3 perliol - C API for Perl's implementation of IO in Layers.
7 /* Defining a layer ... */
12 This document describes the behavior and implementation of the PerlIO
13 abstraction described in L<perlapio> when C<USE_PERLIO> is defined (and
16 =head2 History and Background
18 The PerlIO abstraction was introduced in perl5.003_02 but languished as
19 just an abstraction until perl5.7.0. However during that time a number
20 of perl extensions switched to using it, so the API is mostly fixed to
21 maintain (source) compatibility.
23 The aim of the implementation is to provide the PerlIO API in a flexible
24 and platform neutral manner. It is also a trial of an "Object Oriented
25 C, with vtables" approach which may be applied to perl6.
27 =head2 Basic Structure
29 PerlIO is as a stack of layers.
31 The low levels of the stack work with the low-level operating system
32 calls (file descriptors in C) getting bytes in and out, the higher
33 layers of the stack buffer, filter, and otherwise manipulate the I/O.
34 Terms I<above> and I<below> are used to refer to the relative
35 positioning of the stack layers.
37 A layer contains a "vtable", the table of I/O operations (at C level
38 a table of function pointers), and status flags. The functions in the
39 vtable implement operations like "open", "read", and "write".
41 When I/O, for example "read", is requested, the request goes from Perl
42 first down the stack using "read" functions of each layer, then at the
43 bottom the input is requested from the operating system services, then
44 the result is returned up the stack, finally being interpreted as Perl
47 When you do an open() and specify extra PerlIO layers to be deployed,
48 the layers you specify are "pushed" on top of the already existing
49 default stack. What exact layers are in this default stack depends on
50 a lot of things: your operating system, Perl version, Perl compile
51 time configuration, and Perl runtime configuration. See L<PerlIO>,
52 L<perlrun/PERLIO>, and L<open> for more information.
54 binmode() operates similarly to open(): by default the specified
55 layers are pushed on top of the existing stack.
57 However, note that even as the specified layers are "pushed on top"
58 for open() and binmode(), this doesn't mean that the effects are
59 limited to the "top": PerlIO layers can be very 'active' and inspect
60 and affect layers also deeper in the stack. As an example there
61 is a layer called "raw" which repeatedly "pops" layers until
62 it reaches the first layer that has declared itself capable of
63 handling binary data. The "pushed" layers are processed in left-to-right
66 sysopen() operates (unsurprisingly) at a lower level in the stack than
67 open(). For example in UNIX or UNIX-like systems sysopen() operates
68 directly at the level of file descriptors: in the terms of PerlIO
69 layers, it uses only the "unix" layer, which is a rather thin wrapper
70 on top of the UNIX file descriptors.
72 =head2 Layers vs Disciplines
74 Initial discussion of the ability to modify IO streams behaviour used
75 the term "discipline" for the entities which were added. This came (I
76 believe) from the use of the term in "sfio", which in turn borrowed it
77 from "line disciplines" on Unix terminals. However, this document (and
78 the C code) uses the term "layer".
80 This is, I hope, a natural term given the implementation, and should
81 avoid connotations that are inherent in earlier uses of "discipline"
82 for things which are rather different.
84 =head2 Data Structures
86 The basic data structure is a PerlIOl:
88 typedef struct _PerlIO PerlIOl;
89 typedef struct _PerlIO_funcs PerlIO_funcs;
90 typedef PerlIOl *PerlIO;
94 PerlIOl * next; /* Lower layer */
95 PerlIO_funcs * tab; /* Functions for this layer */
96 IV flags; /* Various flags for state */
99 A C<PerlIOl *> is a pointer to the struct, and the I<application>
100 level C<PerlIO *> is a pointer to a C<PerlIOl *> - i.e. a pointer
101 to a pointer to the struct. This allows the application level C<PerlIO *>
102 to remain constant while the actual C<PerlIOl *> underneath
103 changes. (Compare perl's C<SV *> which remains constant while its
104 C<sv_any> field changes as the scalar's type changes.) An IO stream is
105 then in general represented as a pointer to this linked-list of
108 It should be noted that because of the double indirection in a C<PerlIO *>,
109 a C<< &(perlio->next) >> "is" a C<PerlIO *>, and so to some degree
110 at least one layer can use the "standard" API on the next layer down.
112 A "layer" is composed of two parts:
118 The functions and attributes of the "layer class".
122 The per-instance data for a particular handle.
126 =head2 Functions and Attributes
128 The functions and attributes are accessed via the "tab" (for table)
129 member of C<PerlIOl>. The functions (methods of the layer "class") are
130 fixed, and are defined by the C<PerlIO_funcs> type. They are broadly the
131 same as the public C<PerlIO_xxxxx> functions:
139 IV (*Pushed)(pTHX_ PerlIO *f,const char *mode,SV *arg, PerlIO_funcs *tab);
140 IV (*Popped)(pTHX_ PerlIO *f);
141 PerlIO * (*Open)(pTHX_ PerlIO_funcs *tab,
144 int fd, int imode, int perm,
146 int narg, SV **args);
147 IV (*Binmode)(pTHX_ PerlIO *f);
148 SV * (*Getarg)(pTHX_ PerlIO *f, CLONE_PARAMS *param, int flags)
149 IV (*Fileno)(pTHX_ PerlIO *f);
150 PerlIO * (*Dup)(pTHX_ PerlIO *f, PerlIO *o, CLONE_PARAMS *param, int flags)
151 /* Unix-like functions - cf sfio line disciplines */
152 SSize_t (*Read)(pTHX_ PerlIO *f, void *vbuf, Size_t count);
153 SSize_t (*Unread)(pTHX_ PerlIO *f, const void *vbuf, Size_t count);
154 SSize_t (*Write)(pTHX_ PerlIO *f, const void *vbuf, Size_t count);
155 IV (*Seek)(pTHX_ PerlIO *f, Off_t offset, int whence);
156 Off_t (*Tell)(pTHX_ PerlIO *f);
157 IV (*Close)(pTHX_ PerlIO *f);
158 /* Stdio-like buffered IO functions */
159 IV (*Flush)(pTHX_ PerlIO *f);
160 IV (*Fill)(pTHX_ PerlIO *f);
161 IV (*Eof)(pTHX_ PerlIO *f);
162 IV (*Error)(pTHX_ PerlIO *f);
163 void (*Clearerr)(pTHX_ PerlIO *f);
164 void (*Setlinebuf)(pTHX_ PerlIO *f);
165 /* Perl's snooping functions */
166 STDCHAR * (*Get_base)(pTHX_ PerlIO *f);
167 Size_t (*Get_bufsiz)(pTHX_ PerlIO *f);
168 STDCHAR * (*Get_ptr)(pTHX_ PerlIO *f);
169 SSize_t (*Get_cnt)(pTHX_ PerlIO *f);
170 void (*Set_ptrcnt)(pTHX_ PerlIO *f,STDCHAR *ptr,SSize_t cnt);
173 The first few members of the struct give a function table size for
174 compatibility check "name" for the layer, the size to C<malloc> for the per-instance data,
175 and some flags which are attributes of the class as whole (such as whether it is a buffering
176 layer), then follow the functions which fall into four basic groups:
182 Opening and setup functions
190 Stdio class buffering options.
194 Functions to support Perl's traditional "fast" access to the buffer.
198 A layer does not have to implement all the functions, but the whole
199 table has to be present. Unimplemented slots can be NULL (which will
200 result in an error when called) or can be filled in with stubs to
201 "inherit" behaviour from a "base class". This "inheritance" is fixed
202 for all instances of the layer, but as the layer chooses which stubs
203 to populate the table, limited "multiple inheritance" is possible.
205 =head2 Per-instance Data
207 The per-instance data are held in memory beyond the basic PerlIOl
208 struct, by making a PerlIOl the first member of the layer's struct
213 struct _PerlIO base; /* Base "class" info */
214 STDCHAR * buf; /* Start of buffer */
215 STDCHAR * end; /* End of valid part of buffer */
216 STDCHAR * ptr; /* Current position in buffer */
217 Off_t posn; /* Offset of buf into the file */
218 Size_t bufsiz; /* Real size of buffer */
219 IV oneword; /* Emergency buffer */
222 In this way (as for perl's scalars) a pointer to a PerlIOBuf can be
223 treated as a pointer to a PerlIOl.
225 =head2 Layers in action.
229 +-----------+ +----------+ +--------+
230 PerlIO ->| |--->| next |--->| NULL |
231 +-----------+ +----------+ +--------+
232 | | | buffer | | fd |
233 +-----------+ | | +--------+
237 The above attempts to show how the layer scheme works in a simple case.
238 The application's C<PerlIO *> points to an entry in the table(s)
239 representing open (allocated) handles. For example the first three slots
240 in the table correspond to C<stdin>,C<stdout> and C<stderr>. The table
241 in turn points to the current "top" layer for the handle - in this case
242 an instance of the generic buffering layer "perlio". That layer in turn
243 points to the next layer down - in this case the lowlevel "unix" layer.
245 The above is roughly equivalent to a "stdio" buffered stream, but with
246 much more flexibility:
252 If Unix level C<read>/C<write>/C<lseek> is not appropriate for (say)
253 sockets then the "unix" layer can be replaced (at open time or even
254 dynamically) with a "socket" layer.
258 Different handles can have different buffering schemes. The "top"
259 layer could be the "mmap" layer if reading disk files was quicker
260 using C<mmap> than C<read>. An "unbuffered" stream can be implemented
261 simply by not having a buffer layer.
265 Extra layers can be inserted to process the data as it flows through.
266 This was the driving need for including the scheme in perl 5.7.0+ - we
267 needed a mechanism to allow data to be translated between perl's
268 internal encoding (conceptually at least Unicode as UTF-8), and the
269 "native" format used by the system. This is provided by the
270 ":encoding(xxxx)" layer which typically sits above the buffering layer.
274 A layer can be added that does "\n" to CRLF translation. This layer
275 can be used on any platform, not just those that normally do such
280 =head2 Per-instance flag bits
282 The generic flag bits are a hybrid of C<O_XXXXX> style flags deduced
283 from the mode string passed to C<PerlIO_open()>, and state bits for
284 typical buffer layers.
292 =item PERLIO_F_CANWRITE
294 Writes are permitted, i.e. opened as "w" or "r+" or "a", etc.
296 =item PERLIO_F_CANREAD
298 Reads are permitted i.e. opened "r" or "w+" (or even "a+" - ick).
302 An error has occurred (for C<PerlIO_error()>).
304 =item PERLIO_F_TRUNCATE
306 Truncate file suggested by open mode.
308 =item PERLIO_F_APPEND
310 All writes should be appends.
314 Layer is performing Win32-like "\n" mapped to CR,LF for output and CR,LF
315 mapped to "\n" for input. Normally the provided "crlf" layer is the only
316 layer that need bother about this. C<PerlIO_binmode()> will mess with this
317 flag rather than add/remove layers if the C<PERLIO_K_CANCRLF> bit is set
318 for the layers class.
322 Data written to this layer should be UTF-8 encoded; data provided
323 by this layer should be considered UTF-8 encoded. Can be set on any layer
324 by ":utf8" dummy layer. Also set on ":encoding" layer.
328 Layer is unbuffered - i.e. write to next layer down should occur for
329 each write to this layer.
333 The buffer for this layer currently holds data written to it but not sent
338 The buffer for this layer currently holds unconsumed data read from
341 =item PERLIO_F_LINEBUF
343 Layer is line buffered. Write data should be passed to next layer down
344 whenever a "\n" is seen. Any data beyond the "\n" should then be
349 File has been C<unlink()>ed, or should be deleted on C<close()>.
355 =item PERLIO_F_FASTGETS
357 This instance of this layer supports the "fast C<gets>" interface.
358 Normally set based on C<PERLIO_K_FASTGETS> for the class and by the
359 existence of the function(s) in the table. However a class that
360 normally provides that interface may need to avoid it on a
361 particular instance. The "pending" layer needs to do this when
362 it is pushed above a layer which does not support the interface.
363 (Perl's C<sv_gets()> does not expect the streams fast C<gets> behaviour
364 to change during one "get".)
368 =head2 Methods in Detail
376 Size of the function table. This is compared against the value PerlIO
377 code "knows" as a compatibility check. Future versions I<may> be able
378 to tolerate layers compiled against an old version of the headers.
384 The name of the layer whose open() method Perl should invoke on
385 open(). For example if the layer is called APR, you will call:
387 open $fh, ">:APR", ...
389 and Perl knows that it has to invoke the PerlIOAPR_open() method
390 implemented by the APR layer.
396 The size of the per-instance data structure, e.g.:
400 If this field is zero then C<PerlIO_pushed> does not malloc anything
401 and assumes layer's Pushed function will do any required layer stack
402 manipulation - used to avoid malloc/free overhead for dummy layers.
403 If the field is non-zero it must be at least the size of C<PerlIOl>,
404 C<PerlIO_pushed> will allocate memory for the layer's data structures
405 and link new layer onto the stream's stack. (If the layer's Pushed
406 method returns an error indication the layer is popped again.)
414 =item * PERLIO_K_BUFFERED
416 The layer is buffered.
420 The layer is acceptable to have in a binmode(FH) stack - i.e. it does not
421 (or will configure itself not to) transform bytes passing through it.
423 =item * PERLIO_K_CANCRLF
425 Layer can translate between "\n" and CRLF line ends.
427 =item * PERLIO_K_FASTGETS
429 Layer allows buffer snooping.
431 =item * PERLIO_K_MULTIARG
433 Used when the layer's open() accepts more arguments than usual. The
434 extra arguments should come not before the C<MODE> argument. When this
435 flag is used it's up to the layer to validate the args.
441 IV (*Pushed)(pTHX_ PerlIO *f,const char *mode, SV *arg);
443 The only absolutely mandatory method. Called when the layer is pushed
444 onto the stack. The C<mode> argument may be NULL if this occurs
445 post-open. The C<arg> will be non-C<NULL> if an argument string was
446 passed. In most cases this should call C<PerlIOBase_pushed()> to
447 convert C<mode> into the appropriate C<PERLIO_F_XXXXX> flags in
448 addition to any actions the layer itself takes. If a layer is not
449 expecting an argument it need neither save the one passed to it, nor
450 provide C<Getarg()> (it could perhaps C<Perl_warn> that the argument
453 Returns 0 on success. On failure returns -1 and should set errno.
457 IV (*Popped)(pTHX_ PerlIO *f);
459 Called when the layer is popped from the stack. A layer will normally
460 be popped after C<Close()> is called. But a layer can be popped
461 without being closed if the program is dynamically managing layers on
462 the stream. In such cases C<Popped()> should free any resources
463 (buffers, translation tables, ...) not held directly in the layer's
464 struct. It should also C<Unread()> any unconsumed data that has been
465 read and buffered from the layer below back to that layer, so that it
466 can be re-provided to what ever is now above.
468 Returns 0 on success and failure. If C<Popped()> returns I<true> then
469 I<perlio.c> assumes that either the layer has popped itself, or the
470 layer is super special and needs to be retained for other reasons.
471 In most cases it should return I<false>.
475 PerlIO * (*Open)(...);
477 The C<Open()> method has lots of arguments because it combines the
478 functions of perl's C<open>, C<PerlIO_open>, perl's C<sysopen>,
479 C<PerlIO_fdopen> and C<PerlIO_reopen>. The full prototype is as
482 PerlIO * (*Open)(pTHX_ PerlIO_funcs *tab,
485 int fd, int imode, int perm,
487 int narg, SV **args);
489 Open should (perhaps indirectly) call C<PerlIO_allocate()> to allocate
490 a slot in the table and associate it with the layers information for
491 the opened file, by calling C<PerlIO_push>. The I<layers> AV is an
492 array of all the layers destined for the C<PerlIO *>, and any
493 arguments passed to them, I<n> is the index into that array of the
494 layer being called. The macro C<PerlIOArg> will return a (possibly
495 C<NULL>) SV * for the argument passed to the layer.
497 The I<mode> string is an "C<fopen()>-like" string which would match
498 the regular expression C</^[I#]?[rwa]\+?[bt]?$/>.
500 The C<'I'> prefix is used during creation of C<stdin>..C<stderr> via
501 special C<PerlIO_fdopen> calls; the C<'#'> prefix means that this is
502 C<sysopen> and that I<imode> and I<perm> should be passed to
503 C<PerlLIO_open3>; C<'r'> means B<r>ead, C<'w'> means B<w>rite and
504 C<'a'> means B<a>ppend. The C<'+'> suffix means that both reading and
505 writing/appending are permitted. The C<'b'> suffix means file should
506 be binary, and C<'t'> means it is text. (Almost all layers should do
507 the IO in binary mode, and ignore the b/t bits. The C<:crlf> layer
508 should be pushed to handle the distinction.)
510 If I<old> is not C<NULL> then this is a C<PerlIO_reopen>. Perl itself
511 does not use this (yet?) and semantics are a little vague.
513 If I<fd> not negative then it is the numeric file descriptor I<fd>,
514 which will be open in a manner compatible with the supplied mode
515 string, the call is thus equivalent to C<PerlIO_fdopen>. In this case
516 I<nargs> will be zero.
518 If I<nargs> is greater than zero then it gives the number of arguments
519 passed to C<open>, otherwise it will be 1 if for example
520 C<PerlIO_open> was called. In simple cases SvPV_nolen(*args) is the
523 Having said all that translation-only layers do not need to provide
524 C<Open()> at all, but rather leave the opening to a lower level layer
525 and wait to be "pushed". If a layer does provide C<Open()> it should
526 normally call the C<Open()> method of next layer down (if any) and
527 then push itself on top if that succeeds.
529 If C<PerlIO_push> was performed and open has failed, it must
530 C<PerlIO_pop> itself, since if it's not, the layer won't be removed
531 and may cause bad problems.
533 Returns C<NULL> on failure.
537 IV (*Binmode)(pTHX_ PerlIO *f);
539 Optional. Used when C<:raw> layer is pushed (explicitly or as a result
540 of binmode(FH)). If not present layer will be popped. If present
541 should configure layer as binary (or pop itself) and return 0.
542 If it returns -1 for error C<binmode> will fail with layer
547 SV * (*Getarg)(pTHX_ PerlIO *f,
548 CLONE_PARAMS *param, int flags);
550 Optional. If present should return an SV * representing the string
551 argument passed to the layer when it was
552 pushed. e.g. ":encoding(ascii)" would return an SvPV with value
553 "ascii". (I<param> and I<flags> arguments can be ignored in most
558 IV (*Fileno)(pTHX_ PerlIO *f);
560 Returns the Unix/Posix numeric file descriptor for the handle. Normally
561 C<PerlIOBase_fileno()> (which just asks next layer down) will suffice
564 Returns -1 on error, which is considered to include the case where the
565 layer cannot provide such a file descriptor.
569 PerlIO * (*Dup)(pTHX_ PerlIO *f, PerlIO *o,
570 CLONE_PARAMS *param, int flags);
572 XXX: Needs more docs.
574 Used as part of the "clone" process when a thread is spawned (in which
575 case param will be non-NULL) and when a stream is being duplicated via
578 Similar to C<Open>, returns PerlIO* on success, C<NULL> on failure.
582 SSize_t (*Read)(pTHX_ PerlIO *f, void *vbuf, Size_t count);
584 Basic read operation.
586 Typically will call C<Fill> and manipulate pointers (possibly via the
587 API). C<PerlIOBuf_read()> may be suitable for derived classes which
588 provide "fast gets" methods.
590 Returns actual bytes read, or -1 on an error.
594 SSize_t (*Unread)(pTHX_ PerlIO *f,
595 const void *vbuf, Size_t count);
597 A superset of stdio's C<ungetc()>. Should arrange for future reads to
598 see the bytes in C<vbuf>. If there is no obviously better implementation
599 then C<PerlIOBase_unread()> provides the function by pushing a "fake"
600 "pending" layer above the calling layer.
602 Returns the number of unread chars.
606 SSize_t (*Write)(PerlIO *f, const void *vbuf, Size_t count);
608 Basic write operation.
610 Returns bytes written or -1 on an error.
614 IV (*Seek)(pTHX_ PerlIO *f, Off_t offset, int whence);
616 Position the file pointer. Should normally call its own C<Flush>
617 method and then the C<Seek> method of next layer down.
619 Returns 0 on success, -1 on failure.
623 Off_t (*Tell)(pTHX_ PerlIO *f);
625 Return the file pointer. May be based on layers cached concept of
626 position to avoid overhead.
628 Returns -1 on failure to get the file pointer.
632 IV (*Close)(pTHX_ PerlIO *f);
634 Close the stream. Should normally call C<PerlIOBase_close()> to flush
635 itself and close layers below, and then deallocate any data structures
636 (buffers, translation tables, ...) not held directly in the data
639 Returns 0 on success, -1 on failure.
643 IV (*Flush)(pTHX_ PerlIO *f);
645 Should make stream's state consistent with layers below. That is, any
646 buffered write data should be written, and file position of lower layers
647 adjusted for data read from below but not actually consumed.
648 (Should perhaps C<Unread()> such data to the lower layer.)
650 Returns 0 on success, -1 on failure.
654 IV (*Fill)(pTHX_ PerlIO *f);
656 The buffer for this layer should be filled (for read) from layer
657 below. When you "subclass" PerlIOBuf layer, you want to use its
658 I<_read> method and to supply your own fill method, which fills the
661 Returns 0 on success, -1 on failure.
665 IV (*Eof)(pTHX_ PerlIO *f);
667 Return end-of-file indicator. C<PerlIOBase_eof()> is normally sufficient.
669 Returns 0 on end-of-file, 1 if not end-of-file, -1 on error.
673 IV (*Error)(pTHX_ PerlIO *f);
675 Return error indicator. C<PerlIOBase_error()> is normally sufficient.
677 Returns 1 if there is an error (usually when C<PERLIO_F_ERROR> is set,
682 void (*Clearerr)(pTHX_ PerlIO *f);
684 Clear end-of-file and error indicators. Should call C<PerlIOBase_clearerr()>
685 to set the C<PERLIO_F_XXXXX> flags, which may suffice.
689 void (*Setlinebuf)(pTHX_ PerlIO *f);
691 Mark the stream as line buffered. C<PerlIOBase_setlinebuf()> sets the
692 PERLIO_F_LINEBUF flag and is normally sufficient.
696 STDCHAR * (*Get_base)(pTHX_ PerlIO *f);
698 Allocate (if not already done so) the read buffer for this layer and
699 return pointer to it. Return NULL on failure.
703 Size_t (*Get_bufsiz)(pTHX_ PerlIO *f);
705 Return the number of bytes that last C<Fill()> put in the buffer.
709 STDCHAR * (*Get_ptr)(pTHX_ PerlIO *f);
711 Return the current read pointer relative to this layer's buffer.
715 SSize_t (*Get_cnt)(pTHX_ PerlIO *f);
717 Return the number of bytes left to be read in the current buffer.
721 void (*Set_ptrcnt)(pTHX_ PerlIO *f,
722 STDCHAR *ptr, SSize_t cnt);
724 Adjust the read pointer and count of bytes to match C<ptr> and/or C<cnt>.
725 The application (or layer above) must ensure they are consistent.
726 (Checking is allowed by the paranoid.)
730 =head2 Implementing PerlIO Layers
732 If you find the implementation document unclear or not sufficient,
733 look at the existing perlio layer implementations, which include:
737 =item * C implementations
739 PerlIO::encoding, PerlIO::scalar, PerlIO::via in the Perl core.
741 PerlIO::gzip and APR::PerlIO (mod_perl 2.0) on CPAN.
743 =item * Perl implementations
745 PerlIO::via::QuotedPrint in the Perl core and PerlIO::via::* on CPAN.
749 If you are creating a PerlIO layer, you may want to be lazy, in other
750 words, implement only the methods that interest you. The other methods
751 you can either replace with the "blank" methods
756 (which do nothing, and return zero and -1, respectively) or for
757 certain methods you may assume a default behaviour by using a NULL
758 method. The Open method looks for help in the 'parent' layer.
759 The following table summarizes the behaviour:
761 method behaviour with NULL
763 Clearerr PerlIOBase_clearerr
764 Close PerlIOBase_close
767 Error PerlIOBase_error
768 Fileno PerlIOBase_fileno
783 Setlinebuf PerlIOBase_setlinebuf
785 Unread PerlIOBase_unread
788 FAILURE Set errno (to EINVAL in UNIXish, to LIB$_INVARG in VMS) and
789 return -1 (for numeric return values) or NULL (for pointers)
790 INHERITED Inherited from the layer below
791 SUCCESS Return 0 (for numeric return values) or a pointer
795 The file C<perlio.c> provides the following layers:
801 A basic non-buffered layer which calls Unix/POSIX C<read()>, C<write()>,
802 C<lseek()>, C<close()>. No buffering. Even on platforms that distinguish
803 between O_TEXT and O_BINARY this layer is always O_BINARY.
807 A very complete generic buffering layer which provides the whole of
808 PerlIO API. It is also intended to be used as a "base class" for other
809 layers. (For example its C<Read()> method is implemented in terms of
810 the C<Get_cnt()>/C<Get_ptr()>/C<Set_ptrcnt()> methods).
812 "perlio" over "unix" provides a complete replacement for stdio as seen
813 via PerlIO API. This is the default for USE_PERLIO when system's stdio
814 does not permit perl's "fast gets" access, and which do not
815 distinguish between C<O_TEXT> and C<O_BINARY>.
819 A layer which provides the PerlIO API via the layer scheme, but
820 implements it by calling system's stdio. This is (currently) the default
821 if system's stdio provides sufficient access to allow perl's "fast gets"
822 access and which do not distinguish between C<O_TEXT> and C<O_BINARY>.
826 A layer derived using "perlio" as a base class. It provides Win32-like
827 "\n" to CR,LF translation. Can either be applied above "perlio" or serve
828 as the buffer layer itself. "crlf" over "unix" is the default if system
829 distinguishes between C<O_TEXT> and C<O_BINARY> opens. (At some point
830 "unix" will be replaced by a "native" Win32 IO layer on that platform,
831 as Win32's read/write layer has various drawbacks.) The "crlf" layer is
832 a reasonable model for a layer which transforms data in some way.
836 If Configure detects C<mmap()> functions this layer is provided (with
837 "perlio" as a "base") which does "read" operations by mmap()ing the
838 file. Performance improvement is marginal on modern systems, so it is
839 mainly there as a proof of concept. It is likely to be unbundled from
840 the core at some point. The "mmap" layer is a reasonable model for a
841 minimalist "derived" layer.
845 An "internal" derivative of "perlio" which can be used to provide
846 Unread() function for layers which have no buffer or cannot be
847 bothered. (Basically this layer's C<Fill()> pops itself off the stack
848 and so resumes reading from layer below.)
852 A dummy layer which never exists on the layer stack. Instead when
853 "pushed" it actually pops the stack removing itself, it then calls
854 Binmode function table entry on all the layers in the stack - normally
855 this (via PerlIOBase_binmode) removes any layers which do not have
856 C<PERLIO_K_RAW> bit set. Layers can modify that behaviour by defining
857 their own Binmode entry.
861 Another dummy layer. When pushed it pops itself and sets the
862 C<PERLIO_F_UTF8> flag on the layer which was (and now is once more)
863 the top of the stack.
867 In addition F<perlio.c> also provides a number of C<PerlIOBase_xxxx()>
868 functions which are intended to be used in the table slots of classes
869 which do not need to do anything special for a particular method.
871 =head2 Extension Layers
873 Layers can made available by extension modules. When an unknown layer
874 is encountered the PerlIO code will perform the equivalent of :
878 Where I<layer> is the unknown layer. F<PerlIO.pm> will then attempt to:
880 require PerlIO::layer;
882 If after that process the layer is still not defined then the C<open>
885 The following extension layers are bundled with perl:
893 makes this layer available, although F<PerlIO.pm> "knows" where to
894 find it. It is an example of a layer which takes an argument as it is
897 open( $fh, "<:encoding(iso-8859-7)", $pathname );
901 Provides support for reading data from and writing data to a scalar.
903 open( $fh, "+<:scalar", \$scalar );
905 When a handle is so opened, then reads get bytes from the string value
906 of I<$scalar>, and writes change the value. In both cases the position
907 in I<$scalar> starts as zero but can be altered via C<seek>, and
908 determined via C<tell>.
910 Please note that this layer is implied when calling open() thus:
912 open( $fh, "+<", \$scalar );
916 Provided to allow layers to be implemented as Perl code. For instance:
918 use PerlIO::via::StripHTML;
919 open( my $fh, "<:via(StripHTML)", "index.html" );
921 See L<PerlIO::via> for details.
927 Things that need to be done to improve this document.
933 Explain how to make a valid fh without going through open()(i.e. apply
934 a layer). For example if the file is not opened through perl, but we
935 want to get back a fh, like it was opened by Perl.
937 How PerlIO_apply_layera fits in, where its docs, was it made public?
939 Currently the example could be something like this:
941 PerlIO *foo_to_PerlIO(pTHX_ char *mode, ...)
943 char *mode; /* "w", "r", etc */
944 const char *layers = ":APR"; /* the layer name */
945 PerlIO *f = PerlIO_allocate(aTHX);
950 PerlIO_apply_layers(aTHX_ f, mode, layers);
953 PerlIOAPR *st = PerlIOSelf(f, PerlIOAPR);
954 /* fill in the st struct, as in _open() */
956 PerlIOBase(f)->flags |= PERLIO_F_OPEN;
965 fix/add the documentation in places marked as XXX.
969 The handling of errors by the layer is not specified. e.g. when $!
970 should be set explicitly, when the error handling should be just
971 delegated to the top layer.
973 Probably give some hints on using SETERRNO() or pointers to where they
978 I think it would help to give some concrete examples to make it easier
979 to understand the API. Of course I agree that the API has to be
980 concise, but since there is no second document that is more of a
981 guide, I think that it'd make it easier to start with the doc which is
982 an API, but has examples in it in places where things are unclear, to
983 a person who is not a PerlIO guru (yet).