ptob_descriptor
2176 @findex SCM_PTAB_ENTRY
2179 There are two main data structures. A port type object (ptob) is of
2180 type @code{scm_ptob_descriptor}. A port instance is of type
2181 @code{scm_port}. Given an @code{SCM} variable which points to a port,
2182 the corresponding C port object can be obtained using the
2183 @code{SCM_PTAB_ENTRY} macro. The ptob can be obtained by using
2184 @code{SCM_PTOBNUM} to give an index into the @code{scm_ptobs}
2187 @subsubheading Port buffers
2189 An input port always has a read buffer and an output port always has a
2190 write buffer. However the size of these buffers is not guaranteed to be
2191 more than one byte (e.g., the @code{shortbuf} field in @code{scm_port}
2192 which is used when no other buffer is allocated). The way in which the
2193 buffers are allocated depends on the implementation of the ptob. For
2194 example in the case of an fport, buffers may be allocated with malloc
2195 when the port is created, but in the case of an strport the underlying
2196 string is used as the buffer.
2198 @subsubheading The @code{rw_random} flag
2200 Special treatment is required for ports which can be seeked at random.
2201 Before various operations, such as seeking the port or changing from
2202 input to output on a bidirectional port or vice versa, the port
2203 implementation must be given a chance to update its state. The write
2204 buffer is updated by calling the @code{flush} ptob procedure and the
2205 input buffer is updated by calling the @code{end_input} ptob procedure.
2206 In the case of an fport, @code{flush} causes buffered output to be
2207 written to the file descriptor, while @code{end_input} causes the
2208 descriptor position to be adjusted to account for buffered input which
2211 The special treatment must be performed if the @code{rw_random} flag in
2212 the port is non-zero.
2214 @subsubheading The @code{rw_active} variable
2216 The @code{rw_active} variable in the port is only used if
2217 @code{rw_random} is set. It's defined as an enum with the following
2222 the read buffer may have unread data.
2224 @item SCM_PORT_WRITE
2225 the write buffer may have unwritten data.
2227 @item SCM_PORT_NEITHER
2228 neither the write nor the read buffer has data.
2231 @subsubheading Reading from a port.
2233 To read from a port, it's possible to either call existing libguile
2234 procedures such as @code{scm_getc} and @code{scm_read_line} or to read
2235 data from the read buffer directly. Reading from the buffer involves
2236 the following steps:
2240 Flush output on the port, if @code{rw_active} is @code{SCM_PORT_WRITE}.
2243 Fill the read buffer, if it's empty, using @code{scm_fill_input}.
2245 @item Read the data from the buffer and update the read position in
2246 the buffer. Steps 2) and 3) may be repeated as many times as required.
2248 @item Set rw_active to @code{SCM_PORT_READ} if @code{rw_random} is set.
2250 @item update the port's line and column counts.
2253 @subsubheading Writing to a port.
2255 To write data to a port, calling @code{scm_lfwrite} should be sufficient for
2256 most purposes. This takes care of the following steps:
2260 End input on the port, if @code{rw_active} is @code{SCM_PORT_READ}.
2263 Pass the data to the ptob implementation using the @code{write} ptob
2264 procedure. The advantage of using the ptob @code{write} instead of
2265 manipulating the write buffer directly is that it allows the data to be
2266 written in one operation even if the port is using the single-byte
2270 Set @code{rw_active} to @code{SCM_PORT_WRITE} if @code{rw_random}
2275 @node Port Implementation
2276 @subsubsection Port Implementation
2277 @cindex Port implementation
2279 This section describes how to implement a new port type in C.
2281 As described in the previous section, a port type object (ptob) is
2282 a structure of type @code{scm_ptob_descriptor}. A ptob is created by
2283 calling @code{scm_make_port_type}.
2285 @deftypefun scm_t_bits scm_make_port_type (char *name, int (*fill_input) (SCM port), void (*write) (SCM port, const void *data, size_t size))
2286 Return a new port type object. The @var{name}, @var{fill_input} and
2287 @var{write} parameters are initial values for those port type fields,
2288 as described below. The other fields are initialized with default
2289 values and can be changed later.
2292 All of the elements of the ptob, apart from @code{name}, are procedures
2293 which collectively implement the port behaviour. Creating a new port
2294 type mostly involves writing these procedures.
2298 A pointer to a NUL terminated string: the name of the port type. This
2299 is the only element of @code{scm_ptob_descriptor} which is not
2300 a procedure. Set via the first argument to @code{scm_make_port_type}.
2303 Called during garbage collection to mark any SCM objects that a port
2304 object may contain. It doesn't need to be set unless the port has
2305 @code{SCM} components. Set using
2307 @deftypefun void scm_set_port_mark (scm_t_bits tc, SCM (*mark) (SCM port))
2311 Called when the port is collected during gc. It
2312 should free any resources used by the port.
2315 @deftypefun void scm_set_port_free (scm_t_bits tc, size_t (*free) (SCM port))
2319 Called when @code{write} is called on the port object, to print a
2320 port description. E.g., for an fport it may produce something like:
2321 @code{#<input: /etc/passwd 3>}. Set using
2323 @deftypefun void scm_set_port_print (scm_t_bits tc, int (*print) (SCM port, SCM dest_port, scm_print_state *pstate))
2324 The first argument @var{port} is the object being printed, the second
2325 argument @var{dest_port} is where its description should go.
2329 Not used at present. Set using
2331 @deftypefun void scm_set_port_equalp (scm_t_bits tc, SCM (*equalp) (SCM, SCM))
2335 Called when the port is closed, unless it was collected during gc. It
2336 should free any resources used by the port.
2339 @deftypefun void scm_set_port_close (scm_t_bits tc, int (*close) (SCM port))
2343 Accept data which is to be written using the port. The port implementation
2344 may choose to buffer the data instead of processing it directly.
2345 Set via the third argument to @code{scm_make_port_type}.
2348 Complete the processing of buffered output data. Reset the value of
2349 @code{rw_active} to @code{SCM_PORT_NEITHER}.
2352 @deftypefun void scm_set_port_flush (scm_t_bits tc, void (*flush) (SCM port))
2356 Perform any synchronization required when switching from input to output
2357 on the port. Reset the value of @code{rw_active} to @code{SCM_PORT_NEITHER}.
2360 @deftypefun void scm_set_port_end_input (scm_t_bits tc, void (*end_input) (SCM port, int offset))
2364 Read new data into the read buffer and return the first character. It
2365 can be assumed that the read buffer is empty when this procedure is called.
2366 Set via the second argument to @code{scm_make_port_type}.
2369 Return a lower bound on the number of bytes that could be read from the
2370 port without blocking. It can be assumed that the current state of
2371 @code{rw_active} is @code{SCM_PORT_NEITHER}.
2374 @deftypefun void scm_set_port_input_waiting (scm_t_bits tc, int (*input_waiting) (SCM port))
2378 Set the current position of the port. The procedure can not make
2379 any assumptions about the value of @code{rw_active} when it's
2380 called. It can reset the buffers first if desired by using something
2384 if (pt->rw_active == SCM_PORT_READ)
2385 scm_end_input (port);
2386 else if (pt->rw_active == SCM_PORT_WRITE)
2390 However note that this will have the side effect of discarding any data
2391 in the unread-char buffer, in addition to any side effects from the
2392 @code{end_input} and @code{flush} ptob procedures. This is undesirable
2393 when seek is called to measure the current position of the port, i.e.,
2394 @code{(seek p 0 SEEK_CUR)}. The libguile fport and string port
2395 implementations take care to avoid this problem.
2397 The procedure is set using
2399 @deftypefun void scm_set_port_seek (scm_t_bits tc, scm_t_off (*seek) (SCM port, scm_t_off offset, int whence))
2403 Truncate the port data to be specified length. It can be assumed that the
2404 current state of @code{rw_active} is @code{SCM_PORT_NEITHER}.
2407 @deftypefun void scm_set_port_truncate (scm_t_bits tc, void (*truncate) (SCM port, scm_t_off length))
2413 @subsection Handling of Unicode byte order marks.
2415 @cindex byte order mark
2417 This section documents the finer points of Guile's handling of Unicode
2418 byte order marks (BOMs). A byte order mark (U+FEFF) is typically found
2419 at the start of a UTF-16 or UTF-32 stream, to allow readers to reliably
2420 determine the byte order. Occasionally, a BOM is found at the start of
2421 a UTF-8 stream, but this is much less common and not generally
2424 Guile attempts to handle BOMs automatically, and in accordance with the
2425 recommendations of the Unicode Standard, when the port encoding is set
2426 to @code{UTF-8}, @code{UTF-16}, or @code{UTF-32}. In brief, Guile
2427 automatically writes a BOM at the start of a UTF-16 or UTF-32 stream,
2428 and automatically consumes one from the start of a UTF-8, UTF-16, or
2431 As specified in the Unicode Standard, a BOM is only handled specially at
2432 the start of a stream, and only if the port encoding is set to
2433 @code{UTF-8}, @code{UTF-16} or @code{UTF-32}. If the port encoding is
2434 set to @code{UTF-16BE}, @code{UTF-16LE}, @code{UTF-32BE}, or
2435 @code{UTF-32LE}, then BOMs are @emph{not} handled specially, and none of
2436 the special handling described in this section applies.
2440 To ensure that Guile will properly detect the byte order of a UTF-16 or
2441 UTF-32 stream, you must perform a textual read before any writes, seeks,
2442 or binary I/O. Guile will not attempt to read a BOM unless a read is
2443 explicitly requested at the start of the stream.
2446 If a textual write is performed before the first read, then an arbitrary
2447 byte order will be chosen. Currently, big endian is the default on all
2448 platforms, but that may change in the future. If you wish to explicitly
2449 control the byte order of an output stream, set the port encoding to
2450 @code{UTF-16BE}, @code{UTF-16LE}, @code{UTF-32BE}, or @code{UTF-32LE},
2451 and explicitly write a BOM (@code{#\xFEFF}) if desired.
2454 If @code{set-port-encoding!} is called in the middle of a stream, Guile
2455 treats this as a new logical ``start of stream'' for purposes of BOM
2456 handling, and will forget about any BOMs that had previously been seen.
2457 Therefore, it may choose a different byte order than had been used
2458 previously. This is intended to support multiple logical text streams
2459 embedded within a larger binary stream.
2462 Binary I/O operations are not guaranteed to update Guile's notion of
2463 whether the port is at the ``start of the stream'', nor are they
2464 guaranteed to produce or consume BOMs.
2467 For ports that support seeking (e.g. normal files), the input and output
2468 streams are considered linked: if the user reads first, then a BOM will
2469 be consumed (if appropriate), but later writes will @emph{not} produce a
2470 BOM. Similarly, if the user writes first, then later reads will
2471 @emph{not} consume a BOM.
2474 For ports that do not support seeking (e.g. pipes, sockets, and
2475 terminals), the input and output streams are considered
2476 @emph{independent} for purposes of BOM handling: the first read will
2477 consume a BOM (if appropriate), and the first write will @emph{also}
2478 produce a BOM (if appropriate). However, the input and output streams
2479 will always use the same byte order.
2482 Seeks to the beginning of a file will set the ``start of stream'' flags.
2483 Therefore, a subsequent textual read or write will consume or produce a
2484 BOM. However, unlike @code{set-port-encoding!}, if a byte order had
2485 already been chosen for the port, it will remain in effect after a seek,
2486 and cannot be changed by the presence of a BOM. Seeks anywhere other
2487 than the beginning of a file clear the ``start of stream'' flags.
2491 @c TeX-master: "guile.texi"