This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Split out study magic from pos magic.
[perl5.git] / pod / perlguts.pod
CommitLineData
a0d0e21e
LW
1=head1 NAME
2
954c1994 3perlguts - Introduction to the Perl API
a0d0e21e
LW
4
5=head1 DESCRIPTION
6
b3b6085d 7This document attempts to describe how to use the Perl API, as well as
06f6df17 8to provide some info on the basic workings of the Perl core. It is far
b3b6085d
PP
9from complete and probably contains many errors. Please refer any
10questions or comments to the author below.
a0d0e21e 11
0a753a76
PP
12=head1 Variables
13
5f05dabc 14=head2 Datatypes
a0d0e21e
LW
15
16Perl has three typedefs that handle Perl's three main data types:
17
18 SV Scalar Value
19 AV Array Value
20 HV Hash Value
21
d1b91892 22Each typedef has specific routines that manipulate the various data types.
a0d0e21e
LW
23
24=head2 What is an "IV"?
25
954c1994 26Perl uses a special typedef IV which is a simple signed integer type that is
5f05dabc 27guaranteed to be large enough to hold a pointer (as well as an integer).
954c1994 28Additionally, there is the UV, which is simply an unsigned IV.
a0d0e21e 29
d1b91892 30Perl also uses two special typedefs, I32 and I16, which will always be at
954c1994 31least 32-bits and 16-bits long, respectively. (Again, there are U32 and U16,
20dbd849
NC
32as well.) They will usually be exactly 32 and 16 bits long, but on Crays
33they will both be 64 bits.
a0d0e21e 34
54310121 35=head2 Working with SVs
a0d0e21e 36
20dbd849
NC
37An SV can be created and loaded with one command. There are five types of
38values that can be loaded: an integer value (IV), an unsigned integer
39value (UV), a double (NV), a string (PV), and another scalar (SV).
a0d0e21e 40
20dbd849 41The seven routines are:
a0d0e21e
LW
42
43 SV* newSViv(IV);
20dbd849 44 SV* newSVuv(UV);
a0d0e21e 45 SV* newSVnv(double);
06f6df17
RGS
46 SV* newSVpv(const char*, STRLEN);
47 SV* newSVpvn(const char*, STRLEN);
46fc3d4c 48 SV* newSVpvf(const char*, ...);
a0d0e21e
LW
49 SV* newSVsv(SV*);
50
06f6df17
RGS
51C<STRLEN> is an integer type (Size_t, usually defined as size_t in
52F<config.h>) guaranteed to be large enough to represent the size of
53any string that perl can handle.
54
55In the unlikely case of a SV requiring more complex initialisation, you
56can create an empty SV with newSV(len). If C<len> is 0 an empty SV of
57type NULL is returned, else an SV of type PV is returned with len + 1 (for
58the NUL) bytes of storage allocated, accessible via SvPVX. In both cases
59the SV has value undef.
20dbd849 60
06f6df17
RGS
61 SV *sv = newSV(0); /* no storage allocated */
62 SV *sv = newSV(10); /* 10 (+1) bytes of uninitialised storage allocated */
20dbd849 63
06f6df17 64To change the value of an I<already-existing> SV, there are eight routines:
a0d0e21e
LW
65
66 void sv_setiv(SV*, IV);
deb3007b 67 void sv_setuv(SV*, UV);
a0d0e21e 68 void sv_setnv(SV*, double);
08105a92 69 void sv_setpv(SV*, const char*);
06f6df17 70 void sv_setpvn(SV*, const char*, STRLEN)
46fc3d4c 71 void sv_setpvf(SV*, const char*, ...);
5b3218b9 72 void sv_vsetpvfn(SV*, const char*, STRLEN, va_list *, SV **, I32, bool *);
a0d0e21e
LW
73 void sv_setsv(SV*, SV*);
74
75Notice that you can choose to specify the length of the string to be
9da1e3b5
MUN
76assigned by using C<sv_setpvn>, C<newSVpvn>, or C<newSVpv>, or you may
77allow Perl to calculate the length by using C<sv_setpv> or by specifying
780 as the second argument to C<newSVpv>. Be warned, though, that Perl will
79determine the string's length by using C<strlen>, which depends on the
9abd00ed
GS
80string terminating with a NUL character.
81
82The arguments of C<sv_setpvf> are processed like C<sprintf>, and the
83formatted output becomes the value.
84
328bf373 85C<sv_vsetpvfn> is an analogue of C<vsprintf>, but it allows you to specify
9abd00ed
GS
86either a pointer to a variable argument list or the address and length of
87an array of SVs. The last argument points to a boolean; on return, if that
88boolean is true, then locale-specific information has been used to format
c2611fb3 89the string, and the string's contents are therefore untrustworthy (see
9abd00ed
GS
90L<perlsec>). This pointer may be NULL if that information is not
91important. Note that this function requires you to specify the length of
92the format.
93
9da1e3b5
MUN
94The C<sv_set*()> functions are not generic enough to operate on values
95that have "magic". See L<Magic Virtual Tables> later in this document.
a0d0e21e 96
a3cb178b
GS
97All SVs that contain strings should be terminated with a NUL character.
98If it is not NUL-terminated there is a risk of
5f05dabc
PP
99core dumps and corruptions from code which passes the string to C
100functions or system calls which expect a NUL-terminated string.
101Perl's own functions typically add a trailing NUL for this reason.
102Nevertheless, you should be very careful when you pass a string stored
103in an SV to a C function or system call.
104
a0d0e21e
LW
105To access the actual value that an SV points to, you can use the macros:
106
107 SvIV(SV*)
954c1994 108 SvUV(SV*)
a0d0e21e
LW
109 SvNV(SV*)
110 SvPV(SV*, STRLEN len)
1fa8b10d 111 SvPV_nolen(SV*)
a0d0e21e 112
954c1994 113which will automatically coerce the actual scalar type into an IV, UV, double,
a0d0e21e
LW
114or string.
115
116In the C<SvPV> macro, the length of the string returned is placed into the
1fa8b10d
JD
117variable C<len> (this is a macro, so you do I<not> use C<&len>). If you do
118not care what the length of the data is, use the C<SvPV_nolen> macro.
119Historically the C<SvPV> macro with the global variable C<PL_na> has been
120used in this case. But that can be quite inefficient because C<PL_na> must
121be accessed in thread-local storage in threaded Perl. In any case, remember
122that Perl allows arbitrary strings of data that may both contain NULs and
123might not be terminated by a NUL.
a0d0e21e 124
ce2f5d8f
KA
125Also remember that C doesn't allow you to safely say C<foo(SvPV(s, len),
126len);>. It might work with your compiler, but it won't work for everyone.
127Break this sort of statement up into separate assignments:
128
1aa6ea50
JC
129 SV *s;
130 STRLEN len;
131 char * ptr;
132 ptr = SvPV(s, len);
133 foo(ptr, len);
ce2f5d8f 134
07fa94a1 135If you want to know if the scalar value is TRUE, you can use:
a0d0e21e
LW
136
137 SvTRUE(SV*)
138
139Although Perl will automatically grow strings for you, if you need to force
140Perl to allocate more memory for your SV, you can use the macro
141
142 SvGROW(SV*, STRLEN newlen)
143
144which will determine if more memory needs to be allocated. If so, it will
145call the function C<sv_grow>. Note that C<SvGROW> can only increase, not
5f05dabc
PP
146decrease, the allocated memory of an SV and that it does not automatically
147add a byte for the a trailing NUL (perl's own string functions typically do
8ebc5c01 148C<SvGROW(sv, len + 1)>).
a0d0e21e
LW
149
150If you have an SV and want to know what kind of data Perl thinks is stored
151in it, you can use the following macros to check the type of SV you have.
152
153 SvIOK(SV*)
154 SvNOK(SV*)
155 SvPOK(SV*)
156
157You can get and set the current length of the string stored in an SV with
158the following macros:
159
160 SvCUR(SV*)
161 SvCUR_set(SV*, I32 val)
162
cb1a09d0
AD
163You can also get a pointer to the end of the string stored in the SV
164with the macro:
165
166 SvEND(SV*)
167
168But note that these last three macros are valid only if C<SvPOK()> is true.
a0d0e21e 169
d1b91892
AD
170If you want to append something to the end of string stored in an C<SV*>,
171you can use the following functions:
172
08105a92 173 void sv_catpv(SV*, const char*);
e65f3abd 174 void sv_catpvn(SV*, const char*, STRLEN);
46fc3d4c 175 void sv_catpvf(SV*, const char*, ...);
328bf373 176 void sv_vcatpvfn(SV*, const char*, STRLEN, va_list *, SV **, I32, bool);
d1b91892
AD
177 void sv_catsv(SV*, SV*);
178
179The first function calculates the length of the string to be appended by
180using C<strlen>. In the second, you specify the length of the string
46fc3d4c 181yourself. The third function processes its arguments like C<sprintf> and
9abd00ed
GS
182appends the formatted output. The fourth function works like C<vsprintf>.
183You can specify the address and length of an array of SVs instead of the
184va_list argument. The fifth function extends the string stored in the first
185SV with the string stored in the second SV. It also forces the second SV
186to be interpreted as a string.
187
188The C<sv_cat*()> functions are not generic enough to operate on values that
189have "magic". See L<Magic Virtual Tables> later in this document.
d1b91892 190
a0d0e21e
LW
191If you know the name of a scalar variable, you can get a pointer to its SV
192by using the following:
193
64ace3f8 194 SV* get_sv("package::varname", 0);
a0d0e21e
LW
195
196This returns NULL if the variable does not exist.
197
d1b91892 198If you want to know if this variable (or any other SV) is actually C<defined>,
a0d0e21e
LW
199you can call:
200
201 SvOK(SV*)
202
06f6df17 203The scalar C<undef> value is stored in an SV instance called C<PL_sv_undef>.
9adebda4
SB
204
205Its address can be used whenever an C<SV*> is needed. Make sure that
206you don't try to compare a random sv with C<&PL_sv_undef>. For example
207when interfacing Perl code, it'll work correctly for:
208
209 foo(undef);
210
211But won't work when called as:
212
213 $x = undef;
214 foo($x);
215
216So to repeat always use SvOK() to check whether an sv is defined.
217
218Also you have to be careful when using C<&PL_sv_undef> as a value in
219AVs or HVs (see L<AVs, HVs and undefined values>).
a0d0e21e 220
06f6df17
RGS
221There are also the two values C<PL_sv_yes> and C<PL_sv_no>, which contain
222boolean TRUE and FALSE values, respectively. Like C<PL_sv_undef>, their
223addresses can be used whenever an C<SV*> is needed.
a0d0e21e 224
9cde0e7f 225Do not be fooled into thinking that C<(SV *) 0> is the same as C<&PL_sv_undef>.
a0d0e21e
LW
226Take this code:
227
228 SV* sv = (SV*) 0;
229 if (I-am-to-return-a-real-value) {
230 sv = sv_2mortal(newSViv(42));
231 }
232 sv_setsv(ST(0), sv);
233
234This code tries to return a new SV (which contains the value 42) if it should
04343c6d 235return a real value, or undef otherwise. Instead it has returned a NULL
a0d0e21e 236pointer which, somewhere down the line, will cause a segmentation violation,
06f6df17
RGS
237bus error, or just weird results. Change the zero to C<&PL_sv_undef> in the
238first line and all will be well.
a0d0e21e
LW
239
240To free an SV that you've created, call C<SvREFCNT_dec(SV*)>. Normally this
3fe9a6f1 241call is not necessary (see L<Reference Counts and Mortality>).
a0d0e21e 242
94dde4fb
SC
243=head2 Offsets
244
245Perl provides the function C<sv_chop> to efficiently remove characters
246from the beginning of a string; you give it an SV and a pointer to
da75cd15 247somewhere inside the PV, and it discards everything before the
94dde4fb
SC
248pointer. The efficiency comes by means of a little hack: instead of
249actually removing the characters, C<sv_chop> sets the flag C<OOK>
250(offset OK) to signal to other functions that the offset hack is in
251effect, and it puts the number of bytes chopped off into the IV field
252of the SV. It then moves the PV pointer (called C<SvPVX>) forward that
00aadd71 253many bytes, and adjusts C<SvCUR> and C<SvLEN>.
94dde4fb
SC
254
255Hence, at this point, the start of the buffer that we allocated lives
256at C<SvPVX(sv) - SvIV(sv)> in memory and the PV pointer is pointing
257into the middle of this allocated storage.
258
259This is best demonstrated by example:
260
261 % ./perl -Ilib -MDevel::Peek -le '$a="12345"; $a=~s/.//; Dump($a)'
262 SV = PVIV(0x8128450) at 0x81340f0
263 REFCNT = 1
264 FLAGS = (POK,OOK,pPOK)
265 IV = 1 (OFFSET)
266 PV = 0x8135781 ( "1" . ) "2345"\0
267 CUR = 4
268 LEN = 5
269
270Here the number of bytes chopped off (1) is put into IV, and
271C<Devel::Peek::Dump> helpfully reminds us that this is an offset. The
272portion of the string between the "real" and the "fake" beginnings is
273shown in parentheses, and the values of C<SvCUR> and C<SvLEN> reflect
274the fake beginning, not the real one.
275
fe854a6f 276Something similar to the offset hack is performed on AVs to enable
319cef53
SC
277efficient shifting and splicing off the beginning of the array; while
278C<AvARRAY> points to the first element in the array that is visible from
279Perl, C<AvALLOC> points to the real start of the C array. These are
280usually the same, but a C<shift> operation can be carried out by
6de131f0 281increasing C<AvARRAY> by one and decreasing C<AvFILL> and C<AvMAX>.
319cef53
SC
282Again, the location of the real start of the C array only comes into
283play when freeing the array. See C<av_shift> in F<av.c>.
284
d1b91892 285=head2 What's Really Stored in an SV?
a0d0e21e
LW
286
287Recall that the usual method of determining the type of scalar you have is
5f05dabc 288to use C<Sv*OK> macros. Because a scalar can be both a number and a string,
d1b91892 289usually these macros will always return TRUE and calling the C<Sv*V>
a0d0e21e
LW
290macros will do the appropriate conversion of string to integer/double or
291integer/double to string.
292
293If you I<really> need to know if you have an integer, double, or string
294pointer in an SV, you can use the following three macros instead:
295
296 SvIOKp(SV*)
297 SvNOKp(SV*)
298 SvPOKp(SV*)
299
300These will tell you if you truly have an integer, double, or string pointer
d1b91892 301stored in your SV. The "p" stands for private.
a0d0e21e 302
9e9796d6
JH
303The are various ways in which the private and public flags may differ.
304For example, a tied SV may have a valid underlying value in the IV slot
305(so SvIOKp is true), but the data should be accessed via the FETCH
306routine rather than directly, so SvIOK is false. Another is when
d7f8936a 307numeric conversion has occurred and precision has been lost: only the
9e9796d6
JH
308private flag is set on 'lossy' values. So when an NV is converted to an
309IV with loss, SvIOKp, SvNOKp and SvNOK will be set, while SvIOK wont be.
310
07fa94a1 311In general, though, it's best to use the C<Sv*V> macros.
a0d0e21e 312
54310121 313=head2 Working with AVs
a0d0e21e 314
07fa94a1
JO
315There are two ways to create and load an AV. The first method creates an
316empty AV:
a0d0e21e
LW
317
318 AV* newAV();
319
54310121 320The second method both creates the AV and initially populates it with SVs:
a0d0e21e
LW
321
322 AV* av_make(I32 num, SV **ptr);
323
5f05dabc 324The second argument points to an array containing C<num> C<SV*>'s. Once the
54310121 325AV has been created, the SVs can be destroyed, if so desired.
a0d0e21e 326
54310121 327Once the AV has been created, the following operations are possible on AVs:
a0d0e21e
LW
328
329 void av_push(AV*, SV*);
330 SV* av_pop(AV*);
331 SV* av_shift(AV*);
332 void av_unshift(AV*, I32 num);
333
334These should be familiar operations, with the exception of C<av_unshift>.
335This routine adds C<num> elements at the front of the array with the C<undef>
336value. You must then use C<av_store> (described below) to assign values
337to these new elements.
338
339Here are some other functions:
340
5f05dabc 341 I32 av_len(AV*);
a0d0e21e 342 SV** av_fetch(AV*, I32 key, I32 lval);
a0d0e21e 343 SV** av_store(AV*, I32 key, SV* val);
a0d0e21e 344
5f05dabc
PP
345The C<av_len> function returns the highest index value in array (just
346like $#array in Perl). If the array is empty, -1 is returned. The
347C<av_fetch> function returns the value at index C<key>, but if C<lval>
348is non-zero, then C<av_fetch> will store an undef value at that index.
04343c6d
GS
349The C<av_store> function stores the value C<val> at index C<key>, and does
350not increment the reference count of C<val>. Thus the caller is responsible
351for taking care of that, and if C<av_store> returns NULL, the caller will
352have to decrement the reference count to avoid a memory leak. Note that
353C<av_fetch> and C<av_store> both return C<SV**>'s, not C<SV*>'s as their
354return value.
d1b91892 355
a0d0e21e 356 void av_clear(AV*);
a0d0e21e 357 void av_undef(AV*);
cb1a09d0 358 void av_extend(AV*, I32 key);
5f05dabc
PP
359
360The C<av_clear> function deletes all the elements in the AV* array, but
361does not actually delete the array itself. The C<av_undef> function will
362delete all the elements in the array plus the array itself. The
adc882cf
GS
363C<av_extend> function extends the array so that it contains at least C<key+1>
364elements. If C<key+1> is less than the currently allocated length of the array,
365then nothing is done.
a0d0e21e
LW
366
367If you know the name of an array variable, you can get a pointer to its AV
368by using the following:
369
cbfd0a87 370 AV* get_av("package::varname", 0);
a0d0e21e
LW
371
372This returns NULL if the variable does not exist.
373
04343c6d
GS
374See L<Understanding the Magic of Tied Hashes and Arrays> for more
375information on how to use the array access functions on tied arrays.
376
54310121 377=head2 Working with HVs
a0d0e21e
LW
378
379To create an HV, you use the following routine:
380
381 HV* newHV();
382
54310121 383Once the HV has been created, the following operations are possible on HVs:
a0d0e21e 384
08105a92
GS
385 SV** hv_store(HV*, const char* key, U32 klen, SV* val, U32 hash);
386 SV** hv_fetch(HV*, const char* key, U32 klen, I32 lval);
a0d0e21e 387
5f05dabc
PP
388The C<klen> parameter is the length of the key being passed in (Note that
389you cannot pass 0 in as a value of C<klen> to tell Perl to measure the
390length of the key). The C<val> argument contains the SV pointer to the
54310121 391scalar being stored, and C<hash> is the precomputed hash value (zero if
5f05dabc
PP
392you want C<hv_store> to calculate it for you). The C<lval> parameter
393indicates whether this fetch is actually a part of a store operation, in
394which case a new undefined value will be added to the HV with the supplied
395key and C<hv_fetch> will return as if the value had already existed.
a0d0e21e 396
5f05dabc
PP
397Remember that C<hv_store> and C<hv_fetch> return C<SV**>'s and not just
398C<SV*>. To access the scalar value, you must first dereference the return
399value. However, you should check to make sure that the return value is
400not NULL before dereferencing it.
a0d0e21e
LW
401
402These two functions check if a hash table entry exists, and deletes it.
403
08105a92
GS
404 bool hv_exists(HV*, const char* key, U32 klen);
405 SV* hv_delete(HV*, const char* key, U32 klen, I32 flags);
a0d0e21e 406
5f05dabc
PP
407If C<flags> does not include the C<G_DISCARD> flag then C<hv_delete> will
408create and return a mortal copy of the deleted value.
409
a0d0e21e
LW
410And more miscellaneous functions:
411
412 void hv_clear(HV*);
a0d0e21e 413 void hv_undef(HV*);
5f05dabc
PP
414
415Like their AV counterparts, C<hv_clear> deletes all the entries in the hash
416table but does not actually delete the hash table. The C<hv_undef> deletes
417both the entries and the hash table itself.
a0d0e21e 418
d1b91892
AD
419Perl keeps the actual data in linked list of structures with a typedef of HE.
420These contain the actual key and value pointers (plus extra administrative
421overhead). The key is a string pointer; the value is an C<SV*>. However,
422once you have an C<HE*>, to get the actual key and value, use the routines
423specified below.
424
a0d0e21e
LW
425 I32 hv_iterinit(HV*);
426 /* Prepares starting point to traverse hash table */
427 HE* hv_iternext(HV*);
428 /* Get the next entry, and return a pointer to a
429 structure that has both the key and value */
430 char* hv_iterkey(HE* entry, I32* retlen);
431 /* Get the key from an HE structure and also return
432 the length of the key string */
cb1a09d0 433 SV* hv_iterval(HV*, HE* entry);
d1be9408 434 /* Return an SV pointer to the value of the HE
a0d0e21e 435 structure */
cb1a09d0 436 SV* hv_iternextsv(HV*, char** key, I32* retlen);
d1b91892
AD
437 /* This convenience routine combines hv_iternext,
438 hv_iterkey, and hv_iterval. The key and retlen
439 arguments are return values for the key and its
440 length. The value is returned in the SV* argument */
a0d0e21e
LW
441
442If you know the name of a hash variable, you can get a pointer to its HV
443by using the following:
444
6673a63c 445 HV* get_hv("package::varname", 0);
a0d0e21e
LW
446
447This returns NULL if the variable does not exist.
448
8ebc5c01 449The hash algorithm is defined in the C<PERL_HASH(hash, key, klen)> macro:
a0d0e21e 450
a0d0e21e 451 hash = 0;
ab192400
GS
452 while (klen--)
453 hash = (hash * 33) + *key++;
87275199 454 hash = hash + (hash >> 5); /* after 5.6 */
ab192400 455
87275199 456The last step was added in version 5.6 to improve distribution of
ab192400 457lower bits in the resulting hash value.
a0d0e21e 458
04343c6d
GS
459See L<Understanding the Magic of Tied Hashes and Arrays> for more
460information on how to use the hash access functions on tied hashes.
461
1e422769
PP
462=head2 Hash API Extensions
463
464Beginning with version 5.004, the following functions are also supported:
465
466 HE* hv_fetch_ent (HV* tb, SV* key, I32 lval, U32 hash);
467 HE* hv_store_ent (HV* tb, SV* key, SV* val, U32 hash);
c47ff5f1 468
1e422769
PP
469 bool hv_exists_ent (HV* tb, SV* key, U32 hash);
470 SV* hv_delete_ent (HV* tb, SV* key, I32 flags, U32 hash);
c47ff5f1 471
1e422769
PP
472 SV* hv_iterkeysv (HE* entry);
473
474Note that these functions take C<SV*> keys, which simplifies writing
475of extension code that deals with hash structures. These functions
476also allow passing of C<SV*> keys to C<tie> functions without forcing
477you to stringify the keys (unlike the previous set of functions).
478
479They also return and accept whole hash entries (C<HE*>), making their
480use more efficient (since the hash number for a particular string
4a4eefd0
GS
481doesn't have to be recomputed every time). See L<perlapi> for detailed
482descriptions.
1e422769
PP
483
484The following macros must always be used to access the contents of hash
485entries. Note that the arguments to these macros must be simple
486variables, since they may get evaluated more than once. See
4a4eefd0 487L<perlapi> for detailed descriptions of these macros.
1e422769
PP
488
489 HePV(HE* he, STRLEN len)
490 HeVAL(HE* he)
491 HeHASH(HE* he)
492 HeSVKEY(HE* he)
493 HeSVKEY_force(HE* he)
494 HeSVKEY_set(HE* he, SV* sv)
495
496These two lower level macros are defined, but must only be used when
497dealing with keys that are not C<SV*>s:
498
499 HeKEY(HE* he)
500 HeKLEN(HE* he)
501
04343c6d
GS
502Note that both C<hv_store> and C<hv_store_ent> do not increment the
503reference count of the stored C<val>, which is the caller's responsibility.
504If these functions return a NULL value, the caller will usually have to
505decrement the reference count of C<val> to avoid a memory leak.
1e422769 506
a9381218
MHM
507=head2 AVs, HVs and undefined values
508
509Sometimes you have to store undefined values in AVs or HVs. Although
510this may be a rare case, it can be tricky. That's because you're
511used to using C<&PL_sv_undef> if you need an undefined SV.
512
513For example, intuition tells you that this XS code:
514
515 AV *av = newAV();
516 av_store( av, 0, &PL_sv_undef );
517
518is equivalent to this Perl code:
519
520 my @av;
521 $av[0] = undef;
522
523Unfortunately, this isn't true. AVs use C<&PL_sv_undef> as a marker
524for indicating that an array element has not yet been initialized.
525Thus, C<exists $av[0]> would be true for the above Perl code, but
526false for the array generated by the XS code.
527
528Other problems can occur when storing C<&PL_sv_undef> in HVs:
529
530 hv_store( hv, "key", 3, &PL_sv_undef, 0 );
531
532This will indeed make the value C<undef>, but if you try to modify
533the value of C<key>, you'll get the following error:
534
535 Modification of non-creatable hash value attempted
536
537In perl 5.8.0, C<&PL_sv_undef> was also used to mark placeholders
538in restricted hashes. This caused such hash entries not to appear
539when iterating over the hash or when checking for the keys
540with the C<hv_exists> function.
541
8abccac8
AR
542You can run into similar problems when you store C<&PL_sv_yes> or
543C<&PL_sv_no> into AVs or HVs. Trying to modify such elements
a9381218
MHM
544will give you the following error:
545
546 Modification of a read-only value attempted
547
548To make a long story short, you can use the special variables
8abccac8 549C<&PL_sv_undef>, C<&PL_sv_yes> and C<&PL_sv_no> with AVs and
a9381218
MHM
550HVs, but you have to make sure you know what you're doing.
551
552Generally, if you want to store an undefined value in an AV
553or HV, you should not use C<&PL_sv_undef>, but rather create a
554new undefined value using the C<newSV> function, for example:
555
556 av_store( av, 42, newSV(0) );
557 hv_store( hv, "foo", 3, newSV(0), 0 );
558
a0d0e21e
LW
559=head2 References
560
d1b91892
AD
561References are a special type of scalar that point to other data types
562(including references).
a0d0e21e 563
07fa94a1 564To create a reference, use either of the following functions:
a0d0e21e 565
5f05dabc
PP
566 SV* newRV_inc((SV*) thing);
567 SV* newRV_noinc((SV*) thing);
a0d0e21e 568
5f05dabc 569The C<thing> argument can be any of an C<SV*>, C<AV*>, or C<HV*>. The
07fa94a1
JO
570functions are identical except that C<newRV_inc> increments the reference
571count of the C<thing>, while C<newRV_noinc> does not. For historical
572reasons, C<newRV> is a synonym for C<newRV_inc>.
573
574Once you have a reference, you can use the following macro to dereference
575the reference:
a0d0e21e
LW
576
577 SvRV(SV*)
578
579then call the appropriate routines, casting the returned C<SV*> to either an
d1b91892 580C<AV*> or C<HV*>, if required.
a0d0e21e 581
d1b91892 582To determine if an SV is a reference, you can use the following macro:
a0d0e21e
LW
583
584 SvROK(SV*)
585
07fa94a1
JO
586To discover what type of value the reference refers to, use the following
587macro and then check the return value.
d1b91892
AD
588
589 SvTYPE(SvRV(SV*))
590
591The most useful types that will be returned are:
592
593 SVt_IV Scalar
594 SVt_NV Scalar
595 SVt_PV Scalar
5f05dabc 596 SVt_RV Scalar
d1b91892
AD
597 SVt_PVAV Array
598 SVt_PVHV Hash
599 SVt_PVCV Code
5f05dabc
PP
600 SVt_PVGV Glob (possible a file handle)
601 SVt_PVMG Blessed or Magical Scalar
602
3f2c382a 603See the F<sv.h> header file for more details.
d1b91892 604
cb1a09d0
AD
605=head2 Blessed References and Class Objects
606
06f6df17 607References are also used to support object-oriented programming. In perl's
cb1a09d0
AD
608OO lexicon, an object is simply a reference that has been blessed into a
609package (or class). Once blessed, the programmer may now use the reference
610to access the various methods in the class.
611
612A reference can be blessed into a package with the following function:
613
614 SV* sv_bless(SV* sv, HV* stash);
615
06f6df17
RGS
616The C<sv> argument must be a reference value. The C<stash> argument
617specifies which class the reference will belong to. See
2ae324a7 618L<Stashes and Globs> for information on converting class names into stashes.
cb1a09d0
AD
619
620/* Still under construction */
621
ddd2cc91
DM
622The following function upgrades rv to reference if not already one.
623Creates a new SV for rv to point to. If C<classname> is non-null, the SV
624is blessed into the specified class. SV is returned.
cb1a09d0 625
08105a92 626 SV* newSVrv(SV* rv, const char* classname);
cb1a09d0 627
ddd2cc91
DM
628The following three functions copy integer, unsigned integer or double
629into an SV whose reference is C<rv>. SV is blessed if C<classname> is
630non-null.
cb1a09d0 631
08105a92 632 SV* sv_setref_iv(SV* rv, const char* classname, IV iv);
e1c57cef 633 SV* sv_setref_uv(SV* rv, const char* classname, UV uv);
08105a92 634 SV* sv_setref_nv(SV* rv, const char* classname, NV iv);
cb1a09d0 635
ddd2cc91
DM
636The following function copies the pointer value (I<the address, not the
637string!>) into an SV whose reference is rv. SV is blessed if C<classname>
638is non-null.
cb1a09d0 639
ddd2cc91 640 SV* sv_setref_pv(SV* rv, const char* classname, void* pv);
cb1a09d0 641
ddd2cc91
DM
642The following function copies string into an SV whose reference is C<rv>.
643Set length to 0 to let Perl calculate the string length. SV is blessed if
644C<classname> is non-null.
cb1a09d0 645
ddd2cc91 646 SV* sv_setref_pvn(SV* rv, const char* classname, char* pv, STRLEN length);
cb1a09d0 647
ddd2cc91
DM
648The following function tests whether the SV is blessed into the specified
649class. It does not check inheritance relationships.
9abd00ed 650
08105a92 651 int sv_isa(SV* sv, const char* name);
9abd00ed 652
ddd2cc91 653The following function tests whether the SV is a reference to a blessed object.
9abd00ed
GS
654
655 int sv_isobject(SV* sv);
656
ddd2cc91
DM
657The following function tests whether the SV is derived from the specified
658class. SV can be either a reference to a blessed object or a string
659containing a class name. This is the function implementing the
660C<UNIVERSAL::isa> functionality.
9abd00ed 661
08105a92 662 bool sv_derived_from(SV* sv, const char* name);
9abd00ed 663
00aadd71 664To check if you've got an object derived from a specific class you have
9abd00ed
GS
665to write:
666
667 if (sv_isobject(sv) && sv_derived_from(sv, class)) { ... }
cb1a09d0 668
5f05dabc 669=head2 Creating New Variables
cb1a09d0 670
5f05dabc
PP
671To create a new Perl variable with an undef value which can be accessed from
672your Perl script, use the following routines, depending on the variable type.
cb1a09d0 673
64ace3f8 674 SV* get_sv("package::varname", GV_ADD);
cbfd0a87 675 AV* get_av("package::varname", GV_ADD);
6673a63c 676 HV* get_hv("package::varname", GV_ADD);
cb1a09d0 677
058a5f6c 678Notice the use of GV_ADD as the second parameter. The new variable can now
cb1a09d0
AD
679be set, using the routines appropriate to the data type.
680
5f05dabc 681There are additional macros whose values may be bitwise OR'ed with the
058a5f6c 682C<GV_ADD> argument to enable certain extra features. Those bits are:
cb1a09d0 683
9a68f1db
SB
684=over
685
686=item GV_ADDMULTI
687
688Marks the variable as multiply defined, thus preventing the:
689
690 Name <varname> used only once: possible typo
691
692warning.
693
9a68f1db
SB
694=item GV_ADDWARN
695
696Issues the warning:
697
698 Had to create <varname> unexpectedly
699
700if the variable did not exist before the function was called.
701
702=back
cb1a09d0 703
07fa94a1
JO
704If you do not specify a package name, the variable is created in the current
705package.
cb1a09d0 706
5f05dabc 707=head2 Reference Counts and Mortality
a0d0e21e 708
d1be9408 709Perl uses a reference count-driven garbage collection mechanism. SVs,
54310121 710AVs, or HVs (xV for short in the following) start their life with a
55497cff 711reference count of 1. If the reference count of an xV ever drops to 0,
07fa94a1 712then it will be destroyed and its memory made available for reuse.
55497cff
PP
713
714This normally doesn't happen at the Perl level unless a variable is
5f05dabc
PP
715undef'ed or the last variable holding a reference to it is changed or
716overwritten. At the internal level, however, reference counts can be
55497cff
PP
717manipulated with the following macros:
718
719 int SvREFCNT(SV* sv);
5f05dabc 720 SV* SvREFCNT_inc(SV* sv);
55497cff
PP
721 void SvREFCNT_dec(SV* sv);
722
723However, there is one other function which manipulates the reference
07fa94a1
JO
724count of its argument. The C<newRV_inc> function, you will recall,
725creates a reference to the specified argument. As a side effect,
726it increments the argument's reference count. If this is not what
727you want, use C<newRV_noinc> instead.
728
729For example, imagine you want to return a reference from an XSUB function.
730Inside the XSUB routine, you create an SV which initially has a reference
731count of one. Then you call C<newRV_inc>, passing it the just-created SV.
5f05dabc
PP
732This returns the reference as a new SV, but the reference count of the
733SV you passed to C<newRV_inc> has been incremented to two. Now you
07fa94a1
JO
734return the reference from the XSUB routine and forget about the SV.
735But Perl hasn't! Whenever the returned reference is destroyed, the
736reference count of the original SV is decreased to one and nothing happens.
737The SV will hang around without any way to access it until Perl itself
738terminates. This is a memory leak.
5f05dabc
PP
739
740The correct procedure, then, is to use C<newRV_noinc> instead of
faed5253
JO
741C<newRV_inc>. Then, if and when the last reference is destroyed,
742the reference count of the SV will go to zero and it will be destroyed,
07fa94a1 743stopping any memory leak.
55497cff 744
5f05dabc 745There are some convenience functions available that can help with the
54310121 746destruction of xVs. These functions introduce the concept of "mortality".
07fa94a1
JO
747An xV that is mortal has had its reference count marked to be decremented,
748but not actually decremented, until "a short time later". Generally the
749term "short time later" means a single Perl statement, such as a call to
54310121 750an XSUB function. The actual determinant for when mortal xVs have their
07fa94a1
JO
751reference count decremented depends on two macros, SAVETMPS and FREETMPS.
752See L<perlcall> and L<perlxs> for more details on these macros.
55497cff
PP
753
754"Mortalization" then is at its simplest a deferred C<SvREFCNT_dec>.
755However, if you mortalize a variable twice, the reference count will
756later be decremented twice.
757
00aadd71
NIS
758"Mortal" SVs are mainly used for SVs that are placed on perl's stack.
759For example an SV which is created just to pass a number to a called sub
06f6df17
RGS
760is made mortal to have it cleaned up automatically when it's popped off
761the stack. Similarly, results returned by XSUBs (which are pushed on the
762stack) are often made mortal.
a0d0e21e
LW
763
764To create a mortal variable, use the functions:
765
766 SV* sv_newmortal()
767 SV* sv_2mortal(SV*)
768 SV* sv_mortalcopy(SV*)
769
00aadd71 770The first call creates a mortal SV (with no value), the second converts an existing
5f05dabc
PP
771SV to a mortal SV (and thus defers a call to C<SvREFCNT_dec>), and the
772third creates a mortal copy of an existing SV.
00aadd71 773Because C<sv_newmortal> gives the new SV no value,it must normally be given one
9a68f1db 774via C<sv_setpv>, C<sv_setiv>, etc. :
00aadd71
NIS
775
776 SV *tmp = sv_newmortal();
777 sv_setiv(tmp, an_integer);
778
779As that is multiple C statements it is quite common so see this idiom instead:
780
781 SV *tmp = sv_2mortal(newSViv(an_integer));
782
783
784You should be careful about creating mortal variables. Strange things
785can happen if you make the same value mortal within multiple contexts,
786or if you make a variable mortal multiple times. Thinking of "Mortalization"
787as deferred C<SvREFCNT_dec> should help to minimize such problems.
788For example if you are passing an SV which you I<know> has high enough REFCNT
789to survive its use on the stack you need not do any mortalization.
790If you are not sure then doing an C<SvREFCNT_inc> and C<sv_2mortal>, or
791making a C<sv_mortalcopy> is safer.
a0d0e21e 792
ac036724 793The mortal routines are not just for SVs; AVs and HVs can be
faed5253 794made mortal by passing their address (type-casted to C<SV*>) to the
07fa94a1 795C<sv_2mortal> or C<sv_mortalcopy> routines.
a0d0e21e 796
5f05dabc 797=head2 Stashes and Globs
a0d0e21e 798
06f6df17
RGS
799A B<stash> is a hash that contains all variables that are defined
800within a package. Each key of the stash is a symbol
aa689395
PP
801name (shared by all the different types of objects that have the same
802name), and each value in the hash table is a GV (Glob Value). This GV
803in turn contains references to the various objects of that name,
804including (but not limited to) the following:
cb1a09d0 805
a0d0e21e
LW
806 Scalar Value
807 Array Value
808 Hash Value
a3cb178b 809 I/O Handle
a0d0e21e
LW
810 Format
811 Subroutine
812
06f6df17
RGS
813There is a single stash called C<PL_defstash> that holds the items that exist
814in the C<main> package. To get at the items in other packages, append the
815string "::" to the package name. The items in the C<Foo> package are in
816the stash C<Foo::> in PL_defstash. The items in the C<Bar::Baz> package are
817in the stash C<Baz::> in C<Bar::>'s stash.
a0d0e21e 818
d1b91892 819To get the stash pointer for a particular package, use the function:
a0d0e21e 820
da51bb9b
NC
821 HV* gv_stashpv(const char* name, I32 flags)
822 HV* gv_stashsv(SV*, I32 flags)
a0d0e21e
LW
823
824The first function takes a literal string, the second uses the string stored
d1b91892 825in the SV. Remember that a stash is just a hash table, so you get back an
da51bb9b 826C<HV*>. The C<flags> flag will create a new package if it is set to GV_ADD.
a0d0e21e
LW
827
828The name that C<gv_stash*v> wants is the name of the package whose symbol table
829you want. The default package is called C<main>. If you have multiply nested
d1b91892
AD
830packages, pass their names to C<gv_stash*v>, separated by C<::> as in the Perl
831language itself.
a0d0e21e
LW
832
833Alternately, if you have an SV that is a blessed reference, you can find
834out the stash pointer by using:
835
836 HV* SvSTASH(SvRV(SV*));
837
838then use the following to get the package name itself:
839
840 char* HvNAME(HV* stash);
841
5f05dabc
PP
842If you need to bless or re-bless an object you can use the following
843function:
a0d0e21e
LW
844
845 SV* sv_bless(SV*, HV* stash)
846
847where the first argument, an C<SV*>, must be a reference, and the second
848argument is a stash. The returned C<SV*> can now be used in the same way
849as any other SV.
850
d1b91892
AD
851For more information on references and blessings, consult L<perlref>.
852
54310121 853=head2 Double-Typed SVs
0a753a76
PP
854
855Scalar variables normally contain only one type of value, an integer,
856double, pointer, or reference. Perl will automatically convert the
857actual scalar data from the stored type into the requested type.
858
859Some scalar variables contain more than one type of scalar data. For
860example, the variable C<$!> contains either the numeric value of C<errno>
861or its string equivalent from either C<strerror> or C<sys_errlist[]>.
862
863To force multiple data values into an SV, you must do two things: use the
864C<sv_set*v> routines to add the additional scalar type, then set a flag
865so that Perl will believe it contains more than one type of data. The
866four macros to set the flags are:
867
868 SvIOK_on
869 SvNOK_on
870 SvPOK_on
871 SvROK_on
872
873The particular macro you must use depends on which C<sv_set*v> routine
874you called first. This is because every C<sv_set*v> routine turns on
875only the bit for the particular type of data being set, and turns off
876all the rest.
877
878For example, to create a new Perl variable called "dberror" that contains
879both the numeric and descriptive string error values, you could use the
880following code:
881
882 extern int dberror;
883 extern char *dberror_list;
884
64ace3f8 885 SV* sv = get_sv("dberror", GV_ADD);
0a753a76
PP
886 sv_setiv(sv, (IV) dberror);
887 sv_setpv(sv, dberror_list[dberror]);
888 SvIOK_on(sv);
889
890If the order of C<sv_setiv> and C<sv_setpv> had been reversed, then the
891macro C<SvPOK_on> would need to be called instead of C<SvIOK_on>.
892
893=head2 Magic Variables
a0d0e21e 894
d1b91892
AD
895[This section still under construction. Ignore everything here. Post no
896bills. Everything not permitted is forbidden.]
897
d1b91892
AD
898Any SV may be magical, that is, it has special features that a normal
899SV does not have. These features are stored in the SV structure in a
5f05dabc 900linked list of C<struct magic>'s, typedef'ed to C<MAGIC>.
d1b91892
AD
901
902 struct magic {
903 MAGIC* mg_moremagic;
904 MGVTBL* mg_virtual;
905 U16 mg_private;
906 char mg_type;
907 U8 mg_flags;
b205eb13 908 I32 mg_len;
d1b91892
AD
909 SV* mg_obj;
910 char* mg_ptr;
d1b91892
AD
911 };
912
913Note this is current as of patchlevel 0, and could change at any time.
914
915=head2 Assigning Magic
916
917Perl adds magic to an SV using the sv_magic function:
918
08105a92 919 void sv_magic(SV* sv, SV* obj, int how, const char* name, I32 namlen);
d1b91892
AD
920
921The C<sv> argument is a pointer to the SV that is to acquire a new magical
922feature.
923
924If C<sv> is not already magical, Perl uses the C<SvUPGRADE> macro to
645c22ef
DM
925convert C<sv> to type C<SVt_PVMG>. Perl then continues by adding new magic
926to the beginning of the linked list of magical features. Any prior entry
927of the same type of magic is deleted. Note that this can be overridden,
928and multiple instances of the same type of magic can be associated with an
929SV.
d1b91892 930
54310121
PP
931The C<name> and C<namlen> arguments are used to associate a string with
932the magic, typically the name of a variable. C<namlen> is stored in the
2d8d5d5a
SH
933C<mg_len> field and if C<name> is non-null then either a C<savepvn> copy of
934C<name> or C<name> itself is stored in the C<mg_ptr> field, depending on
935whether C<namlen> is greater than zero or equal to zero respectively. As a
936special case, if C<(name && namlen == HEf_SVKEY)> then C<name> is assumed
937to contain an C<SV*> and is stored as-is with its REFCNT incremented.
d1b91892
AD
938
939The sv_magic function uses C<how> to determine which, if any, predefined
940"Magic Virtual Table" should be assigned to the C<mg_virtual> field.
06f6df17 941See the L<Magic Virtual Tables> section below. The C<how> argument is also
14befaf4 942stored in the C<mg_type> field. The value of C<how> should be chosen
06f6df17 943from the set of macros C<PERL_MAGIC_foo> found in F<perl.h>. Note that before
645c22ef 944these macros were added, Perl internals used to directly use character
14befaf4 945literals, so you may occasionally come across old code or documentation
75d0f26d 946referring to 'U' magic rather than C<PERL_MAGIC_uvar> for example.
d1b91892
AD
947
948The C<obj> argument is stored in the C<mg_obj> field of the C<MAGIC>
949structure. If it is not the same as the C<sv> argument, the reference
950count of the C<obj> object is incremented. If it is the same, or if
645c22ef 951the C<how> argument is C<PERL_MAGIC_arylen>, or if it is a NULL pointer,
14befaf4 952then C<obj> is merely stored, without the reference count being incremented.
d1b91892 953
2d8d5d5a
SH
954See also C<sv_magicext> in L<perlapi> for a more flexible way to add magic
955to an SV.
956
cb1a09d0
AD
957There is also a function to add magic to an C<HV>:
958
959 void hv_magic(HV *hv, GV *gv, int how);
960
961This simply calls C<sv_magic> and coerces the C<gv> argument into an C<SV>.
962
963To remove the magic from an SV, call the function sv_unmagic:
964
70a53b35 965 int sv_unmagic(SV *sv, int type);
cb1a09d0
AD
966
967The C<type> argument should be equal to the C<how> value when the C<SV>
968was initially made magical.
969
f6ee7b17
FR
970However, note that C<sv_unmagic> removes all magic of a certain C<type> from the
971C<SV>. If you want to remove only certain magic of a C<type> based on the magic
972virtual table, use C<sv_unmagicext> instead:
973
974 int sv_unmagicext(SV *sv, int type, MGVTBL *vtbl);
975
d1b91892
AD
976=head2 Magic Virtual Tables
977
d1be9408 978The C<mg_virtual> field in the C<MAGIC> structure is a pointer to an
d1b91892
AD
979C<MGVTBL>, which is a structure of function pointers and stands for
980"Magic Virtual Table" to handle the various operations that might be
981applied to that variable.
982
301cb7e8
DM
983The C<MGVTBL> has five (or sometimes eight) pointers to the following
984routine types:
d1b91892
AD
985
986 int (*svt_get)(SV* sv, MAGIC* mg);
987 int (*svt_set)(SV* sv, MAGIC* mg);
988 U32 (*svt_len)(SV* sv, MAGIC* mg);
989 int (*svt_clear)(SV* sv, MAGIC* mg);
990 int (*svt_free)(SV* sv, MAGIC* mg);
991
3468c7ea 992 int (*svt_copy)(SV *sv, MAGIC* mg, SV *nsv, const char *name, I32 namlen);
301cb7e8
DM
993 int (*svt_dup)(MAGIC *mg, CLONE_PARAMS *param);
994 int (*svt_local)(SV *nsv, MAGIC *mg);
995
996
06f6df17 997This MGVTBL structure is set at compile-time in F<perl.h> and there are
b7a0f54c
SM
998currently 32 types. These different structures contain pointers to various
999routines that perform additional actions depending on which function is
1000being called.
d1b91892
AD
1001
1002 Function pointer Action taken
1003 ---------------- ------------
8b0711c3 1004 svt_get Do something before the value of the SV is retrieved.
d1b91892
AD
1005 svt_set Do something after the SV is assigned a value.
1006 svt_len Report on the SV's length.
805b34a4 1007 svt_clear Clear something the SV represents.
d1b91892
AD
1008 svt_free Free any extra storage associated with the SV.
1009
805b34a4
AB
1010 svt_copy copy tied variable magic to a tied element
1011 svt_dup duplicate a magic structure during thread cloning
1012 svt_local copy magic to local value during 'local'
301cb7e8 1013
d1b91892 1014For instance, the MGVTBL structure called C<vtbl_sv> (which corresponds
14befaf4 1015to an C<mg_type> of C<PERL_MAGIC_sv>) contains:
d1b91892
AD
1016
1017 { magic_get, magic_set, magic_len, 0, 0 }
1018
14befaf4
DM
1019Thus, when an SV is determined to be magical and of type C<PERL_MAGIC_sv>,
1020if a get operation is being performed, the routine C<magic_get> is
1021called. All the various routines for the various magical types begin
1022with C<magic_>. NOTE: the magic routines are not considered part of
1023the Perl API, and may not be exported by the Perl library.
d1b91892 1024
301cb7e8
DM
1025The last three slots are a recent addition, and for source code
1026compatibility they are only checked for if one of the three flags
1027MGf_COPY, MGf_DUP or MGf_LOCAL is set in mg_flags. This means that most
1028code can continue declaring a vtable as a 5-element value. These three are
1029currently used exclusively by the threading code, and are highly subject
1030to change.
1031
d1b91892
AD
1032The current kinds of Magic Virtual Tables are:
1033
14befaf4 1034 mg_type
5633a428
NC
1035 (old-style char and macro) MGVTBL Type of magic
1036 -------------------------- ------ -------------
805b34a4 1037 \0 PERL_MAGIC_sv vtbl_sv Special scalar variable
63e77aaf
NC
1038 # PERL_MAGIC_arylen vtbl_arylen Array length ($#ary)
1039 % PERL_MAGIC_rhash (none) extra data for restricted
1040 hashes
1041 . PERL_MAGIC_pos vtbl_pos pos() lvalue
1042 : PERL_MAGIC_symtab (none) extra data for symbol tables
1043 < PERL_MAGIC_backref vtbl_backref for weak ref data
1044 @ PERL_MAGIC_arylen_p (none) to move arylen out of XPVAV
805b34a4 1045 A PERL_MAGIC_overload vtbl_amagic %OVERLOAD hash
14befaf4 1046 a PERL_MAGIC_overload_elem vtbl_amagicelem %OVERLOAD hash element
63e77aaf
NC
1047 B PERL_MAGIC_bm vtbl_regexp Boyer-Moore
1048 (fast string search)
1049 c PERL_MAGIC_overload_table vtbl_ovrld Holds overload table
1050 (AMT) on stash
1051 D PERL_MAGIC_regdata vtbl_regdata Regex match position data
805b34a4
AB
1052 (@+ and @- vars)
1053 d PERL_MAGIC_regdatum vtbl_regdatum Regex match position data
1054 element
1055 E PERL_MAGIC_env vtbl_env %ENV hash
1056 e PERL_MAGIC_envelem vtbl_envelem %ENV hash element
63e77aaf 1057 f PERL_MAGIC_fm vtbl_regdata Formline ('compiled' format)
0177730e
NC
1058 G PERL_MAGIC_study vtbl_regdata study()ed string
1059 g PERL_MAGIC_regex_global vtbl_mglob m//g target
f747ebd6 1060 H PERL_MAGIC_hints vtbl_hints %^H hash
805b34a4
AB
1061 h PERL_MAGIC_hintselem vtbl_hintselem %^H hash element
1062 I PERL_MAGIC_isa vtbl_isa @ISA array
1063 i PERL_MAGIC_isaelem vtbl_isaelem @ISA array element
1064 k PERL_MAGIC_nkeys vtbl_nkeys scalar(keys()) lvalue
a418c3da
NC
1065 L PERL_MAGIC_dbfile (none) Debugger %_<filename
1066 l PERL_MAGIC_dbline vtbl_dbline Debugger %_<filename element
63e77aaf
NC
1067 N PERL_MAGIC_shared (none) Shared between threads
1068 n PERL_MAGIC_shared_scalar (none) Shared between threads
1069 o PERL_MAGIC_collxfrm vtbl_collxfrm Locale transformation
805b34a4
AB
1070 P PERL_MAGIC_tied vtbl_pack Tied array or hash
1071 p PERL_MAGIC_tiedelem vtbl_packelem Tied array or hash element
1072 q PERL_MAGIC_tiedscalar vtbl_packelem Tied scalar or handle
63e77aaf 1073 r PERL_MAGIC_qr vtbl_regexp precompiled qr// regex
4de01b54 1074 S PERL_MAGIC_sig (none) %SIG hash
805b34a4
AB
1075 s PERL_MAGIC_sigelem vtbl_sigelem %SIG hash element
1076 t PERL_MAGIC_taint vtbl_taint Taintedness
1077 U PERL_MAGIC_uvar vtbl_uvar Available for use by extensions
63e77aaf
NC
1078 u PERL_MAGIC_uvar_elem (none) Reserved for use by extensions
1079 V PERL_MAGIC_vstring (none) SV was vstring literal
805b34a4 1080 v PERL_MAGIC_vec vtbl_vec vec() lvalue
63e77aaf 1081 w PERL_MAGIC_utf8 vtbl_utf8 Cached UTF-8 information
805b34a4
AB
1082 x PERL_MAGIC_substr vtbl_substr substr() lvalue
1083 y PERL_MAGIC_defelem vtbl_defelem Shadow "foreach" iterator
1084 variable / smart parameter
1085 vivification
63e77aaf
NC
1086 ] PERL_MAGIC_checkcall (none) inlining/mutation of call to
1087 this CV
805b34a4 1088 ~ PERL_MAGIC_ext (none) Available for use by extensions
0cbee0a4 1089
d1b91892 1090
68dc0745 1091When an uppercase and lowercase letter both exist in the table, then the
92f0c265
JP
1092uppercase letter is typically used to represent some kind of composite type
1093(a list or a hash), and the lowercase letter is used to represent an element
1094of that composite type. Some internals code makes use of this case
1095relationship. However, 'v' and 'V' (vec and v-string) are in no way related.
14befaf4
DM
1096
1097The C<PERL_MAGIC_ext> and C<PERL_MAGIC_uvar> magic types are defined
1098specifically for use by extensions and will not be used by perl itself.
1099Extensions can use C<PERL_MAGIC_ext> magic to 'attach' private information
1100to variables (typically objects). This is especially useful because
1101there is no way for normal perl code to corrupt this private information
1102(unlike using extra elements of a hash object).
1103
1104Similarly, C<PERL_MAGIC_uvar> magic can be used much like tie() to call a
1105C function any time a scalar's value is used or changed. The C<MAGIC>'s
bdbeb323
SM
1106C<mg_ptr> field points to a C<ufuncs> structure:
1107
1108 struct ufuncs {
a9402793
AB
1109 I32 (*uf_val)(pTHX_ IV, SV*);
1110 I32 (*uf_set)(pTHX_ IV, SV*);
bdbeb323
SM
1111 IV uf_index;
1112 };
1113
1114When the SV is read from or written to, the C<uf_val> or C<uf_set>
14befaf4
DM
1115function will be called with C<uf_index> as the first arg and a pointer to
1116the SV as the second. A simple example of how to add C<PERL_MAGIC_uvar>
1526ead6
AB
1117magic is shown below. Note that the ufuncs structure is copied by
1118sv_magic, so you can safely allocate it on the stack.
1119
1120 void
1121 Umagic(sv)
1122 SV *sv;
1123 PREINIT:
1124 struct ufuncs uf;
1125 CODE:
1126 uf.uf_val = &my_get_fn;
1127 uf.uf_set = &my_set_fn;
1128 uf.uf_index = 0;
14befaf4 1129 sv_magic(sv, 0, PERL_MAGIC_uvar, (char*)&uf, sizeof(uf));
5f05dabc 1130
1e73acc8
AS
1131Attaching C<PERL_MAGIC_uvar> to arrays is permissible but has no effect.
1132
1133For hashes there is a specialized hook that gives control over hash
1134keys (but not values). This hook calls C<PERL_MAGIC_uvar> 'get' magic
1135if the "set" function in the C<ufuncs> structure is NULL. The hook
1136is activated whenever the hash is accessed with a key specified as
1137an C<SV> through the functions C<hv_store_ent>, C<hv_fetch_ent>,
1138C<hv_delete_ent>, and C<hv_exists_ent>. Accessing the key as a string
1139through the functions without the C<..._ent> suffix circumvents the
4509d391 1140hook. See L<Hash::Util::FieldHash/GUTS> for a detailed description.
1e73acc8 1141
14befaf4
DM
1142Note that because multiple extensions may be using C<PERL_MAGIC_ext>
1143or C<PERL_MAGIC_uvar> magic, it is important for extensions to take
1144extra care to avoid conflict. Typically only using the magic on
1145objects blessed into the same class as the extension is sufficient.
2f07f21a
FR
1146For C<PERL_MAGIC_ext> magic, it is usually a good idea to define an
1147C<MGVTBL>, even if all its fields will be C<0>, so that individual
1148C<MAGIC> pointers can be identified as a particular kind of magic
f6ee7b17
FR
1149using their magic virtual table. C<mg_findext> provides an easy way
1150to do that:
2f07f21a
FR
1151
1152 STATIC MGVTBL my_vtbl = { 0, 0, 0, 0, 0, 0, 0, 0 };
1153
1154 MAGIC *mg;
f6ee7b17
FR
1155 if ((mg = mg_findext(sv, PERL_MAGIC_ext, &my_vtbl))) {
1156 /* this is really ours, not another module's PERL_MAGIC_ext */
1157 my_priv_data_t *priv = (my_priv_data_t *)mg->mg_ptr;
1158 ...
2f07f21a 1159 }
5f05dabc 1160
ef50df4b
GS
1161Also note that the C<sv_set*()> and C<sv_cat*()> functions described
1162earlier do B<not> invoke 'set' magic on their targets. This must
1163be done by the user either by calling the C<SvSETMAGIC()> macro after
1164calling these functions, or by using one of the C<sv_set*_mg()> or
1165C<sv_cat*_mg()> functions. Similarly, generic C code must call the
1166C<SvGETMAGIC()> macro to invoke any 'get' magic if they use an SV
1167obtained from external sources in functions that don't handle magic.
4a4eefd0 1168See L<perlapi> for a description of these functions.
189b2af5
GS
1169For example, calls to the C<sv_cat*()> functions typically need to be
1170followed by C<SvSETMAGIC()>, but they don't need a prior C<SvGETMAGIC()>
1171since their implementation handles 'get' magic.
1172
d1b91892
AD
1173=head2 Finding Magic
1174
f6ee7b17
FR
1175 MAGIC *mg_find(SV *sv, int type); /* Finds the magic pointer of that type */
1176
1177This routine returns a pointer to a C<MAGIC> structure stored in the SV.
1178If the SV does not have that magical feature, C<NULL> is returned. If the
1179SV has multiple instances of that magical feature, the first one will be
1180returned. C<mg_findext> can be used to find a C<MAGIC> structure of an SV
1181based on both it's magic type and it's magic virtual table:
1182
1183 MAGIC *mg_findext(SV *sv, int type, MGVTBL *vtbl);
d1b91892 1184
f6ee7b17
FR
1185Also, if the SV passed to C<mg_find> or C<mg_findext> is not of type
1186SVt_PVMG, Perl may core dump.
d1b91892 1187
08105a92 1188 int mg_copy(SV* sv, SV* nsv, const char* key, STRLEN klen);
d1b91892
AD
1189
1190This routine checks to see what types of magic C<sv> has. If the mg_type
68dc0745
PP
1191field is an uppercase letter, then the mg_obj is copied to C<nsv>, but
1192the mg_type field is changed to be the lowercase letter.
a0d0e21e 1193
04343c6d
GS
1194=head2 Understanding the Magic of Tied Hashes and Arrays
1195
14befaf4
DM
1196Tied hashes and arrays are magical beasts of the C<PERL_MAGIC_tied>
1197magic type.
9edb2b46
GS
1198
1199WARNING: As of the 5.004 release, proper usage of the array and hash
1200access functions requires understanding a few caveats. Some
1201of these caveats are actually considered bugs in the API, to be fixed
1202in later releases, and are bracketed with [MAYCHANGE] below. If
1203you find yourself actually applying such information in this section, be
1204aware that the behavior may change in the future, umm, without warning.
04343c6d 1205
1526ead6 1206The perl tie function associates a variable with an object that implements
9a68f1db 1207the various GET, SET, etc methods. To perform the equivalent of the perl
1526ead6
AB
1208tie function from an XSUB, you must mimic this behaviour. The code below
1209carries out the necessary steps - firstly it creates a new hash, and then
1210creates a second hash which it blesses into the class which will implement
1211the tie methods. Lastly it ties the two hashes together, and returns a
1212reference to the new tied hash. Note that the code below does NOT call the
1213TIEHASH method in the MyTie class -
1214see L<Calling Perl Routines from within C Programs> for details on how
1215to do this.
1216
1217 SV*
1218 mytie()
1219 PREINIT:
1220 HV *hash;
1221 HV *stash;
1222 SV *tie;
1223 CODE:
1224 hash = newHV();
1225 tie = newRV_noinc((SV*)newHV());
da51bb9b 1226 stash = gv_stashpv("MyTie", GV_ADD);
1526ead6 1227 sv_bless(tie, stash);
899e16d0 1228 hv_magic(hash, (GV*)tie, PERL_MAGIC_tied);
1526ead6
AB
1229 RETVAL = newRV_noinc(hash);
1230 OUTPUT:
1231 RETVAL
1232
04343c6d
GS
1233The C<av_store> function, when given a tied array argument, merely
1234copies the magic of the array onto the value to be "stored", using
1235C<mg_copy>. It may also return NULL, indicating that the value did not
9edb2b46
GS
1236actually need to be stored in the array. [MAYCHANGE] After a call to
1237C<av_store> on a tied array, the caller will usually need to call
1238C<mg_set(val)> to actually invoke the perl level "STORE" method on the
1239TIEARRAY object. If C<av_store> did return NULL, a call to
1240C<SvREFCNT_dec(val)> will also be usually necessary to avoid a memory
1241leak. [/MAYCHANGE]
04343c6d
GS
1242
1243The previous paragraph is applicable verbatim to tied hash access using the
1244C<hv_store> and C<hv_store_ent> functions as well.
1245
1246C<av_fetch> and the corresponding hash functions C<hv_fetch> and
1247C<hv_fetch_ent> actually return an undefined mortal value whose magic
1248has been initialized using C<mg_copy>. Note the value so returned does not
9edb2b46
GS
1249need to be deallocated, as it is already mortal. [MAYCHANGE] But you will
1250need to call C<mg_get()> on the returned value in order to actually invoke
1251the perl level "FETCH" method on the underlying TIE object. Similarly,
04343c6d
GS
1252you may also call C<mg_set()> on the return value after possibly assigning
1253a suitable value to it using C<sv_setsv>, which will invoke the "STORE"
9edb2b46 1254method on the TIE object. [/MAYCHANGE]
04343c6d 1255
9edb2b46 1256[MAYCHANGE]
04343c6d
GS
1257In other words, the array or hash fetch/store functions don't really
1258fetch and store actual values in the case of tied arrays and hashes. They
1259merely call C<mg_copy> to attach magic to the values that were meant to be
1260"stored" or "fetched". Later calls to C<mg_get> and C<mg_set> actually
1261do the job of invoking the TIE methods on the underlying objects. Thus
9edb2b46 1262the magic mechanism currently implements a kind of lazy access to arrays
04343c6d
GS
1263and hashes.
1264
1265Currently (as of perl version 5.004), use of the hash and array access
1266functions requires the user to be aware of whether they are operating on
9edb2b46
GS
1267"normal" hashes and arrays, or on their tied variants. The API may be
1268changed to provide more transparent access to both tied and normal data
1269types in future versions.
1270[/MAYCHANGE]
04343c6d
GS
1271
1272You would do well to understand that the TIEARRAY and TIEHASH interfaces
1273are mere sugar to invoke some perl method calls while using the uniform hash
1274and array syntax. The use of this sugar imposes some overhead (typically
1275about two to four extra opcodes per FETCH/STORE operation, in addition to
1276the creation of all the mortal variables required to invoke the methods).
1277This overhead will be comparatively small if the TIE methods are themselves
1278substantial, but if they are only a few statements long, the overhead
1279will not be insignificant.
1280
d1c897a1
IZ
1281=head2 Localizing changes
1282
1283Perl has a very handy construction
1284
1285 {
1286 local $var = 2;
1287 ...
1288 }
1289
1290This construction is I<approximately> equivalent to
1291
1292 {
1293 my $oldvar = $var;
1294 $var = 2;
1295 ...
1296 $var = $oldvar;
1297 }
1298
1299The biggest difference is that the first construction would
1300reinstate the initial value of $var, irrespective of how control exits
9a68f1db 1301the block: C<goto>, C<return>, C<die>/C<eval>, etc. It is a little bit
d1c897a1
IZ
1302more efficient as well.
1303
1304There is a way to achieve a similar task from C via Perl API: create a
1305I<pseudo-block>, and arrange for some changes to be automatically
1306undone at the end of it, either explicit, or via a non-local exit (via
1307die()). A I<block>-like construct is created by a pair of
b687b08b
TC
1308C<ENTER>/C<LEAVE> macros (see L<perlcall/"Returning a Scalar">).
1309Such a construct may be created specially for some important localized
1310task, or an existing one (like boundaries of enclosing Perl
1311subroutine/block, or an existing pair for freeing TMPs) may be
1312used. (In the second case the overhead of additional localization must
1313be almost negligible.) Note that any XSUB is automatically enclosed in
1314an C<ENTER>/C<LEAVE> pair.
d1c897a1
IZ
1315
1316Inside such a I<pseudo-block> the following service is available:
1317
13a2d996 1318=over 4
d1c897a1
IZ
1319
1320=item C<SAVEINT(int i)>
1321
1322=item C<SAVEIV(IV i)>
1323
1324=item C<SAVEI32(I32 i)>
1325
1326=item C<SAVELONG(long i)>
1327
1328These macros arrange things to restore the value of integer variable
1329C<i> at the end of enclosing I<pseudo-block>.
1330
1331=item C<SAVESPTR(s)>
1332
1333=item C<SAVEPPTR(p)>
1334
1335These macros arrange things to restore the value of pointers C<s> and
1336C<p>. C<s> must be a pointer of a type which survives conversion to
1337C<SV*> and back, C<p> should be able to survive conversion to C<char*>
1338and back.
1339
1340=item C<SAVEFREESV(SV *sv)>
1341
1342The refcount of C<sv> would be decremented at the end of
26d9b02f
JH
1343I<pseudo-block>. This is similar to C<sv_2mortal> in that it is also a
1344mechanism for doing a delayed C<SvREFCNT_dec>. However, while C<sv_2mortal>
1345extends the lifetime of C<sv> until the beginning of the next statement,
1346C<SAVEFREESV> extends it until the end of the enclosing scope. These
1347lifetimes can be wildly different.
1348
1349Also compare C<SAVEMORTALIZESV>.
1350
1351=item C<SAVEMORTALIZESV(SV *sv)>
1352
1353Just like C<SAVEFREESV>, but mortalizes C<sv> at the end of the current
1354scope instead of decrementing its reference count. This usually has the
1355effect of keeping C<sv> alive until the statement that called the currently
1356live scope has finished executing.
d1c897a1
IZ
1357
1358=item C<SAVEFREEOP(OP *op)>
1359
1360The C<OP *> is op_free()ed at the end of I<pseudo-block>.
1361
1362=item C<SAVEFREEPV(p)>
1363
1364The chunk of memory which is pointed to by C<p> is Safefree()ed at the
1365end of I<pseudo-block>.
1366
1367=item C<SAVECLEARSV(SV *sv)>
1368
1369Clears a slot in the current scratchpad which corresponds to C<sv> at
1370the end of I<pseudo-block>.
1371
1372=item C<SAVEDELETE(HV *hv, char *key, I32 length)>
1373
1374The key C<key> of C<hv> is deleted at the end of I<pseudo-block>. The
1375string pointed to by C<key> is Safefree()ed. If one has a I<key> in
1376short-lived storage, the corresponding string may be reallocated like
1377this:
1378
9cde0e7f 1379 SAVEDELETE(PL_defstash, savepv(tmpbuf), strlen(tmpbuf));
d1c897a1 1380
c76ac1ee 1381=item C<SAVEDESTRUCTOR(DESTRUCTORFUNC_NOCONTEXT_t f, void *p)>
d1c897a1
IZ
1382
1383At the end of I<pseudo-block> the function C<f> is called with the
c76ac1ee
GS
1384only argument C<p>.
1385
1386=item C<SAVEDESTRUCTOR_X(DESTRUCTORFUNC_t f, void *p)>
1387
1388At the end of I<pseudo-block> the function C<f> is called with the
1389implicit context argument (if any), and C<p>.
d1c897a1
IZ
1390
1391=item C<SAVESTACK_POS()>
1392
1393The current offset on the Perl internal stack (cf. C<SP>) is restored
1394at the end of I<pseudo-block>.
1395
1396=back
1397
1398The following API list contains functions, thus one needs to
1399provide pointers to the modifiable data explicitly (either C pointers,
00aadd71 1400or Perlish C<GV *>s). Where the above macros take C<int>, a similar
d1c897a1
IZ
1401function takes C<int *>.
1402
13a2d996 1403=over 4
d1c897a1
IZ
1404
1405=item C<SV* save_scalar(GV *gv)>
1406
1407Equivalent to Perl code C<local $gv>.
1408
1409=item C<AV* save_ary(GV *gv)>
1410
1411=item C<HV* save_hash(GV *gv)>
1412
1413Similar to C<save_scalar>, but localize C<@gv> and C<%gv>.
1414
1415=item C<void save_item(SV *item)>
1416
1417Duplicates the current value of C<SV>, on the exit from the current
1418C<ENTER>/C<LEAVE> I<pseudo-block> will restore the value of C<SV>
038fcae3
SB
1419using the stored value. It doesn't handle magic. Use C<save_scalar> if
1420magic is affected.
d1c897a1
IZ
1421
1422=item C<void save_list(SV **sarg, I32 maxsarg)>
1423
1424A variant of C<save_item> which takes multiple arguments via an array
1425C<sarg> of C<SV*> of length C<maxsarg>.
1426
1427=item C<SV* save_svref(SV **sptr)>
1428
d1be9408 1429Similar to C<save_scalar>, but will reinstate an C<SV *>.
d1c897a1
IZ
1430
1431=item C<void save_aptr(AV **aptr)>
1432
1433=item C<void save_hptr(HV **hptr)>
1434
1435Similar to C<save_svref>, but localize C<AV *> and C<HV *>.
1436
1437=back
1438
1439The C<Alias> module implements localization of the basic types within the
1440I<caller's scope>. People who are interested in how to localize things in
1441the containing scope should take a look there too.
1442
0a753a76 1443=head1 Subroutines
a0d0e21e 1444
68dc0745 1445=head2 XSUBs and the Argument Stack
5f05dabc
PP
1446
1447The XSUB mechanism is a simple way for Perl programs to access C subroutines.
1448An XSUB routine will have a stack that contains the arguments from the Perl
1449program, and a way to map from the Perl data structures to a C equivalent.
1450
1451The stack arguments are accessible through the C<ST(n)> macro, which returns
1452the C<n>'th stack argument. Argument 0 is the first argument passed in the
1453Perl subroutine call. These arguments are C<SV*>, and can be used anywhere
1454an C<SV*> is used.
1455
1456Most of the time, output from the C routine can be handled through use of
1457the RETVAL and OUTPUT directives. However, there are some cases where the
1458argument stack is not already long enough to handle all the return values.
1459An example is the POSIX tzname() call, which takes no arguments, but returns
1460two, the local time zone's standard and summer time abbreviations.
1461
1462To handle this situation, the PPCODE directive is used and the stack is
1463extended using the macro:
1464
924508f0 1465 EXTEND(SP, num);
5f05dabc 1466
924508f0
GS
1467where C<SP> is the macro that represents the local copy of the stack pointer,
1468and C<num> is the number of elements the stack should be extended by.
5f05dabc 1469
00aadd71 1470Now that there is room on the stack, values can be pushed on it using C<PUSHs>
06f6df17 1471macro. The pushed values will often need to be "mortal" (See
d82b684c 1472L</Reference Counts and Mortality>):
5f05dabc 1473
00aadd71 1474 PUSHs(sv_2mortal(newSViv(an_integer)))
d82b684c
SH
1475 PUSHs(sv_2mortal(newSVuv(an_unsigned_integer)))
1476 PUSHs(sv_2mortal(newSVnv(a_double)))
00aadd71 1477 PUSHs(sv_2mortal(newSVpv("Some String",0)))
a3179684
NC
1478 /* Although the last example is better written as the more efficient: */
1479 PUSHs(newSVpvs_flags("Some String", SVs_TEMP))
5f05dabc
PP
1480
1481And now the Perl program calling C<tzname>, the two values will be assigned
1482as in:
1483
1484 ($standard_abbrev, $summer_abbrev) = POSIX::tzname;
1485
1486An alternate (and possibly simpler) method to pushing values on the stack is
00aadd71 1487to use the macro:
5f05dabc 1488
5f05dabc
PP
1489 XPUSHs(SV*)
1490
00aadd71 1491This macro automatically adjust the stack for you, if needed. Thus, you
5f05dabc 1492do not need to call C<EXTEND> to extend the stack.
00aadd71
NIS
1493
1494Despite their suggestions in earlier versions of this document the macros
d82b684c
SH
1495C<(X)PUSH[iunp]> are I<not> suited to XSUBs which return multiple results.
1496For that, either stick to the C<(X)PUSHs> macros shown above, or use the new
1497C<m(X)PUSH[iunp]> macros instead; see L</Putting a C value on Perl stack>.
5f05dabc
PP
1498
1499For more information, consult L<perlxs> and L<perlxstut>.
1500
1501=head2 Calling Perl Routines from within C Programs
a0d0e21e
LW
1502
1503There are four routines that can be used to call a Perl subroutine from
1504within a C program. These four are:
1505
954c1994
GS
1506 I32 call_sv(SV*, I32);
1507 I32 call_pv(const char*, I32);
1508 I32 call_method(const char*, I32);
1509 I32 call_argv(const char*, I32, register char**);
a0d0e21e 1510
954c1994 1511The routine most often used is C<call_sv>. The C<SV*> argument
d1b91892
AD
1512contains either the name of the Perl subroutine to be called, or a
1513reference to the subroutine. The second argument consists of flags
1514that control the context in which the subroutine is called, whether
1515or not the subroutine is being passed arguments, how errors should be
1516trapped, and how to treat return values.
a0d0e21e
LW
1517
1518All four routines return the number of arguments that the subroutine returned
1519on the Perl stack.
1520
9a68f1db 1521These routines used to be called C<perl_call_sv>, etc., before Perl v5.6.0,
954c1994
GS
1522but those names are now deprecated; macros of the same name are provided for
1523compatibility.
1524
1525When using any of these routines (except C<call_argv>), the programmer
d1b91892
AD
1526must manipulate the Perl stack. These include the following macros and
1527functions:
a0d0e21e
LW
1528
1529 dSP
924508f0 1530 SP
a0d0e21e
LW
1531 PUSHMARK()
1532 PUTBACK
1533 SPAGAIN
1534 ENTER
1535 SAVETMPS
1536 FREETMPS
1537 LEAVE
1538 XPUSH*()
cb1a09d0 1539 POP*()
a0d0e21e 1540
5f05dabc
PP
1541For a detailed description of calling conventions from C to Perl,
1542consult L<perlcall>.
a0d0e21e 1543
5f05dabc 1544=head2 Memory Allocation
a0d0e21e 1545
06f6df17
RGS
1546=head3 Allocation
1547
86058a2d
GS
1548All memory meant to be used with the Perl API functions should be manipulated
1549using the macros described in this section. The macros provide the necessary
1550transparency between differences in the actual malloc implementation that is
1551used within perl.
1552
1553It is suggested that you enable the version of malloc that is distributed
5f05dabc 1554with Perl. It keeps pools of various sizes of unallocated memory in
07fa94a1
JO
1555order to satisfy allocation requests more quickly. However, on some
1556platforms, it may cause spurious malloc or free errors.
d1b91892 1557
06f6df17
RGS
1558The following three macros are used to initially allocate memory :
1559
9f653bb5
SH
1560 Newx(pointer, number, type);
1561 Newxc(pointer, number, type, cast);
1562 Newxz(pointer, number, type);
d1b91892 1563
9f653bb5 1564The first argument C<pointer> should be the name of a variable that will
5f05dabc 1565point to the newly allocated memory.
d1b91892 1566
9f653bb5 1567The second and third arguments C<number> and C<type> specify how many of
d1b91892 1568the specified type of data structure should be allocated. The argument
9f653bb5 1569C<type> is passed to C<sizeof>. The final argument to C<Newxc>, C<cast>,
d1b91892
AD
1570should be used if the C<pointer> argument is different from the C<type>
1571argument.
1572
9f653bb5 1573Unlike the C<Newx> and C<Newxc> macros, the C<Newxz> macro calls C<memzero>
d1b91892
AD
1574to zero out all the newly allocated memory.
1575
06f6df17
RGS
1576=head3 Reallocation
1577
d1b91892
AD
1578 Renew(pointer, number, type);
1579 Renewc(pointer, number, type, cast);
1580 Safefree(pointer)
1581
1582These three macros are used to change a memory buffer size or to free a
1583piece of memory no longer needed. The arguments to C<Renew> and C<Renewc>
1584match those of C<New> and C<Newc> with the exception of not needing the
1585"magic cookie" argument.
1586
06f6df17
RGS
1587=head3 Moving
1588
d1b91892
AD
1589 Move(source, dest, number, type);
1590 Copy(source, dest, number, type);
1591 Zero(dest, number, type);
1592
1593These three macros are used to move, copy, or zero out previously allocated
1594memory. The C<source> and C<dest> arguments point to the source and
1595destination starting points. Perl will move, copy, or zero out C<number>
1596instances of the size of the C<type> data structure (using the C<sizeof>
1597function).
a0d0e21e 1598
5f05dabc 1599=head2 PerlIO
ce3d39e2 1600
5f05dabc
PP
1601The most recent development releases of Perl has been experimenting with
1602removing Perl's dependency on the "normal" standard I/O suite and allowing
1603other stdio implementations to be used. This involves creating a new
1604abstraction layer that then calls whichever implementation of stdio Perl
68dc0745 1605was compiled with. All XSUBs should now use the functions in the PerlIO
5f05dabc
PP
1606abstraction layer and not make any assumptions about what kind of stdio
1607is being used.
1608
1609For a complete description of the PerlIO abstraction, consult L<perlapio>.
1610
8ebc5c01 1611=head2 Putting a C value on Perl stack
ce3d39e2
IZ
1612
1613A lot of opcodes (this is an elementary operation in the internal perl
1614stack machine) put an SV* on the stack. However, as an optimization
1615the corresponding SV is (usually) not recreated each time. The opcodes
1616reuse specially assigned SVs (I<target>s) which are (as a corollary)
1617not constantly freed/created.
1618
0a753a76 1619Each of the targets is created only once (but see
ce3d39e2
IZ
1620L<Scratchpads and recursion> below), and when an opcode needs to put
1621an integer, a double, or a string on stack, it just sets the
1622corresponding parts of its I<target> and puts the I<target> on stack.
1623
1624The macro to put this target on stack is C<PUSHTARG>, and it is
1625directly used in some opcodes, as well as indirectly in zillions of
d82b684c 1626others, which use it via C<(X)PUSH[iunp]>.
ce3d39e2 1627
1bd1c0d5
SC
1628Because the target is reused, you must be careful when pushing multiple
1629values on the stack. The following code will not do what you think:
1630
1631 XPUSHi(10);
1632 XPUSHi(20);
1633
1634This translates as "set C<TARG> to 10, push a pointer to C<TARG> onto
1635the stack; set C<TARG> to 20, push a pointer to C<TARG> onto the stack".
1636At the end of the operation, the stack does not contain the values 10
1637and 20, but actually contains two pointers to C<TARG>, which we have set
d82b684c 1638to 20.
1bd1c0d5 1639
d82b684c
SH
1640If you need to push multiple different values then you should either use
1641the C<(X)PUSHs> macros, or else use the new C<m(X)PUSH[iunp]> macros,
1642none of which make use of C<TARG>. The C<(X)PUSHs> macros simply push an
1643SV* on the stack, which, as noted under L</XSUBs and the Argument Stack>,
1644will often need to be "mortal". The new C<m(X)PUSH[iunp]> macros make
1645this a little easier to achieve by creating a new mortal for you (via
1646C<(X)PUSHmortal>), pushing that onto the stack (extending it if necessary
1647in the case of the C<mXPUSH[iunp]> macros), and then setting its value.
1648Thus, instead of writing this to "fix" the example above:
1649
1650 XPUSHs(sv_2mortal(newSViv(10)))
1651 XPUSHs(sv_2mortal(newSViv(20)))
1652
1653you can simply write:
1654
1655 mXPUSHi(10)
1656 mXPUSHi(20)
1657
1658On a related note, if you do use C<(X)PUSH[iunp]>, then you're going to
1bd1c0d5 1659need a C<dTARG> in your variable declarations so that the C<*PUSH*>
d82b684c
SH
1660macros can make use of the local variable C<TARG>. See also C<dTARGET>
1661and C<dXSTARG>.
1bd1c0d5 1662
8ebc5c01 1663=head2 Scratchpads
ce3d39e2 1664
54310121 1665The question remains on when the SVs which are I<target>s for opcodes
ac036724 1666are created. The answer is that they are created when the current
1667unit--a subroutine or a file (for opcodes for statements outside of
1668subroutines)--is compiled. During this time a special anonymous Perl
1669array is created, which is called a scratchpad for the current unit.
ce3d39e2 1670
54310121 1671A scratchpad keeps SVs which are lexicals for the current unit and are
ce3d39e2
IZ
1672targets for opcodes. One can deduce that an SV lives on a scratchpad
1673by looking on its flags: lexicals have C<SVs_PADMY> set, and
1674I<target>s have C<SVs_PADTMP> set.
1675
54310121
PP
1676The correspondence between OPs and I<target>s is not 1-to-1. Different
1677OPs in the compile tree of the unit can use the same target, if this
ce3d39e2
IZ
1678would not conflict with the expected life of the temporary.
1679
2ae324a7 1680=head2 Scratchpads and recursion
ce3d39e2
IZ
1681
1682In fact it is not 100% true that a compiled unit contains a pointer to
1683the scratchpad AV. In fact it contains a pointer to an AV of
1684(initially) one element, and this element is the scratchpad AV. Why do
1685we need an extra level of indirection?
1686
9a68f1db 1687The answer is B<recursion>, and maybe B<threads>. Both
ce3d39e2
IZ
1688these can create several execution pointers going into the same
1689subroutine. For the subroutine-child not write over the temporaries
1690for the subroutine-parent (lifespan of which covers the call to the
1691child), the parent and the child should have different
1692scratchpads. (I<And> the lexicals should be separate anyway!)
1693
5f05dabc
PP
1694So each subroutine is born with an array of scratchpads (of length 1).
1695On each entry to the subroutine it is checked that the current
ce3d39e2
IZ
1696depth of the recursion is not more than the length of this array, and
1697if it is, new scratchpad is created and pushed into the array.
1698
1699The I<target>s on this scratchpad are C<undef>s, but they are already
1700marked with correct flags.
1701
0a753a76
PP
1702=head1 Compiled code
1703
1704=head2 Code tree
1705
1706Here we describe the internal form your code is converted to by
1707Perl. Start with a simple example:
1708
1709 $a = $b + $c;
1710
1711This is converted to a tree similar to this one:
1712
1713 assign-to
1714 / \
1715 + $a
1716 / \
1717 $b $c
1718
7b8d334a 1719(but slightly more complicated). This tree reflects the way Perl
0a753a76
PP
1720parsed your code, but has nothing to do with the execution order.
1721There is an additional "thread" going through the nodes of the tree
1722which shows the order of execution of the nodes. In our simplified
1723example above it looks like:
1724
1725 $b ---> $c ---> + ---> $a ---> assign-to
1726
1727But with the actual compile tree for C<$a = $b + $c> it is different:
1728some nodes I<optimized away>. As a corollary, though the actual tree
1729contains more nodes than our simplified example, the execution order
1730is the same as in our example.
1731
1732=head2 Examining the tree
1733
06f6df17
RGS
1734If you have your perl compiled for debugging (usually done with
1735C<-DDEBUGGING> on the C<Configure> command line), you may examine the
0a753a76
PP
1736compiled tree by specifying C<-Dx> on the Perl command line. The
1737output takes several lines per node, and for C<$b+$c> it looks like
1738this:
1739
1740 5 TYPE = add ===> 6
1741 TARG = 1
1742 FLAGS = (SCALAR,KIDS)
1743 {
1744 TYPE = null ===> (4)
1745 (was rv2sv)
1746 FLAGS = (SCALAR,KIDS)
1747 {
1748 3 TYPE = gvsv ===> 4
1749 FLAGS = (SCALAR)
1750 GV = main::b
1751 }
1752 }
1753 {
1754 TYPE = null ===> (5)
1755 (was rv2sv)
1756 FLAGS = (SCALAR,KIDS)
1757 {
1758 4 TYPE = gvsv ===> 5
1759 FLAGS = (SCALAR)
1760 GV = main::c
1761 }
1762 }
1763
1764This tree has 5 nodes (one per C<TYPE> specifier), only 3 of them are
1765not optimized away (one per number in the left column). The immediate
1766children of the given node correspond to C<{}> pairs on the same level
1767of indentation, thus this listing corresponds to the tree:
1768
1769 add
1770 / \
1771 null null
1772 | |
1773 gvsv gvsv
1774
1775The execution order is indicated by C<===E<gt>> marks, thus it is C<3
17764 5 6> (node C<6> is not included into above listing), i.e.,
1777C<gvsv gvsv add whatever>.
1778
9afa14e3
SC
1779Each of these nodes represents an op, a fundamental operation inside the
1780Perl core. The code which implements each operation can be found in the
1781F<pp*.c> files; the function which implements the op with type C<gvsv>
1782is C<pp_gvsv>, and so on. As the tree above shows, different ops have
1783different numbers of children: C<add> is a binary operator, as one would
1784expect, and so has two children. To accommodate the various different
1785numbers of children, there are various types of op data structure, and
1786they link together in different ways.
1787
1788The simplest type of op structure is C<OP>: this has no children. Unary
1789operators, C<UNOP>s, have one child, and this is pointed to by the
1790C<op_first> field. Binary operators (C<BINOP>s) have not only an
1791C<op_first> field but also an C<op_last> field. The most complex type of
1792op is a C<LISTOP>, which has any number of children. In this case, the
1793first child is pointed to by C<op_first> and the last child by
1794C<op_last>. The children in between can be found by iteratively
1795following the C<op_sibling> pointer from the first child to the last.
1796
1797There are also two other op types: a C<PMOP> holds a regular expression,
1798and has no children, and a C<LOOP> may or may not have children. If the
1799C<op_children> field is non-zero, it behaves like a C<LISTOP>. To
1800complicate matters, if a C<UNOP> is actually a C<null> op after
1801optimization (see L</Compile pass 2: context propagation>) it will still
1802have children in accordance with its former type.
1803
06f6df17
RGS
1804Another way to examine the tree is to use a compiler back-end module, such
1805as L<B::Concise>.
1806
0a753a76
PP
1807=head2 Compile pass 1: check routines
1808
8870b5c7
GS
1809The tree is created by the compiler while I<yacc> code feeds it
1810the constructions it recognizes. Since I<yacc> works bottom-up, so does
0a753a76
PP
1811the first pass of perl compilation.
1812
1813What makes this pass interesting for perl developers is that some
1814optimization may be performed on this pass. This is optimization by
8870b5c7 1815so-called "check routines". The correspondence between node names
0a753a76
PP
1816and corresponding check routines is described in F<opcode.pl> (do not
1817forget to run C<make regen_headers> if you modify this file).
1818
1819A check routine is called when the node is fully constructed except
7b8d334a 1820for the execution-order thread. Since at this time there are no
0a753a76
PP
1821back-links to the currently constructed node, one can do most any
1822operation to the top-level node, including freeing it and/or creating
1823new nodes above/below it.
1824
1825The check routine returns the node which should be inserted into the
1826tree (if the top-level node was not modified, check routine returns
1827its argument).
1828
1829By convention, check routines have names C<ck_*>. They are usually
1830called from C<new*OP> subroutines (or C<convert>) (which in turn are
1831called from F<perly.y>).
1832
1833=head2 Compile pass 1a: constant folding
1834
1835Immediately after the check routine is called the returned node is
1836checked for being compile-time executable. If it is (the value is
1837judged to be constant) it is immediately executed, and a I<constant>
1838node with the "return value" of the corresponding subtree is
1839substituted instead. The subtree is deleted.
1840
1841If constant folding was not performed, the execution-order thread is
1842created.
1843
1844=head2 Compile pass 2: context propagation
1845
1846When a context for a part of compile tree is known, it is propagated
a3cb178b 1847down through the tree. At this time the context can have 5 values
0a753a76
PP
1848(instead of 2 for runtime context): void, boolean, scalar, list, and
1849lvalue. In contrast with the pass 1 this pass is processed from top
1850to bottom: a node's context determines the context for its children.
1851
1852Additional context-dependent optimizations are performed at this time.
1853Since at this moment the compile tree contains back-references (via
1854"thread" pointers), nodes cannot be free()d now. To allow
1855optimized-away nodes at this stage, such nodes are null()ified instead
1856of free()ing (i.e. their type is changed to OP_NULL).
1857
1858=head2 Compile pass 3: peephole optimization
1859
1860After the compile tree for a subroutine (or for an C<eval> or a file)
1861is created, an additional pass over the code is performed. This pass
1862is neither top-down or bottom-up, but in the execution order (with
9ea12537
Z
1863additional complications for conditionals). Optimizations performed
1864at this stage are subject to the same restrictions as in the pass 2.
1865
1866Peephole optimizations are done by calling the function pointed to
1867by the global variable C<PL_peepp>. By default, C<PL_peepp> just
1868calls the function pointed to by the global variable C<PL_rpeepp>.
1869By default, that performs some basic op fixups and optimisations along
1870the execution-order op chain, and recursively calls C<PL_rpeepp> for
1871each side chain of ops (resulting from conditionals). Extensions may
1872provide additional optimisations or fixups, hooking into either the
1873per-subroutine or recursive stage, like this:
1874
1875 static peep_t prev_peepp;
1876 static void my_peep(pTHX_ OP *o)
1877 {
1878 /* custom per-subroutine optimisation goes here */
1879 prev_peepp(o);
1880 /* custom per-subroutine optimisation may also go here */
1881 }
1882 BOOT:
1883 prev_peepp = PL_peepp;
1884 PL_peepp = my_peep;
1885
1886 static peep_t prev_rpeepp;
1887 static void my_rpeep(pTHX_ OP *o)
1888 {
1889 OP *orig_o = o;
1890 for(; o; o = o->op_next) {
1891 /* custom per-op optimisation goes here */
1892 }
1893 prev_rpeepp(orig_o);
1894 }
1895 BOOT:
1896 prev_rpeepp = PL_rpeepp;
1897 PL_rpeepp = my_rpeep;
0a753a76 1898
1ba7f851
PJ
1899=head2 Pluggable runops
1900
1901The compile tree is executed in a runops function. There are two runops
1388f78e
RGS
1902functions, in F<run.c> and in F<dump.c>. C<Perl_runops_debug> is used
1903with DEBUGGING and C<Perl_runops_standard> is used otherwise. For fine
1904control over the execution of the compile tree it is possible to provide
1905your own runops function.
1ba7f851
PJ
1906
1907It's probably best to copy one of the existing runops functions and
1908change it to suit your needs. Then, in the BOOT section of your XS
1909file, add the line:
1910
1911 PL_runops = my_runops;
1912
1913This function should be as efficient as possible to keep your programs
1914running as fast as possible.
1915
fd85fad2
BM
1916=head2 Compile-time scope hooks
1917
1918As of perl 5.14 it is possible to hook into the compile-time lexical
1919scope mechanism using C<Perl_blockhook_register>. This is used like
1920this:
1921
1922 STATIC void my_start_hook(pTHX_ int full);
1923 STATIC BHK my_hooks;
1924
1925 BOOT:
a88d97bf 1926 BhkENTRY_set(&my_hooks, bhk_start, my_start_hook);
fd85fad2
BM
1927 Perl_blockhook_register(aTHX_ &my_hooks);
1928
1929This will arrange to have C<my_start_hook> called at the start of
1930compiling every lexical scope. The available hooks are:
1931
1932=over 4
1933
a88d97bf 1934=item C<void bhk_start(pTHX_ int full)>
fd85fad2
BM
1935
1936This is called just after starting a new lexical scope. Note that Perl
1937code like
1938
1939 if ($x) { ... }
1940
1941creates two scopes: the first starts at the C<(> and has C<full == 1>,
1942the second starts at the C<{> and has C<full == 0>. Both end at the
1943C<}>, so calls to C<start> and C<pre/post_end> will match. Anything
1944pushed onto the save stack by this hook will be popped just before the
1945scope ends (between the C<pre_> and C<post_end> hooks, in fact).
1946
a88d97bf 1947=item C<void bhk_pre_end(pTHX_ OP **o)>
fd85fad2
BM
1948
1949This is called at the end of a lexical scope, just before unwinding the
1950stack. I<o> is the root of the optree representing the scope; it is a
1951double pointer so you can replace the OP if you need to.
1952
a88d97bf 1953=item C<void bhk_post_end(pTHX_ OP **o)>
fd85fad2
BM
1954
1955This is called at the end of a lexical scope, just after unwinding the
1956stack. I<o> is as above. Note that it is possible for calls to C<pre_>
1957and C<post_end> to nest, if there is something on the save stack that
1958calls string eval.
1959
a88d97bf 1960=item C<void bhk_eval(pTHX_ OP *const o)>
fd85fad2
BM
1961
1962This is called just before starting to compile an C<eval STRING>, C<do
1963FILE>, C<require> or C<use>, after the eval has been set up. I<o> is the
1964OP that requested the eval, and will normally be an C<OP_ENTEREVAL>,
1965C<OP_DOFILE> or C<OP_REQUIRE>.
1966
1967=back
1968
1969Once you have your hook functions, you need a C<BHK> structure to put
1970them in. It's best to allocate it statically, since there is no way to
1971free it once it's registered. The function pointers should be inserted
1972into this structure using the C<BhkENTRY_set> macro, which will also set
1973flags indicating which entries are valid. If you do need to allocate
1974your C<BHK> dynamically for some reason, be sure to zero it before you
1975start.
1976
1977Once registered, there is no mechanism to switch these hooks off, so if
1978that is necessary you will need to do this yourself. An entry in C<%^H>
a3e07c87
BM
1979is probably the best way, so the effect is lexically scoped; however it
1980is also possible to use the C<BhkDISABLE> and C<BhkENABLE> macros to
1981temporarily switch entries on and off. You should also be aware that
1982generally speaking at least one scope will have opened before your
1983extension is loaded, so you will see some C<pre/post_end> pairs that
1984didn't have a matching C<start>.
fd85fad2 1985
9afa14e3
SC
1986=head1 Examining internal data structures with the C<dump> functions
1987
1988To aid debugging, the source file F<dump.c> contains a number of
1989functions which produce formatted output of internal data structures.
1990
1991The most commonly used of these functions is C<Perl_sv_dump>; it's used
1992for dumping SVs, AVs, HVs, and CVs. The C<Devel::Peek> module calls
1993C<sv_dump> to produce debugging output from Perl-space, so users of that
00aadd71 1994module should already be familiar with its format.
9afa14e3
SC
1995
1996C<Perl_op_dump> can be used to dump an C<OP> structure or any of its
210b36aa 1997derivatives, and produces output similar to C<perl -Dx>; in fact,
9afa14e3
SC
1998C<Perl_dump_eval> will dump the main root of the code being evaluated,
1999exactly like C<-Dx>.
2000
2001Other useful functions are C<Perl_dump_sub>, which turns a C<GV> into an
2002op tree, C<Perl_dump_packsubs> which calls C<Perl_dump_sub> on all the
2003subroutines in a package like so: (Thankfully, these are all xsubs, so
2004there is no op tree)
2005
2006 (gdb) print Perl_dump_packsubs(PL_defstash)
2007
2008 SUB attributes::bootstrap = (xsub 0x811fedc 0)
2009
2010 SUB UNIVERSAL::can = (xsub 0x811f50c 0)
2011
2012 SUB UNIVERSAL::isa = (xsub 0x811f304 0)
2013
2014 SUB UNIVERSAL::VERSION = (xsub 0x811f7ac 0)
2015
2016 SUB DynaLoader::boot_DynaLoader = (xsub 0x805b188 0)
2017
2018and C<Perl_dump_all>, which dumps all the subroutines in the stash and
2019the op tree of the main root.
2020
954c1994 2021=head1 How multiple interpreters and concurrency are supported
ee072b34 2022
ee072b34
GS
2023=head2 Background and PERL_IMPLICIT_CONTEXT
2024
2025The Perl interpreter can be regarded as a closed box: it has an API
2026for feeding it code or otherwise making it do things, but it also has
2027functions for its own use. This smells a lot like an object, and
2028there are ways for you to build Perl so that you can have multiple
acfe0abc
GS
2029interpreters, with one interpreter represented either as a C structure,
2030or inside a thread-specific structure. These structures contain all
2031the context, the state of that interpreter.
2032
7b52221d
RGS
2033One macro controls the major Perl build flavor: MULTIPLICITY. The
2034MULTIPLICITY build has a C structure that packages all the interpreter
2035state. With multiplicity-enabled perls, PERL_IMPLICIT_CONTEXT is also
2036normally defined, and enables the support for passing in a "hidden" first
2037argument that represents all three data structures. MULTIPLICITY makes
1a64a5e6 2038multi-threaded perls possible (with the ithreads threading model, related
7b52221d 2039to the macro USE_ITHREADS.)
54aff467 2040
27da23d5
JH
2041Two other "encapsulation" macros are the PERL_GLOBAL_STRUCT and
2042PERL_GLOBAL_STRUCT_PRIVATE (the latter turns on the former, and the
2043former turns on MULTIPLICITY.) The PERL_GLOBAL_STRUCT causes all the
2044internal variables of Perl to be wrapped inside a single global struct,
2045struct perl_vars, accessible as (globals) &PL_Vars or PL_VarsPtr or
2046the function Perl_GetVars(). The PERL_GLOBAL_STRUCT_PRIVATE goes
2047one step further, there is still a single struct (allocated in main()
2048either from heap or from stack) but there are no global data symbols
2049pointing to it. In either case the global struct should be initialised
2050as the very first thing in main() using Perl_init_global_struct() and
2051correspondingly tear it down after perl_free() using Perl_free_global_struct(),
2052please see F<miniperlmain.c> for usage details. You may also need
2053to use C<dVAR> in your coding to "declare the global variables"
2054when you are using them. dTHX does this for you automatically.
2055
bc028b6b
JH
2056To see whether you have non-const data you can use a BSD-compatible C<nm>:
2057
2058 nm libperl.a | grep -v ' [TURtr] '
2059
2060If this displays any C<D> or C<d> symbols, you have non-const data.
2061
27da23d5
JH
2062For backward compatibility reasons defining just PERL_GLOBAL_STRUCT
2063doesn't actually hide all symbols inside a big global struct: some
2064PerlIO_xxx vtables are left visible. The PERL_GLOBAL_STRUCT_PRIVATE
2065then hides everything (see how the PERLIO_FUNCS_DECL is used).
2066
54aff467 2067All this obviously requires a way for the Perl internal functions to be
acfe0abc 2068either subroutines taking some kind of structure as the first
ee072b34 2069argument, or subroutines taking nothing as the first argument. To
acfe0abc 2070enable these two very different ways of building the interpreter,
ee072b34
GS
2071the Perl source (as it does in so many other situations) makes heavy
2072use of macros and subroutine naming conventions.
2073
54aff467 2074First problem: deciding which functions will be public API functions and
00aadd71 2075which will be private. All functions whose names begin C<S_> are private
954c1994
GS
2076(think "S" for "secret" or "static"). All other functions begin with
2077"Perl_", but just because a function begins with "Perl_" does not mean it is
00aadd71
NIS
2078part of the API. (See L</Internal Functions>.) The easiest way to be B<sure> a
2079function is part of the API is to find its entry in L<perlapi>.
2080If it exists in L<perlapi>, it's part of the API. If it doesn't, and you
2081think it should be (i.e., you need it for your extension), send mail via
a422fd2d 2082L<perlbug> explaining why you think it should be.
ee072b34
GS
2083
2084Second problem: there must be a syntax so that the same subroutine
2085declarations and calls can pass a structure as their first argument,
2086or pass nothing. To solve this, the subroutines are named and
2087declared in a particular way. Here's a typical start of a static
2088function used within the Perl guts:
2089
2090 STATIC void
2091 S_incline(pTHX_ char *s)
2092
acfe0abc
GS
2093STATIC becomes "static" in C, and may be #define'd to nothing in some
2094configurations in future.
ee072b34 2095
651a3225
GS
2096A public function (i.e. part of the internal API, but not necessarily
2097sanctioned for use in extensions) begins like this:
ee072b34
GS
2098
2099 void
2307c6d0 2100 Perl_sv_setiv(pTHX_ SV* dsv, IV num)
ee072b34 2101
0147cd53 2102C<pTHX_> is one of a number of macros (in F<perl.h>) that hide the
ee072b34
GS
2103details of the interpreter's context. THX stands for "thread", "this",
2104or "thingy", as the case may be. (And no, George Lucas is not involved. :-)
2105The first character could be 'p' for a B<p>rototype, 'a' for B<a>rgument,
a7486cbb
JH
2106or 'd' for B<d>eclaration, so we have C<pTHX>, C<aTHX> and C<dTHX>, and
2107their variants.
ee072b34 2108
a7486cbb
JH
2109When Perl is built without options that set PERL_IMPLICIT_CONTEXT, there is no
2110first argument containing the interpreter's context. The trailing underscore
ee072b34
GS
2111in the pTHX_ macro indicates that the macro expansion needs a comma
2112after the context argument because other arguments follow it. If
2113PERL_IMPLICIT_CONTEXT is not defined, pTHX_ will be ignored, and the
54aff467
GS
2114subroutine is not prototyped to take the extra argument. The form of the
2115macro without the trailing underscore is used when there are no additional
ee072b34
GS
2116explicit arguments.
2117
54aff467 2118When a core function calls another, it must pass the context. This
2307c6d0 2119is normally hidden via macros. Consider C<sv_setiv>. It expands into
ee072b34
GS
2120something like this:
2121
2307c6d0
SB
2122 #ifdef PERL_IMPLICIT_CONTEXT
2123 #define sv_setiv(a,b) Perl_sv_setiv(aTHX_ a, b)
ee072b34 2124 /* can't do this for vararg functions, see below */
2307c6d0
SB
2125 #else
2126 #define sv_setiv Perl_sv_setiv
2127 #endif
ee072b34
GS
2128
2129This works well, and means that XS authors can gleefully write:
2130
2307c6d0 2131 sv_setiv(foo, bar);
ee072b34
GS
2132
2133and still have it work under all the modes Perl could have been
2134compiled with.
2135
ee072b34
GS
2136This doesn't work so cleanly for varargs functions, though, as macros
2137imply that the number of arguments is known in advance. Instead we
2138either need to spell them out fully, passing C<aTHX_> as the first
2139argument (the Perl core tends to do this with functions like
2140Perl_warner), or use a context-free version.
2141
2142The context-free version of Perl_warner is called
2143Perl_warner_nocontext, and does not take the extra argument. Instead
2144it does dTHX; to get the context from thread-local storage. We
2145C<#define warner Perl_warner_nocontext> so that extensions get source
2146compatibility at the expense of performance. (Passing an arg is
2147cheaper than grabbing it from thread-local storage.)
2148
acfe0abc 2149You can ignore [pad]THXx when browsing the Perl headers/sources.
ee072b34
GS
2150Those are strictly for use within the core. Extensions and embedders
2151need only be aware of [pad]THX.
2152
a7486cbb
JH
2153=head2 So what happened to dTHR?
2154
2155C<dTHR> was introduced in perl 5.005 to support the older thread model.
2156The older thread model now uses the C<THX> mechanism to pass context
2157pointers around, so C<dTHR> is not useful any more. Perl 5.6.0 and
2158later still have it for backward source compatibility, but it is defined
2159to be a no-op.
2160
ee072b34
GS
2161=head2 How do I use all this in extensions?
2162
2163When Perl is built with PERL_IMPLICIT_CONTEXT, extensions that call
2164any functions in the Perl API will need to pass the initial context
2165argument somehow. The kicker is that you will need to write it in
2166such a way that the extension still compiles when Perl hasn't been
2167built with PERL_IMPLICIT_CONTEXT enabled.
2168
2169There are three ways to do this. First, the easy but inefficient way,
2170which is also the default, in order to maintain source compatibility
0147cd53 2171with extensions: whenever F<XSUB.h> is #included, it redefines the aTHX
ee072b34
GS
2172and aTHX_ macros to call a function that will return the context.
2173Thus, something like:
2174
2307c6d0 2175 sv_setiv(sv, num);
ee072b34 2176
4375e838 2177in your extension will translate to this when PERL_IMPLICIT_CONTEXT is
54aff467 2178in effect:
ee072b34 2179
2307c6d0 2180 Perl_sv_setiv(Perl_get_context(), sv, num);
ee072b34 2181
54aff467 2182or to this otherwise:
ee072b34 2183
2307c6d0 2184 Perl_sv_setiv(sv, num);
ee072b34
GS
2185
2186You have to do nothing new in your extension to get this; since
2fa86c13 2187the Perl library provides Perl_get_context(), it will all just
ee072b34
GS
2188work.
2189
2190The second, more efficient way is to use the following template for
2191your Foo.xs:
2192
c52f9dcd
JH
2193 #define PERL_NO_GET_CONTEXT /* we want efficiency */
2194 #include "EXTERN.h"
2195 #include "perl.h"
2196 #include "XSUB.h"
ee072b34 2197
fd061412 2198 STATIC void my_private_function(int arg1, int arg2);
ee072b34 2199
fd061412 2200 STATIC void
c52f9dcd
JH
2201 my_private_function(int arg1, int arg2)
2202 {
2203 dTHX; /* fetch context */
2204 ... call many Perl API functions ...
2205 }
ee072b34
GS
2206
2207 [... etc ...]
2208
c52f9dcd 2209 MODULE = Foo PACKAGE = Foo
ee072b34 2210
c52f9dcd 2211 /* typical XSUB */
ee072b34 2212
c52f9dcd
JH
2213 void
2214 my_xsub(arg)
2215 int arg
2216 CODE:
2217 my_private_function(arg, 10);
ee072b34
GS
2218
2219Note that the only two changes from the normal way of writing an
2220extension is the addition of a C<#define PERL_NO_GET_CONTEXT> before
2221including the Perl headers, followed by a C<dTHX;> declaration at
2222the start of every function that will call the Perl API. (You'll
2223know which functions need this, because the C compiler will complain
2224that there's an undeclared identifier in those functions.) No changes
2225are needed for the XSUBs themselves, because the XS() macro is
2226correctly defined to pass in the implicit context if needed.
2227
2228The third, even more efficient way is to ape how it is done within
2229the Perl guts:
2230
2231
c52f9dcd
JH
2232 #define PERL_NO_GET_CONTEXT /* we want efficiency */
2233 #include "EXTERN.h"
2234 #include "perl.h"
2235 #include "XSUB.h"
ee072b34
GS
2236
2237 /* pTHX_ only needed for functions that call Perl API */
fd061412 2238 STATIC void my_private_function(pTHX_ int arg1, int arg2);
ee072b34 2239
fd061412 2240 STATIC void
c52f9dcd
JH
2241 my_private_function(pTHX_ int arg1, int arg2)
2242 {
2243 /* dTHX; not needed here, because THX is an argument */
2244 ... call Perl API functions ...
2245 }
ee072b34
GS
2246
2247 [... etc ...]
2248
c52f9dcd 2249 MODULE = Foo PACKAGE = Foo
ee072b34 2250
c52f9dcd 2251 /* typical XSUB */
ee072b34 2252
c52f9dcd
JH
2253 void
2254 my_xsub(arg)
2255 int arg
2256 CODE:
2257 my_private_function(aTHX_ arg, 10);
ee072b34
GS
2258
2259This implementation never has to fetch the context using a function
2260call, since it is always passed as an extra argument. Depending on
2261your needs for simplicity or efficiency, you may mix the previous
2262two approaches freely.
2263
651a3225
GS
2264Never add a comma after C<pTHX> yourself--always use the form of the
2265macro with the underscore for functions that take explicit arguments,
2266or the form without the argument for functions with no explicit arguments.
ee072b34 2267
27da23d5
JH
2268If one is compiling Perl with the C<-DPERL_GLOBAL_STRUCT> the C<dVAR>
2269definition is needed if the Perl global variables (see F<perlvars.h>
2270or F<globvar.sym>) are accessed in the function and C<dTHX> is not
2271used (the C<dTHX> includes the C<dVAR> if necessary). One notices
2272the need for C<dVAR> only with the said compile-time define, because
2273otherwise the Perl global variables are visible as-is.
2274
a7486cbb
JH
2275=head2 Should I do anything special if I call perl from multiple threads?
2276
2277If you create interpreters in one thread and then proceed to call them in
2278another, you need to make sure perl's own Thread Local Storage (TLS) slot is
2279initialized correctly in each of those threads.
2280
2281The C<perl_alloc> and C<perl_clone> API functions will automatically set
2282the TLS slot to the interpreter they created, so that there is no need to do
2283anything special if the interpreter is always accessed in the same thread that
2284created it, and that thread did not create or call any other interpreters
2285afterwards. If that is not the case, you have to set the TLS slot of the
2286thread before calling any functions in the Perl API on that particular
2287interpreter. This is done by calling the C<PERL_SET_CONTEXT> macro in that
2288thread as the first thing you do:
2289
2290 /* do this before doing anything else with some_perl */
2291 PERL_SET_CONTEXT(some_perl);
2292
2293 ... other Perl API calls on some_perl go here ...
2294
ee072b34
GS
2295=head2 Future Plans and PERL_IMPLICIT_SYS
2296
2297Just as PERL_IMPLICIT_CONTEXT provides a way to bundle up everything
2298that the interpreter knows about itself and pass it around, so too are
2299there plans to allow the interpreter to bundle up everything it knows
2300about the environment it's running on. This is enabled with the
7b52221d
RGS
2301PERL_IMPLICIT_SYS macro. Currently it only works with USE_ITHREADS on
2302Windows.
ee072b34
GS
2303
2304This allows the ability to provide an extra pointer (called the "host"
2305environment) for all the system calls. This makes it possible for
2306all the system stuff to maintain their own state, broken down into
2307seven C structures. These are thin wrappers around the usual system
0147cd53 2308calls (see F<win32/perllib.c>) for the default perl executable, but for a
ee072b34
GS
2309more ambitious host (like the one that would do fork() emulation) all
2310the extra work needed to pretend that different interpreters are
2311actually different "processes", would be done here.
2312
2313The Perl engine/interpreter and the host are orthogonal entities.
2314There could be one or more interpreters in a process, and one or
2315more "hosts", with free association between them.
2316
a422fd2d
SC
2317=head1 Internal Functions
2318
2319All of Perl's internal functions which will be exposed to the outside
06f6df17 2320world are prefixed by C<Perl_> so that they will not conflict with XS
a422fd2d
SC
2321functions or functions used in a program in which Perl is embedded.
2322Similarly, all global variables begin with C<PL_>. (By convention,
06f6df17 2323static functions start with C<S_>.)
a422fd2d 2324
0972ecdf
DM
2325Inside the Perl core (C<PERL_CORE> defined), you can get at the functions
2326either with or without the C<Perl_> prefix, thanks to a bunch of defines
2327that live in F<embed.h>. Note that extension code should I<not> set
2328C<PERL_CORE>; this exposes the full perl internals, and is likely to cause
2329breakage of the XS in each new perl release.
2330
2331The file F<embed.h> is generated automatically from
dc9b1d22
MHM
2332F<embed.pl> and F<embed.fnc>. F<embed.pl> also creates the prototyping
2333header files for the internal functions, generates the documentation
2334and a lot of other bits and pieces. It's important that when you add
2335a new function to the core or change an existing one, you change the
2336data in the table in F<embed.fnc> as well. Here's a sample entry from
2337that table:
a422fd2d
SC
2338
2339 Apd |SV** |av_fetch |AV* ar|I32 key|I32 lval
2340
2341The second column is the return type, the third column the name. Columns
2342after that are the arguments. The first column is a set of flags:
2343
2344=over 3
2345
2346=item A
2347
1aa6ea50
JC
2348This function is a part of the public API. All such functions should also
2349have 'd', very few do not.
a422fd2d
SC
2350
2351=item p
2352
1aa6ea50
JC
2353This function has a C<Perl_> prefix; i.e. it is defined as
2354C<Perl_av_fetch>.
a422fd2d
SC
2355
2356=item d
2357
2358This function has documentation using the C<apidoc> feature which we'll
1aa6ea50 2359look at in a second. Some functions have 'd' but not 'A'; docs are good.
a422fd2d
SC
2360
2361=back
2362
2363Other available flags are:
2364
2365=over 3
2366
2367=item s
2368
1aa6ea50
JC
2369This is a static function and is defined as C<STATIC S_whatever>, and
2370usually called within the sources as C<whatever(...)>.
a422fd2d
SC
2371
2372=item n
2373
1aa6ea50
JC
2374This does not need a interpreter context, so the definition has no
2375C<pTHX>, and it follows that callers don't use C<aTHX>. (See
d3a43cd8 2376L</Background and PERL_IMPLICIT_CONTEXT>.)
a422fd2d
SC
2377
2378=item r
2379
2380This function never returns; C<croak>, C<exit> and friends.
2381
2382=item f
2383
2384This function takes a variable number of arguments, C<printf> style.
2385The argument list should end with C<...>, like this:
2386
2387 Afprd |void |croak |const char* pat|...
2388
a7486cbb 2389=item M
a422fd2d 2390
00aadd71 2391This function is part of the experimental development API, and may change
a422fd2d
SC
2392or disappear without notice.
2393
2394=item o
2395
2396This function should not have a compatibility macro to define, say,
2397C<Perl_parse> to C<parse>. It must be called as C<Perl_parse>.
2398
a422fd2d
SC
2399=item x
2400
2401This function isn't exported out of the Perl core.
2402
dc9b1d22
MHM
2403=item m
2404
2405This is implemented as a macro.
2406
2407=item X
2408
2409This function is explicitly exported.
2410
2411=item E
2412
2413This function is visible to extensions included in the Perl core.
2414
2415=item b
2416
2417Binary backward compatibility; this function is a macro but also has
2418a C<Perl_> implementation (which is exported).
2419
1aa6ea50
JC
2420=item others
2421
2422See the comments at the top of C<embed.fnc> for others.
2423
a422fd2d
SC
2424=back
2425
dc9b1d22
MHM
2426If you edit F<embed.pl> or F<embed.fnc>, you will need to run
2427C<make regen_headers> to force a rebuild of F<embed.h> and other
2428auto-generated files.
a422fd2d 2429
6b4667fc 2430=head2 Formatted Printing of IVs, UVs, and NVs
9dd9db0b 2431
6b4667fc
A
2432If you are printing IVs, UVs, or NVS instead of the stdio(3) style
2433formatting codes like C<%d>, C<%ld>, C<%f>, you should use the
2434following macros for portability
9dd9db0b 2435
c52f9dcd
JH
2436 IVdf IV in decimal
2437 UVuf UV in decimal
2438 UVof UV in octal
2439 UVxf UV in hexadecimal
2440 NVef NV %e-like
2441 NVff NV %f-like
2442 NVgf NV %g-like
9dd9db0b 2443
6b4667fc
A
2444These will take care of 64-bit integers and long doubles.
2445For example:
2446
c52f9dcd 2447 printf("IV is %"IVdf"\n", iv);
6b4667fc
A
2448
2449The IVdf will expand to whatever is the correct format for the IVs.
9dd9db0b 2450
8908e76d
JH
2451If you are printing addresses of pointers, use UVxf combined
2452with PTR2UV(), do not use %lx or %p.
2453
2454=head2 Pointer-To-Integer and Integer-To-Pointer
2455
2456Because pointer size does not necessarily equal integer size,
2457use the follow macros to do it right.
2458
c52f9dcd
JH
2459 PTR2UV(pointer)
2460 PTR2IV(pointer)
2461 PTR2NV(pointer)
2462 INT2PTR(pointertotype, integer)
8908e76d
JH
2463
2464For example:
2465
c52f9dcd
JH
2466 IV iv = ...;
2467 SV *sv = INT2PTR(SV*, iv);
8908e76d
JH
2468
2469and
2470
c52f9dcd
JH
2471 AV *av = ...;
2472 UV uv = PTR2UV(av);
8908e76d 2473
0ca3a874
MHM
2474=head2 Exception Handling
2475
9b5c3821
MHM
2476There are a couple of macros to do very basic exception handling in XS
2477modules. You have to define C<NO_XSLOCKS> before including F<XSUB.h> to
2478be able to use these macros:
2479
2480 #define NO_XSLOCKS
2481 #include "XSUB.h"
2482
2483You can use these macros if you call code that may croak, but you need
2484to do some cleanup before giving control back to Perl. For example:
0ca3a874 2485
d7f8936a 2486 dXCPT; /* set up necessary variables */
0ca3a874
MHM
2487
2488 XCPT_TRY_START {
2489 code_that_may_croak();
2490 } XCPT_TRY_END
2491
2492 XCPT_CATCH
2493 {
2494 /* do cleanup here */
2495 XCPT_RETHROW;
2496 }
2497
2498Note that you always have to rethrow an exception that has been
2499caught. Using these macros, it is not possible to just catch the
2500exception and ignore it. If you have to ignore the exception, you
2501have to use the C<call_*> function.
2502
2503The advantage of using the above macros is that you don't have
2504to setup an extra function for C<call_*>, and that using these
2505macros is faster than using C<call_*>.
2506
a422fd2d
SC
2507=head2 Source Documentation
2508
2509There's an effort going on to document the internal functions and
2510automatically produce reference manuals from them - L<perlapi> is one
2511such manual which details all the functions which are available to XS
2512writers. L<perlintern> is the autogenerated manual for the functions
2513which are not part of the API and are supposedly for internal use only.
2514
2515Source documentation is created by putting POD comments into the C
2516source, like this:
2517
2518 /*
2519 =for apidoc sv_setiv
2520
2521 Copies an integer into the given SV. Does not handle 'set' magic. See
2522 C<sv_setiv_mg>.
2523
2524 =cut
2525 */
2526
2527Please try and supply some documentation if you add functions to the
2528Perl core.
2529
0d098d33
MHM
2530=head2 Backwards compatibility
2531
2532The Perl API changes over time. New functions are added or the interfaces
2533of existing functions are changed. The C<Devel::PPPort> module tries to
2534provide compatibility code for some of these changes, so XS writers don't
2535have to code it themselves when supporting multiple versions of Perl.
2536
2537C<Devel::PPPort> generates a C header file F<ppport.h> that can also
2538be run as a Perl script. To generate F<ppport.h>, run:
2539
2540 perl -MDevel::PPPort -eDevel::PPPort::WriteFile
2541
2542Besides checking existing XS code, the script can also be used to retrieve
2543compatibility information for various API calls using the C<--api-info>
2544command line switch. For example:
2545
2546 % perl ppport.h --api-info=sv_magicext
2547
2548For details, see C<perldoc ppport.h>.
2549
a422fd2d
SC
2550=head1 Unicode Support
2551
2552Perl 5.6.0 introduced Unicode support. It's important for porters and XS
2553writers to understand this support and make sure that the code they
2554write does not corrupt Unicode data.
2555
2556=head2 What B<is> Unicode, anyway?
2557
2558In the olden, less enlightened times, we all used to use ASCII. Most of
2559us did, anyway. The big problem with ASCII is that it's American. Well,
2560no, that's not actually the problem; the problem is that it's not
2561particularly useful for people who don't use the Roman alphabet. What
2562used to happen was that particular languages would stick their own
2563alphabet in the upper range of the sequence, between 128 and 255. Of
2564course, we then ended up with plenty of variants that weren't quite
2565ASCII, and the whole point of it being a standard was lost.
2566
2567Worse still, if you've got a language like Chinese or
2568Japanese that has hundreds or thousands of characters, then you really
2569can't fit them into a mere 256, so they had to forget about ASCII
2570altogether, and build their own systems using pairs of numbers to refer
2571to one character.
2572
2573To fix this, some people formed Unicode, Inc. and
2574produced a new character set containing all the characters you can
2575possibly think of and more. There are several ways of representing these
1e54db1a 2576characters, and the one Perl uses is called UTF-8. UTF-8 uses
2575c402
JW
2577a variable number of bytes to represent a character. You can learn more
2578about Unicode and Perl's Unicode model in L<perlunicode>.
a422fd2d 2579
1e54db1a 2580=head2 How can I recognise a UTF-8 string?
a422fd2d 2581
1e54db1a
JH
2582You can't. This is because UTF-8 data is stored in bytes just like
2583non-UTF-8 data. The Unicode character 200, (C<0xC8> for you hex types)
a422fd2d
SC
2584capital E with a grave accent, is represented by the two bytes
2585C<v196.172>. Unfortunately, the non-Unicode string C<chr(196).chr(172)>
2586has that byte sequence as well. So you can't tell just by looking - this
2587is what makes Unicode input an interesting problem.
2588
2575c402
JW
2589In general, you either have to know what you're dealing with, or you
2590have to guess. The API function C<is_utf8_string> can help; it'll tell
2591you if a string contains only valid UTF-8 characters. However, it can't
2592do the work for you. On a character-by-character basis, C<is_utf8_char>
2593will tell you whether the current character in a string is valid UTF-8.
a422fd2d 2594
1e54db1a 2595=head2 How does UTF-8 represent Unicode characters?
a422fd2d 2596
1e54db1a 2597As mentioned above, UTF-8 uses a variable number of bytes to store a
2575c402
JW
2598character. Characters with values 0...127 are stored in one byte, just
2599like good ol' ASCII. Character 128 is stored as C<v194.128>; this
a31a806a 2600continues up to character 191, which is C<v194.191>. Now we've run out of
a422fd2d
SC
2601bits (191 is binary C<10111111>) so we move on; 192 is C<v195.128>. And
2602so it goes on, moving to three bytes at character 2048.
2603
1e54db1a 2604Assuming you know you're dealing with a UTF-8 string, you can find out
a422fd2d
SC
2605how long the first character in it is with the C<UTF8SKIP> macro:
2606
2607 char *utf = "\305\233\340\240\201";
2608 I32 len;
2609
2610 len = UTF8SKIP(utf); /* len is 2 here */
2611 utf += len;
2612 len = UTF8SKIP(utf); /* len is 3 here */
2613
1e54db1a 2614Another way to skip over characters in a UTF-8 string is to use
a422fd2d
SC
2615C<utf8_hop>, which takes a string and a number of characters to skip
2616over. You're on your own about bounds checking, though, so don't use it
2617lightly.
2618
1e54db1a 2619All bytes in a multi-byte UTF-8 character will have the high bit set,
3a2263fe
RGS
2620so you can test if you need to do something special with this
2621character like this (the UTF8_IS_INVARIANT() is a macro that tests
2622whether the byte can be encoded as a single byte even in UTF-8):
a422fd2d 2623
3a2263fe
RGS
2624 U8 *utf;
2625 UV uv; /* Note: a UV, not a U8, not a char */
a422fd2d 2626
3a2263fe 2627 if (!UTF8_IS_INVARIANT(*utf))
1e54db1a 2628 /* Must treat this as UTF-8 */
a422fd2d
SC
2629 uv = utf8_to_uv(utf);
2630 else
2631 /* OK to treat this character as a byte */
2632 uv = *utf;
2633
2634You can also see in that example that we use C<utf8_to_uv> to get the
2635value of the character; the inverse function C<uv_to_utf8> is available
1e54db1a 2636for putting a UV into UTF-8:
a422fd2d 2637
3a2263fe 2638 if (!UTF8_IS_INVARIANT(uv))
a422fd2d
SC
2639 /* Must treat this as UTF8 */
2640 utf8 = uv_to_utf8(utf8, uv);
2641 else
2642 /* OK to treat this character as a byte */
2643 *utf8++ = uv;
2644
2645You B<must> convert characters to UVs using the above functions if
1e54db1a
JH
2646you're ever in a situation where you have to match UTF-8 and non-UTF-8
2647characters. You may not skip over UTF-8 characters in this case. If you
2648do this, you'll lose the ability to match hi-bit non-UTF-8 characters;
2649for instance, if your UTF-8 string contains C<v196.172>, and you skip
2650that character, you can never match a C<chr(200)> in a non-UTF-8 string.
a422fd2d
SC
2651So don't do that!
2652
1e54db1a 2653=head2 How does Perl store UTF-8 strings?
a422fd2d
SC
2654
2655Currently, Perl deals with Unicode strings and non-Unicode strings
2575c402
JW
2656slightly differently. A flag in the SV, C<SVf_UTF8>, indicates that the
2657string is internally encoded as UTF-8. Without it, the byte value is the
2658codepoint number and vice versa (in other words, the string is encoded
e1b711da
KW
2659as iso-8859-1, but C<use feature 'unicode_strings'> is needed to get iso-8859-1
2660semantics). You can check and manipulate this flag with the
2575c402 2661following macros:
a422fd2d
SC
2662
2663 SvUTF8(sv)
2664 SvUTF8_on(sv)
2665 SvUTF8_off(sv)
2666
2667This flag has an important effect on Perl's treatment of the string: if
2668Unicode data is not properly distinguished, regular expressions,
2669C<length>, C<substr> and other string handling operations will have
2670undesirable results.
2671
2672The problem comes when you have, for instance, a string that isn't
2575c402 2673flagged as UTF-8, and contains a byte sequence that could be UTF-8 -
1e54db1a 2674especially when combining non-UTF-8 and UTF-8 strings.
a422fd2d
SC
2675
2676Never forget that the C<SVf_UTF8> flag is separate to the PV value; you
2677need be sure you don't accidentally knock it off while you're
2678manipulating SVs. More specifically, you cannot expect to do this:
2679
2680 SV *sv;
2681 SV *nsv;
2682 STRLEN len;
2683 char *p;
2684
2685 p = SvPV(sv, len);
2686 frobnicate(p);
2687 nsv = newSVpvn(p, len);
2688
2689The C<char*> string does not tell you the whole story, and you can't
2690copy or reconstruct an SV just by copying the string value. Check if the
2575c402 2691old SV has the UTF8 flag set, and act accordingly:
a422fd2d
SC
2692
2693 p = SvPV(sv, len);
2694 frobnicate(p);
2695 nsv = newSVpvn(p, len);
2696 if (SvUTF8(sv))
2697 SvUTF8_on(nsv);
2698
2699In fact, your C<frobnicate> function should be made aware of whether or
1e54db1a 2700not it's dealing with UTF-8 data, so that it can handle the string
a422fd2d
SC
2701appropriately.
2702
3a2263fe 2703Since just passing an SV to an XS function and copying the data of
2575c402 2704the SV is not enough to copy the UTF8 flags, even less right is just
3a2263fe
RGS
2705passing a C<char *> to an XS function.
2706
1e54db1a 2707=head2 How do I convert a string to UTF-8?
a422fd2d 2708
2575c402
JW
2709If you're mixing UTF-8 and non-UTF-8 strings, it is necessary to upgrade
2710one of the strings to UTF-8. If you've got an SV, the easiest way to do
2711this is:
a422fd2d
SC
2712
2713 sv_utf8_upgrade(sv);
2714
2715However, you must not do this, for example:
2716
2717 if (!SvUTF8(left))
2718 sv_utf8_upgrade(left);
2719
2720If you do this in a binary operator, you will actually change one of the
b1866b2d 2721strings that came into the operator, and, while it shouldn't be noticeable
2575c402 2722by the end user, it can cause problems in deficient code.
a422fd2d 2723
1e54db1a 2724Instead, C<bytes_to_utf8> will give you a UTF-8-encoded B<copy> of its
a422fd2d 2725string argument. This is useful for having the data available for
b1866b2d 2726comparisons and so on, without harming the original SV. There's also
a422fd2d
SC
2727C<utf8_to_bytes> to go the other way, but naturally, this will fail if
2728the string contains any characters above 255 that can't be represented
2729in a single byte.
2730
2731=head2 Is there anything else I need to know?
2732
2733Not really. Just remember these things:
2734
2735=over 3
2736
2737=item *
2738
1e54db1a
JH
2739There's no way to tell if a string is UTF-8 or not. You can tell if an SV
2740is UTF-8 by looking at is C<SvUTF8> flag. Don't forget to set the flag if
2741something should be UTF-8. Treat the flag as part of the PV, even though
a422fd2d
SC
2742it's not - if you pass on the PV to somewhere, pass on the flag too.
2743
2744=item *
2745
1e54db1a 2746If a string is UTF-8, B<always> use C<utf8_to_uv> to get at the value,
3a2263fe 2747unless C<UTF8_IS_INVARIANT(*s)> in which case you can use C<*s>.
a422fd2d
SC
2748
2749=item *
2750
1e54db1a 2751When writing a character C<uv> to a UTF-8 string, B<always> use
3a2263fe
RGS
2752C<uv_to_utf8>, unless C<UTF8_IS_INVARIANT(uv))> in which case
2753you can use C<*s = uv>.
a422fd2d
SC
2754
2755=item *
2756
1e54db1a 2757Mixing UTF-8 and non-UTF-8 strings is tricky. Use C<bytes_to_utf8> to get
2bbc8d55 2758a new string which is UTF-8 encoded, and then combine them.
a422fd2d
SC
2759
2760=back
2761
53e06cf0
SC
2762=head1 Custom Operators
2763
9a68f1db 2764Custom operator support is a new experimental feature that allows you to
53e06cf0
SC
2765define your own ops. This is primarily to allow the building of
2766interpreters for other languages in the Perl core, but it also allows
2767optimizations through the creation of "macro-ops" (ops which perform the
2768functions of multiple ops which are usually executed together, such as
1aa6ea50 2769C<gvsv, gvsv, add>.)
53e06cf0 2770
b455bf3f 2771This feature is implemented as a new op type, C<OP_CUSTOM>. The Perl
53e06cf0
SC
2772core does not "know" anything special about this op type, and so it will
2773not be involved in any optimizations. This also means that you can
2774define your custom ops to be any op structure - unary, binary, list and
2775so on - you like.
2776
2777It's important to know what custom operators won't do for you. They
2778won't let you add new syntax to Perl, directly. They won't even let you
2779add new keywords, directly. In fact, they won't change the way Perl
2780compiles a program at all. You have to do those changes yourself, after
2781Perl has compiled the program. You do this either by manipulating the op
2782tree using a C<CHECK> block and the C<B::Generate> module, or by adding
2783a custom peephole optimizer with the C<optimize> module.
2784
2785When you do this, you replace ordinary Perl ops with custom ops by
2786creating ops with the type C<OP_CUSTOM> and the C<pp_addr> of your own
2787PP function. This should be defined in XS code, and should look like
2788the PP ops in C<pp_*.c>. You are responsible for ensuring that your op
2789takes the appropriate number of values from the stack, and you are
2790responsible for adding stack marks if necessary.
2791
2792You should also "register" your op with the Perl interpreter so that it
2793can produce sensible error and warning messages. Since it is possible to
2794have multiple custom ops within the one "logical" op type C<OP_CUSTOM>,
9733086d
BM
2795Perl uses the value of C<< o->op_ppaddr >> to determine which custom op
2796it is dealing with. You should create an C<XOP> structure for each
2797ppaddr you use, set the properties of the custom op with
2798C<XopENTRY_set>, and register the structure against the ppaddr using
2799C<Perl_custom_op_register>. A trivial example might look like:
2800
2801 static XOP my_xop;
2802 static OP *my_pp(pTHX);
2803
2804 BOOT:
2805 XopENTRY_set(&my_xop, xop_name, "myxop");
2806 XopENTRY_set(&my_xop, xop_desc, "Useless custom op");
2807 Perl_custom_op_register(aTHX_ my_pp, &my_xop);
2808
2809The available fields in the structure are:
2810
2811=over 4
2812
2813=item xop_name
2814
2815A short name for your op. This will be included in some error messages,
2816and will also be returned as C<< $op->name >> by the L<B|B> module, so
2817it will appear in the output of module like L<B::Concise|B::Concise>.
2818
2819=item xop_desc
2820
2821A short description of the function of the op.
2822
2823=item xop_class
2824
2825Which of the various C<*OP> structures this op uses. This should be one of
2826the C<OA_*> constants from F<op.h>, namely
2827
2828=over 4
2829
2830=item OA_BASEOP
2831
2832=item OA_UNOP
2833
2834=item OA_BINOP
2835
2836=item OA_LOGOP
2837
2838=item OA_LISTOP
2839
2840=item OA_PMOP
2841
2842=item OA_SVOP
2843
2844=item OA_PADOP
2845
2846=item OA_PVOP_OR_SVOP
2847
2848This should be interpreted as 'C<PVOP>' only. The C<_OR_SVOP> is because
2849the only core C<PVOP>, C<OP_TRANS>, can sometimes be a C<SVOP> instead.
2850
2851=item OA_LOOP
2852
2853=item OA_COP
2854
2855=back
2856
2857The other C<OA_*> constants should not be used.
2858
2859=item xop_peep
2860
2861This member is of type C<Perl_cpeep_t>, which expands to C<void
2862(*Perl_cpeep_t)(aTHX_ OP *o, OP *oldop)>. If it is set, this function
2863will be called from C<Perl_rpeep> when ops of this type are encountered
2864by the peephole optimizer. I<o> is the OP that needs optimizing;
2865I<oldop> is the previous OP optimized, whose C<op_next> points to I<o>.
2866
2867=back
53e06cf0 2868
e7d4c058 2869C<B::Generate> directly supports the creation of custom ops by name.
53e06cf0 2870
954c1994 2871=head1 AUTHORS
e89caa19 2872
954c1994 2873Until May 1997, this document was maintained by Jeff Okamoto
9b5bb84f
SB
2874E<lt>okamoto@corp.hp.comE<gt>. It is now maintained as part of Perl
2875itself by the Perl 5 Porters E<lt>perl5-porters@perl.orgE<gt>.
cb1a09d0 2876
954c1994
GS
2877With lots of help and suggestions from Dean Roehrich, Malcolm Beattie,
2878Andreas Koenig, Paul Hudson, Ilya Zakharevich, Paul Marquess, Neil
2879Bowers, Matthew Green, Tim Bunce, Spider Boardman, Ulrich Pfeifer,
2880Stephen McCamant, and Gurusamy Sarathy.
cb1a09d0 2881
954c1994 2882=head1 SEE ALSO
cb1a09d0 2883
ba555bf5 2884L<perlapi>, L<perlintern>, L<perlxs>, L<perlembed>