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 Layers vs Disciplines
29 Initial discussion of the ability to modify IO streams behaviour used
30 the term "discipline" for the entities which were added. This came (I
31 believe) from the use of the term in "sfio", which in turn borrowed it
32 from "line disciplines" on Unix terminals. However, this document (and
33 the C code) uses the term "layer".
35 This is, I hope, a natural term given the implementation, and should
36 avoid connotations that are inherent in earlier uses of "discipline"
37 for things which are rather different.
39 =head2 Data Structures
41 The basic data structure is a PerlIOl:
43 typedef struct _PerlIO PerlIOl;
44 typedef struct _PerlIO_funcs PerlIO_funcs;
45 typedef PerlIOl *PerlIO;
49 PerlIOl * next; /* Lower layer */
50 PerlIO_funcs * tab; /* Functions for this layer */
51 IV flags; /* Various flags for state */
54 A C<PerlIOl *> is a pointer to the struct, and the I<application>
55 level C<PerlIO *> is a pointer to a C<PerlIOl *> - i.e. a pointer
56 to a pointer to the struct. This allows the application level C<PerlIO *>
57 to remain constant while the actual C<PerlIOl *> underneath
58 changes. (Compare perl's C<SV *> which remains constant while its
59 C<sv_any> field changes as the scalar's type changes.) An IO stream is
60 then in general represented as a pointer to this linked-list of
63 It should be noted that because of the double indirection in a C<PerlIO *>,
64 a C<< &(perlio->next) >> "is" a C<PerlIO *>, and so to some degree
65 at least one layer can use the "standard" API on the next layer down.
67 A "layer" is composed of two parts:
73 The functions and attributes of the "layer class".
77 The per-instance data for a particular handle.
81 =head2 Functions and Attributes
83 The functions and attributes are accessed via the "tab" (for table)
84 member of C<PerlIOl>. The functions (methods of the layer "class") are
85 fixed, and are defined by the C<PerlIO_funcs> type. They are broadly the
86 same as the public C<PerlIO_xxxxx> functions:
94 IV (*Pushed)(pTHX_ PerlIO *f,const char *mode,SV *arg, PerlIO_funcs *tab);
95 IV (*Popped)(pTHX_ PerlIO *f);
96 PerlIO * (*Open)(pTHX_ PerlIO_funcs *tab,
99 int fd, int imode, int perm,
101 int narg, SV **args);
102 IV (*Binmode)(pTHX_ PerlIO *f);
103 SV * (*Getarg)(pTHX_ PerlIO *f, CLONE_PARAMS *param, int flags)
104 IV (*Fileno)(pTHX_ PerlIO *f);
105 PerlIO * (*Dup)(pTHX_ PerlIO *f, PerlIO *o, CLONE_PARAMS *param, int flags)
106 /* Unix-like functions - cf sfio line disciplines */
107 SSize_t (*Read)(pTHX_ PerlIO *f, void *vbuf, Size_t count);
108 SSize_t (*Unread)(pTHX_ PerlIO *f, const void *vbuf, Size_t count);
109 SSize_t (*Write)(pTHX_ PerlIO *f, const void *vbuf, Size_t count);
110 IV (*Seek)(pTHX_ PerlIO *f, Off_t offset, int whence);
111 Off_t (*Tell)(pTHX_ PerlIO *f);
112 IV (*Close)(pTHX_ PerlIO *f);
113 /* Stdio-like buffered IO functions */
114 IV (*Flush)(pTHX_ PerlIO *f);
115 IV (*Fill)(pTHX_ PerlIO *f);
116 IV (*Eof)(pTHX_ PerlIO *f);
117 IV (*Error)(pTHX_ PerlIO *f);
118 void (*Clearerr)(pTHX_ PerlIO *f);
119 void (*Setlinebuf)(pTHX_ PerlIO *f);
120 /* Perl's snooping functions */
121 STDCHAR * (*Get_base)(pTHX_ PerlIO *f);
122 Size_t (*Get_bufsiz)(pTHX_ PerlIO *f);
123 STDCHAR * (*Get_ptr)(pTHX_ PerlIO *f);
124 SSize_t (*Get_cnt)(pTHX_ PerlIO *f);
125 void (*Set_ptrcnt)(pTHX_ PerlIO *f,STDCHAR *ptr,SSize_t cnt);
128 The first few members of the struct give a function table size for
129 compatibility check "name" for the layer, the size to C<malloc> for the per-instance data,
130 and some flags which are attributes of the class as whole (such as whether it is a buffering
131 layer), then follow the functions which fall into four basic groups:
137 Opening and setup functions
145 Stdio class buffering options.
149 Functions to support Perl's traditional "fast" access to the buffer.
153 A layer does not have to implement all the functions, but the whole
154 table has to be present. Unimplemented slots can be NULL (which will
155 result in an error when called) or can be filled in with stubs to
156 "inherit" behaviour from a "base class". This "inheritance" is fixed
157 for all instances of the layer, but as the layer chooses which stubs
158 to populate the table, limited "multiple inheritance" is possible.
160 =head2 Per-instance Data
162 The per-instance data are held in memory beyond the basic PerlIOl
163 struct, by making a PerlIOl the first member of the layer's struct
168 struct _PerlIO base; /* Base "class" info */
169 STDCHAR * buf; /* Start of buffer */
170 STDCHAR * end; /* End of valid part of buffer */
171 STDCHAR * ptr; /* Current position in buffer */
172 Off_t posn; /* Offset of buf into the file */
173 Size_t bufsiz; /* Real size of buffer */
174 IV oneword; /* Emergency buffer */
177 In this way (as for perl's scalars) a pointer to a PerlIOBuf can be
178 treated as a pointer to a PerlIOl.
180 =head2 Layers in action.
184 +-----------+ +----------+ +--------+
185 PerlIO ->| |--->| next |--->| NULL |
186 +-----------+ +----------+ +--------+
187 | | | buffer | | fd |
188 +-----------+ | | +--------+
192 The above attempts to show how the layer scheme works in a simple case.
193 The application's C<PerlIO *> points to an entry in the table(s)
194 representing open (allocated) handles. For example the first three slots
195 in the table correspond to C<stdin>,C<stdout> and C<stderr>. The table
196 in turn points to the current "top" layer for the handle - in this case
197 an instance of the generic buffering layer "perlio". That layer in turn
198 points to the next layer down - in this case the lowlevel "unix" layer.
200 The above is roughly equivalent to a "stdio" buffered stream, but with
201 much more flexibility:
207 If Unix level C<read>/C<write>/C<lseek> is not appropriate for (say)
208 sockets then the "unix" layer can be replaced (at open time or even
209 dynamically) with a "socket" layer.
213 Different handles can have different buffering schemes. The "top"
214 layer could be the "mmap" layer if reading disk files was quicker
215 using C<mmap> than C<read>. An "unbuffered" stream can be implemented
216 simply by not having a buffer layer.
220 Extra layers can be inserted to process the data as it flows through.
221 This was the driving need for including the scheme in perl 5.7.0+ - we
222 needed a mechanism to allow data to be translated between perl's
223 internal encoding (conceptually at least Unicode as UTF-8), and the
224 "native" format used by the system. This is provided by the
225 ":encoding(xxxx)" layer which typically sits above the buffering layer.
229 A layer can be added that does "\n" to CRLF translation. This layer
230 can be used on any platform, not just those that normally do such
235 =head2 Per-instance flag bits
237 The generic flag bits are a hybrid of C<O_XXXXX> style flags deduced
238 from the mode string passed to C<PerlIO_open()>, and state bits for
239 typical buffer layers.
247 =item PERLIO_F_CANWRITE
249 Writes are permitted, i.e. opened as "w" or "r+" or "a", etc.
251 =item PERLIO_F_CANREAD
253 Reads are permitted i.e. opened "r" or "w+" (or even "a+" - ick).
257 An error has occurred (for C<PerlIO_error()>).
259 =item PERLIO_F_TRUNCATE
261 Truncate file suggested by open mode.
263 =item PERLIO_F_APPEND
265 All writes should be appends.
269 Layer is performing Win32-like "\n" mapped to CR,LF for output and CR,LF
270 mapped to "\n" for input. Normally the provided "crlf" layer is the only
271 layer that need bother about this. C<PerlIO_binmode()> will mess with this
272 flag rather than add/remove layers if the C<PERLIO_K_CANCRLF> bit is set
273 for the layers class.
277 Data written to this layer should be UTF-8 encoded; data provided
278 by this layer should be considered UTF-8 encoded. Can be set on any layer
279 by ":utf8" dummy layer. Also set on ":encoding" layer.
283 Layer is unbuffered - i.e. write to next layer down should occur for
284 each write to this layer.
288 The buffer for this layer currently holds data written to it but not sent
293 The buffer for this layer currently holds unconsumed data read from
296 =item PERLIO_F_LINEBUF
298 Layer is line buffered. Write data should be passed to next layer down
299 whenever a "\n" is seen. Any data beyond the "\n" should then be
304 File has been C<unlink()>ed, or should be deleted on C<close()>.
310 =item PERLIO_F_FASTGETS
312 This instance of this layer supports the "fast C<gets>" interface.
313 Normally set based on C<PERLIO_K_FASTGETS> for the class and by the
314 existence of the function(s) in the table. However a class that
315 normally provides that interface may need to avoid it on a
316 particular instance. The "pending" layer needs to do this when
317 it is pushed above a layer which does not support the interface.
318 (Perl's C<sv_gets()> does not expect the streams fast C<gets> behaviour
319 to change during one "get".)
323 =head2 Methods in Detail
331 Size of the function table. This is compared against the value PerlIO
332 code "knows" as a compatibility check. Future versions I<may> be able
333 to tolerate layers compiled against an old version of the headers.
339 The name of the layer whose open() method Perl should invoke on
340 open(). For example if the layer is called APR, you will call:
342 open $fh, ">:APR", ...
344 and Perl knows that it has to invoke the PerlIOAPR_open() method
345 implemented by the APR layer.
351 The size of the per-instance data structure, e.g.:
355 If this field is zero then C<PerlIO_pushed> does not malloc anything
356 and assumes layer's Pushed function will do any required layer stack
357 manipulation - used to avoid malloc/free overhead for dummy layers.
358 If the field is non-zero it must be at least the size of C<PerlIOl>,
359 C<PerlIO_pushed> will allocate memory for the layer's data structures
360 and link new layer onto the stream's stack. (If the layer's Pushed
361 method returns an error indication the layer is popped again.)
369 =item * PERLIO_K_BUFFERED
371 The layer is buffered.
375 The layer is acceptable to have in a binmode(FH) stack - i.e. it does not
376 (or will configure itself not to) transform bytes passing through it.
378 =item * PERLIO_K_CANCRLF
380 Layer can translate between "\n" and CRLF line ends.
382 =item * PERLIO_K_FASTGETS
384 Layer allows buffer snooping.
386 =item * PERLIO_K_MULTIARG
388 Used when the layer's open() accepts more arguments than usual. The
389 extra arguments should come not before the C<MODE> argument. When this
390 flag is used it's up to the layer to validate the args.
396 IV (*Pushed)(pTHX_ PerlIO *f,const char *mode, SV *arg);
398 The only absolutely mandatory method. Called when the layer is pushed
399 onto the stack. The C<mode> argument may be NULL if this occurs
400 post-open. The C<arg> will be non-C<NULL> if an argument string was
401 passed. In most cases this should call C<PerlIOBase_pushed()> to
402 convert C<mode> into the appropriate C<PERLIO_F_XXXXX> flags in
403 addition to any actions the layer itself takes. If a layer is not
404 expecting an argument it need neither save the one passed to it, nor
405 provide C<Getarg()> (it could perhaps C<Perl_warn> that the argument
408 Returns 0 on success. On failure returns -1 and should set errno.
412 IV (*Popped)(pTHX_ PerlIO *f);
414 Called when the layer is popped from the stack. A layer will normally
415 be popped after C<Close()> is called. But a layer can be popped
416 without being closed if the program is dynamically managing layers on
417 the stream. In such cases C<Popped()> should free any resources
418 (buffers, translation tables, ...) not held directly in the layer's
419 struct. It should also C<Unread()> any unconsumed data that has been
420 read and buffered from the layer below back to that layer, so that it
421 can be re-provided to what ever is now above.
423 When a PerlIO * is destructed all the layers should pop back to
424 leave a NULL. However if ->Popped() returns true then perlio.c
425 assumes that either the layer has popped itself, or the layer is
426 super special and needs to be retained for other reasons.
428 Returns 0 on success and failure.
432 PerlIO * (*Open)(...);
434 The C<Open()> method has lots of arguments because it combines the
435 functions of perl's C<open>, C<PerlIO_open>, perl's C<sysopen>,
436 C<PerlIO_fdopen> and C<PerlIO_reopen>. The full prototype is as
439 PerlIO * (*Open)(pTHX_ PerlIO_funcs *tab,
442 int fd, int imode, int perm,
444 int narg, SV **args);
446 Open should (perhaps indirectly) call C<PerlIO_allocate()> to allocate
447 a slot in the table and associate it with the layers information for
448 the opened file, by calling C<PerlIO_push>. The I<layers> AV is an
449 array of all the layers destined for the C<PerlIO *>, and any
450 arguments passed to them, I<n> is the index into that array of the
451 layer being called. The macro C<PerlIOArg> will return a (possibly
452 C<NULL>) SV * for the argument passed to the layer.
454 The I<mode> string is an "C<fopen()>-like" string which would match
455 the regular expression C</^[I#]?[rwa]\+?[bt]?$/>.
457 The C<'I'> prefix is used during creation of C<stdin>..C<stderr> via
458 special C<PerlIO_fdopen> calls; the C<'#'> prefix means that this is
459 C<sysopen> and that I<imode> and I<perm> should be passed to
460 C<PerlLIO_open3>; C<'r'> means B<r>ead, C<'w'> means B<w>rite and
461 C<'a'> means B<a>ppend. The C<'+'> suffix means that both reading and
462 writing/appending are permitted. The C<'b'> suffix means file should
463 be binary, and C<'t'> means it is text. (Almost all layers should do
464 the IO in binary mode, and ignore the b/t bits. The C<:crlf> layer
465 should be pushed to handle the distinction.)
467 If I<old> is not C<NULL> then this is a C<PerlIO_reopen>. Perl itself
468 does not use this (yet?) and semantics are a little vague.
470 If I<fd> not negative then it is the numeric file descriptor I<fd>,
471 which will be open in a manner compatible with the supplied mode
472 string, the call is thus equivalent to C<PerlIO_fdopen>. In this case
473 I<nargs> will be zero.
475 If I<nargs> is greater than zero then it gives the number of arguments
476 passed to C<open>, otherwise it will be 1 if for example
477 C<PerlIO_open> was called. In simple cases SvPV_nolen(*args) is the
480 Having said all that translation-only layers do not need to provide
481 C<Open()> at all, but rather leave the opening to a lower level layer
482 and wait to be "pushed". If a layer does provide C<Open()> it should
483 normally call the C<Open()> method of next layer down (if any) and
484 then push itself on top if that succeeds.
486 Returns C<NULL> on failure.
490 IV (*Binmode)(pTHX_ PerlIO *f);
492 Optional. Used when C<:raw> layer is pushed (explicitly or as a result
493 of binmode(FH)). If not present layer will be popped. If present
494 should configure layer as binary (or pop itself) and return 0.
495 If it returns -1 for error C<binmode> will fail with layer
500 SV * (*Getarg)(pTHX_ PerlIO *f,
501 CLONE_PARAMS *param, int flags);
503 Optional. If present should return an SV * representing the string
504 argument passed to the layer when it was
505 pushed. e.g. ":encoding(ascii)" would return an SvPV with value
506 "ascii". (I<param> and I<flags> arguments can be ignored in most
511 IV (*Fileno)(pTHX_ PerlIO *f);
513 Returns the Unix/Posix numeric file descriptor for the handle. Normally
514 C<PerlIOBase_fileno()> (which just asks next layer down) will suffice
517 Returns -1 on error, which is considered to include the case where the
518 layer cannot provide such a file descriptor.
522 PerlIO * (*Dup)(pTHX_ PerlIO *f, PerlIO *o,
523 CLONE_PARAMS *param, int flags);
525 XXX: Needs more docs.
527 Used as part of the "clone" process when a thread is spawned (in which
528 case param will be non-NULL) and when a stream is being duplicated via
531 Similar to C<Open>, returns PerlIO* on success, C<NULL> on failure.
535 SSize_t (*Read)(pTHX_ PerlIO *f, void *vbuf, Size_t count);
537 Basic read operation.
539 Typically will call C<Fill> and manipulate pointers (possibly via the
540 API). C<PerlIOBuf_read()> may be suitable for derived classes which
541 provide "fast gets" methods.
543 Returns actual bytes read, or -1 on an error.
547 SSize_t (*Unread)(pTHX_ PerlIO *f,
548 const void *vbuf, Size_t count);
550 A superset of stdio's C<ungetc()>. Should arrange for future reads to
551 see the bytes in C<vbuf>. If there is no obviously better implementation
552 then C<PerlIOBase_unread()> provides the function by pushing a "fake"
553 "pending" layer above the calling layer.
555 Returns the number of unread chars.
559 SSize_t (*Write)(PerlIO *f, const void *vbuf, Size_t count);
561 Basic write operation.
563 Returns bytes written or -1 on an error.
567 IV (*Seek)(pTHX_ PerlIO *f, Off_t offset, int whence);
569 Position the file pointer. Should normally call its own C<Flush>
570 method and then the C<Seek> method of next layer down.
572 Returns 0 on success, -1 on failure.
576 Off_t (*Tell)(pTHX_ PerlIO *f);
578 Return the file pointer. May be based on layers cached concept of
579 position to avoid overhead.
581 Returns -1 on failure to get the file pointer.
585 IV (*Close)(pTHX_ PerlIO *f);
587 Close the stream. Should normally call C<PerlIOBase_close()> to flush
588 itself and close layers below, and then deallocate any data structures
589 (buffers, translation tables, ...) not held directly in the data
592 Returns 0 on success, -1 on failure.
596 IV (*Flush)(pTHX_ PerlIO *f);
598 Should make stream's state consistent with layers below. That is, any
599 buffered write data should be written, and file position of lower layers
600 adjusted for data read from below but not actually consumed.
601 (Should perhaps C<Unread()> such data to the lower layer.)
603 Returns 0 on success, -1 on failure.
607 IV (*Fill)(pTHX_ PerlIO *f);
609 The buffer for this layer should be filled (for read) from layer
610 below. When you "subclass" PerlIOBuf layer, you want to use its
611 I<_read> method and to supply your own fill method, which fills the
614 Returns 0 on success, -1 on failure.
618 IV (*Eof)(pTHX_ PerlIO *f);
620 Return end-of-file indicator. C<PerlIOBase_eof()> is normally sufficient.
622 Returns 0 on end-of-file, 1 if not end-of-file, -1 on error.
626 IV (*Error)(pTHX_ PerlIO *f);
628 Return error indicator. C<PerlIOBase_error()> is normally sufficient.
630 Returns 1 if there is an error (usually when C<PERLIO_F_ERROR> is set,
635 void (*Clearerr)(pTHX_ PerlIO *f);
637 Clear end-of-file and error indicators. Should call C<PerlIOBase_clearerr()>
638 to set the C<PERLIO_F_XXXXX> flags, which may suffice.
642 void (*Setlinebuf)(pTHX_ PerlIO *f);
644 Mark the stream as line buffered. C<PerlIOBase_setlinebuf()> sets the
645 PERLIO_F_LINEBUF flag and is normally sufficient.
649 STDCHAR * (*Get_base)(pTHX_ PerlIO *f);
651 Allocate (if not already done so) the read buffer for this layer and
652 return pointer to it. Return NULL on failure.
656 Size_t (*Get_bufsiz)(pTHX_ PerlIO *f);
658 Return the number of bytes that last C<Fill()> put in the buffer.
662 STDCHAR * (*Get_ptr)(pTHX_ PerlIO *f);
664 Return the current read pointer relative to this layer's buffer.
668 SSize_t (*Get_cnt)(pTHX_ PerlIO *f);
670 Return the number of bytes left to be read in the current buffer.
674 void (*Set_ptrcnt)(pTHX_ PerlIO *f,
675 STDCHAR *ptr, SSize_t cnt);
677 Adjust the read pointer and count of bytes to match C<ptr> and/or C<cnt>.
678 The application (or layer above) must ensure they are consistent.
679 (Checking is allowed by the paranoid.)
683 =head2 Implementing PerlIO Layers
685 If you are creating a PerlIO layer, you may want to be lazy, in other
686 words, implement only the methods that interest you. The other methods
687 you can either replace with the "blank" methods
692 (which do nothing, and return zero and -1, respectively) or for
693 certain methods you may assume a default behaviour by using a NULL
694 method. The Open method looks for help in the 'parent' layer.
695 The following table summarizes the behaviour:
697 method behaviour with NULL
699 Clearerr PerlIOBase_clearerr
700 Close PerlIOBase_close
703 Error PerlIOBase_error
704 Fileno PerlIOBase_fileno
719 Setlinebuf PerlIOBase_setlinebuf
721 Unread PerlIOBase_unread
724 FAILURE Set errno (to EINVAL in UNIXish, to LIB$_INVARG in VMS) and
725 return -1 (for numeric return values) or NULL (for pointers)
726 INHERITED Inherited from the layer below
727 SUCCESS Return 0 (for numeric return values) or a pointer
731 The file C<perlio.c> provides the following layers:
737 A basic non-buffered layer which calls Unix/POSIX C<read()>, C<write()>,
738 C<lseek()>, C<close()>. No buffering. Even on platforms that distinguish
739 between O_TEXT and O_BINARY this layer is always O_BINARY.
743 A very complete generic buffering layer which provides the whole of
744 PerlIO API. It is also intended to be used as a "base class" for other
745 layers. (For example its C<Read()> method is implemented in terms of
746 the C<Get_cnt()>/C<Get_ptr()>/C<Set_ptrcnt()> methods).
748 "perlio" over "unix" provides a complete replacement for stdio as seen
749 via PerlIO API. This is the default for USE_PERLIO when system's stdio
750 does not permit perl's "fast gets" access, and which do not
751 distinguish between C<O_TEXT> and C<O_BINARY>.
755 A layer which provides the PerlIO API via the layer scheme, but
756 implements it by calling system's stdio. This is (currently) the default
757 if system's stdio provides sufficient access to allow perl's "fast gets"
758 access and which do not distinguish between C<O_TEXT> and C<O_BINARY>.
762 A layer derived using "perlio" as a base class. It provides Win32-like
763 "\n" to CR,LF translation. Can either be applied above "perlio" or serve
764 as the buffer layer itself. "crlf" over "unix" is the default if system
765 distinguishes between C<O_TEXT> and C<O_BINARY> opens. (At some point
766 "unix" will be replaced by a "native" Win32 IO layer on that platform,
767 as Win32's read/write layer has various drawbacks.) The "crlf" layer is
768 a reasonable model for a layer which transforms data in some way.
772 If Configure detects C<mmap()> functions this layer is provided (with
773 "perlio" as a "base") which does "read" operations by mmap()ing the
774 file. Performance improvement is marginal on modern systems, so it is
775 mainly there as a proof of concept. It is likely to be unbundled from
776 the core at some point. The "mmap" layer is a reasonable model for a
777 minimalist "derived" layer.
781 An "internal" derivative of "perlio" which can be used to provide
782 Unread() function for layers which have no buffer or cannot be
783 bothered. (Basically this layer's C<Fill()> pops itself off the stack
784 and so resumes reading from layer below.)
788 A dummy layer which never exists on the layer stack. Instead when
789 "pushed" it actually pops the stack removing itself, it then calls
790 Binmode function table entry on all the layers in the stack - normally
791 this (via PerlIOBase_binmode) removes any layers which do not have
792 C<PERLIO_K_RAW> bit set. Layers can modify that behaviour by defining
793 their own Binmode entry.
797 Another dummy layer. When pushed it pops itself and sets the
798 C<PERLIO_F_UTF8> flag on the layer which was (and now is once more)
799 the top of the stack.
803 In addition F<perlio.c> also provides a number of C<PerlIOBase_xxxx()>
804 functions which are intended to be used in the table slots of classes
805 which do not need to do anything special for a particular method.
807 =head2 Extension Layers
809 Layers can made available by extension modules. When an unknown layer
810 is encountered the PerlIO code will perform the equivalent of :
814 Where I<layer> is the unknown layer. F<PerlIO.pm> will then attempt to:
816 require PerlIO::layer;
818 If after that process the layer is still not defined then the C<open>
821 The following extension layers are bundled with perl:
829 makes this layer available, although F<PerlIO.pm> "knows" where to
830 find it. It is an example of a layer which takes an argument as it is
833 open( $fh, "<:encoding(iso-8859-7)", $pathname );
837 Provides support for reading data from and writing data to a scalar.
839 open( $fh, "+<:scalar", \$scalar );
841 When a handle is so opened, then reads get bytes from the string value
842 of I<$scalar>, and writes change the value. In both cases the position
843 in I<$scalar> starts as zero but can be altered via C<seek>, and
844 determined via C<tell>.
846 Please note that this layer is implied when calling open() thus:
848 open( $fh, "+<", \$scalar );
852 Provided to allow layers to be implemented as Perl code. For instance:
854 use PerlIO::via::StripHTML;
855 open( my $fh, "<:via(StripHTML)", "index.html" );
857 See L<PerlIO::via> for details.
863 Things that need to be done to improve this document.
869 Explain how to make a valid fh without going through open()(i.e. apply
870 a layer). For example if the file is not opened through perl, but we
871 want to get back a fh, like it was opened by Perl.
873 How PerlIO_apply_layera fits in, where its docs, was it made public?
875 Currently the example could be something like this:
877 PerlIO *foo_to_PerlIO(pTHX_ char *mode, ...)
879 char *mode; /* "w", "r", etc */
880 const char *layers = ":APR"; /* the layer name */
881 PerlIO *f = PerlIO_allocate(aTHX);
886 PerlIO_apply_layers(aTHX_ f, mode, layers);
889 PerlIOAPR *st = PerlIOSelf(f, PerlIOAPR);
890 /* fill in the st struct, as in _open() */
892 PerlIOBase(f)->flags |= PERLIO_F_OPEN;
901 fix/add the documentation in places marked as XXX.
905 The handling of errors by the layer is not specified. e.g. when $!
906 should be set explicitly, when the error handling should be just
907 delegated to the top layer.
909 Probably give some hints on using SETERRNO() or pointers to where they
914 I think it would help to give some concrete examples to make it easier
915 to understand the API. Of course I agree that the API has to be
916 concise, but since there is no second document that is more of a
917 guide, I think that it'd make it easier to start with the doc which is
918 an API, but has examples in it in places where things are unclear, to
919 a person who is not a PerlIO guru (yet).