Commit | Line | Data |
---|---|---|
a0d0e21e | 1 | /* sv.c |
79072805 | 2 | * |
1129b882 | 3 | * Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, |
83706693 RGS |
4 | * 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 by Larry Wall |
5 | * and others | |
79072805 LW |
6 | * |
7 | * You may distribute under the terms of either the GNU General Public | |
8 | * License or the Artistic License, as specified in the README file. | |
9 | * | |
4ac71550 TC |
10 | */ |
11 | ||
12 | /* | |
13 | * 'I wonder what the Entish is for "yes" and "no",' he thought. | |
14 | * --Pippin | |
15 | * | |
16 | * [p.480 of _The Lord of the Rings_, III/iv: "Treebeard"] | |
17 | */ | |
18 | ||
19 | /* | |
645c22ef DM |
20 | * |
21 | * | |
5e045b90 AMS |
22 | * This file contains the code that creates, manipulates and destroys |
23 | * scalar values (SVs). The other types (AV, HV, GV, etc.) reuse the | |
24 | * structure of an SV, so their creation and destruction is handled | |
25 | * here; higher-level functions are in av.c, hv.c, and so on. Opcode | |
26 | * level functions (eg. substr, split, join) for each of the types are | |
27 | * in the pp*.c files. | |
79072805 LW |
28 | */ |
29 | ||
30 | #include "EXTERN.h" | |
864dbfa3 | 31 | #define PERL_IN_SV_C |
79072805 | 32 | #include "perl.h" |
d2f185dc | 33 | #include "regcomp.h" |
79072805 | 34 | |
51371543 | 35 | #define FCALL *f |
2c5424a7 | 36 | |
2f8ed50e OS |
37 | #ifdef __Lynx__ |
38 | /* Missing proto on LynxOS */ | |
39 | char *gconvert(double, int, int, char *); | |
40 | #endif | |
41 | ||
e23c8137 | 42 | #ifdef PERL_UTF8_CACHE_ASSERT |
ab455f60 | 43 | /* if adding more checks watch out for the following tests: |
e23c8137 JH |
44 | * t/op/index.t t/op/length.t t/op/pat.t t/op/substr.t |
45 | * lib/utf8.t lib/Unicode/Collate/t/index.t | |
46 | * --jhi | |
47 | */ | |
6f207bd3 | 48 | # define ASSERT_UTF8_CACHE(cache) \ |
ab455f60 NC |
49 | STMT_START { if (cache) { assert((cache)[0] <= (cache)[1]); \ |
50 | assert((cache)[2] <= (cache)[3]); \ | |
51 | assert((cache)[3] <= (cache)[1]);} \ | |
52 | } STMT_END | |
e23c8137 | 53 | #else |
6f207bd3 | 54 | # define ASSERT_UTF8_CACHE(cache) NOOP |
e23c8137 JH |
55 | #endif |
56 | ||
f8c7b90f | 57 | #ifdef PERL_OLD_COPY_ON_WRITE |
765f542d | 58 | #define SV_COW_NEXT_SV(sv) INT2PTR(SV *,SvUVX(sv)) |
607fa7f2 | 59 | #define SV_COW_NEXT_SV_SET(current,next) SvUV_set(current, PTR2UV(next)) |
b5ccf5f2 | 60 | /* This is a pessimistic view. Scalar must be purely a read-write PV to copy- |
765f542d | 61 | on-write. */ |
765f542d | 62 | #endif |
645c22ef DM |
63 | |
64 | /* ============================================================================ | |
65 | ||
66 | =head1 Allocation and deallocation of SVs. | |
67 | ||
d2a0f284 JC |
68 | An SV (or AV, HV, etc.) is allocated in two parts: the head (struct |
69 | sv, av, hv...) contains type and reference count information, and for | |
70 | many types, a pointer to the body (struct xrv, xpv, xpviv...), which | |
71 | contains fields specific to each type. Some types store all they need | |
72 | in the head, so don't have a body. | |
73 | ||
74 | In all but the most memory-paranoid configuations (ex: PURIFY), heads | |
75 | and bodies are allocated out of arenas, which by default are | |
76 | approximately 4K chunks of memory parcelled up into N heads or bodies. | |
93e68bfb JC |
77 | Sv-bodies are allocated by their sv-type, guaranteeing size |
78 | consistency needed to allocate safely from arrays. | |
79 | ||
d2a0f284 JC |
80 | For SV-heads, the first slot in each arena is reserved, and holds a |
81 | link to the next arena, some flags, and a note of the number of slots. | |
82 | Snaked through each arena chain is a linked list of free items; when | |
83 | this becomes empty, an extra arena is allocated and divided up into N | |
84 | items which are threaded into the free list. | |
85 | ||
86 | SV-bodies are similar, but they use arena-sets by default, which | |
87 | separate the link and info from the arena itself, and reclaim the 1st | |
88 | slot in the arena. SV-bodies are further described later. | |
645c22ef DM |
89 | |
90 | The following global variables are associated with arenas: | |
91 | ||
92 | PL_sv_arenaroot pointer to list of SV arenas | |
93 | PL_sv_root pointer to list of free SV structures | |
94 | ||
d2a0f284 JC |
95 | PL_body_arenas head of linked-list of body arenas |
96 | PL_body_roots[] array of pointers to list of free bodies of svtype | |
97 | arrays are indexed by the svtype needed | |
93e68bfb | 98 | |
d2a0f284 JC |
99 | A few special SV heads are not allocated from an arena, but are |
100 | instead directly created in the interpreter structure, eg PL_sv_undef. | |
93e68bfb JC |
101 | The size of arenas can be changed from the default by setting |
102 | PERL_ARENA_SIZE appropriately at compile time. | |
645c22ef DM |
103 | |
104 | The SV arena serves the secondary purpose of allowing still-live SVs | |
105 | to be located and destroyed during final cleanup. | |
106 | ||
107 | At the lowest level, the macros new_SV() and del_SV() grab and free | |
108 | an SV head. (If debugging with -DD, del_SV() calls the function S_del_sv() | |
109 | to return the SV to the free list with error checking.) new_SV() calls | |
110 | more_sv() / sv_add_arena() to add an extra arena if the free list is empty. | |
111 | SVs in the free list have their SvTYPE field set to all ones. | |
112 | ||
ff276b08 | 113 | At the time of very final cleanup, sv_free_arenas() is called from |
645c22ef | 114 | perl_destruct() to physically free all the arenas allocated since the |
6a93a7e5 | 115 | start of the interpreter. |
645c22ef | 116 | |
645c22ef DM |
117 | The function visit() scans the SV arenas list, and calls a specified |
118 | function for each SV it finds which is still live - ie which has an SvTYPE | |
119 | other than all 1's, and a non-zero SvREFCNT. visit() is used by the | |
120 | following functions (specified as [function that calls visit()] / [function | |
121 | called by visit() for each SV]): | |
122 | ||
123 | sv_report_used() / do_report_used() | |
f2524eef | 124 | dump all remaining SVs (debugging aid) |
645c22ef DM |
125 | |
126 | sv_clean_objs() / do_clean_objs(),do_clean_named_objs() | |
127 | Attempt to free all objects pointed to by RVs, | |
128 | and, unless DISABLE_DESTRUCTOR_KLUDGE is defined, | |
129 | try to do the same for all objects indirectly | |
130 | referenced by typeglobs too. Called once from | |
131 | perl_destruct(), prior to calling sv_clean_all() | |
132 | below. | |
133 | ||
134 | sv_clean_all() / do_clean_all() | |
135 | SvREFCNT_dec(sv) each remaining SV, possibly | |
136 | triggering an sv_free(). It also sets the | |
137 | SVf_BREAK flag on the SV to indicate that the | |
138 | refcnt has been artificially lowered, and thus | |
139 | stopping sv_free() from giving spurious warnings | |
140 | about SVs which unexpectedly have a refcnt | |
141 | of zero. called repeatedly from perl_destruct() | |
142 | until there are no SVs left. | |
143 | ||
93e68bfb | 144 | =head2 Arena allocator API Summary |
645c22ef DM |
145 | |
146 | Private API to rest of sv.c | |
147 | ||
148 | new_SV(), del_SV(), | |
149 | ||
150 | new_XIV(), del_XIV(), | |
151 | new_XNV(), del_XNV(), | |
152 | etc | |
153 | ||
154 | Public API: | |
155 | ||
8cf8f3d1 | 156 | sv_report_used(), sv_clean_objs(), sv_clean_all(), sv_free_arenas() |
645c22ef | 157 | |
645c22ef DM |
158 | =cut |
159 | ||
3e8320cc | 160 | * ========================================================================= */ |
645c22ef | 161 | |
4561caa4 CS |
162 | /* |
163 | * "A time to plant, and a time to uproot what was planted..." | |
164 | */ | |
165 | ||
77354fb4 | 166 | void |
de37a194 | 167 | Perl_offer_nice_chunk(pTHX_ void *const chunk, const U32 chunk_size) |
77354fb4 | 168 | { |
97aff369 | 169 | dVAR; |
77354fb4 NC |
170 | void *new_chunk; |
171 | U32 new_chunk_size; | |
7918f24d NC |
172 | |
173 | PERL_ARGS_ASSERT_OFFER_NICE_CHUNK; | |
174 | ||
77354fb4 NC |
175 | new_chunk = (void *)(chunk); |
176 | new_chunk_size = (chunk_size); | |
177 | if (new_chunk_size > PL_nice_chunk_size) { | |
178 | Safefree(PL_nice_chunk); | |
179 | PL_nice_chunk = (char *) new_chunk; | |
180 | PL_nice_chunk_size = new_chunk_size; | |
181 | } else { | |
182 | Safefree(chunk); | |
183 | } | |
77354fb4 | 184 | } |
cac9b346 | 185 | |
d7a2c63c MHM |
186 | #ifdef PERL_MEM_LOG |
187 | # define MEM_LOG_NEW_SV(sv, file, line, func) \ | |
188 | Perl_mem_log_new_sv(sv, file, line, func) | |
189 | # define MEM_LOG_DEL_SV(sv, file, line, func) \ | |
190 | Perl_mem_log_del_sv(sv, file, line, func) | |
191 | #else | |
192 | # define MEM_LOG_NEW_SV(sv, file, line, func) NOOP | |
193 | # define MEM_LOG_DEL_SV(sv, file, line, func) NOOP | |
194 | #endif | |
195 | ||
fd0854ff | 196 | #ifdef DEBUG_LEAKING_SCALARS |
22162ca8 | 197 | # define FREE_SV_DEBUG_FILE(sv) Safefree((sv)->sv_debug_file) |
d7a2c63c MHM |
198 | # define DEBUG_SV_SERIAL(sv) \ |
199 | DEBUG_m(PerlIO_printf(Perl_debug_log, "0x%"UVxf": (%05ld) del_SV\n", \ | |
200 | PTR2UV(sv), (long)(sv)->sv_debug_serial)) | |
fd0854ff DM |
201 | #else |
202 | # define FREE_SV_DEBUG_FILE(sv) | |
d7a2c63c | 203 | # define DEBUG_SV_SERIAL(sv) NOOP |
fd0854ff DM |
204 | #endif |
205 | ||
48614a46 NC |
206 | #ifdef PERL_POISON |
207 | # define SvARENA_CHAIN(sv) ((sv)->sv_u.svu_rv) | |
daba3364 | 208 | # define SvARENA_CHAIN_SET(sv,val) (sv)->sv_u.svu_rv = MUTABLE_SV((val)) |
48614a46 NC |
209 | /* Whilst I'd love to do this, it seems that things like to check on |
210 | unreferenced scalars | |
7e337ee0 | 211 | # define POSION_SV_HEAD(sv) PoisonNew(sv, 1, struct STRUCT_SV) |
48614a46 | 212 | */ |
7e337ee0 JH |
213 | # define POSION_SV_HEAD(sv) PoisonNew(&SvANY(sv), 1, void *), \ |
214 | PoisonNew(&SvREFCNT(sv), 1, U32) | |
48614a46 NC |
215 | #else |
216 | # define SvARENA_CHAIN(sv) SvANY(sv) | |
3eef1deb | 217 | # define SvARENA_CHAIN_SET(sv,val) SvANY(sv) = (void *)(val) |
48614a46 NC |
218 | # define POSION_SV_HEAD(sv) |
219 | #endif | |
220 | ||
990198f0 DM |
221 | /* Mark an SV head as unused, and add to free list. |
222 | * | |
223 | * If SVf_BREAK is set, skip adding it to the free list, as this SV had | |
224 | * its refcount artificially decremented during global destruction, so | |
225 | * there may be dangling pointers to it. The last thing we want in that | |
226 | * case is for it to be reused. */ | |
227 | ||
053fc874 GS |
228 | #define plant_SV(p) \ |
229 | STMT_START { \ | |
990198f0 | 230 | const U32 old_flags = SvFLAGS(p); \ |
d7a2c63c MHM |
231 | MEM_LOG_DEL_SV(p, __FILE__, __LINE__, FUNCTION__); \ |
232 | DEBUG_SV_SERIAL(p); \ | |
fd0854ff | 233 | FREE_SV_DEBUG_FILE(p); \ |
48614a46 | 234 | POSION_SV_HEAD(p); \ |
053fc874 | 235 | SvFLAGS(p) = SVTYPEMASK; \ |
990198f0 | 236 | if (!(old_flags & SVf_BREAK)) { \ |
3eef1deb | 237 | SvARENA_CHAIN_SET(p, PL_sv_root); \ |
990198f0 DM |
238 | PL_sv_root = (p); \ |
239 | } \ | |
053fc874 GS |
240 | --PL_sv_count; \ |
241 | } STMT_END | |
a0d0e21e | 242 | |
053fc874 GS |
243 | #define uproot_SV(p) \ |
244 | STMT_START { \ | |
245 | (p) = PL_sv_root; \ | |
daba3364 | 246 | PL_sv_root = MUTABLE_SV(SvARENA_CHAIN(p)); \ |
053fc874 GS |
247 | ++PL_sv_count; \ |
248 | } STMT_END | |
249 | ||
645c22ef | 250 | |
cac9b346 NC |
251 | /* make some more SVs by adding another arena */ |
252 | ||
cac9b346 NC |
253 | STATIC SV* |
254 | S_more_sv(pTHX) | |
255 | { | |
97aff369 | 256 | dVAR; |
cac9b346 NC |
257 | SV* sv; |
258 | ||
259 | if (PL_nice_chunk) { | |
260 | sv_add_arena(PL_nice_chunk, PL_nice_chunk_size, 0); | |
bd61b366 | 261 | PL_nice_chunk = NULL; |
cac9b346 NC |
262 | PL_nice_chunk_size = 0; |
263 | } | |
264 | else { | |
265 | char *chunk; /* must use New here to match call to */ | |
d2a0f284 | 266 | Newx(chunk,PERL_ARENA_SIZE,char); /* Safefree() in sv_free_arenas() */ |
2e7ed132 | 267 | sv_add_arena(chunk, PERL_ARENA_SIZE, 0); |
cac9b346 NC |
268 | } |
269 | uproot_SV(sv); | |
270 | return sv; | |
271 | } | |
272 | ||
645c22ef DM |
273 | /* new_SV(): return a new, empty SV head */ |
274 | ||
eba0f806 DM |
275 | #ifdef DEBUG_LEAKING_SCALARS |
276 | /* provide a real function for a debugger to play with */ | |
277 | STATIC SV* | |
d7a2c63c | 278 | S_new_SV(pTHX_ const char *file, int line, const char *func) |
eba0f806 DM |
279 | { |
280 | SV* sv; | |
281 | ||
eba0f806 DM |
282 | if (PL_sv_root) |
283 | uproot_SV(sv); | |
284 | else | |
cac9b346 | 285 | sv = S_more_sv(aTHX); |
eba0f806 DM |
286 | SvANY(sv) = 0; |
287 | SvREFCNT(sv) = 1; | |
288 | SvFLAGS(sv) = 0; | |
fd0854ff | 289 | sv->sv_debug_optype = PL_op ? PL_op->op_type : 0; |
e385c3bf DM |
290 | sv->sv_debug_line = (U16) (PL_parser && PL_parser->copline != NOLINE |
291 | ? PL_parser->copline | |
292 | : PL_curcop | |
f24aceb1 DM |
293 | ? CopLINE(PL_curcop) |
294 | : 0 | |
e385c3bf | 295 | ); |
fd0854ff DM |
296 | sv->sv_debug_inpad = 0; |
297 | sv->sv_debug_cloned = 0; | |
fd0854ff | 298 | sv->sv_debug_file = PL_curcop ? savepv(CopFILE(PL_curcop)): NULL; |
d7a2c63c MHM |
299 | |
300 | sv->sv_debug_serial = PL_sv_serial++; | |
301 | ||
302 | MEM_LOG_NEW_SV(sv, file, line, func); | |
303 | DEBUG_m(PerlIO_printf(Perl_debug_log, "0x%"UVxf": (%05ld) new_SV (from %s:%d [%s])\n", | |
304 | PTR2UV(sv), (long)sv->sv_debug_serial, file, line, func)); | |
305 | ||
eba0f806 DM |
306 | return sv; |
307 | } | |
d7a2c63c | 308 | # define new_SV(p) (p)=S_new_SV(aTHX_ __FILE__, __LINE__, FUNCTION__) |
eba0f806 DM |
309 | |
310 | #else | |
311 | # define new_SV(p) \ | |
053fc874 | 312 | STMT_START { \ |
053fc874 GS |
313 | if (PL_sv_root) \ |
314 | uproot_SV(p); \ | |
315 | else \ | |
cac9b346 | 316 | (p) = S_more_sv(aTHX); \ |
053fc874 GS |
317 | SvANY(p) = 0; \ |
318 | SvREFCNT(p) = 1; \ | |
319 | SvFLAGS(p) = 0; \ | |
d7a2c63c | 320 | MEM_LOG_NEW_SV(p, __FILE__, __LINE__, FUNCTION__); \ |
053fc874 | 321 | } STMT_END |
eba0f806 | 322 | #endif |
463ee0b2 | 323 | |
645c22ef DM |
324 | |
325 | /* del_SV(): return an empty SV head to the free list */ | |
326 | ||
a0d0e21e | 327 | #ifdef DEBUGGING |
4561caa4 | 328 | |
053fc874 GS |
329 | #define del_SV(p) \ |
330 | STMT_START { \ | |
aea4f609 | 331 | if (DEBUG_D_TEST) \ |
053fc874 GS |
332 | del_sv(p); \ |
333 | else \ | |
334 | plant_SV(p); \ | |
053fc874 | 335 | } STMT_END |
a0d0e21e | 336 | |
76e3520e | 337 | STATIC void |
cea2e8a9 | 338 | S_del_sv(pTHX_ SV *p) |
463ee0b2 | 339 | { |
97aff369 | 340 | dVAR; |
7918f24d NC |
341 | |
342 | PERL_ARGS_ASSERT_DEL_SV; | |
343 | ||
aea4f609 | 344 | if (DEBUG_D_TEST) { |
4633a7c4 | 345 | SV* sva; |
a3b680e6 | 346 | bool ok = 0; |
daba3364 | 347 | for (sva = PL_sv_arenaroot; sva; sva = MUTABLE_SV(SvANY(sva))) { |
53c1dcc0 AL |
348 | const SV * const sv = sva + 1; |
349 | const SV * const svend = &sva[SvREFCNT(sva)]; | |
c0ff570e | 350 | if (p >= sv && p < svend) { |
a0d0e21e | 351 | ok = 1; |
c0ff570e NC |
352 | break; |
353 | } | |
a0d0e21e LW |
354 | } |
355 | if (!ok) { | |
9b387841 NC |
356 | Perl_ck_warner_d(aTHX_ packWARN(WARN_INTERNAL), |
357 | "Attempt to free non-arena SV: 0x%"UVxf | |
358 | pTHX__FORMAT, PTR2UV(p) pTHX__VALUE); | |
a0d0e21e LW |
359 | return; |
360 | } | |
361 | } | |
4561caa4 | 362 | plant_SV(p); |
463ee0b2 | 363 | } |
a0d0e21e | 364 | |
4561caa4 CS |
365 | #else /* ! DEBUGGING */ |
366 | ||
367 | #define del_SV(p) plant_SV(p) | |
368 | ||
369 | #endif /* DEBUGGING */ | |
463ee0b2 | 370 | |
645c22ef DM |
371 | |
372 | /* | |
ccfc67b7 JH |
373 | =head1 SV Manipulation Functions |
374 | ||
645c22ef DM |
375 | =for apidoc sv_add_arena |
376 | ||
377 | Given a chunk of memory, link it to the head of the list of arenas, | |
378 | and split it into a list of free SVs. | |
379 | ||
380 | =cut | |
381 | */ | |
382 | ||
d2bd4e7f NC |
383 | static void |
384 | S_sv_add_arena(pTHX_ char *const ptr, const U32 size, const U32 flags) | |
463ee0b2 | 385 | { |
97aff369 | 386 | dVAR; |
daba3364 | 387 | SV *const sva = MUTABLE_SV(ptr); |
463ee0b2 LW |
388 | register SV* sv; |
389 | register SV* svend; | |
4633a7c4 | 390 | |
7918f24d NC |
391 | PERL_ARGS_ASSERT_SV_ADD_ARENA; |
392 | ||
4633a7c4 | 393 | /* The first SV in an arena isn't an SV. */ |
3280af22 | 394 | SvANY(sva) = (void *) PL_sv_arenaroot; /* ptr to next arena */ |
4633a7c4 LW |
395 | SvREFCNT(sva) = size / sizeof(SV); /* number of SV slots */ |
396 | SvFLAGS(sva) = flags; /* FAKE if not to be freed */ | |
397 | ||
3280af22 NIS |
398 | PL_sv_arenaroot = sva; |
399 | PL_sv_root = sva + 1; | |
4633a7c4 LW |
400 | |
401 | svend = &sva[SvREFCNT(sva) - 1]; | |
402 | sv = sva + 1; | |
463ee0b2 | 403 | while (sv < svend) { |
3eef1deb | 404 | SvARENA_CHAIN_SET(sv, (sv + 1)); |
03e36789 | 405 | #ifdef DEBUGGING |
978b032e | 406 | SvREFCNT(sv) = 0; |
03e36789 | 407 | #endif |
4b69cbe3 | 408 | /* Must always set typemask because it's always checked in on cleanup |
03e36789 | 409 | when the arenas are walked looking for objects. */ |
8990e307 | 410 | SvFLAGS(sv) = SVTYPEMASK; |
463ee0b2 LW |
411 | sv++; |
412 | } | |
3eef1deb | 413 | SvARENA_CHAIN_SET(sv, 0); |
03e36789 NC |
414 | #ifdef DEBUGGING |
415 | SvREFCNT(sv) = 0; | |
416 | #endif | |
4633a7c4 LW |
417 | SvFLAGS(sv) = SVTYPEMASK; |
418 | } | |
419 | ||
055972dc DM |
420 | /* visit(): call the named function for each non-free SV in the arenas |
421 | * whose flags field matches the flags/mask args. */ | |
645c22ef | 422 | |
5226ed68 | 423 | STATIC I32 |
de37a194 | 424 | S_visit(pTHX_ SVFUNC_t f, const U32 flags, const U32 mask) |
8990e307 | 425 | { |
97aff369 | 426 | dVAR; |
4633a7c4 | 427 | SV* sva; |
5226ed68 | 428 | I32 visited = 0; |
8990e307 | 429 | |
7918f24d NC |
430 | PERL_ARGS_ASSERT_VISIT; |
431 | ||
daba3364 | 432 | for (sva = PL_sv_arenaroot; sva; sva = MUTABLE_SV(SvANY(sva))) { |
53c1dcc0 | 433 | register const SV * const svend = &sva[SvREFCNT(sva)]; |
a3b680e6 | 434 | register SV* sv; |
4561caa4 | 435 | for (sv = sva + 1; sv < svend; ++sv) { |
055972dc DM |
436 | if (SvTYPE(sv) != SVTYPEMASK |
437 | && (sv->sv_flags & mask) == flags | |
438 | && SvREFCNT(sv)) | |
439 | { | |
acfe0abc | 440 | (FCALL)(aTHX_ sv); |
5226ed68 JH |
441 | ++visited; |
442 | } | |
8990e307 LW |
443 | } |
444 | } | |
5226ed68 | 445 | return visited; |
8990e307 LW |
446 | } |
447 | ||
758a08c3 JH |
448 | #ifdef DEBUGGING |
449 | ||
645c22ef DM |
450 | /* called by sv_report_used() for each live SV */ |
451 | ||
452 | static void | |
5fa45a31 | 453 | do_report_used(pTHX_ SV *const sv) |
645c22ef DM |
454 | { |
455 | if (SvTYPE(sv) != SVTYPEMASK) { | |
456 | PerlIO_printf(Perl_debug_log, "****\n"); | |
457 | sv_dump(sv); | |
458 | } | |
459 | } | |
758a08c3 | 460 | #endif |
645c22ef DM |
461 | |
462 | /* | |
463 | =for apidoc sv_report_used | |
464 | ||
465 | Dump the contents of all SVs not yet freed. (Debugging aid). | |
466 | ||
467 | =cut | |
468 | */ | |
469 | ||
8990e307 | 470 | void |
864dbfa3 | 471 | Perl_sv_report_used(pTHX) |
4561caa4 | 472 | { |
ff270d3a | 473 | #ifdef DEBUGGING |
055972dc | 474 | visit(do_report_used, 0, 0); |
96a5add6 AL |
475 | #else |
476 | PERL_UNUSED_CONTEXT; | |
ff270d3a | 477 | #endif |
4561caa4 CS |
478 | } |
479 | ||
645c22ef DM |
480 | /* called by sv_clean_objs() for each live SV */ |
481 | ||
482 | static void | |
de37a194 | 483 | do_clean_objs(pTHX_ SV *const ref) |
645c22ef | 484 | { |
97aff369 | 485 | dVAR; |
ea724faa NC |
486 | assert (SvROK(ref)); |
487 | { | |
823a54a3 AL |
488 | SV * const target = SvRV(ref); |
489 | if (SvOBJECT(target)) { | |
490 | DEBUG_D((PerlIO_printf(Perl_debug_log, "Cleaning object ref:\n "), sv_dump(ref))); | |
491 | if (SvWEAKREF(ref)) { | |
492 | sv_del_backref(target, ref); | |
493 | SvWEAKREF_off(ref); | |
494 | SvRV_set(ref, NULL); | |
495 | } else { | |
496 | SvROK_off(ref); | |
497 | SvRV_set(ref, NULL); | |
498 | SvREFCNT_dec(target); | |
499 | } | |
645c22ef DM |
500 | } |
501 | } | |
502 | ||
503 | /* XXX Might want to check arrays, etc. */ | |
504 | } | |
505 | ||
506 | /* called by sv_clean_objs() for each live SV */ | |
507 | ||
508 | #ifndef DISABLE_DESTRUCTOR_KLUDGE | |
509 | static void | |
f30de749 | 510 | do_clean_named_objs(pTHX_ SV *const sv) |
645c22ef | 511 | { |
97aff369 | 512 | dVAR; |
ea724faa | 513 | assert(SvTYPE(sv) == SVt_PVGV); |
d011219a NC |
514 | assert(isGV_with_GP(sv)); |
515 | if (GvGP(sv)) { | |
c69033f2 NC |
516 | if (( |
517 | #ifdef PERL_DONT_CREATE_GVSV | |
518 | GvSV(sv) && | |
519 | #endif | |
520 | SvOBJECT(GvSV(sv))) || | |
645c22ef DM |
521 | (GvAV(sv) && SvOBJECT(GvAV(sv))) || |
522 | (GvHV(sv) && SvOBJECT(GvHV(sv))) || | |
9c12f1e5 RGS |
523 | /* In certain rare cases GvIOp(sv) can be NULL, which would make SvOBJECT(GvIO(sv)) dereference NULL. */ |
524 | (GvIO(sv) ? (SvFLAGS(GvIOp(sv)) & SVs_OBJECT) : 0) || | |
645c22ef DM |
525 | (GvCV(sv) && SvOBJECT(GvCV(sv))) ) |
526 | { | |
527 | DEBUG_D((PerlIO_printf(Perl_debug_log, "Cleaning named glob object:\n "), sv_dump(sv))); | |
ec5f3c78 | 528 | SvFLAGS(sv) |= SVf_BREAK; |
645c22ef DM |
529 | SvREFCNT_dec(sv); |
530 | } | |
531 | } | |
532 | } | |
533 | #endif | |
534 | ||
535 | /* | |
536 | =for apidoc sv_clean_objs | |
537 | ||
538 | Attempt to destroy all objects not yet freed | |
539 | ||
540 | =cut | |
541 | */ | |
542 | ||
4561caa4 | 543 | void |
864dbfa3 | 544 | Perl_sv_clean_objs(pTHX) |
4561caa4 | 545 | { |
97aff369 | 546 | dVAR; |
3280af22 | 547 | PL_in_clean_objs = TRUE; |
055972dc | 548 | visit(do_clean_objs, SVf_ROK, SVf_ROK); |
4561caa4 | 549 | #ifndef DISABLE_DESTRUCTOR_KLUDGE |
2d0f3c12 | 550 | /* some barnacles may yet remain, clinging to typeglobs */ |
d011219a | 551 | visit(do_clean_named_objs, SVt_PVGV|SVpgv_GP, SVTYPEMASK|SVp_POK|SVpgv_GP); |
4561caa4 | 552 | #endif |
3280af22 | 553 | PL_in_clean_objs = FALSE; |
4561caa4 CS |
554 | } |
555 | ||
645c22ef DM |
556 | /* called by sv_clean_all() for each live SV */ |
557 | ||
558 | static void | |
de37a194 | 559 | do_clean_all(pTHX_ SV *const sv) |
645c22ef | 560 | { |
97aff369 | 561 | dVAR; |
daba3364 | 562 | if (sv == (const SV *) PL_fdpid || sv == (const SV *)PL_strtab) { |
cddfcddc | 563 | /* don't clean pid table and strtab */ |
d17ea597 | 564 | return; |
cddfcddc | 565 | } |
645c22ef DM |
566 | DEBUG_D((PerlIO_printf(Perl_debug_log, "Cleaning loops: SV at 0x%"UVxf"\n", PTR2UV(sv)) )); |
567 | SvFLAGS(sv) |= SVf_BREAK; | |
568 | SvREFCNT_dec(sv); | |
569 | } | |
570 | ||
571 | /* | |
572 | =for apidoc sv_clean_all | |
573 | ||
574 | Decrement the refcnt of each remaining SV, possibly triggering a | |
575 | cleanup. This function may have to be called multiple times to free | |
ff276b08 | 576 | SVs which are in complex self-referential hierarchies. |
645c22ef DM |
577 | |
578 | =cut | |
579 | */ | |
580 | ||
5226ed68 | 581 | I32 |
864dbfa3 | 582 | Perl_sv_clean_all(pTHX) |
8990e307 | 583 | { |
97aff369 | 584 | dVAR; |
5226ed68 | 585 | I32 cleaned; |
3280af22 | 586 | PL_in_clean_all = TRUE; |
055972dc | 587 | cleaned = visit(do_clean_all, 0,0); |
3280af22 | 588 | PL_in_clean_all = FALSE; |
5226ed68 | 589 | return cleaned; |
8990e307 | 590 | } |
463ee0b2 | 591 | |
5e258f8c JC |
592 | /* |
593 | ARENASETS: a meta-arena implementation which separates arena-info | |
594 | into struct arena_set, which contains an array of struct | |
595 | arena_descs, each holding info for a single arena. By separating | |
596 | the meta-info from the arena, we recover the 1st slot, formerly | |
597 | borrowed for list management. The arena_set is about the size of an | |
39244528 | 598 | arena, avoiding the needless malloc overhead of a naive linked-list. |
5e258f8c JC |
599 | |
600 | The cost is 1 arena-set malloc per ~320 arena-mallocs, + the unused | |
601 | memory in the last arena-set (1/2 on average). In trade, we get | |
602 | back the 1st slot in each arena (ie 1.7% of a CV-arena, less for | |
d2a0f284 | 603 | smaller types). The recovery of the wasted space allows use of |
e15dad31 JC |
604 | small arenas for large, rare body types, by changing array* fields |
605 | in body_details_by_type[] below. | |
5e258f8c | 606 | */ |
5e258f8c | 607 | struct arena_desc { |
398c677b NC |
608 | char *arena; /* the raw storage, allocated aligned */ |
609 | size_t size; /* its size ~4k typ */ | |
e5973ed5 | 610 | svtype utype; /* bodytype stored in arena */ |
5e258f8c JC |
611 | }; |
612 | ||
e6148039 NC |
613 | struct arena_set; |
614 | ||
615 | /* Get the maximum number of elements in set[] such that struct arena_set | |
e15dad31 | 616 | will fit within PERL_ARENA_SIZE, which is probably just under 4K, and |
e6148039 NC |
617 | therefore likely to be 1 aligned memory page. */ |
618 | ||
619 | #define ARENAS_PER_SET ((PERL_ARENA_SIZE - sizeof(struct arena_set*) \ | |
620 | - 2 * sizeof(int)) / sizeof (struct arena_desc)) | |
5e258f8c JC |
621 | |
622 | struct arena_set { | |
623 | struct arena_set* next; | |
0a848332 NC |
624 | unsigned int set_size; /* ie ARENAS_PER_SET */ |
625 | unsigned int curr; /* index of next available arena-desc */ | |
5e258f8c JC |
626 | struct arena_desc set[ARENAS_PER_SET]; |
627 | }; | |
628 | ||
645c22ef DM |
629 | /* |
630 | =for apidoc sv_free_arenas | |
631 | ||
632 | Deallocate the memory used by all arenas. Note that all the individual SV | |
633 | heads and bodies within the arenas must already have been freed. | |
634 | ||
635 | =cut | |
636 | */ | |
4633a7c4 | 637 | void |
864dbfa3 | 638 | Perl_sv_free_arenas(pTHX) |
4633a7c4 | 639 | { |
97aff369 | 640 | dVAR; |
4633a7c4 LW |
641 | SV* sva; |
642 | SV* svanext; | |
0a848332 | 643 | unsigned int i; |
4633a7c4 LW |
644 | |
645 | /* Free arenas here, but be careful about fake ones. (We assume | |
646 | contiguity of the fake ones with the corresponding real ones.) */ | |
647 | ||
3280af22 | 648 | for (sva = PL_sv_arenaroot; sva; sva = svanext) { |
daba3364 | 649 | svanext = MUTABLE_SV(SvANY(sva)); |
4633a7c4 | 650 | while (svanext && SvFAKE(svanext)) |
daba3364 | 651 | svanext = MUTABLE_SV(SvANY(svanext)); |
4633a7c4 LW |
652 | |
653 | if (!SvFAKE(sva)) | |
1df70142 | 654 | Safefree(sva); |
4633a7c4 | 655 | } |
93e68bfb | 656 | |
5e258f8c | 657 | { |
0a848332 NC |
658 | struct arena_set *aroot = (struct arena_set*) PL_body_arenas; |
659 | ||
660 | while (aroot) { | |
661 | struct arena_set *current = aroot; | |
662 | i = aroot->curr; | |
663 | while (i--) { | |
5e258f8c JC |
664 | assert(aroot->set[i].arena); |
665 | Safefree(aroot->set[i].arena); | |
666 | } | |
0a848332 NC |
667 | aroot = aroot->next; |
668 | Safefree(current); | |
5e258f8c JC |
669 | } |
670 | } | |
dc8220bf | 671 | PL_body_arenas = 0; |
fdda85ca | 672 | |
0a848332 NC |
673 | i = PERL_ARENA_ROOTS_SIZE; |
674 | while (i--) | |
93e68bfb | 675 | PL_body_roots[i] = 0; |
93e68bfb | 676 | |
43c5f42d | 677 | Safefree(PL_nice_chunk); |
bd61b366 | 678 | PL_nice_chunk = NULL; |
3280af22 NIS |
679 | PL_nice_chunk_size = 0; |
680 | PL_sv_arenaroot = 0; | |
681 | PL_sv_root = 0; | |
4633a7c4 LW |
682 | } |
683 | ||
bd81e77b NC |
684 | /* |
685 | Here are mid-level routines that manage the allocation of bodies out | |
686 | of the various arenas. There are 5 kinds of arenas: | |
29489e7c | 687 | |
bd81e77b NC |
688 | 1. SV-head arenas, which are discussed and handled above |
689 | 2. regular body arenas | |
690 | 3. arenas for reduced-size bodies | |
691 | 4. Hash-Entry arenas | |
29489e7c | 692 | |
bd81e77b NC |
693 | Arena types 2 & 3 are chained by body-type off an array of |
694 | arena-root pointers, which is indexed by svtype. Some of the | |
695 | larger/less used body types are malloced singly, since a large | |
696 | unused block of them is wasteful. Also, several svtypes dont have | |
697 | bodies; the data fits into the sv-head itself. The arena-root | |
698 | pointer thus has a few unused root-pointers (which may be hijacked | |
699 | later for arena types 4,5) | |
29489e7c | 700 | |
bd81e77b NC |
701 | 3 differs from 2 as an optimization; some body types have several |
702 | unused fields in the front of the structure (which are kept in-place | |
703 | for consistency). These bodies can be allocated in smaller chunks, | |
704 | because the leading fields arent accessed. Pointers to such bodies | |
705 | are decremented to point at the unused 'ghost' memory, knowing that | |
706 | the pointers are used with offsets to the real memory. | |
29489e7c | 707 | |
bd81e77b NC |
708 | HE, HEK arenas are managed separately, with separate code, but may |
709 | be merge-able later.. | |
bd81e77b NC |
710 | */ |
711 | ||
bd26d9a3 | 712 | /* get_arena(size): this creates custom-sized arenas |
5e258f8c JC |
713 | TBD: export properly for hv.c: S_more_he(). |
714 | */ | |
715 | void* | |
e5973ed5 | 716 | Perl_get_arena(pTHX_ const size_t arena_size, const svtype bodytype) |
5e258f8c | 717 | { |
7a89be66 | 718 | dVAR; |
5e258f8c | 719 | struct arena_desc* adesc; |
39244528 | 720 | struct arena_set *aroot = (struct arena_set*) PL_body_arenas; |
0a848332 | 721 | unsigned int curr; |
5e258f8c | 722 | |
476a1e16 JC |
723 | /* shouldnt need this |
724 | if (!arena_size) arena_size = PERL_ARENA_SIZE; | |
725 | */ | |
5e258f8c JC |
726 | |
727 | /* may need new arena-set to hold new arena */ | |
39244528 NC |
728 | if (!aroot || aroot->curr >= aroot->set_size) { |
729 | struct arena_set *newroot; | |
5e258f8c JC |
730 | Newxz(newroot, 1, struct arena_set); |
731 | newroot->set_size = ARENAS_PER_SET; | |
39244528 NC |
732 | newroot->next = aroot; |
733 | aroot = newroot; | |
734 | PL_body_arenas = (void *) newroot; | |
52944de8 | 735 | DEBUG_m(PerlIO_printf(Perl_debug_log, "new arenaset %p\n", (void*)aroot)); |
5e258f8c JC |
736 | } |
737 | ||
738 | /* ok, now have arena-set with at least 1 empty/available arena-desc */ | |
39244528 NC |
739 | curr = aroot->curr++; |
740 | adesc = &(aroot->set[curr]); | |
5e258f8c JC |
741 | assert(!adesc->arena); |
742 | ||
89086707 | 743 | Newx(adesc->arena, arena_size, char); |
5e258f8c | 744 | adesc->size = arena_size; |
e5973ed5 | 745 | adesc->utype = bodytype; |
d67b3c53 JH |
746 | DEBUG_m(PerlIO_printf(Perl_debug_log, "arena %d added: %p size %"UVuf"\n", |
747 | curr, (void*)adesc->arena, (UV)arena_size)); | |
5e258f8c JC |
748 | |
749 | return adesc->arena; | |
5e258f8c JC |
750 | } |
751 | ||
53c1dcc0 | 752 | |
bd81e77b | 753 | /* return a thing to the free list */ |
29489e7c | 754 | |
bd81e77b NC |
755 | #define del_body(thing, root) \ |
756 | STMT_START { \ | |
00b6aa41 | 757 | void ** const thing_copy = (void **)thing;\ |
bd81e77b NC |
758 | *thing_copy = *root; \ |
759 | *root = (void*)thing_copy; \ | |
bd81e77b | 760 | } STMT_END |
29489e7c | 761 | |
bd81e77b | 762 | /* |
d2a0f284 JC |
763 | |
764 | =head1 SV-Body Allocation | |
765 | ||
766 | Allocation of SV-bodies is similar to SV-heads, differing as follows; | |
767 | the allocation mechanism is used for many body types, so is somewhat | |
768 | more complicated, it uses arena-sets, and has no need for still-live | |
769 | SV detection. | |
770 | ||
771 | At the outermost level, (new|del)_X*V macros return bodies of the | |
772 | appropriate type. These macros call either (new|del)_body_type or | |
773 | (new|del)_body_allocated macro pairs, depending on specifics of the | |
774 | type. Most body types use the former pair, the latter pair is used to | |
775 | allocate body types with "ghost fields". | |
776 | ||
777 | "ghost fields" are fields that are unused in certain types, and | |
69ba284b | 778 | consequently don't need to actually exist. They are declared because |
d2a0f284 JC |
779 | they're part of a "base type", which allows use of functions as |
780 | methods. The simplest examples are AVs and HVs, 2 aggregate types | |
781 | which don't use the fields which support SCALAR semantics. | |
782 | ||
69ba284b | 783 | For these types, the arenas are carved up into appropriately sized |
d2a0f284 JC |
784 | chunks, we thus avoid wasted memory for those unaccessed members. |
785 | When bodies are allocated, we adjust the pointer back in memory by the | |
69ba284b | 786 | size of the part not allocated, so it's as if we allocated the full |
d2a0f284 JC |
787 | structure. (But things will all go boom if you write to the part that |
788 | is "not there", because you'll be overwriting the last members of the | |
789 | preceding structure in memory.) | |
790 | ||
69ba284b NC |
791 | We calculate the correction using the STRUCT_OFFSET macro on the first |
792 | member present. If the allocated structure is smaller (no initial NV | |
793 | actually allocated) then the net effect is to subtract the size of the NV | |
794 | from the pointer, to return a new pointer as if an initial NV were actually | |
795 | allocated. (We were using structures named *_allocated for this, but | |
796 | this turned out to be a subtle bug, because a structure without an NV | |
797 | could have a lower alignment constraint, but the compiler is allowed to | |
798 | optimised accesses based on the alignment constraint of the actual pointer | |
799 | to the full structure, for example, using a single 64 bit load instruction | |
800 | because it "knows" that two adjacent 32 bit members will be 8-byte aligned.) | |
d2a0f284 JC |
801 | |
802 | This is the same trick as was used for NV and IV bodies. Ironically it | |
803 | doesn't need to be used for NV bodies any more, because NV is now at | |
804 | the start of the structure. IV bodies don't need it either, because | |
805 | they are no longer allocated. | |
806 | ||
807 | In turn, the new_body_* allocators call S_new_body(), which invokes | |
808 | new_body_inline macro, which takes a lock, and takes a body off the | |
809 | linked list at PL_body_roots[sv_type], calling S_more_bodies() if | |
810 | necessary to refresh an empty list. Then the lock is released, and | |
811 | the body is returned. | |
812 | ||
813 | S_more_bodies calls get_arena(), and carves it up into an array of N | |
814 | bodies, which it strings into a linked list. It looks up arena-size | |
815 | and body-size from the body_details table described below, thus | |
816 | supporting the multiple body-types. | |
817 | ||
818 | If PURIFY is defined, or PERL_ARENA_SIZE=0, arenas are not used, and | |
819 | the (new|del)_X*V macros are mapped directly to malloc/free. | |
820 | ||
821 | */ | |
822 | ||
823 | /* | |
824 | ||
825 | For each sv-type, struct body_details bodies_by_type[] carries | |
826 | parameters which control these aspects of SV handling: | |
827 | ||
828 | Arena_size determines whether arenas are used for this body type, and if | |
829 | so, how big they are. PURIFY or PERL_ARENA_SIZE=0 set this field to | |
830 | zero, forcing individual mallocs and frees. | |
831 | ||
832 | Body_size determines how big a body is, and therefore how many fit into | |
833 | each arena. Offset carries the body-pointer adjustment needed for | |
69ba284b | 834 | "ghost fields", and is used in *_allocated macros. |
d2a0f284 JC |
835 | |
836 | But its main purpose is to parameterize info needed in | |
837 | Perl_sv_upgrade(). The info here dramatically simplifies the function | |
69ba284b | 838 | vs the implementation in 5.8.8, making it table-driven. All fields |
d2a0f284 JC |
839 | are used for this, except for arena_size. |
840 | ||
841 | For the sv-types that have no bodies, arenas are not used, so those | |
842 | PL_body_roots[sv_type] are unused, and can be overloaded. In | |
843 | something of a special case, SVt_NULL is borrowed for HE arenas; | |
c6f8b1d0 | 844 | PL_body_roots[HE_SVSLOT=SVt_NULL] is filled by S_more_he, but the |
d2a0f284 | 845 | bodies_by_type[SVt_NULL] slot is not used, as the table is not |
c6f8b1d0 | 846 | available in hv.c. |
d2a0f284 | 847 | |
29489e7c DM |
848 | */ |
849 | ||
bd81e77b | 850 | struct body_details { |
0fb58b32 | 851 | U8 body_size; /* Size to allocate */ |
10666ae3 | 852 | U8 copy; /* Size of structure to copy (may be shorter) */ |
0fb58b32 | 853 | U8 offset; |
10666ae3 NC |
854 | unsigned int type : 4; /* We have space for a sanity check. */ |
855 | unsigned int cant_upgrade : 1; /* Cannot upgrade this type */ | |
856 | unsigned int zero_nv : 1; /* zero the NV when upgrading from this */ | |
857 | unsigned int arena : 1; /* Allocated from an arena */ | |
858 | size_t arena_size; /* Size of arena to allocate */ | |
bd81e77b | 859 | }; |
29489e7c | 860 | |
bd81e77b NC |
861 | #define HADNV FALSE |
862 | #define NONV TRUE | |
29489e7c | 863 | |
d2a0f284 | 864 | |
bd81e77b NC |
865 | #ifdef PURIFY |
866 | /* With -DPURFIY we allocate everything directly, and don't use arenas. | |
867 | This seems a rather elegant way to simplify some of the code below. */ | |
868 | #define HASARENA FALSE | |
869 | #else | |
870 | #define HASARENA TRUE | |
871 | #endif | |
872 | #define NOARENA FALSE | |
29489e7c | 873 | |
d2a0f284 JC |
874 | /* Size the arenas to exactly fit a given number of bodies. A count |
875 | of 0 fits the max number bodies into a PERL_ARENA_SIZE.block, | |
876 | simplifying the default. If count > 0, the arena is sized to fit | |
877 | only that many bodies, allowing arenas to be used for large, rare | |
878 | bodies (XPVFM, XPVIO) without undue waste. The arena size is | |
879 | limited by PERL_ARENA_SIZE, so we can safely oversize the | |
880 | declarations. | |
881 | */ | |
95db5f15 MB |
882 | #define FIT_ARENA0(body_size) \ |
883 | ((size_t)(PERL_ARENA_SIZE / body_size) * body_size) | |
884 | #define FIT_ARENAn(count,body_size) \ | |
885 | ( count * body_size <= PERL_ARENA_SIZE) \ | |
886 | ? count * body_size \ | |
887 | : FIT_ARENA0 (body_size) | |
888 | #define FIT_ARENA(count,body_size) \ | |
889 | count \ | |
890 | ? FIT_ARENAn (count, body_size) \ | |
891 | : FIT_ARENA0 (body_size) | |
d2a0f284 | 892 | |
bd81e77b NC |
893 | /* Calculate the length to copy. Specifically work out the length less any |
894 | final padding the compiler needed to add. See the comment in sv_upgrade | |
895 | for why copying the padding proved to be a bug. */ | |
29489e7c | 896 | |
bd81e77b NC |
897 | #define copy_length(type, last_member) \ |
898 | STRUCT_OFFSET(type, last_member) \ | |
daba3364 | 899 | + sizeof (((type*)SvANY((const SV *)0))->last_member) |
29489e7c | 900 | |
bd81e77b | 901 | static const struct body_details bodies_by_type[] = { |
10666ae3 NC |
902 | { sizeof(HE), 0, 0, SVt_NULL, |
903 | FALSE, NONV, NOARENA, FIT_ARENA(0, sizeof(HE)) }, | |
d2a0f284 | 904 | |
1cb9cd50 | 905 | /* The bind placeholder pretends to be an RV for now. |
c6f8b1d0 | 906 | Also it's marked as "can't upgrade" to stop anyone using it before it's |
1cb9cd50 NC |
907 | implemented. */ |
908 | { 0, 0, 0, SVt_BIND, TRUE, NONV, NOARENA, 0 }, | |
909 | ||
db93c0c4 NC |
910 | /* IVs are in the head, so the allocation size is 0. */ |
911 | { 0, | |
d2a0f284 | 912 | sizeof(IV), /* This is used to copy out the IV body. */ |
10666ae3 | 913 | STRUCT_OFFSET(XPVIV, xiv_iv), SVt_IV, FALSE, NONV, |
db93c0c4 | 914 | NOARENA /* IVS don't need an arena */, 0 |
d2a0f284 JC |
915 | }, |
916 | ||
bd81e77b | 917 | /* 8 bytes on most ILP32 with IEEE doubles */ |
10666ae3 | 918 | { sizeof(NV), sizeof(NV), 0, SVt_NV, FALSE, HADNV, HASARENA, |
d2a0f284 JC |
919 | FIT_ARENA(0, sizeof(NV)) }, |
920 | ||
bd81e77b | 921 | /* 8 bytes on most ILP32 with IEEE doubles */ |
69ba284b NC |
922 | { sizeof(XPV) - STRUCT_OFFSET(XPV, xpv_cur), |
923 | copy_length(XPV, xpv_len) - STRUCT_OFFSET(XPV, xpv_cur), | |
924 | + STRUCT_OFFSET(XPV, xpv_cur), | |
925 | SVt_PV, FALSE, NONV, HASARENA, | |
926 | FIT_ARENA(0, sizeof(XPV) - STRUCT_OFFSET(XPV, xpv_cur)) }, | |
d2a0f284 | 927 | |
bd81e77b | 928 | /* 12 */ |
69ba284b NC |
929 | { sizeof(XPVIV) - STRUCT_OFFSET(XPV, xpv_cur), |
930 | copy_length(XPVIV, xiv_u) - STRUCT_OFFSET(XPV, xpv_cur), | |
931 | + STRUCT_OFFSET(XPVIV, xpv_cur), | |
932 | SVt_PVIV, FALSE, NONV, HASARENA, | |
933 | FIT_ARENA(0, sizeof(XPV) - STRUCT_OFFSET(XPV, xpv_cur)) }, | |
d2a0f284 | 934 | |
bd81e77b | 935 | /* 20 */ |
10666ae3 | 936 | { sizeof(XPVNV), copy_length(XPVNV, xiv_u), 0, SVt_PVNV, FALSE, HADNV, |
d2a0f284 JC |
937 | HASARENA, FIT_ARENA(0, sizeof(XPVNV)) }, |
938 | ||
bd81e77b | 939 | /* 28 */ |
10666ae3 | 940 | { sizeof(XPVMG), copy_length(XPVMG, xmg_stash), 0, SVt_PVMG, FALSE, HADNV, |
d2a0f284 | 941 | HASARENA, FIT_ARENA(0, sizeof(XPVMG)) }, |
4df7f6af | 942 | |
288b8c02 | 943 | /* something big */ |
b6f60916 NC |
944 | { sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur), |
945 | sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur), | |
946 | + STRUCT_OFFSET(regexp, xpv_cur), | |
08e44740 | 947 | SVt_REGEXP, FALSE, NONV, HASARENA, |
b6f60916 | 948 | FIT_ARENA(0, sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur)) |
5c35adbb | 949 | }, |
4df7f6af | 950 | |
bd81e77b | 951 | /* 48 */ |
10666ae3 | 952 | { sizeof(XPVGV), sizeof(XPVGV), 0, SVt_PVGV, TRUE, HADNV, |
d2a0f284 JC |
953 | HASARENA, FIT_ARENA(0, sizeof(XPVGV)) }, |
954 | ||
bd81e77b | 955 | /* 64 */ |
10666ae3 | 956 | { sizeof(XPVLV), sizeof(XPVLV), 0, SVt_PVLV, TRUE, HADNV, |
d2a0f284 JC |
957 | HASARENA, FIT_ARENA(0, sizeof(XPVLV)) }, |
958 | ||
69ba284b NC |
959 | { sizeof(XPVAV) - STRUCT_OFFSET(XPVAV, xav_fill), |
960 | copy_length(XPVAV, xmg_stash) - STRUCT_OFFSET(XPVAV, xav_fill), | |
961 | + STRUCT_OFFSET(XPVAV, xav_fill), | |
962 | SVt_PVAV, TRUE, NONV, HASARENA, | |
963 | FIT_ARENA(0, sizeof(XPVAV) - STRUCT_OFFSET(XPVAV, xav_fill)) }, | |
d2a0f284 | 964 | |
69ba284b NC |
965 | { sizeof(XPVHV) - STRUCT_OFFSET(XPVHV, xhv_fill), |
966 | copy_length(XPVHV, xmg_stash) - STRUCT_OFFSET(XPVHV, xhv_fill), | |
967 | + STRUCT_OFFSET(XPVHV, xhv_fill), | |
968 | SVt_PVHV, TRUE, NONV, HASARENA, | |
969 | FIT_ARENA(0, sizeof(XPVHV) - STRUCT_OFFSET(XPVHV, xhv_fill)) }, | |
d2a0f284 | 970 | |
c84c4652 | 971 | /* 56 */ |
69ba284b NC |
972 | { sizeof(XPVCV) - STRUCT_OFFSET(XPVCV, xpv_cur), |
973 | sizeof(XPVCV) - STRUCT_OFFSET(XPVCV, xpv_cur), | |
974 | + STRUCT_OFFSET(XPVCV, xpv_cur), | |
975 | SVt_PVCV, TRUE, NONV, HASARENA, | |
976 | FIT_ARENA(0, sizeof(XPVCV) - STRUCT_OFFSET(XPVCV, xpv_cur)) }, | |
977 | ||
978 | { sizeof(XPVFM) - STRUCT_OFFSET(XPVFM, xpv_cur), | |
979 | sizeof(XPVFM) - STRUCT_OFFSET(XPVFM, xpv_cur), | |
980 | + STRUCT_OFFSET(XPVFM, xpv_cur), | |
981 | SVt_PVFM, TRUE, NONV, NOARENA, | |
982 | FIT_ARENA(20, sizeof(XPVFM) - STRUCT_OFFSET(XPVFM, xpv_cur)) }, | |
d2a0f284 JC |
983 | |
984 | /* XPVIO is 84 bytes, fits 48x */ | |
b6f60916 NC |
985 | { sizeof(XPVIO) - STRUCT_OFFSET(XPVIO, xpv_cur), |
986 | sizeof(XPVIO) - STRUCT_OFFSET(XPVIO, xpv_cur), | |
987 | + STRUCT_OFFSET(XPVIO, xpv_cur), | |
988 | SVt_PVIO, TRUE, NONV, HASARENA, | |
989 | FIT_ARENA(24, sizeof(XPVIO) - STRUCT_OFFSET(XPVIO, xpv_cur)) }, | |
bd81e77b | 990 | }; |
29489e7c | 991 | |
d2a0f284 JC |
992 | #define new_body_type(sv_type) \ |
993 | (void *)((char *)S_new_body(aTHX_ sv_type)) | |
29489e7c | 994 | |
bd81e77b NC |
995 | #define del_body_type(p, sv_type) \ |
996 | del_body(p, &PL_body_roots[sv_type]) | |
29489e7c | 997 | |
29489e7c | 998 | |
bd81e77b | 999 | #define new_body_allocated(sv_type) \ |
d2a0f284 | 1000 | (void *)((char *)S_new_body(aTHX_ sv_type) \ |
bd81e77b | 1001 | - bodies_by_type[sv_type].offset) |
29489e7c | 1002 | |
bd81e77b NC |
1003 | #define del_body_allocated(p, sv_type) \ |
1004 | del_body(p + bodies_by_type[sv_type].offset, &PL_body_roots[sv_type]) | |
29489e7c | 1005 | |
29489e7c | 1006 | |
bd81e77b NC |
1007 | #define my_safemalloc(s) (void*)safemalloc(s) |
1008 | #define my_safecalloc(s) (void*)safecalloc(s, 1) | |
1009 | #define my_safefree(p) safefree((char*)p) | |
29489e7c | 1010 | |
bd81e77b | 1011 | #ifdef PURIFY |
29489e7c | 1012 | |
bd81e77b NC |
1013 | #define new_XNV() my_safemalloc(sizeof(XPVNV)) |
1014 | #define del_XNV(p) my_safefree(p) | |
29489e7c | 1015 | |
bd81e77b NC |
1016 | #define new_XPVNV() my_safemalloc(sizeof(XPVNV)) |
1017 | #define del_XPVNV(p) my_safefree(p) | |
29489e7c | 1018 | |
bd81e77b NC |
1019 | #define new_XPVAV() my_safemalloc(sizeof(XPVAV)) |
1020 | #define del_XPVAV(p) my_safefree(p) | |
29489e7c | 1021 | |
bd81e77b NC |
1022 | #define new_XPVHV() my_safemalloc(sizeof(XPVHV)) |
1023 | #define del_XPVHV(p) my_safefree(p) | |
29489e7c | 1024 | |
bd81e77b NC |
1025 | #define new_XPVMG() my_safemalloc(sizeof(XPVMG)) |
1026 | #define del_XPVMG(p) my_safefree(p) | |
29489e7c | 1027 | |
bd81e77b NC |
1028 | #define new_XPVGV() my_safemalloc(sizeof(XPVGV)) |
1029 | #define del_XPVGV(p) my_safefree(p) | |
29489e7c | 1030 | |
bd81e77b | 1031 | #else /* !PURIFY */ |
29489e7c | 1032 | |
bd81e77b NC |
1033 | #define new_XNV() new_body_type(SVt_NV) |
1034 | #define del_XNV(p) del_body_type(p, SVt_NV) | |
29489e7c | 1035 | |
bd81e77b NC |
1036 | #define new_XPVNV() new_body_type(SVt_PVNV) |
1037 | #define del_XPVNV(p) del_body_type(p, SVt_PVNV) | |
29489e7c | 1038 | |
bd81e77b NC |
1039 | #define new_XPVAV() new_body_allocated(SVt_PVAV) |
1040 | #define del_XPVAV(p) del_body_allocated(p, SVt_PVAV) | |
645c22ef | 1041 | |
bd81e77b NC |
1042 | #define new_XPVHV() new_body_allocated(SVt_PVHV) |
1043 | #define del_XPVHV(p) del_body_allocated(p, SVt_PVHV) | |
645c22ef | 1044 | |
bd81e77b NC |
1045 | #define new_XPVMG() new_body_type(SVt_PVMG) |
1046 | #define del_XPVMG(p) del_body_type(p, SVt_PVMG) | |
645c22ef | 1047 | |
bd81e77b NC |
1048 | #define new_XPVGV() new_body_type(SVt_PVGV) |
1049 | #define del_XPVGV(p) del_body_type(p, SVt_PVGV) | |
1d7c1841 | 1050 | |
bd81e77b | 1051 | #endif /* PURIFY */ |
93e68bfb | 1052 | |
bd81e77b | 1053 | /* no arena for you! */ |
93e68bfb | 1054 | |
bd81e77b | 1055 | #define new_NOARENA(details) \ |
d2a0f284 | 1056 | my_safemalloc((details)->body_size + (details)->offset) |
bd81e77b | 1057 | #define new_NOARENAZ(details) \ |
d2a0f284 JC |
1058 | my_safecalloc((details)->body_size + (details)->offset) |
1059 | ||
1060 | STATIC void * | |
de37a194 | 1061 | S_more_bodies (pTHX_ const svtype sv_type) |
d2a0f284 JC |
1062 | { |
1063 | dVAR; | |
1064 | void ** const root = &PL_body_roots[sv_type]; | |
96a5add6 | 1065 | const struct body_details * const bdp = &bodies_by_type[sv_type]; |
d2a0f284 JC |
1066 | const size_t body_size = bdp->body_size; |
1067 | char *start; | |
1068 | const char *end; | |
d8fca402 | 1069 | const size_t arena_size = Perl_malloc_good_size(bdp->arena_size); |
0b2d3faa | 1070 | #if defined(DEBUGGING) && !defined(PERL_GLOBAL_STRUCT_PRIVATE) |
23e9d66c NC |
1071 | static bool done_sanity_check; |
1072 | ||
0b2d3faa JH |
1073 | /* PERL_GLOBAL_STRUCT_PRIVATE cannot coexist with global |
1074 | * variables like done_sanity_check. */ | |
10666ae3 | 1075 | if (!done_sanity_check) { |
ea471437 | 1076 | unsigned int i = SVt_LAST; |
10666ae3 NC |
1077 | |
1078 | done_sanity_check = TRUE; | |
1079 | ||
1080 | while (i--) | |
1081 | assert (bodies_by_type[i].type == i); | |
1082 | } | |
1083 | #endif | |
1084 | ||
23e9d66c NC |
1085 | assert(bdp->arena_size); |
1086 | ||
d8fca402 | 1087 | start = (char*) Perl_get_arena(aTHX_ arena_size, sv_type); |
d2a0f284 | 1088 | |
d8fca402 | 1089 | end = start + arena_size - 2 * body_size; |
d2a0f284 | 1090 | |
d2a0f284 | 1091 | /* computed count doesnt reflect the 1st slot reservation */ |
d8fca402 NC |
1092 | #if defined(MYMALLOC) || defined(HAS_MALLOC_GOOD_SIZE) |
1093 | DEBUG_m(PerlIO_printf(Perl_debug_log, | |
1094 | "arena %p end %p arena-size %d (from %d) type %d " | |
1095 | "size %d ct %d\n", | |
1096 | (void*)start, (void*)end, (int)arena_size, | |
1097 | (int)bdp->arena_size, sv_type, (int)body_size, | |
1098 | (int)arena_size / (int)body_size)); | |
1099 | #else | |
d2a0f284 JC |
1100 | DEBUG_m(PerlIO_printf(Perl_debug_log, |
1101 | "arena %p end %p arena-size %d type %d size %d ct %d\n", | |
6c9570dc | 1102 | (void*)start, (void*)end, |
0e84aef4 JH |
1103 | (int)bdp->arena_size, sv_type, (int)body_size, |
1104 | (int)bdp->arena_size / (int)body_size)); | |
d8fca402 | 1105 | #endif |
d2a0f284 JC |
1106 | *root = (void *)start; |
1107 | ||
d8fca402 | 1108 | while (start <= end) { |
d2a0f284 JC |
1109 | char * const next = start + body_size; |
1110 | *(void**) start = (void *)next; | |
1111 | start = next; | |
1112 | } | |
1113 | *(void **)start = 0; | |
1114 | ||
1115 | return *root; | |
1116 | } | |
1117 | ||
1118 | /* grab a new thing from the free list, allocating more if necessary. | |
1119 | The inline version is used for speed in hot routines, and the | |
1120 | function using it serves the rest (unless PURIFY). | |
1121 | */ | |
1122 | #define new_body_inline(xpv, sv_type) \ | |
1123 | STMT_START { \ | |
1124 | void ** const r3wt = &PL_body_roots[sv_type]; \ | |
11b79775 DD |
1125 | xpv = (PTR_TBL_ENT_t*) (*((void **)(r3wt)) \ |
1126 | ? *((void **)(r3wt)) : more_bodies(sv_type)); \ | |
d2a0f284 | 1127 | *(r3wt) = *(void**)(xpv); \ |
d2a0f284 JC |
1128 | } STMT_END |
1129 | ||
1130 | #ifndef PURIFY | |
1131 | ||
1132 | STATIC void * | |
de37a194 | 1133 | S_new_body(pTHX_ const svtype sv_type) |
d2a0f284 JC |
1134 | { |
1135 | dVAR; | |
1136 | void *xpv; | |
1137 | new_body_inline(xpv, sv_type); | |
1138 | return xpv; | |
1139 | } | |
1140 | ||
1141 | #endif | |
93e68bfb | 1142 | |
238b27b3 NC |
1143 | static const struct body_details fake_rv = |
1144 | { 0, 0, 0, SVt_IV, FALSE, NONV, NOARENA, 0 }; | |
1145 | ||
bd81e77b NC |
1146 | /* |
1147 | =for apidoc sv_upgrade | |
93e68bfb | 1148 | |
bd81e77b NC |
1149 | Upgrade an SV to a more complex form. Generally adds a new body type to the |
1150 | SV, then copies across as much information as possible from the old body. | |
1151 | You generally want to use the C<SvUPGRADE> macro wrapper. See also C<svtype>. | |
93e68bfb | 1152 | |
bd81e77b | 1153 | =cut |
93e68bfb | 1154 | */ |
93e68bfb | 1155 | |
bd81e77b | 1156 | void |
aad570aa | 1157 | Perl_sv_upgrade(pTHX_ register SV *const sv, svtype new_type) |
cac9b346 | 1158 | { |
97aff369 | 1159 | dVAR; |
bd81e77b NC |
1160 | void* old_body; |
1161 | void* new_body; | |
42d0e0b7 | 1162 | const svtype old_type = SvTYPE(sv); |
d2a0f284 | 1163 | const struct body_details *new_type_details; |
238b27b3 | 1164 | const struct body_details *old_type_details |
bd81e77b | 1165 | = bodies_by_type + old_type; |
4df7f6af | 1166 | SV *referant = NULL; |
cac9b346 | 1167 | |
7918f24d NC |
1168 | PERL_ARGS_ASSERT_SV_UPGRADE; |
1169 | ||
1776cbe8 NC |
1170 | if (old_type == new_type) |
1171 | return; | |
1172 | ||
1173 | /* This clause was purposefully added ahead of the early return above to | |
1174 | the shared string hackery for (sort {$a <=> $b} keys %hash), with the | |
1175 | inference by Nick I-S that it would fix other troublesome cases. See | |
1176 | changes 7162, 7163 (f130fd4589cf5fbb24149cd4db4137c8326f49c1 and parent) | |
1177 | ||
1178 | Given that shared hash key scalars are no longer PVIV, but PV, there is | |
1179 | no longer need to unshare so as to free up the IVX slot for its proper | |
1180 | purpose. So it's safe to move the early return earlier. */ | |
1181 | ||
bd81e77b NC |
1182 | if (new_type != SVt_PV && SvIsCOW(sv)) { |
1183 | sv_force_normal_flags(sv, 0); | |
1184 | } | |
cac9b346 | 1185 | |
bd81e77b | 1186 | old_body = SvANY(sv); |
de042e1d | 1187 | |
bd81e77b NC |
1188 | /* Copying structures onto other structures that have been neatly zeroed |
1189 | has a subtle gotcha. Consider XPVMG | |
cac9b346 | 1190 | |
bd81e77b NC |
1191 | +------+------+------+------+------+-------+-------+ |
1192 | | NV | CUR | LEN | IV | MAGIC | STASH | | |
1193 | +------+------+------+------+------+-------+-------+ | |
1194 | 0 4 8 12 16 20 24 28 | |
645c22ef | 1195 | |
bd81e77b NC |
1196 | where NVs are aligned to 8 bytes, so that sizeof that structure is |
1197 | actually 32 bytes long, with 4 bytes of padding at the end: | |
08742458 | 1198 | |
bd81e77b NC |
1199 | +------+------+------+------+------+-------+-------+------+ |
1200 | | NV | CUR | LEN | IV | MAGIC | STASH | ??? | | |
1201 | +------+------+------+------+------+-------+-------+------+ | |
1202 | 0 4 8 12 16 20 24 28 32 | |
08742458 | 1203 | |
bd81e77b | 1204 | so what happens if you allocate memory for this structure: |
30f9da9e | 1205 | |
bd81e77b NC |
1206 | +------+------+------+------+------+-------+-------+------+------+... |
1207 | | NV | CUR | LEN | IV | MAGIC | STASH | GP | NAME | | |
1208 | +------+------+------+------+------+-------+-------+------+------+... | |
1209 | 0 4 8 12 16 20 24 28 32 36 | |
bfc44f79 | 1210 | |
bd81e77b NC |
1211 | zero it, then copy sizeof(XPVMG) bytes on top of it? Not quite what you |
1212 | expect, because you copy the area marked ??? onto GP. Now, ??? may have | |
1213 | started out as zero once, but it's quite possible that it isn't. So now, | |
1214 | rather than a nicely zeroed GP, you have it pointing somewhere random. | |
1215 | Bugs ensue. | |
bfc44f79 | 1216 | |
bd81e77b NC |
1217 | (In fact, GP ends up pointing at a previous GP structure, because the |
1218 | principle cause of the padding in XPVMG getting garbage is a copy of | |
6c9e42f7 NC |
1219 | sizeof(XPVMG) bytes from a XPVGV structure in sv_unglob. Right now |
1220 | this happens to be moot because XPVGV has been re-ordered, with GP | |
1221 | no longer after STASH) | |
30f9da9e | 1222 | |
bd81e77b NC |
1223 | So we are careful and work out the size of used parts of all the |
1224 | structures. */ | |
bfc44f79 | 1225 | |
bd81e77b NC |
1226 | switch (old_type) { |
1227 | case SVt_NULL: | |
1228 | break; | |
1229 | case SVt_IV: | |
4df7f6af NC |
1230 | if (SvROK(sv)) { |
1231 | referant = SvRV(sv); | |
238b27b3 NC |
1232 | old_type_details = &fake_rv; |
1233 | if (new_type == SVt_NV) | |
1234 | new_type = SVt_PVNV; | |
4df7f6af NC |
1235 | } else { |
1236 | if (new_type < SVt_PVIV) { | |
1237 | new_type = (new_type == SVt_NV) | |
1238 | ? SVt_PVNV : SVt_PVIV; | |
1239 | } | |
bd81e77b NC |
1240 | } |
1241 | break; | |
1242 | case SVt_NV: | |
1243 | if (new_type < SVt_PVNV) { | |
1244 | new_type = SVt_PVNV; | |
bd81e77b NC |
1245 | } |
1246 | break; | |
bd81e77b NC |
1247 | case SVt_PV: |
1248 | assert(new_type > SVt_PV); | |
1249 | assert(SVt_IV < SVt_PV); | |
1250 | assert(SVt_NV < SVt_PV); | |
1251 | break; | |
1252 | case SVt_PVIV: | |
1253 | break; | |
1254 | case SVt_PVNV: | |
1255 | break; | |
1256 | case SVt_PVMG: | |
1257 | /* Because the XPVMG of PL_mess_sv isn't allocated from the arena, | |
1258 | there's no way that it can be safely upgraded, because perl.c | |
1259 | expects to Safefree(SvANY(PL_mess_sv)) */ | |
1260 | assert(sv != PL_mess_sv); | |
1261 | /* This flag bit is used to mean other things in other scalar types. | |
1262 | Given that it only has meaning inside the pad, it shouldn't be set | |
1263 | on anything that can get upgraded. */ | |
00b1698f | 1264 | assert(!SvPAD_TYPED(sv)); |
bd81e77b NC |
1265 | break; |
1266 | default: | |
1267 | if (old_type_details->cant_upgrade) | |
c81225bc NC |
1268 | Perl_croak(aTHX_ "Can't upgrade %s (%" UVuf ") to %" UVuf, |
1269 | sv_reftype(sv, 0), (UV) old_type, (UV) new_type); | |
bd81e77b | 1270 | } |
3376de98 NC |
1271 | |
1272 | if (old_type > new_type) | |
1273 | Perl_croak(aTHX_ "sv_upgrade from type %d down to type %d", | |
1274 | (int)old_type, (int)new_type); | |
1275 | ||
2fa1109b | 1276 | new_type_details = bodies_by_type + new_type; |
645c22ef | 1277 | |
bd81e77b NC |
1278 | SvFLAGS(sv) &= ~SVTYPEMASK; |
1279 | SvFLAGS(sv) |= new_type; | |
932e9ff9 | 1280 | |
ab4416c0 NC |
1281 | /* This can't happen, as SVt_NULL is <= all values of new_type, so one of |
1282 | the return statements above will have triggered. */ | |
1283 | assert (new_type != SVt_NULL); | |
bd81e77b | 1284 | switch (new_type) { |
bd81e77b NC |
1285 | case SVt_IV: |
1286 | assert(old_type == SVt_NULL); | |
1287 | SvANY(sv) = (XPVIV*)((char*)&(sv->sv_u.svu_iv) - STRUCT_OFFSET(XPVIV, xiv_iv)); | |
1288 | SvIV_set(sv, 0); | |
1289 | return; | |
1290 | case SVt_NV: | |
1291 | assert(old_type == SVt_NULL); | |
1292 | SvANY(sv) = new_XNV(); | |
1293 | SvNV_set(sv, 0); | |
1294 | return; | |
bd81e77b | 1295 | case SVt_PVHV: |
bd81e77b | 1296 | case SVt_PVAV: |
d2a0f284 | 1297 | assert(new_type_details->body_size); |
c1ae03ae NC |
1298 | |
1299 | #ifndef PURIFY | |
1300 | assert(new_type_details->arena); | |
d2a0f284 | 1301 | assert(new_type_details->arena_size); |
c1ae03ae | 1302 | /* This points to the start of the allocated area. */ |
d2a0f284 JC |
1303 | new_body_inline(new_body, new_type); |
1304 | Zero(new_body, new_type_details->body_size, char); | |
c1ae03ae NC |
1305 | new_body = ((char *)new_body) - new_type_details->offset; |
1306 | #else | |
1307 | /* We always allocated the full length item with PURIFY. To do this | |
1308 | we fake things so that arena is false for all 16 types.. */ | |
1309 | new_body = new_NOARENAZ(new_type_details); | |
1310 | #endif | |
1311 | SvANY(sv) = new_body; | |
1312 | if (new_type == SVt_PVAV) { | |
1313 | AvMAX(sv) = -1; | |
1314 | AvFILLp(sv) = -1; | |
1315 | AvREAL_only(sv); | |
64484faa | 1316 | if (old_type_details->body_size) { |
ac572bf4 NC |
1317 | AvALLOC(sv) = 0; |
1318 | } else { | |
1319 | /* It will have been zeroed when the new body was allocated. | |
1320 | Lets not write to it, in case it confuses a write-back | |
1321 | cache. */ | |
1322 | } | |
78ac7dd9 NC |
1323 | } else { |
1324 | assert(!SvOK(sv)); | |
1325 | SvOK_off(sv); | |
1326 | #ifndef NODEFAULT_SHAREKEYS | |
1327 | HvSHAREKEYS_on(sv); /* key-sharing on by default */ | |
1328 | #endif | |
1329 | HvMAX(sv) = 7; /* (start with 8 buckets) */ | |
64484faa | 1330 | if (old_type_details->body_size) { |
78ac7dd9 NC |
1331 | HvFILL(sv) = 0; |
1332 | } else { | |
1333 | /* It will have been zeroed when the new body was allocated. | |
1334 | Lets not write to it, in case it confuses a write-back | |
1335 | cache. */ | |
1336 | } | |
c1ae03ae | 1337 | } |
aeb18a1e | 1338 | |
bd81e77b NC |
1339 | /* SVt_NULL isn't the only thing upgraded to AV or HV. |
1340 | The target created by newSVrv also is, and it can have magic. | |
1341 | However, it never has SvPVX set. | |
1342 | */ | |
4df7f6af NC |
1343 | if (old_type == SVt_IV) { |
1344 | assert(!SvROK(sv)); | |
1345 | } else if (old_type >= SVt_PV) { | |
bd81e77b NC |
1346 | assert(SvPVX_const(sv) == 0); |
1347 | } | |
aeb18a1e | 1348 | |
bd81e77b | 1349 | if (old_type >= SVt_PVMG) { |
e736a858 | 1350 | SvMAGIC_set(sv, ((XPVMG*)old_body)->xmg_u.xmg_magic); |
bd81e77b | 1351 | SvSTASH_set(sv, ((XPVMG*)old_body)->xmg_stash); |
797c7171 NC |
1352 | } else { |
1353 | sv->sv_u.svu_array = NULL; /* or svu_hash */ | |
bd81e77b NC |
1354 | } |
1355 | break; | |
93e68bfb | 1356 | |
93e68bfb | 1357 | |
b9ad13ac NC |
1358 | case SVt_REGEXP: |
1359 | /* This ensures that SvTHINKFIRST(sv) is true, and hence that | |
1360 | sv_force_normal_flags(sv) is called. */ | |
1361 | SvFAKE_on(sv); | |
bd81e77b NC |
1362 | case SVt_PVIV: |
1363 | /* XXX Is this still needed? Was it ever needed? Surely as there is | |
1364 | no route from NV to PVIV, NOK can never be true */ | |
1365 | assert(!SvNOKp(sv)); | |
1366 | assert(!SvNOK(sv)); | |
1367 | case SVt_PVIO: | |
1368 | case SVt_PVFM: | |
bd81e77b NC |
1369 | case SVt_PVGV: |
1370 | case SVt_PVCV: | |
1371 | case SVt_PVLV: | |
1372 | case SVt_PVMG: | |
1373 | case SVt_PVNV: | |
1374 | case SVt_PV: | |
93e68bfb | 1375 | |
d2a0f284 | 1376 | assert(new_type_details->body_size); |
bd81e77b NC |
1377 | /* We always allocated the full length item with PURIFY. To do this |
1378 | we fake things so that arena is false for all 16 types.. */ | |
1379 | if(new_type_details->arena) { | |
1380 | /* This points to the start of the allocated area. */ | |
d2a0f284 JC |
1381 | new_body_inline(new_body, new_type); |
1382 | Zero(new_body, new_type_details->body_size, char); | |
bd81e77b NC |
1383 | new_body = ((char *)new_body) - new_type_details->offset; |
1384 | } else { | |
1385 | new_body = new_NOARENAZ(new_type_details); | |
1386 | } | |
1387 | SvANY(sv) = new_body; | |
5e2fc214 | 1388 | |
bd81e77b | 1389 | if (old_type_details->copy) { |
f9ba3d20 NC |
1390 | /* There is now the potential for an upgrade from something without |
1391 | an offset (PVNV or PVMG) to something with one (PVCV, PVFM) */ | |
1392 | int offset = old_type_details->offset; | |
1393 | int length = old_type_details->copy; | |
1394 | ||
1395 | if (new_type_details->offset > old_type_details->offset) { | |
d4c19fe8 | 1396 | const int difference |
f9ba3d20 NC |
1397 | = new_type_details->offset - old_type_details->offset; |
1398 | offset += difference; | |
1399 | length -= difference; | |
1400 | } | |
1401 | assert (length >= 0); | |
1402 | ||
1403 | Copy((char *)old_body + offset, (char *)new_body + offset, length, | |
1404 | char); | |
bd81e77b NC |
1405 | } |
1406 | ||
1407 | #ifndef NV_ZERO_IS_ALLBITS_ZERO | |
f2524eef | 1408 | /* If NV 0.0 is stores as all bits 0 then Zero() already creates a |
e5ce394c NC |
1409 | * correct 0.0 for us. Otherwise, if the old body didn't have an |
1410 | * NV slot, but the new one does, then we need to initialise the | |
1411 | * freshly created NV slot with whatever the correct bit pattern is | |
1412 | * for 0.0 */ | |
e22a937e NC |
1413 | if (old_type_details->zero_nv && !new_type_details->zero_nv |
1414 | && !isGV_with_GP(sv)) | |
bd81e77b | 1415 | SvNV_set(sv, 0); |
82048762 | 1416 | #endif |
5e2fc214 | 1417 | |
85dca89a NC |
1418 | if (new_type == SVt_PVIO) { |
1419 | IO * const io = MUTABLE_IO(sv); | |
d963bf01 | 1420 | GV *iogv = gv_fetchpvs("IO::File::", GV_ADD, SVt_PVHV); |
85dca89a NC |
1421 | |
1422 | SvOBJECT_on(io); | |
1423 | /* Clear the stashcache because a new IO could overrule a package | |
1424 | name */ | |
1425 | hv_clear(PL_stashcache); | |
1426 | ||
85dca89a | 1427 | SvSTASH_set(io, MUTABLE_HV(SvREFCNT_inc(GvHV(iogv)))); |
f2524eef | 1428 | IoPAGE_LEN(sv) = 60; |
85dca89a | 1429 | } |
4df7f6af NC |
1430 | if (old_type < SVt_PV) { |
1431 | /* referant will be NULL unless the old type was SVt_IV emulating | |
1432 | SVt_RV */ | |
1433 | sv->sv_u.svu_rv = referant; | |
1434 | } | |
bd81e77b NC |
1435 | break; |
1436 | default: | |
afd78fd5 JH |
1437 | Perl_croak(aTHX_ "panic: sv_upgrade to unknown type %lu", |
1438 | (unsigned long)new_type); | |
bd81e77b | 1439 | } |
73171d91 | 1440 | |
db93c0c4 | 1441 | if (old_type > SVt_IV) { |
bd81e77b NC |
1442 | #ifdef PURIFY |
1443 | my_safefree(old_body); | |
1444 | #else | |
bc786448 GG |
1445 | /* Note that there is an assumption that all bodies of types that |
1446 | can be upgraded came from arenas. Only the more complex non- | |
1447 | upgradable types are allowed to be directly malloc()ed. */ | |
1448 | assert(old_type_details->arena); | |
bd81e77b NC |
1449 | del_body((void*)((char*)old_body + old_type_details->offset), |
1450 | &PL_body_roots[old_type]); | |
1451 | #endif | |
1452 | } | |
1453 | } | |
73171d91 | 1454 | |
bd81e77b NC |
1455 | /* |
1456 | =for apidoc sv_backoff | |
73171d91 | 1457 | |
bd81e77b NC |
1458 | Remove any string offset. You should normally use the C<SvOOK_off> macro |
1459 | wrapper instead. | |
73171d91 | 1460 | |
bd81e77b | 1461 | =cut |
73171d91 NC |
1462 | */ |
1463 | ||
bd81e77b | 1464 | int |
aad570aa | 1465 | Perl_sv_backoff(pTHX_ register SV *const sv) |
bd81e77b | 1466 | { |
69240efd | 1467 | STRLEN delta; |
7a4bba22 | 1468 | const char * const s = SvPVX_const(sv); |
7918f24d NC |
1469 | |
1470 | PERL_ARGS_ASSERT_SV_BACKOFF; | |
96a5add6 | 1471 | PERL_UNUSED_CONTEXT; |
7918f24d | 1472 | |
bd81e77b NC |
1473 | assert(SvOOK(sv)); |
1474 | assert(SvTYPE(sv) != SVt_PVHV); | |
1475 | assert(SvTYPE(sv) != SVt_PVAV); | |
7a4bba22 | 1476 | |
69240efd NC |
1477 | SvOOK_offset(sv, delta); |
1478 | ||
7a4bba22 NC |
1479 | SvLEN_set(sv, SvLEN(sv) + delta); |
1480 | SvPV_set(sv, SvPVX(sv) - delta); | |
1481 | Move(s, SvPVX(sv), SvCUR(sv)+1, char); | |
bd81e77b NC |
1482 | SvFLAGS(sv) &= ~SVf_OOK; |
1483 | return 0; | |
1484 | } | |
73171d91 | 1485 | |
bd81e77b NC |
1486 | /* |
1487 | =for apidoc sv_grow | |
73171d91 | 1488 | |
bd81e77b NC |
1489 | Expands the character buffer in the SV. If necessary, uses C<sv_unref> and |
1490 | upgrades the SV to C<SVt_PV>. Returns a pointer to the character buffer. | |
1491 | Use the C<SvGROW> wrapper instead. | |
93e68bfb | 1492 | |
bd81e77b NC |
1493 | =cut |
1494 | */ | |
93e68bfb | 1495 | |
bd81e77b | 1496 | char * |
aad570aa | 1497 | Perl_sv_grow(pTHX_ register SV *const sv, register STRLEN newlen) |
bd81e77b NC |
1498 | { |
1499 | register char *s; | |
93e68bfb | 1500 | |
7918f24d NC |
1501 | PERL_ARGS_ASSERT_SV_GROW; |
1502 | ||
5db06880 NC |
1503 | if (PL_madskills && newlen >= 0x100000) { |
1504 | PerlIO_printf(Perl_debug_log, | |
1505 | "Allocation too large: %"UVxf"\n", (UV)newlen); | |
1506 | } | |
bd81e77b NC |
1507 | #ifdef HAS_64K_LIMIT |
1508 | if (newlen >= 0x10000) { | |
1509 | PerlIO_printf(Perl_debug_log, | |
1510 | "Allocation too large: %"UVxf"\n", (UV)newlen); | |
1511 | my_exit(1); | |
1512 | } | |
1513 | #endif /* HAS_64K_LIMIT */ | |
1514 | if (SvROK(sv)) | |
1515 | sv_unref(sv); | |
1516 | if (SvTYPE(sv) < SVt_PV) { | |
1517 | sv_upgrade(sv, SVt_PV); | |
1518 | s = SvPVX_mutable(sv); | |
1519 | } | |
1520 | else if (SvOOK(sv)) { /* pv is offset? */ | |
1521 | sv_backoff(sv); | |
1522 | s = SvPVX_mutable(sv); | |
1523 | if (newlen > SvLEN(sv)) | |
1524 | newlen += 10 * (newlen - SvCUR(sv)); /* avoid copy each time */ | |
1525 | #ifdef HAS_64K_LIMIT | |
1526 | if (newlen >= 0x10000) | |
1527 | newlen = 0xFFFF; | |
1528 | #endif | |
1529 | } | |
1530 | else | |
1531 | s = SvPVX_mutable(sv); | |
aeb18a1e | 1532 | |
bd81e77b | 1533 | if (newlen > SvLEN(sv)) { /* need more room? */ |
aedff202 | 1534 | #ifndef Perl_safesysmalloc_size |
bd81e77b | 1535 | newlen = PERL_STRLEN_ROUNDUP(newlen); |
bd81e77b | 1536 | #endif |
98653f18 | 1537 | if (SvLEN(sv) && s) { |
10edeb5d | 1538 | s = (char*)saferealloc(s, newlen); |
bd81e77b NC |
1539 | } |
1540 | else { | |
10edeb5d | 1541 | s = (char*)safemalloc(newlen); |
bd81e77b NC |
1542 | if (SvPVX_const(sv) && SvCUR(sv)) { |
1543 | Move(SvPVX_const(sv), s, (newlen < SvCUR(sv)) ? newlen : SvCUR(sv), char); | |
1544 | } | |
1545 | } | |
1546 | SvPV_set(sv, s); | |
ca7c1a29 | 1547 | #ifdef Perl_safesysmalloc_size |
98653f18 NC |
1548 | /* Do this here, do it once, do it right, and then we will never get |
1549 | called back into sv_grow() unless there really is some growing | |
1550 | needed. */ | |
ca7c1a29 | 1551 | SvLEN_set(sv, Perl_safesysmalloc_size(s)); |
98653f18 | 1552 | #else |
bd81e77b | 1553 | SvLEN_set(sv, newlen); |
98653f18 | 1554 | #endif |
bd81e77b NC |
1555 | } |
1556 | return s; | |
1557 | } | |
aeb18a1e | 1558 | |
bd81e77b NC |
1559 | /* |
1560 | =for apidoc sv_setiv | |
932e9ff9 | 1561 | |
bd81e77b NC |
1562 | Copies an integer into the given SV, upgrading first if necessary. |
1563 | Does not handle 'set' magic. See also C<sv_setiv_mg>. | |
463ee0b2 | 1564 | |
bd81e77b NC |
1565 | =cut |
1566 | */ | |
463ee0b2 | 1567 | |
bd81e77b | 1568 | void |
aad570aa | 1569 | Perl_sv_setiv(pTHX_ register SV *const sv, const IV i) |
bd81e77b | 1570 | { |
97aff369 | 1571 | dVAR; |
7918f24d NC |
1572 | |
1573 | PERL_ARGS_ASSERT_SV_SETIV; | |
1574 | ||
bd81e77b NC |
1575 | SV_CHECK_THINKFIRST_COW_DROP(sv); |
1576 | switch (SvTYPE(sv)) { | |
1577 | case SVt_NULL: | |
bd81e77b | 1578 | case SVt_NV: |
3376de98 | 1579 | sv_upgrade(sv, SVt_IV); |
bd81e77b | 1580 | break; |
bd81e77b NC |
1581 | case SVt_PV: |
1582 | sv_upgrade(sv, SVt_PVIV); | |
1583 | break; | |
463ee0b2 | 1584 | |
bd81e77b | 1585 | case SVt_PVGV: |
6e592b3a BM |
1586 | if (!isGV_with_GP(sv)) |
1587 | break; | |
bd81e77b NC |
1588 | case SVt_PVAV: |
1589 | case SVt_PVHV: | |
1590 | case SVt_PVCV: | |
1591 | case SVt_PVFM: | |
1592 | case SVt_PVIO: | |
1593 | Perl_croak(aTHX_ "Can't coerce %s to integer in %s", sv_reftype(sv,0), | |
1594 | OP_DESC(PL_op)); | |
42d0e0b7 | 1595 | default: NOOP; |
bd81e77b NC |
1596 | } |
1597 | (void)SvIOK_only(sv); /* validate number */ | |
1598 | SvIV_set(sv, i); | |
1599 | SvTAINT(sv); | |
1600 | } | |
932e9ff9 | 1601 | |
bd81e77b NC |
1602 | /* |
1603 | =for apidoc sv_setiv_mg | |
d33b2eba | 1604 | |
bd81e77b | 1605 | Like C<sv_setiv>, but also handles 'set' magic. |
1c846c1f | 1606 | |
bd81e77b NC |
1607 | =cut |
1608 | */ | |
d33b2eba | 1609 | |
bd81e77b | 1610 | void |
aad570aa | 1611 | Perl_sv_setiv_mg(pTHX_ register SV *const sv, const IV i) |
bd81e77b | 1612 | { |
7918f24d NC |
1613 | PERL_ARGS_ASSERT_SV_SETIV_MG; |
1614 | ||
bd81e77b NC |
1615 | sv_setiv(sv,i); |
1616 | SvSETMAGIC(sv); | |
1617 | } | |
727879eb | 1618 | |
bd81e77b NC |
1619 | /* |
1620 | =for apidoc sv_setuv | |
d33b2eba | 1621 | |
bd81e77b NC |
1622 | Copies an unsigned integer into the given SV, upgrading first if necessary. |
1623 | Does not handle 'set' magic. See also C<sv_setuv_mg>. | |
9b94d1dd | 1624 | |
bd81e77b NC |
1625 | =cut |
1626 | */ | |
d33b2eba | 1627 | |
bd81e77b | 1628 | void |
aad570aa | 1629 | Perl_sv_setuv(pTHX_ register SV *const sv, const UV u) |
bd81e77b | 1630 | { |
7918f24d NC |
1631 | PERL_ARGS_ASSERT_SV_SETUV; |
1632 | ||
bd81e77b NC |
1633 | /* With these two if statements: |
1634 | u=1.49 s=0.52 cu=72.49 cs=10.64 scripts=270 tests=20865 | |
d33b2eba | 1635 | |
bd81e77b NC |
1636 | without |
1637 | u=1.35 s=0.47 cu=73.45 cs=11.43 scripts=270 tests=20865 | |
1c846c1f | 1638 | |
bd81e77b NC |
1639 | If you wish to remove them, please benchmark to see what the effect is |
1640 | */ | |
1641 | if (u <= (UV)IV_MAX) { | |
1642 | sv_setiv(sv, (IV)u); | |
1643 | return; | |
1644 | } | |
1645 | sv_setiv(sv, 0); | |
1646 | SvIsUV_on(sv); | |
1647 | SvUV_set(sv, u); | |
1648 | } | |
d33b2eba | 1649 | |
bd81e77b NC |
1650 | /* |
1651 | =for apidoc sv_setuv_mg | |
727879eb | 1652 | |
bd81e77b | 1653 | Like C<sv_setuv>, but also handles 'set' magic. |
9b94d1dd | 1654 | |
bd81e77b NC |
1655 | =cut |
1656 | */ | |
5e2fc214 | 1657 | |
bd81e77b | 1658 | void |
aad570aa | 1659 | Perl_sv_setuv_mg(pTHX_ register SV *const sv, const UV u) |
bd81e77b | 1660 | { |
7918f24d NC |
1661 | PERL_ARGS_ASSERT_SV_SETUV_MG; |
1662 | ||
bd81e77b NC |
1663 | sv_setuv(sv,u); |
1664 | SvSETMAGIC(sv); | |
1665 | } | |
5e2fc214 | 1666 | |
954c1994 | 1667 | /* |
bd81e77b | 1668 | =for apidoc sv_setnv |
954c1994 | 1669 | |
bd81e77b NC |
1670 | Copies a double into the given SV, upgrading first if necessary. |
1671 | Does not handle 'set' magic. See also C<sv_setnv_mg>. | |
954c1994 GS |
1672 | |
1673 | =cut | |
1674 | */ | |
1675 | ||
63f97190 | 1676 | void |
aad570aa | 1677 | Perl_sv_setnv(pTHX_ register SV *const sv, const NV num) |
79072805 | 1678 | { |
97aff369 | 1679 | dVAR; |
7918f24d NC |
1680 | |
1681 | PERL_ARGS_ASSERT_SV_SETNV; | |
1682 | ||
bd81e77b NC |
1683 | SV_CHECK_THINKFIRST_COW_DROP(sv); |
1684 | switch (SvTYPE(sv)) { | |
79072805 | 1685 | case SVt_NULL: |
79072805 | 1686 | case SVt_IV: |
bd81e77b | 1687 | sv_upgrade(sv, SVt_NV); |
79072805 LW |
1688 | break; |
1689 | case SVt_PV: | |
79072805 | 1690 | case SVt_PVIV: |
bd81e77b | 1691 | sv_upgrade(sv, SVt_PVNV); |
79072805 | 1692 | break; |
bd4b1eb5 | 1693 | |
bd4b1eb5 | 1694 | case SVt_PVGV: |
6e592b3a BM |
1695 | if (!isGV_with_GP(sv)) |
1696 | break; | |
bd81e77b NC |
1697 | case SVt_PVAV: |
1698 | case SVt_PVHV: | |
79072805 | 1699 | case SVt_PVCV: |
bd81e77b NC |
1700 | case SVt_PVFM: |
1701 | case SVt_PVIO: | |
1702 | Perl_croak(aTHX_ "Can't coerce %s to number in %s", sv_reftype(sv,0), | |
94bbb3f4 | 1703 | OP_DESC(PL_op)); |
42d0e0b7 | 1704 | default: NOOP; |
2068cd4d | 1705 | } |
bd81e77b NC |
1706 | SvNV_set(sv, num); |
1707 | (void)SvNOK_only(sv); /* validate number */ | |
1708 | SvTAINT(sv); | |
79072805 LW |
1709 | } |
1710 | ||
645c22ef | 1711 | /* |
bd81e77b | 1712 | =for apidoc sv_setnv_mg |
645c22ef | 1713 | |
bd81e77b | 1714 | Like C<sv_setnv>, but also handles 'set' magic. |
645c22ef DM |
1715 | |
1716 | =cut | |
1717 | */ | |
1718 | ||
bd81e77b | 1719 | void |
aad570aa | 1720 | Perl_sv_setnv_mg(pTHX_ register SV *const sv, const NV num) |
79072805 | 1721 | { |
7918f24d NC |
1722 | PERL_ARGS_ASSERT_SV_SETNV_MG; |
1723 | ||
bd81e77b NC |
1724 | sv_setnv(sv,num); |
1725 | SvSETMAGIC(sv); | |
79072805 LW |
1726 | } |
1727 | ||
bd81e77b NC |
1728 | /* Print an "isn't numeric" warning, using a cleaned-up, |
1729 | * printable version of the offending string | |
1730 | */ | |
954c1994 | 1731 | |
bd81e77b | 1732 | STATIC void |
aad570aa | 1733 | S_not_a_number(pTHX_ SV *const sv) |
79072805 | 1734 | { |
97aff369 | 1735 | dVAR; |
bd81e77b NC |
1736 | SV *dsv; |
1737 | char tmpbuf[64]; | |
1738 | const char *pv; | |
94463019 | 1739 | |
7918f24d NC |
1740 | PERL_ARGS_ASSERT_NOT_A_NUMBER; |
1741 | ||
94463019 | 1742 | if (DO_UTF8(sv)) { |
84bafc02 | 1743 | dsv = newSVpvs_flags("", SVs_TEMP); |
94463019 JH |
1744 | pv = sv_uni_display(dsv, sv, 10, 0); |
1745 | } else { | |
1746 | char *d = tmpbuf; | |
551405c4 | 1747 | const char * const limit = tmpbuf + sizeof(tmpbuf) - 8; |
94463019 JH |
1748 | /* each *s can expand to 4 chars + "...\0", |
1749 | i.e. need room for 8 chars */ | |
ecdeb87c | 1750 | |
00b6aa41 AL |
1751 | const char *s = SvPVX_const(sv); |
1752 | const char * const end = s + SvCUR(sv); | |
1753 | for ( ; s < end && d < limit; s++ ) { | |
94463019 JH |
1754 | int ch = *s & 0xFF; |
1755 | if (ch & 128 && !isPRINT_LC(ch)) { | |
1756 | *d++ = 'M'; | |
1757 | *d++ = '-'; | |
1758 | ch &= 127; | |
1759 | } | |
1760 | if (ch == '\n') { | |
1761 | *d++ = '\\'; | |
1762 | *d++ = 'n'; | |
1763 | } | |
1764 | else if (ch == '\r') { | |
1765 | *d++ = '\\'; | |
1766 | *d++ = 'r'; | |
1767 | } | |
1768 | else if (ch == '\f') { | |
1769 | *d++ = '\\'; | |
1770 | *d++ = 'f'; | |
1771 | } | |
1772 | else if (ch == '\\') { | |
1773 | *d++ = '\\'; | |
1774 | *d++ = '\\'; | |
1775 | } | |
1776 | else if (ch == '\0') { | |
1777 | *d++ = '\\'; | |
1778 | *d++ = '0'; | |
1779 | } | |
1780 | else if (isPRINT_LC(ch)) | |
1781 | *d++ = ch; | |
1782 | else { | |
1783 | *d++ = '^'; | |
1784 | *d++ = toCTRL(ch); | |
1785 | } | |
1786 | } | |
1787 | if (s < end) { | |
1788 | *d++ = '.'; | |
1789 | *d++ = '.'; | |
1790 | *d++ = '.'; | |
1791 | } | |
1792 | *d = '\0'; | |
1793 | pv = tmpbuf; | |
a0d0e21e | 1794 | } |
a0d0e21e | 1795 | |
533c011a | 1796 | if (PL_op) |
9014280d | 1797 | Perl_warner(aTHX_ packWARN(WARN_NUMERIC), |
94463019 JH |
1798 | "Argument \"%s\" isn't numeric in %s", pv, |
1799 | OP_DESC(PL_op)); | |
a0d0e21e | 1800 | else |
9014280d | 1801 | Perl_warner(aTHX_ packWARN(WARN_NUMERIC), |
94463019 | 1802 | "Argument \"%s\" isn't numeric", pv); |
a0d0e21e LW |
1803 | } |
1804 | ||
c2988b20 NC |
1805 | /* |
1806 | =for apidoc looks_like_number | |
1807 | ||
645c22ef DM |
1808 | Test if the content of an SV looks like a number (or is a number). |
1809 | C<Inf> and C<Infinity> are treated as numbers (so will not issue a | |
1810 | non-numeric warning), even if your atof() doesn't grok them. | |
c2988b20 NC |
1811 | |
1812 | =cut | |
1813 | */ | |
1814 | ||
1815 | I32 | |
aad570aa | 1816 | Perl_looks_like_number(pTHX_ SV *const sv) |
c2988b20 | 1817 | { |
a3b680e6 | 1818 | register const char *sbegin; |
c2988b20 NC |
1819 | STRLEN len; |
1820 | ||
7918f24d NC |
1821 | PERL_ARGS_ASSERT_LOOKS_LIKE_NUMBER; |
1822 | ||
c2988b20 | 1823 | if (SvPOK(sv)) { |
3f7c398e | 1824 | sbegin = SvPVX_const(sv); |
c2988b20 NC |
1825 | len = SvCUR(sv); |
1826 | } | |
1827 | else if (SvPOKp(sv)) | |
83003860 | 1828 | sbegin = SvPV_const(sv, len); |
c2988b20 | 1829 | else |
e0ab1c0e | 1830 | return SvFLAGS(sv) & (SVf_NOK|SVp_NOK|SVf_IOK|SVp_IOK); |
c2988b20 NC |
1831 | return grok_number(sbegin, len, NULL); |
1832 | } | |
25da4f38 | 1833 | |
19f6321d NC |
1834 | STATIC bool |
1835 | S_glob_2number(pTHX_ GV * const gv) | |
180488f8 NC |
1836 | { |
1837 | const U32 wasfake = SvFLAGS(gv) & SVf_FAKE; | |
1838 | SV *const buffer = sv_newmortal(); | |
1839 | ||
7918f24d NC |
1840 | PERL_ARGS_ASSERT_GLOB_2NUMBER; |
1841 | ||
180488f8 NC |
1842 | /* FAKE globs can get coerced, so need to turn this off temporarily if it |
1843 | is on. */ | |
1844 | SvFAKE_off(gv); | |
1845 | gv_efullname3(buffer, gv, "*"); | |
1846 | SvFLAGS(gv) |= wasfake; | |
1847 | ||
675c862f AL |
1848 | /* We know that all GVs stringify to something that is not-a-number, |
1849 | so no need to test that. */ | |
1850 | if (ckWARN(WARN_NUMERIC)) | |
1851 | not_a_number(buffer); | |
1852 | /* We just want something true to return, so that S_sv_2iuv_common | |
1853 | can tail call us and return true. */ | |
19f6321d | 1854 | return TRUE; |
675c862f AL |
1855 | } |
1856 | ||
25da4f38 IZ |
1857 | /* Actually, ISO C leaves conversion of UV to IV undefined, but |
1858 | until proven guilty, assume that things are not that bad... */ | |
1859 | ||
645c22ef DM |
1860 | /* |
1861 | NV_PRESERVES_UV: | |
1862 | ||
1863 | As 64 bit platforms often have an NV that doesn't preserve all bits of | |
28e5dec8 JH |
1864 | an IV (an assumption perl has been based on to date) it becomes necessary |
1865 | to remove the assumption that the NV always carries enough precision to | |
1866 | recreate the IV whenever needed, and that the NV is the canonical form. | |
1867 | Instead, IV/UV and NV need to be given equal rights. So as to not lose | |
645c22ef | 1868 | precision as a side effect of conversion (which would lead to insanity |
28e5dec8 JH |
1869 | and the dragon(s) in t/op/numconvert.t getting very angry) the intent is |
1870 | 1) to distinguish between IV/UV/NV slots that have cached a valid | |
1871 | conversion where precision was lost and IV/UV/NV slots that have a | |
1872 | valid conversion which has lost no precision | |
645c22ef | 1873 | 2) to ensure that if a numeric conversion to one form is requested that |
28e5dec8 JH |
1874 | would lose precision, the precise conversion (or differently |
1875 | imprecise conversion) is also performed and cached, to prevent | |
1876 | requests for different numeric formats on the same SV causing | |
1877 | lossy conversion chains. (lossless conversion chains are perfectly | |
1878 | acceptable (still)) | |
1879 | ||
1880 | ||
1881 | flags are used: | |
1882 | SvIOKp is true if the IV slot contains a valid value | |
1883 | SvIOK is true only if the IV value is accurate (UV if SvIOK_UV true) | |
1884 | SvNOKp is true if the NV slot contains a valid value | |
1885 | SvNOK is true only if the NV value is accurate | |
1886 | ||
1887 | so | |
645c22ef | 1888 | while converting from PV to NV, check to see if converting that NV to an |
28e5dec8 JH |
1889 | IV(or UV) would lose accuracy over a direct conversion from PV to |
1890 | IV(or UV). If it would, cache both conversions, return NV, but mark | |
1891 | SV as IOK NOKp (ie not NOK). | |
1892 | ||
645c22ef | 1893 | While converting from PV to IV, check to see if converting that IV to an |
28e5dec8 JH |
1894 | NV would lose accuracy over a direct conversion from PV to NV. If it |
1895 | would, cache both conversions, flag similarly. | |
1896 | ||
1897 | Before, the SV value "3.2" could become NV=3.2 IV=3 NOK, IOK quite | |
1898 | correctly because if IV & NV were set NV *always* overruled. | |
645c22ef DM |
1899 | Now, "3.2" will become NV=3.2 IV=3 NOK, IOKp, because the flag's meaning |
1900 | changes - now IV and NV together means that the two are interchangeable: | |
28e5dec8 | 1901 | SvIVX == (IV) SvNVX && SvNVX == (NV) SvIVX; |
d460ef45 | 1902 | |
645c22ef DM |
1903 | The benefit of this is that operations such as pp_add know that if |
1904 | SvIOK is true for both left and right operands, then integer addition | |
1905 | can be used instead of floating point (for cases where the result won't | |
1906 | overflow). Before, floating point was always used, which could lead to | |
28e5dec8 JH |
1907 | loss of precision compared with integer addition. |
1908 | ||
1909 | * making IV and NV equal status should make maths accurate on 64 bit | |
1910 | platforms | |
1911 | * may speed up maths somewhat if pp_add and friends start to use | |
645c22ef | 1912 | integers when possible instead of fp. (Hopefully the overhead in |
28e5dec8 JH |
1913 | looking for SvIOK and checking for overflow will not outweigh the |
1914 | fp to integer speedup) | |
1915 | * will slow down integer operations (callers of SvIV) on "inaccurate" | |
1916 | values, as the change from SvIOK to SvIOKp will cause a call into | |
1917 | sv_2iv each time rather than a macro access direct to the IV slot | |
1918 | * should speed up number->string conversion on integers as IV is | |
645c22ef | 1919 | favoured when IV and NV are equally accurate |
28e5dec8 JH |
1920 | |
1921 | #################################################################### | |
645c22ef DM |
1922 | You had better be using SvIOK_notUV if you want an IV for arithmetic: |
1923 | SvIOK is true if (IV or UV), so you might be getting (IV)SvUV. | |
1924 | On the other hand, SvUOK is true iff UV. | |
28e5dec8 JH |
1925 | #################################################################### |
1926 | ||
645c22ef | 1927 | Your mileage will vary depending your CPU's relative fp to integer |
28e5dec8 JH |
1928 | performance ratio. |
1929 | */ | |
1930 | ||
1931 | #ifndef NV_PRESERVES_UV | |
645c22ef DM |
1932 | # define IS_NUMBER_UNDERFLOW_IV 1 |
1933 | # define IS_NUMBER_UNDERFLOW_UV 2 | |
1934 | # define IS_NUMBER_IV_AND_UV 2 | |
1935 | # define IS_NUMBER_OVERFLOW_IV 4 | |
1936 | # define IS_NUMBER_OVERFLOW_UV 5 | |
1937 | ||
1938 | /* sv_2iuv_non_preserve(): private routine for use by sv_2iv() and sv_2uv() */ | |
28e5dec8 JH |
1939 | |
1940 | /* For sv_2nv these three cases are "SvNOK and don't bother casting" */ | |
1941 | STATIC int | |
5de3775c | 1942 | S_sv_2iuv_non_preserve(pTHX_ register SV *const sv |
47031da6 NC |
1943 | # ifdef DEBUGGING |
1944 | , I32 numtype | |
1945 | # endif | |
1946 | ) | |
28e5dec8 | 1947 | { |
97aff369 | 1948 | dVAR; |
7918f24d NC |
1949 | |
1950 | PERL_ARGS_ASSERT_SV_2IUV_NON_PRESERVE; | |
1951 | ||
3f7c398e | 1952 | DEBUG_c(PerlIO_printf(Perl_debug_log,"sv_2iuv_non '%s', IV=0x%"UVxf" NV=%"NVgf" inttype=%"UVXf"\n", SvPVX_const(sv), SvIVX(sv), SvNVX(sv), (UV)numtype)); |
28e5dec8 JH |
1953 | if (SvNVX(sv) < (NV)IV_MIN) { |
1954 | (void)SvIOKp_on(sv); | |
1955 | (void)SvNOK_on(sv); | |
45977657 | 1956 | SvIV_set(sv, IV_MIN); |
28e5dec8 JH |
1957 | return IS_NUMBER_UNDERFLOW_IV; |
1958 | } | |
1959 | if (SvNVX(sv) > (NV)UV_MAX) { | |
1960 | (void)SvIOKp_on(sv); | |
1961 | (void)SvNOK_on(sv); | |
1962 | SvIsUV_on(sv); | |
607fa7f2 | 1963 | SvUV_set(sv, UV_MAX); |
28e5dec8 JH |
1964 | return IS_NUMBER_OVERFLOW_UV; |
1965 | } | |
c2988b20 NC |
1966 | (void)SvIOKp_on(sv); |
1967 | (void)SvNOK_on(sv); | |
1968 | /* Can't use strtol etc to convert this string. (See truth table in | |
1969 | sv_2iv */ | |
1970 | if (SvNVX(sv) <= (UV)IV_MAX) { | |
45977657 | 1971 | SvIV_set(sv, I_V(SvNVX(sv))); |
c2988b20 NC |
1972 | if ((NV)(SvIVX(sv)) == SvNVX(sv)) { |
1973 | SvIOK_on(sv); /* Integer is precise. NOK, IOK */ | |
1974 | } else { | |
1975 | /* Integer is imprecise. NOK, IOKp */ | |
1976 | } | |
1977 | return SvNVX(sv) < 0 ? IS_NUMBER_UNDERFLOW_UV : IS_NUMBER_IV_AND_UV; | |
1978 | } | |
1979 | SvIsUV_on(sv); | |
607fa7f2 | 1980 | SvUV_set(sv, U_V(SvNVX(sv))); |
c2988b20 NC |
1981 | if ((NV)(SvUVX(sv)) == SvNVX(sv)) { |
1982 | if (SvUVX(sv) == UV_MAX) { | |
1983 | /* As we know that NVs don't preserve UVs, UV_MAX cannot | |
1984 | possibly be preserved by NV. Hence, it must be overflow. | |
1985 | NOK, IOKp */ | |
1986 | return IS_NUMBER_OVERFLOW_UV; | |
1987 | } | |
1988 | SvIOK_on(sv); /* Integer is precise. NOK, UOK */ | |
1989 | } else { | |
1990 | /* Integer is imprecise. NOK, IOKp */ | |
28e5dec8 | 1991 | } |
c2988b20 | 1992 | return IS_NUMBER_OVERFLOW_IV; |
28e5dec8 | 1993 | } |
645c22ef DM |
1994 | #endif /* !NV_PRESERVES_UV*/ |
1995 | ||
af359546 | 1996 | STATIC bool |
7918f24d NC |
1997 | S_sv_2iuv_common(pTHX_ SV *const sv) |
1998 | { | |
97aff369 | 1999 | dVAR; |
7918f24d NC |
2000 | |
2001 | PERL_ARGS_ASSERT_SV_2IUV_COMMON; | |
2002 | ||
af359546 | 2003 | if (SvNOKp(sv)) { |
28e5dec8 JH |
2004 | /* erm. not sure. *should* never get NOKp (without NOK) from sv_2nv |
2005 | * without also getting a cached IV/UV from it at the same time | |
2006 | * (ie PV->NV conversion should detect loss of accuracy and cache | |
af359546 NC |
2007 | * IV or UV at same time to avoid this. */ |
2008 | /* IV-over-UV optimisation - choose to cache IV if possible */ | |
25da4f38 IZ |
2009 | |
2010 | if (SvTYPE(sv) == SVt_NV) | |
2011 | sv_upgrade(sv, SVt_PVNV); | |
2012 | ||
28e5dec8 JH |
2013 | (void)SvIOKp_on(sv); /* Must do this first, to clear any SvOOK */ |
2014 | /* < not <= as for NV doesn't preserve UV, ((NV)IV_MAX+1) will almost | |
2015 | certainly cast into the IV range at IV_MAX, whereas the correct | |
2016 | answer is the UV IV_MAX +1. Hence < ensures that dodgy boundary | |
2017 | cases go to UV */ | |
cab190d4 JD |
2018 | #if defined(NAN_COMPARE_BROKEN) && defined(Perl_isnan) |
2019 | if (Perl_isnan(SvNVX(sv))) { | |
2020 | SvUV_set(sv, 0); | |
2021 | SvIsUV_on(sv); | |
fdbe6d7c | 2022 | return FALSE; |
cab190d4 | 2023 | } |
cab190d4 | 2024 | #endif |
28e5dec8 | 2025 | if (SvNVX(sv) < (NV)IV_MAX + 0.5) { |
45977657 | 2026 | SvIV_set(sv, I_V(SvNVX(sv))); |
28e5dec8 JH |
2027 | if (SvNVX(sv) == (NV) SvIVX(sv) |
2028 | #ifndef NV_PRESERVES_UV | |
2029 | && (((UV)1 << NV_PRESERVES_UV_BITS) > | |
2030 | (UV)(SvIVX(sv) > 0 ? SvIVX(sv) : -SvIVX(sv))) | |
2031 | /* Don't flag it as "accurately an integer" if the number | |
2032 | came from a (by definition imprecise) NV operation, and | |
2033 | we're outside the range of NV integer precision */ | |
2034 | #endif | |
2035 | ) { | |
a43d94f2 NC |
2036 | if (SvNOK(sv)) |
2037 | SvIOK_on(sv); /* Can this go wrong with rounding? NWC */ | |
2038 | else { | |
2039 | /* scalar has trailing garbage, eg "42a" */ | |
2040 | } | |
28e5dec8 | 2041 | DEBUG_c(PerlIO_printf(Perl_debug_log, |
7234c960 | 2042 | "0x%"UVxf" iv(%"NVgf" => %"IVdf") (precise)\n", |
28e5dec8 JH |
2043 | PTR2UV(sv), |
2044 | SvNVX(sv), | |
2045 | SvIVX(sv))); | |
2046 | ||
2047 | } else { | |
2048 | /* IV not precise. No need to convert from PV, as NV | |
2049 | conversion would already have cached IV if it detected | |
2050 | that PV->IV would be better than PV->NV->IV | |
2051 | flags already correct - don't set public IOK. */ | |
2052 | DEBUG_c(PerlIO_printf(Perl_debug_log, | |
7234c960 | 2053 | "0x%"UVxf" iv(%"NVgf" => %"IVdf") (imprecise)\n", |
28e5dec8 JH |
2054 | PTR2UV(sv), |
2055 | SvNVX(sv), | |
2056 | SvIVX(sv))); | |
2057 | } | |
2058 | /* Can the above go wrong if SvIVX == IV_MIN and SvNVX < IV_MIN, | |
2059 | but the cast (NV)IV_MIN rounds to a the value less (more | |
2060 | negative) than IV_MIN which happens to be equal to SvNVX ?? | |
2061 | Analogous to 0xFFFFFFFFFFFFFFFF rounding up to NV (2**64) and | |
2062 | NV rounding back to 0xFFFFFFFFFFFFFFFF, so UVX == UV(NVX) and | |
2063 | (NV)UVX == NVX are both true, but the values differ. :-( | |
2064 | Hopefully for 2s complement IV_MIN is something like | |
2065 | 0x8000000000000000 which will be exact. NWC */ | |
d460ef45 | 2066 | } |
25da4f38 | 2067 | else { |
607fa7f2 | 2068 | SvUV_set(sv, U_V(SvNVX(sv))); |
28e5dec8 JH |
2069 | if ( |
2070 | (SvNVX(sv) == (NV) SvUVX(sv)) | |
2071 | #ifndef NV_PRESERVES_UV | |
2072 | /* Make sure it's not 0xFFFFFFFFFFFFFFFF */ | |
2073 | /*&& (SvUVX(sv) != UV_MAX) irrelevant with code below */ | |
2074 | && (((UV)1 << NV_PRESERVES_UV_BITS) > SvUVX(sv)) | |
2075 | /* Don't flag it as "accurately an integer" if the number | |
2076 | came from a (by definition imprecise) NV operation, and | |
2077 | we're outside the range of NV integer precision */ | |
2078 | #endif | |
a43d94f2 | 2079 | && SvNOK(sv) |
28e5dec8 JH |
2080 | ) |
2081 | SvIOK_on(sv); | |
25da4f38 | 2082 | SvIsUV_on(sv); |
1c846c1f | 2083 | DEBUG_c(PerlIO_printf(Perl_debug_log, |
57def98f | 2084 | "0x%"UVxf" 2iv(%"UVuf" => %"IVdf") (as unsigned)\n", |
56431972 | 2085 | PTR2UV(sv), |
57def98f JH |
2086 | SvUVX(sv), |
2087 | SvUVX(sv))); | |
25da4f38 | 2088 | } |
748a9306 LW |
2089 | } |
2090 | else if (SvPOKp(sv) && SvLEN(sv)) { | |
c2988b20 | 2091 | UV value; |
504618e9 | 2092 | const int numtype = grok_number(SvPVX_const(sv), SvCUR(sv), &value); |
af359546 | 2093 | /* We want to avoid a possible problem when we cache an IV/ a UV which |
25da4f38 | 2094 | may be later translated to an NV, and the resulting NV is not |
c2988b20 NC |
2095 | the same as the direct translation of the initial string |
2096 | (eg 123.456 can shortcut to the IV 123 with atol(), but we must | |
2097 | be careful to ensure that the value with the .456 is around if the | |
2098 | NV value is requested in the future). | |
1c846c1f | 2099 | |
af359546 | 2100 | This means that if we cache such an IV/a UV, we need to cache the |
25da4f38 | 2101 | NV as well. Moreover, we trade speed for space, and do not |
28e5dec8 | 2102 | cache the NV if we are sure it's not needed. |
25da4f38 | 2103 | */ |
16b7a9a4 | 2104 | |
c2988b20 NC |
2105 | /* SVt_PVNV is one higher than SVt_PVIV, hence this order */ |
2106 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) | |
2107 | == IS_NUMBER_IN_UV) { | |
5e045b90 | 2108 | /* It's definitely an integer, only upgrade to PVIV */ |
28e5dec8 JH |
2109 | if (SvTYPE(sv) < SVt_PVIV) |
2110 | sv_upgrade(sv, SVt_PVIV); | |
f7bbb42a | 2111 | (void)SvIOK_on(sv); |
c2988b20 NC |
2112 | } else if (SvTYPE(sv) < SVt_PVNV) |
2113 | sv_upgrade(sv, SVt_PVNV); | |
28e5dec8 | 2114 | |
f2524eef | 2115 | /* If NVs preserve UVs then we only use the UV value if we know that |
c2988b20 NC |
2116 | we aren't going to call atof() below. If NVs don't preserve UVs |
2117 | then the value returned may have more precision than atof() will | |
2118 | return, even though value isn't perfectly accurate. */ | |
2119 | if ((numtype & (IS_NUMBER_IN_UV | |
2120 | #ifdef NV_PRESERVES_UV | |
2121 | | IS_NUMBER_NOT_INT | |
2122 | #endif | |
2123 | )) == IS_NUMBER_IN_UV) { | |
2124 | /* This won't turn off the public IOK flag if it was set above */ | |
2125 | (void)SvIOKp_on(sv); | |
2126 | ||
2127 | if (!(numtype & IS_NUMBER_NEG)) { | |
2128 | /* positive */; | |
2129 | if (value <= (UV)IV_MAX) { | |
45977657 | 2130 | SvIV_set(sv, (IV)value); |
c2988b20 | 2131 | } else { |
af359546 | 2132 | /* it didn't overflow, and it was positive. */ |
607fa7f2 | 2133 | SvUV_set(sv, value); |
c2988b20 NC |
2134 | SvIsUV_on(sv); |
2135 | } | |
2136 | } else { | |
2137 | /* 2s complement assumption */ | |
2138 | if (value <= (UV)IV_MIN) { | |
45977657 | 2139 | SvIV_set(sv, -(IV)value); |
c2988b20 NC |
2140 | } else { |
2141 | /* Too negative for an IV. This is a double upgrade, but | |
d1be9408 | 2142 | I'm assuming it will be rare. */ |
c2988b20 NC |
2143 | if (SvTYPE(sv) < SVt_PVNV) |
2144 | sv_upgrade(sv, SVt_PVNV); | |
2145 | SvNOK_on(sv); | |
2146 | SvIOK_off(sv); | |
2147 | SvIOKp_on(sv); | |
9d6ce603 | 2148 | SvNV_set(sv, -(NV)value); |
45977657 | 2149 | SvIV_set(sv, IV_MIN); |
c2988b20 NC |
2150 | } |
2151 | } | |
2152 | } | |
2153 | /* For !NV_PRESERVES_UV and IS_NUMBER_IN_UV and IS_NUMBER_NOT_INT we | |
2154 | will be in the previous block to set the IV slot, and the next | |
2155 | block to set the NV slot. So no else here. */ | |
2156 | ||
2157 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) | |
2158 | != IS_NUMBER_IN_UV) { | |
2159 | /* It wasn't an (integer that doesn't overflow the UV). */ | |
3f7c398e | 2160 | SvNV_set(sv, Atof(SvPVX_const(sv))); |
28e5dec8 | 2161 | |
c2988b20 NC |
2162 | if (! numtype && ckWARN(WARN_NUMERIC)) |
2163 | not_a_number(sv); | |
28e5dec8 | 2164 | |
65202027 | 2165 | #if defined(USE_LONG_DOUBLE) |
c2988b20 NC |
2166 | DEBUG_c(PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2iv(%" PERL_PRIgldbl ")\n", |
2167 | PTR2UV(sv), SvNVX(sv))); | |
65202027 | 2168 | #else |
1779d84d | 2169 | DEBUG_c(PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2iv(%"NVgf")\n", |
c2988b20 | 2170 | PTR2UV(sv), SvNVX(sv))); |
65202027 | 2171 | #endif |
28e5dec8 | 2172 | |
28e5dec8 | 2173 | #ifdef NV_PRESERVES_UV |
af359546 NC |
2174 | (void)SvIOKp_on(sv); |
2175 | (void)SvNOK_on(sv); | |
2176 | if (SvNVX(sv) < (NV)IV_MAX + 0.5) { | |
2177 | SvIV_set(sv, I_V(SvNVX(sv))); | |
2178 | if ((NV)(SvIVX(sv)) == SvNVX(sv)) { | |
2179 | SvIOK_on(sv); | |
2180 | } else { | |
6f207bd3 | 2181 | NOOP; /* Integer is imprecise. NOK, IOKp */ |
af359546 NC |
2182 | } |
2183 | /* UV will not work better than IV */ | |
2184 | } else { | |
2185 | if (SvNVX(sv) > (NV)UV_MAX) { | |
2186 | SvIsUV_on(sv); | |
2187 | /* Integer is inaccurate. NOK, IOKp, is UV */ | |
2188 | SvUV_set(sv, UV_MAX); | |
af359546 NC |
2189 | } else { |
2190 | SvUV_set(sv, U_V(SvNVX(sv))); | |
2191 | /* 0xFFFFFFFFFFFFFFFF not an issue in here, NVs | |
2192 | NV preservse UV so can do correct comparison. */ | |
2193 | if ((NV)(SvUVX(sv)) == SvNVX(sv)) { | |
2194 | SvIOK_on(sv); | |
af359546 | 2195 | } else { |
6f207bd3 | 2196 | NOOP; /* Integer is imprecise. NOK, IOKp, is UV */ |
af359546 NC |
2197 | } |
2198 | } | |
4b0c9573 | 2199 | SvIsUV_on(sv); |
af359546 | 2200 | } |
28e5dec8 | 2201 | #else /* NV_PRESERVES_UV */ |
c2988b20 NC |
2202 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) |
2203 | == (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) { | |
af359546 | 2204 | /* The IV/UV slot will have been set from value returned by |
c2988b20 NC |
2205 | grok_number above. The NV slot has just been set using |
2206 | Atof. */ | |
560b0c46 | 2207 | SvNOK_on(sv); |
c2988b20 NC |
2208 | assert (SvIOKp(sv)); |
2209 | } else { | |
2210 | if (((UV)1 << NV_PRESERVES_UV_BITS) > | |
2211 | U_V(SvNVX(sv) > 0 ? SvNVX(sv) : -SvNVX(sv))) { | |
2212 | /* Small enough to preserve all bits. */ | |
2213 | (void)SvIOKp_on(sv); | |
2214 | SvNOK_on(sv); | |
45977657 | 2215 | SvIV_set(sv, I_V(SvNVX(sv))); |
c2988b20 NC |
2216 | if ((NV)(SvIVX(sv)) == SvNVX(sv)) |
2217 | SvIOK_on(sv); | |
2218 | /* Assumption: first non-preserved integer is < IV_MAX, | |
2219 | this NV is in the preserved range, therefore: */ | |
2220 | if (!(U_V(SvNVX(sv) > 0 ? SvNVX(sv) : -SvNVX(sv)) | |
2221 | < (UV)IV_MAX)) { | |
32fdb065 | 2222 | Perl_croak(aTHX_ "sv_2iv assumed (U_V(fabs((double)SvNVX(sv))) < (UV)IV_MAX) but SvNVX(sv)=%"NVgf" U_V is 0x%"UVxf", IV_MAX is 0x%"UVxf"\n", SvNVX(sv), U_V(SvNVX(sv)), (UV)IV_MAX); |
c2988b20 NC |
2223 | } |
2224 | } else { | |
2225 | /* IN_UV NOT_INT | |
2226 | 0 0 already failed to read UV. | |
2227 | 0 1 already failed to read UV. | |
2228 | 1 0 you won't get here in this case. IV/UV | |
2229 | slot set, public IOK, Atof() unneeded. | |
2230 | 1 1 already read UV. | |
2231 | so there's no point in sv_2iuv_non_preserve() attempting | |
2232 | to use atol, strtol, strtoul etc. */ | |
47031da6 | 2233 | # ifdef DEBUGGING |
40a17c4c | 2234 | sv_2iuv_non_preserve (sv, numtype); |
47031da6 NC |
2235 | # else |
2236 | sv_2iuv_non_preserve (sv); | |
2237 | # endif | |
c2988b20 NC |
2238 | } |
2239 | } | |
28e5dec8 | 2240 | #endif /* NV_PRESERVES_UV */ |
a43d94f2 NC |
2241 | /* It might be more code efficient to go through the entire logic above |
2242 | and conditionally set with SvIOKp_on() rather than SvIOK(), but it | |
2243 | gets complex and potentially buggy, so more programmer efficient | |
2244 | to do it this way, by turning off the public flags: */ | |
2245 | if (!numtype) | |
2246 | SvFLAGS(sv) &= ~(SVf_IOK|SVf_NOK); | |
25da4f38 | 2247 | } |
af359546 NC |
2248 | } |
2249 | else { | |
675c862f | 2250 | if (isGV_with_GP(sv)) |
159b6efe | 2251 | return glob_2number(MUTABLE_GV(sv)); |
180488f8 | 2252 | |
af359546 NC |
2253 | if (!(SvFLAGS(sv) & SVs_PADTMP)) { |
2254 | if (!PL_localizing && ckWARN(WARN_UNINITIALIZED)) | |
2255 | report_uninit(sv); | |
2256 | } | |
25da4f38 IZ |
2257 | if (SvTYPE(sv) < SVt_IV) |
2258 | /* Typically the caller expects that sv_any is not NULL now. */ | |
2259 | sv_upgrade(sv, SVt_IV); | |
af359546 NC |
2260 | /* Return 0 from the caller. */ |
2261 | return TRUE; | |
2262 | } | |
2263 | return FALSE; | |
2264 | } | |
2265 | ||
2266 | /* | |
2267 | =for apidoc sv_2iv_flags | |
2268 | ||
2269 | Return the integer value of an SV, doing any necessary string | |
2270 | conversion. If flags includes SV_GMAGIC, does an mg_get() first. | |
2271 | Normally used via the C<SvIV(sv)> and C<SvIVx(sv)> macros. | |
2272 | ||
2273 | =cut | |
2274 | */ | |
2275 | ||
2276 | IV | |
5de3775c | 2277 | Perl_sv_2iv_flags(pTHX_ register SV *const sv, const I32 flags) |
af359546 | 2278 | { |
97aff369 | 2279 | dVAR; |
af359546 | 2280 | if (!sv) |
a0d0e21e | 2281 | return 0; |
cecf5685 NC |
2282 | if (SvGMAGICAL(sv) || (SvTYPE(sv) == SVt_PVGV && SvVALID(sv))) { |
2283 | /* FBMs use the same flag bit as SVf_IVisUV, so must let them | |
50caf62e NC |
2284 | cache IVs just in case. In practice it seems that they never |
2285 | actually anywhere accessible by user Perl code, let alone get used | |
2286 | in anything other than a string context. */ | |
af359546 NC |
2287 | if (flags & SV_GMAGIC) |
2288 | mg_get(sv); | |
2289 | if (SvIOKp(sv)) | |
2290 | return SvIVX(sv); | |
2291 | if (SvNOKp(sv)) { | |
2292 | return I_V(SvNVX(sv)); | |
2293 | } | |
71c558c3 NC |
2294 | if (SvPOKp(sv) && SvLEN(sv)) { |
2295 | UV value; | |
2296 | const int numtype | |
2297 | = grok_number(SvPVX_const(sv), SvCUR(sv), &value); | |
2298 | ||
2299 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) | |
2300 | == IS_NUMBER_IN_UV) { | |
2301 | /* It's definitely an integer */ | |
2302 | if (numtype & IS_NUMBER_NEG) { | |
2303 | if (value < (UV)IV_MIN) | |
2304 | return -(IV)value; | |
2305 | } else { | |
2306 | if (value < (UV)IV_MAX) | |
2307 | return (IV)value; | |
2308 | } | |
2309 | } | |
2310 | if (!numtype) { | |
2311 | if (ckWARN(WARN_NUMERIC)) | |
2312 | not_a_number(sv); | |
2313 | } | |
2314 | return I_V(Atof(SvPVX_const(sv))); | |
2315 | } | |
1c7ff15e NC |
2316 | if (SvROK(sv)) { |
2317 | goto return_rok; | |
af359546 | 2318 | } |
1c7ff15e NC |
2319 | assert(SvTYPE(sv) >= SVt_PVMG); |
2320 | /* This falls through to the report_uninit inside S_sv_2iuv_common. */ | |
4cb1ec55 | 2321 | } else if (SvTHINKFIRST(sv)) { |
af359546 | 2322 | if (SvROK(sv)) { |
1c7ff15e | 2323 | return_rok: |
af359546 NC |
2324 | if (SvAMAGIC(sv)) { |
2325 | SV * const tmpstr=AMG_CALLun(sv,numer); | |
2326 | if (tmpstr && (!SvROK(tmpstr) || (SvRV(tmpstr) != SvRV(sv)))) { | |
2327 | return SvIV(tmpstr); | |
2328 | } | |
2329 | } | |
2330 | return PTR2IV(SvRV(sv)); | |
2331 | } | |
2332 | if (SvIsCOW(sv)) { | |
2333 | sv_force_normal_flags(sv, 0); | |
2334 | } | |
2335 | if (SvREADONLY(sv) && !SvOK(sv)) { | |
2336 | if (ckWARN(WARN_UNINITIALIZED)) | |
2337 | report_uninit(sv); | |
2338 | return 0; | |
2339 | } | |
2340 | } | |
2341 | if (!SvIOKp(sv)) { | |
2342 | if (S_sv_2iuv_common(aTHX_ sv)) | |
2343 | return 0; | |
79072805 | 2344 | } |
1d7c1841 GS |
2345 | DEBUG_c(PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2iv(%"IVdf")\n", |
2346 | PTR2UV(sv),SvIVX(sv))); | |
25da4f38 | 2347 | return SvIsUV(sv) ? (IV)SvUVX(sv) : SvIVX(sv); |
79072805 LW |
2348 | } |
2349 | ||
645c22ef | 2350 | /* |
891f9566 | 2351 | =for apidoc sv_2uv_flags |
645c22ef DM |
2352 | |
2353 | Return the unsigned integer value of an SV, doing any necessary string | |
891f9566 YST |
2354 | conversion. If flags includes SV_GMAGIC, does an mg_get() first. |
2355 | Normally used via the C<SvUV(sv)> and C<SvUVx(sv)> macros. | |
645c22ef DM |
2356 | |
2357 | =cut | |
2358 | */ | |
2359 | ||
ff68c719 | 2360 | UV |
5de3775c | 2361 | Perl_sv_2uv_flags(pTHX_ register SV *const sv, const I32 flags) |
ff68c719 | 2362 | { |
97aff369 | 2363 | dVAR; |
ff68c719 | 2364 | if (!sv) |
2365 | return 0; | |
cecf5685 NC |
2366 | if (SvGMAGICAL(sv) || (SvTYPE(sv) == SVt_PVGV && SvVALID(sv))) { |
2367 | /* FBMs use the same flag bit as SVf_IVisUV, so must let them | |
50caf62e | 2368 | cache IVs just in case. */ |
891f9566 YST |
2369 | if (flags & SV_GMAGIC) |
2370 | mg_get(sv); | |
ff68c719 | 2371 | if (SvIOKp(sv)) |
2372 | return SvUVX(sv); | |
2373 | if (SvNOKp(sv)) | |
2374 | return U_V(SvNVX(sv)); | |
71c558c3 NC |
2375 | if (SvPOKp(sv) && SvLEN(sv)) { |
2376 | UV value; | |
2377 | const int numtype | |
2378 | = grok_number(SvPVX_const(sv), SvCUR(sv), &value); | |
2379 | ||
2380 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) | |
2381 | == IS_NUMBER_IN_UV) { | |
2382 | /* It's definitely an integer */ | |
2383 | if (!(numtype & IS_NUMBER_NEG)) | |
2384 | return value; | |
2385 | } | |
2386 | if (!numtype) { | |
2387 | if (ckWARN(WARN_NUMERIC)) | |
2388 | not_a_number(sv); | |
2389 | } | |
2390 | return U_V(Atof(SvPVX_const(sv))); | |
2391 | } | |
1c7ff15e NC |
2392 | if (SvROK(sv)) { |
2393 | goto return_rok; | |
3fe9a6f1 | 2394 | } |
1c7ff15e NC |
2395 | assert(SvTYPE(sv) >= SVt_PVMG); |
2396 | /* This falls through to the report_uninit inside S_sv_2iuv_common. */ | |
4cb1ec55 | 2397 | } else if (SvTHINKFIRST(sv)) { |
ff68c719 | 2398 | if (SvROK(sv)) { |
1c7ff15e | 2399 | return_rok: |
deb46114 NC |
2400 | if (SvAMAGIC(sv)) { |
2401 | SV *const tmpstr = AMG_CALLun(sv,numer); | |
2402 | if (tmpstr && (!SvROK(tmpstr) || (SvRV(tmpstr) != SvRV(sv)))) { | |
2403 | return SvUV(tmpstr); | |
2404 | } | |
2405 | } | |
2406 | return PTR2UV(SvRV(sv)); | |
ff68c719 | 2407 | } |
765f542d NC |
2408 | if (SvIsCOW(sv)) { |
2409 | sv_force_normal_flags(sv, 0); | |
8a818333 | 2410 | } |
0336b60e | 2411 | if (SvREADONLY(sv) && !SvOK(sv)) { |
0336b60e | 2412 | if (ckWARN(WARN_UNINITIALIZED)) |
29489e7c | 2413 | report_uninit(sv); |
ff68c719 | 2414 | return 0; |
2415 | } | |
2416 | } | |
af359546 NC |
2417 | if (!SvIOKp(sv)) { |
2418 | if (S_sv_2iuv_common(aTHX_ sv)) | |
2419 | return 0; | |
ff68c719 | 2420 | } |
25da4f38 | 2421 | |
1d7c1841 GS |
2422 | DEBUG_c(PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2uv(%"UVuf")\n", |
2423 | PTR2UV(sv),SvUVX(sv))); | |
25da4f38 | 2424 | return SvIsUV(sv) ? SvUVX(sv) : (UV)SvIVX(sv); |
ff68c719 | 2425 | } |
2426 | ||
645c22ef DM |
2427 | /* |
2428 | =for apidoc sv_2nv | |
2429 | ||
2430 | Return the num value of an SV, doing any necessary string or integer | |
39d5de13 DM |
2431 | conversion. If flags includes SV_GMAGIC, does an mg_get() first. |
2432 | Normally used via the C<SvNV(sv)> and C<SvNVx(sv)> macros. | |
645c22ef DM |
2433 | |
2434 | =cut | |
2435 | */ | |
2436 | ||
65202027 | 2437 | NV |
39d5de13 | 2438 | Perl_sv_2nv_flags(pTHX_ register SV *const sv, const I32 flags) |
79072805 | 2439 | { |
97aff369 | 2440 | dVAR; |
79072805 LW |
2441 | if (!sv) |
2442 | return 0.0; | |
cecf5685 NC |
2443 | if (SvGMAGICAL(sv) || (SvTYPE(sv) == SVt_PVGV && SvVALID(sv))) { |
2444 | /* FBMs use the same flag bit as SVf_IVisUV, so must let them | |
50caf62e | 2445 | cache IVs just in case. */ |
39d5de13 DM |
2446 | if (flags & SV_GMAGIC) |
2447 | mg_get(sv); | |
463ee0b2 LW |
2448 | if (SvNOKp(sv)) |
2449 | return SvNVX(sv); | |
0aa395f8 | 2450 | if ((SvPOKp(sv) && SvLEN(sv)) && !SvIOKp(sv)) { |
041457d9 | 2451 | if (!SvIOKp(sv) && ckWARN(WARN_NUMERIC) && |
504618e9 | 2452 | !grok_number(SvPVX_const(sv), SvCUR(sv), NULL)) |
a0d0e21e | 2453 | not_a_number(sv); |
3f7c398e | 2454 | return Atof(SvPVX_const(sv)); |
a0d0e21e | 2455 | } |
25da4f38 | 2456 | if (SvIOKp(sv)) { |
1c846c1f | 2457 | if (SvIsUV(sv)) |
65202027 | 2458 | return (NV)SvUVX(sv); |
25da4f38 | 2459 | else |
65202027 | 2460 | return (NV)SvIVX(sv); |
47a72cb8 NC |
2461 | } |
2462 | if (SvROK(sv)) { | |
2463 | goto return_rok; | |
2464 | } | |
2465 | assert(SvTYPE(sv) >= SVt_PVMG); | |
2466 | /* This falls through to the report_uninit near the end of the | |
2467 | function. */ | |
2468 | } else if (SvTHINKFIRST(sv)) { | |
a0d0e21e | 2469 | if (SvROK(sv)) { |
47a72cb8 | 2470 | return_rok: |
deb46114 NC |
2471 | if (SvAMAGIC(sv)) { |
2472 | SV *const tmpstr = AMG_CALLun(sv,numer); | |
2473 | if (tmpstr && (!SvROK(tmpstr) || (SvRV(tmpstr) != SvRV(sv)))) { | |
2474 | return SvNV(tmpstr); | |
2475 | } | |
2476 | } | |
2477 | return PTR2NV(SvRV(sv)); | |
a0d0e21e | 2478 | } |
765f542d NC |
2479 | if (SvIsCOW(sv)) { |
2480 | sv_force_normal_flags(sv, 0); | |
8a818333 | 2481 | } |
0336b60e | 2482 | if (SvREADONLY(sv) && !SvOK(sv)) { |
599cee73 | 2483 | if (ckWARN(WARN_UNINITIALIZED)) |
29489e7c | 2484 | report_uninit(sv); |
ed6116ce LW |
2485 | return 0.0; |
2486 | } | |
79072805 LW |
2487 | } |
2488 | if (SvTYPE(sv) < SVt_NV) { | |
7e25a7e9 NC |
2489 | /* The logic to use SVt_PVNV if necessary is in sv_upgrade. */ |
2490 | sv_upgrade(sv, SVt_NV); | |
906f284f | 2491 | #ifdef USE_LONG_DOUBLE |
097ee67d | 2492 | DEBUG_c({ |
f93f4e46 | 2493 | STORE_NUMERIC_LOCAL_SET_STANDARD(); |
1d7c1841 GS |
2494 | PerlIO_printf(Perl_debug_log, |
2495 | "0x%"UVxf" num(%" PERL_PRIgldbl ")\n", | |
2496 | PTR2UV(sv), SvNVX(sv)); | |
572bbb43 GS |
2497 | RESTORE_NUMERIC_LOCAL(); |
2498 | }); | |
65202027 | 2499 | #else |
572bbb43 | 2500 | DEBUG_c({ |
f93f4e46 | 2501 | STORE_NUMERIC_LOCAL_SET_STANDARD(); |
1779d84d | 2502 | PerlIO_printf(Perl_debug_log, "0x%"UVxf" num(%"NVgf")\n", |
1d7c1841 | 2503 | PTR2UV(sv), SvNVX(sv)); |
097ee67d JH |
2504 | RESTORE_NUMERIC_LOCAL(); |
2505 | }); | |
572bbb43 | 2506 | #endif |
79072805 LW |
2507 | } |
2508 | else if (SvTYPE(sv) < SVt_PVNV) | |
2509 | sv_upgrade(sv, SVt_PVNV); | |
59d8ce62 NC |
2510 | if (SvNOKp(sv)) { |
2511 | return SvNVX(sv); | |
61604483 | 2512 | } |
59d8ce62 | 2513 | if (SvIOKp(sv)) { |
9d6ce603 | 2514 | SvNV_set(sv, SvIsUV(sv) ? (NV)SvUVX(sv) : (NV)SvIVX(sv)); |
28e5dec8 | 2515 | #ifdef NV_PRESERVES_UV |
a43d94f2 NC |
2516 | if (SvIOK(sv)) |
2517 | SvNOK_on(sv); | |
2518 | else | |
2519 | SvNOKp_on(sv); | |
28e5dec8 JH |
2520 | #else |
2521 | /* Only set the public NV OK flag if this NV preserves the IV */ | |
2522 | /* Check it's not 0xFFFFFFFFFFFFFFFF */ | |
a43d94f2 NC |
2523 | if (SvIOK(sv) && |
2524 | SvIsUV(sv) ? ((SvUVX(sv) != UV_MAX)&&(SvUVX(sv) == U_V(SvNVX(sv)))) | |
28e5dec8 JH |
2525 | : (SvIVX(sv) == I_V(SvNVX(sv)))) |
2526 | SvNOK_on(sv); | |
2527 | else | |
2528 | SvNOKp_on(sv); | |
2529 | #endif | |
93a17b20 | 2530 | } |
748a9306 | 2531 | else if (SvPOKp(sv) && SvLEN(sv)) { |
c2988b20 | 2532 | UV value; |
3f7c398e | 2533 | const int numtype = grok_number(SvPVX_const(sv), SvCUR(sv), &value); |
041457d9 | 2534 | if (!SvIOKp(sv) && !numtype && ckWARN(WARN_NUMERIC)) |
a0d0e21e | 2535 | not_a_number(sv); |
28e5dec8 | 2536 | #ifdef NV_PRESERVES_UV |
c2988b20 NC |
2537 | if ((numtype & (IS_NUMBER_IN_UV | IS_NUMBER_NOT_INT)) |
2538 | == IS_NUMBER_IN_UV) { | |
5e045b90 | 2539 | /* It's definitely an integer */ |
9d6ce603 | 2540 | SvNV_set(sv, (numtype & IS_NUMBER_NEG) ? -(NV)value : (NV)value); |
c2988b20 | 2541 | } else |
3f7c398e | 2542 | SvNV_set(sv, Atof(SvPVX_const(sv))); |
a43d94f2 NC |
2543 | if (numtype) |
2544 | SvNOK_on(sv); | |
2545 | else | |
2546 | SvNOKp_on(sv); | |
28e5dec8 | 2547 | #else |
3f7c398e | 2548 | SvNV_set(sv, Atof(SvPVX_const(sv))); |
28e5dec8 JH |
2549 | /* Only set the public NV OK flag if this NV preserves the value in |
2550 | the PV at least as well as an IV/UV would. | |
2551 | Not sure how to do this 100% reliably. */ | |
2552 | /* if that shift count is out of range then Configure's test is | |
2553 | wonky. We shouldn't be in here with NV_PRESERVES_UV_BITS == | |
2554 | UV_BITS */ | |
2555 | if (((UV)1 << NV_PRESERVES_UV_BITS) > | |
c2988b20 | 2556 | U_V(SvNVX(sv) > 0 ? SvNVX(sv) : -SvNVX(sv))) { |
28e5dec8 | 2557 | SvNOK_on(sv); /* Definitely small enough to preserve all bits */ |
c2988b20 NC |
2558 | } else if (!(numtype & IS_NUMBER_IN_UV)) { |
2559 | /* Can't use strtol etc to convert this string, so don't try. | |
2560 | sv_2iv and sv_2uv will use the NV to convert, not the PV. */ | |
2561 | SvNOK_on(sv); | |
2562 | } else { | |
2563 | /* value has been set. It may not be precise. */ | |
2564 | if ((numtype & IS_NUMBER_NEG) && (value > (UV)IV_MIN)) { | |
2565 | /* 2s complement assumption for (UV)IV_MIN */ | |
2566 | SvNOK_on(sv); /* Integer is too negative. */ | |
2567 | } else { | |
2568 | SvNOKp_on(sv); | |
2569 | SvIOKp_on(sv); | |
6fa402ec | 2570 | |
c2988b20 | 2571 | if (numtype & IS_NUMBER_NEG) { |
45977657 | 2572 | SvIV_set(sv, -(IV)value); |
c2988b20 | 2573 | } else if (value <= (UV)IV_MAX) { |
45977657 | 2574 | SvIV_set(sv, (IV)value); |
c2988b20 | 2575 | } else { |
607fa7f2 | 2576 | SvUV_set(sv, value); |
c2988b20 NC |
2577 | SvIsUV_on(sv); |
2578 | } | |
2579 | ||
2580 | if (numtype & IS_NUMBER_NOT_INT) { | |
2581 | /* I believe that even if the original PV had decimals, | |
2582 | they are lost beyond the limit of the FP precision. | |
2583 | However, neither is canonical, so both only get p | |
2584 | flags. NWC, 2000/11/25 */ | |
2585 | /* Both already have p flags, so do nothing */ | |
2586 | } else { | |
66a1b24b | 2587 | const NV nv = SvNVX(sv); |
c2988b20 NC |
2588 | if (SvNVX(sv) < (NV)IV_MAX + 0.5) { |
2589 | if (SvIVX(sv) == I_V(nv)) { | |
2590 | SvNOK_on(sv); | |
c2988b20 | 2591 | } else { |
c2988b20 NC |
2592 | /* It had no "." so it must be integer. */ |
2593 | } | |
00b6aa41 | 2594 | SvIOK_on(sv); |
c2988b20 NC |
2595 | } else { |
2596 | /* between IV_MAX and NV(UV_MAX). | |
2597 | Could be slightly > UV_MAX */ | |
6fa402ec | 2598 | |
c2988b20 NC |
2599 | if (numtype & IS_NUMBER_NOT_INT) { |
2600 | /* UV and NV both imprecise. */ | |
2601 | } else { | |
66a1b24b | 2602 | const UV nv_as_uv = U_V(nv); |
c2988b20 NC |
2603 | |
2604 | if (value == nv_as_uv && SvUVX(sv) != UV_MAX) { | |
2605 | SvNOK_on(sv); | |
c2988b20 | 2606 | } |
00b6aa41 | 2607 | SvIOK_on(sv); |
c2988b20 NC |
2608 | } |
2609 | } | |
2610 | } | |
2611 | } | |
2612 | } | |
a43d94f2 NC |
2613 | /* It might be more code efficient to go through the entire logic above |
2614 | and conditionally set with SvNOKp_on() rather than SvNOK(), but it | |
2615 | gets complex and potentially buggy, so more programmer efficient | |
2616 | to do it this way, by turning off the public flags: */ | |
2617 | if (!numtype) | |
2618 | SvFLAGS(sv) &= ~(SVf_IOK|SVf_NOK); | |
28e5dec8 | 2619 | #endif /* NV_PRESERVES_UV */ |
93a17b20 | 2620 | } |
79072805 | 2621 | else { |
f7877b28 | 2622 | if (isGV_with_GP(sv)) { |
159b6efe | 2623 | glob_2number(MUTABLE_GV(sv)); |
180488f8 NC |
2624 | return 0.0; |
2625 | } | |
2626 | ||
041457d9 | 2627 | if (!PL_localizing && !(SvFLAGS(sv) & SVs_PADTMP) && ckWARN(WARN_UNINITIALIZED)) |
29489e7c | 2628 | report_uninit(sv); |
7e25a7e9 NC |
2629 | assert (SvTYPE(sv) >= SVt_NV); |
2630 | /* Typically the caller expects that sv_any is not NULL now. */ | |
2631 | /* XXX Ilya implies that this is a bug in callers that assume this | |
2632 | and ideally should be fixed. */ | |
a0d0e21e | 2633 | return 0.0; |
79072805 | 2634 | } |
572bbb43 | 2635 | #if defined(USE_LONG_DOUBLE) |
097ee67d | 2636 | DEBUG_c({ |
f93f4e46 | 2637 | STORE_NUMERIC_LOCAL_SET_STANDARD(); |
1d7c1841 GS |
2638 | PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2nv(%" PERL_PRIgldbl ")\n", |
2639 | PTR2UV(sv), SvNVX(sv)); | |
572bbb43 GS |
2640 | RESTORE_NUMERIC_LOCAL(); |
2641 | }); | |
65202027 | 2642 | #else |
572bbb43 | 2643 | DEBUG_c({ |
f93f4e46 | 2644 | STORE_NUMERIC_LOCAL_SET_STANDARD(); |
1779d84d | 2645 | PerlIO_printf(Perl_debug_log, "0x%"UVxf" 1nv(%"NVgf")\n", |
1d7c1841 | 2646 | PTR2UV(sv), SvNVX(sv)); |
097ee67d JH |
2647 | RESTORE_NUMERIC_LOCAL(); |
2648 | }); | |
572bbb43 | 2649 | #endif |
463ee0b2 | 2650 | return SvNVX(sv); |
79072805 LW |
2651 | } |
2652 | ||
800401ee JH |
2653 | /* |
2654 | =for apidoc sv_2num | |
2655 | ||
2656 | Return an SV with the numeric value of the source SV, doing any necessary | |
a196a5fa JH |
2657 | reference or overload conversion. You must use the C<SvNUM(sv)> macro to |
2658 | access this function. | |
800401ee JH |
2659 | |
2660 | =cut | |
2661 | */ | |
2662 | ||
2663 | SV * | |
5de3775c | 2664 | Perl_sv_2num(pTHX_ register SV *const sv) |
800401ee | 2665 | { |
7918f24d NC |
2666 | PERL_ARGS_ASSERT_SV_2NUM; |
2667 | ||
b9ee0594 RGS |
2668 | if (!SvROK(sv)) |
2669 | return sv; | |
800401ee JH |
2670 | if (SvAMAGIC(sv)) { |
2671 | SV * const tmpsv = AMG_CALLun(sv,numer); | |
2672 | if (tmpsv && (!SvROK(tmpsv) || (SvRV(tmpsv) != SvRV(sv)))) | |
2673 | return sv_2num(tmpsv); | |
2674 | } | |
2675 | return sv_2mortal(newSVuv(PTR2UV(SvRV(sv)))); | |
2676 | } | |
2677 | ||
645c22ef DM |
2678 | /* uiv_2buf(): private routine for use by sv_2pv_flags(): print an IV or |
2679 | * UV as a string towards the end of buf, and return pointers to start and | |
2680 | * end of it. | |
2681 | * | |
2682 | * We assume that buf is at least TYPE_CHARS(UV) long. | |
2683 | */ | |
2684 | ||
864dbfa3 | 2685 | static char * |
5de3775c | 2686 | S_uiv_2buf(char *const buf, const IV iv, UV uv, const int is_uv, char **const peob) |
25da4f38 | 2687 | { |
25da4f38 | 2688 | char *ptr = buf + TYPE_CHARS(UV); |
823a54a3 | 2689 | char * const ebuf = ptr; |
25da4f38 | 2690 | int sign; |
25da4f38 | 2691 | |
7918f24d NC |
2692 | PERL_ARGS_ASSERT_UIV_2BUF; |
2693 | ||
25da4f38 IZ |
2694 | if (is_uv) |
2695 | sign = 0; | |
2696 | else if (iv >= 0) { | |
2697 | uv = iv; | |
2698 | sign = 0; | |
2699 | } else { | |
2700 | uv = -iv; | |
2701 | sign = 1; | |
2702 | } | |
2703 | do { | |
eb160463 | 2704 | *--ptr = '0' + (char)(uv % 10); |
25da4f38 IZ |
2705 | } while (uv /= 10); |
2706 | if (sign) | |
2707 | *--ptr = '-'; | |
2708 | *peob = ebuf; | |
2709 | return ptr; | |
2710 | } | |
2711 | ||
645c22ef DM |
2712 | /* |
2713 | =for apidoc sv_2pv_flags | |
2714 | ||
ff276b08 | 2715 | Returns a pointer to the string value of an SV, and sets *lp to its length. |
645c22ef DM |
2716 | If flags includes SV_GMAGIC, does an mg_get() first. Coerces sv to a string |
2717 | if necessary. | |
2718 | Normally invoked via the C<SvPV_flags> macro. C<sv_2pv()> and C<sv_2pv_nomg> | |
2719 | usually end up here too. | |
2720 | ||
2721 | =cut | |
2722 | */ | |
2723 | ||
8d6d96c1 | 2724 | char * |
5de3775c | 2725 | Perl_sv_2pv_flags(pTHX_ register SV *const sv, STRLEN *const lp, const I32 flags) |
8d6d96c1 | 2726 | { |
97aff369 | 2727 | dVAR; |
79072805 | 2728 | register char *s; |
79072805 | 2729 | |
463ee0b2 | 2730 | if (!sv) { |
cdb061a3 NC |
2731 | if (lp) |
2732 | *lp = 0; | |
73d840c0 | 2733 | return (char *)""; |
463ee0b2 | 2734 | } |
8990e307 | 2735 | if (SvGMAGICAL(sv)) { |
8d6d96c1 HS |
2736 | if (flags & SV_GMAGIC) |
2737 | mg_get(sv); | |
463ee0b2 | 2738 | if (SvPOKp(sv)) { |
cdb061a3 NC |
2739 | if (lp) |
2740 | *lp = SvCUR(sv); | |
10516c54 NC |
2741 | if (flags & SV_MUTABLE_RETURN) |
2742 | return SvPVX_mutable(sv); | |
4d84ee25 NC |
2743 | if (flags & SV_CONST_RETURN) |
2744 | return (char *)SvPVX_const(sv); | |
463ee0b2 LW |
2745 | return SvPVX(sv); |
2746 | } | |
75dfc8ec NC |
2747 | if (SvIOKp(sv) || SvNOKp(sv)) { |
2748 | char tbuf[64]; /* Must fit sprintf/Gconvert of longest IV/NV */ | |
75dfc8ec NC |
2749 | STRLEN len; |
2750 | ||
2751 | if (SvIOKp(sv)) { | |
e80fed9d | 2752 | len = SvIsUV(sv) |
d9fad198 JH |
2753 | ? my_snprintf(tbuf, sizeof(tbuf), "%"UVuf, (UV)SvUVX(sv)) |
2754 | : my_snprintf(tbuf, sizeof(tbuf), "%"IVdf, (IV)SvIVX(sv)); | |
75dfc8ec | 2755 | } else { |
e8ada2d0 NC |
2756 | Gconvert(SvNVX(sv), NV_DIG, 0, tbuf); |
2757 | len = strlen(tbuf); | |
75dfc8ec | 2758 | } |
b5b886f0 NC |
2759 | assert(!SvROK(sv)); |
2760 | { | |
75dfc8ec NC |
2761 | dVAR; |
2762 | ||
2763 | #ifdef FIXNEGATIVEZERO | |
e8ada2d0 NC |
2764 | if (len == 2 && tbuf[0] == '-' && tbuf[1] == '0') { |
2765 | tbuf[0] = '0'; | |
2766 | tbuf[1] = 0; | |
75dfc8ec NC |
2767 | len = 1; |
2768 | } | |
2769 | #endif | |
2770 | SvUPGRADE(sv, SVt_PV); | |
2771 | if (lp) | |
2772 | *lp = len; | |
2773 | s = SvGROW_mutable(sv, len + 1); | |
2774 | SvCUR_set(sv, len); | |
2775 | SvPOKp_on(sv); | |
10edeb5d | 2776 | return (char*)memcpy(s, tbuf, len + 1); |
75dfc8ec | 2777 | } |
463ee0b2 | 2778 | } |
1c7ff15e NC |
2779 | if (SvROK(sv)) { |
2780 | goto return_rok; | |
2781 | } | |
2782 | assert(SvTYPE(sv) >= SVt_PVMG); | |
2783 | /* This falls through to the report_uninit near the end of the | |
2784 | function. */ | |
2785 | } else if (SvTHINKFIRST(sv)) { | |
ed6116ce | 2786 | if (SvROK(sv)) { |
1c7ff15e | 2787 | return_rok: |
deb46114 NC |
2788 | if (SvAMAGIC(sv)) { |
2789 | SV *const tmpstr = AMG_CALLun(sv,string); | |
2790 | if (tmpstr && (!SvROK(tmpstr) || (SvRV(tmpstr) != SvRV(sv)))) { | |
2791 | /* Unwrap this: */ | |
2792 | /* char *pv = lp ? SvPV(tmpstr, *lp) : SvPV_nolen(tmpstr); | |
2793 | */ | |
2794 | ||
2795 | char *pv; | |
2796 | if ((SvFLAGS(tmpstr) & (SVf_POK)) == SVf_POK) { | |
2797 | if (flags & SV_CONST_RETURN) { | |
2798 | pv = (char *) SvPVX_const(tmpstr); | |
2799 | } else { | |
2800 | pv = (flags & SV_MUTABLE_RETURN) | |
2801 | ? SvPVX_mutable(tmpstr) : SvPVX(tmpstr); | |
2802 | } | |
2803 | if (lp) | |
2804 | *lp = SvCUR(tmpstr); | |
50adf7d2 | 2805 | } else { |
deb46114 | 2806 | pv = sv_2pv_flags(tmpstr, lp, flags); |
50adf7d2 | 2807 | } |
deb46114 NC |
2808 | if (SvUTF8(tmpstr)) |
2809 | SvUTF8_on(sv); | |
2810 | else | |
2811 | SvUTF8_off(sv); | |
2812 | return pv; | |
50adf7d2 | 2813 | } |
deb46114 NC |
2814 | } |
2815 | { | |
fafee734 NC |
2816 | STRLEN len; |
2817 | char *retval; | |
2818 | char *buffer; | |
d2c6dc5e | 2819 | SV *const referent = SvRV(sv); |
d8eae41e NC |
2820 | |
2821 | if (!referent) { | |
fafee734 NC |
2822 | len = 7; |
2823 | retval = buffer = savepvn("NULLREF", len); | |
5c35adbb | 2824 | } else if (SvTYPE(referent) == SVt_REGEXP) { |
d2c6dc5e | 2825 | REGEXP * const re = (REGEXP *)MUTABLE_PTR(referent); |
67d2d14d AB |
2826 | I32 seen_evals = 0; |
2827 | ||
2828 | assert(re); | |
2829 | ||
2830 | /* If the regex is UTF-8 we want the containing scalar to | |
2831 | have an UTF-8 flag too */ | |
2832 | if (RX_UTF8(re)) | |
2833 | SvUTF8_on(sv); | |
2834 | else | |
2835 | SvUTF8_off(sv); | |
2836 | ||
2837 | if ((seen_evals = RX_SEEN_EVALS(re))) | |
2838 | PL_reginterp_cnt += seen_evals; | |
2839 | ||
2840 | if (lp) | |
2841 | *lp = RX_WRAPLEN(re); | |
2842 | ||
2843 | return RX_WRAPPED(re); | |
d8eae41e NC |
2844 | } else { |
2845 | const char *const typestr = sv_reftype(referent, 0); | |
fafee734 NC |
2846 | const STRLEN typelen = strlen(typestr); |
2847 | UV addr = PTR2UV(referent); | |
2848 | const char *stashname = NULL; | |
2849 | STRLEN stashnamelen = 0; /* hush, gcc */ | |
2850 | const char *buffer_end; | |
d8eae41e | 2851 | |
d8eae41e | 2852 | if (SvOBJECT(referent)) { |
fafee734 NC |
2853 | const HEK *const name = HvNAME_HEK(SvSTASH(referent)); |
2854 | ||
2855 | if (name) { | |
2856 | stashname = HEK_KEY(name); | |
2857 | stashnamelen = HEK_LEN(name); | |
2858 | ||
2859 | if (HEK_UTF8(name)) { | |
2860 | SvUTF8_on(sv); | |
2861 | } else { | |
2862 | SvUTF8_off(sv); | |
2863 | } | |
2864 | } else { | |
2865 | stashname = "__ANON__"; | |
2866 | stashnamelen = 8; | |
2867 | } | |
2868 | len = stashnamelen + 1 /* = */ + typelen + 3 /* (0x */ | |
2869 | + 2 * sizeof(UV) + 2 /* )\0 */; | |
2870 | } else { | |
2871 | len = typelen + 3 /* (0x */ | |
2872 | + 2 * sizeof(UV) + 2 /* )\0 */; | |
d8eae41e | 2873 | } |
fafee734 NC |
2874 | |
2875 | Newx(buffer, len, char); | |
2876 | buffer_end = retval = buffer + len; | |
2877 | ||
2878 | /* Working backwards */ | |
2879 | *--retval = '\0'; | |
2880 | *--retval = ')'; | |
2881 | do { | |
2882 | *--retval = PL_hexdigit[addr & 15]; | |
2883 | } while (addr >>= 4); | |
2884 | *--retval = 'x'; | |
2885 | *--retval = '0'; | |
2886 | *--retval = '('; | |
2887 | ||
2888 | retval -= typelen; | |
2889 | memcpy(retval, typestr, typelen); | |
2890 | ||
2891 | if (stashname) { | |
2892 | *--retval = '='; | |
2893 | retval -= stashnamelen; | |
2894 | memcpy(retval, stashname, stashnamelen); | |
2895 | } | |
2896 | /* retval may not neccesarily have reached the start of the | |
2897 | buffer here. */ | |
2898 | assert (retval >= buffer); | |
2899 | ||
2900 | len = buffer_end - retval - 1; /* -1 for that \0 */ | |
c080367d | 2901 | } |
042dae7a | 2902 | if (lp) |
fafee734 NC |
2903 | *lp = len; |
2904 | SAVEFREEPV(buffer); | |
2905 | return retval; | |
463ee0b2 | 2906 | } |
79072805 | 2907 | } |
0336b60e | 2908 | if (SvREADONLY(sv) && !SvOK(sv)) { |
cdb061a3 NC |
2909 | if (lp) |
2910 | *lp = 0; | |
9f621bb0 NC |
2911 | if (flags & SV_UNDEF_RETURNS_NULL) |
2912 | return NULL; | |
2913 | if (ckWARN(WARN_UNINITIALIZED)) | |
2914 | report_uninit(sv); | |
73d840c0 | 2915 | return (char *)""; |
79072805 | 2916 | } |
79072805 | 2917 | } |
28e5dec8 JH |
2918 | if (SvIOK(sv) || ((SvIOKp(sv) && !SvNOKp(sv)))) { |
2919 | /* I'm assuming that if both IV and NV are equally valid then | |
2920 | converting the IV is going to be more efficient */ | |
e1ec3a88 | 2921 | const U32 isUIOK = SvIsUV(sv); |
28e5dec8 JH |
2922 | char buf[TYPE_CHARS(UV)]; |
2923 | char *ebuf, *ptr; | |
97a130b8 | 2924 | STRLEN len; |
28e5dec8 JH |
2925 | |
2926 | if (SvTYPE(sv) < SVt_PVIV) | |
2927 | sv_upgrade(sv, SVt_PVIV); | |
4ea1d550 | 2928 | ptr = uiv_2buf(buf, SvIVX(sv), SvUVX(sv), isUIOK, &ebuf); |
97a130b8 | 2929 | len = ebuf - ptr; |
5902b6a9 | 2930 | /* inlined from sv_setpvn */ |
97a130b8 NC |
2931 | s = SvGROW_mutable(sv, len + 1); |
2932 | Move(ptr, s, len, char); | |
2933 | s += len; | |
28e5dec8 | 2934 | *s = '\0'; |
28e5dec8 JH |
2935 | } |
2936 | else if (SvNOKp(sv)) { | |
4ee39169 | 2937 | dSAVE_ERRNO; |
79072805 LW |
2938 | if (SvTYPE(sv) < SVt_PVNV) |
2939 | sv_upgrade(sv, SVt_PVNV); | |
1c846c1f | 2940 | /* The +20 is pure guesswork. Configure test needed. --jhi */ |
5902b6a9 | 2941 | s = SvGROW_mutable(sv, NV_DIG + 20); |
c81271c3 | 2942 | /* some Xenix systems wipe out errno here */ |
79072805 | 2943 | #ifdef apollo |
463ee0b2 | 2944 | if (SvNVX(sv) == 0.0) |
d1307786 | 2945 | my_strlcpy(s, "0", SvLEN(sv)); |
79072805 LW |
2946 | else |
2947 | #endif /*apollo*/ | |
bbce6d69 | 2948 | { |
2d4389e4 | 2949 | Gconvert(SvNVX(sv), NV_DIG, 0, s); |
bbce6d69 | 2950 | } |
4ee39169 | 2951 | RESTORE_ERRNO; |
a0d0e21e | 2952 | #ifdef FIXNEGATIVEZERO |
20773dcd NC |
2953 | if (*s == '-' && s[1] == '0' && !s[2]) { |
2954 | s[0] = '0'; | |
2955 | s[1] = 0; | |
2956 | } | |
a0d0e21e | 2957 | #endif |
79072805 LW |
2958 | while (*s) s++; |
2959 | #ifdef hcx | |
2960 | if (s[-1] == '.') | |
46fc3d4c | 2961 | *--s = '\0'; |
79072805 LW |
2962 | #endif |
2963 | } | |
79072805 | 2964 | else { |
8d1c3e26 NC |
2965 | if (isGV_with_GP(sv)) { |
2966 | GV *const gv = MUTABLE_GV(sv); | |
2967 | const U32 wasfake = SvFLAGS(gv) & SVf_FAKE; | |
2968 | SV *const buffer = sv_newmortal(); | |
2969 | ||
2970 | /* FAKE globs can get coerced, so need to turn this off temporarily | |
2971 | if it is on. */ | |
2972 | SvFAKE_off(gv); | |
2973 | gv_efullname3(buffer, gv, "*"); | |
2974 | SvFLAGS(gv) |= wasfake; | |
2975 | ||
1809c940 DM |
2976 | if (SvPOK(buffer)) { |
2977 | if (lp) { | |
2978 | *lp = SvCUR(buffer); | |
2979 | } | |
2980 | return SvPVX(buffer); | |
2981 | } | |
2982 | else { | |
2983 | if (lp) | |
2984 | *lp = 0; | |
2985 | return (char *)""; | |
8d1c3e26 | 2986 | } |
8d1c3e26 | 2987 | } |
180488f8 | 2988 | |
cdb061a3 | 2989 | if (lp) |
00b6aa41 | 2990 | *lp = 0; |
9f621bb0 NC |
2991 | if (flags & SV_UNDEF_RETURNS_NULL) |
2992 | return NULL; | |
2993 | if (!PL_localizing && !(SvFLAGS(sv) & SVs_PADTMP) && ckWARN(WARN_UNINITIALIZED)) | |
2994 | report_uninit(sv); | |
25da4f38 IZ |
2995 | if (SvTYPE(sv) < SVt_PV) |
2996 | /* Typically the caller expects that sv_any is not NULL now. */ | |
2997 | sv_upgrade(sv, SVt_PV); | |
73d840c0 | 2998 | return (char *)""; |
79072805 | 2999 | } |
cdb061a3 | 3000 | { |
823a54a3 | 3001 | const STRLEN len = s - SvPVX_const(sv); |
cdb061a3 NC |
3002 | if (lp) |
3003 | *lp = len; | |
3004 | SvCUR_set(sv, len); | |
3005 | } | |
79072805 | 3006 | SvPOK_on(sv); |
1d7c1841 | 3007 | DEBUG_c(PerlIO_printf(Perl_debug_log, "0x%"UVxf" 2pv(%s)\n", |
3f7c398e | 3008 | PTR2UV(sv),SvPVX_const(sv))); |
4d84ee25 NC |
3009 | if (flags & SV_CONST_RETURN) |
3010 | return (char *)SvPVX_const(sv); | |
10516c54 NC |
3011 | if (flags & SV_MUTABLE_RETURN) |
3012 | return SvPVX_mutable(sv); | |
463ee0b2 LW |
3013 | return SvPVX(sv); |
3014 | } | |
3015 | ||
645c22ef | 3016 | /* |
6050d10e JP |
3017 | =for apidoc sv_copypv |
3018 | ||
3019 | Copies a stringified representation of the source SV into the | |
3020 | destination SV. Automatically performs any necessary mg_get and | |
54f0641b | 3021 | coercion of numeric values into strings. Guaranteed to preserve |
2575c402 | 3022 | UTF8 flag even from overloaded objects. Similar in nature to |
54f0641b NIS |
3023 | sv_2pv[_flags] but operates directly on an SV instead of just the |
3024 | string. Mostly uses sv_2pv_flags to do its work, except when that | |
6050d10e JP |
3025 | would lose the UTF-8'ness of the PV. |
3026 | ||
3027 | =cut | |
3028 | */ | |
3029 | ||
3030 | void | |
5de3775c | 3031 | Perl_sv_copypv(pTHX_ SV *const dsv, register SV *const ssv) |
6050d10e | 3032 | { |
446eaa42 | 3033 | STRLEN len; |
53c1dcc0 | 3034 | const char * const s = SvPV_const(ssv,len); |
7918f24d NC |
3035 | |
3036 | PERL_ARGS_ASSERT_SV_COPYPV; | |
3037 | ||
cb50f42d | 3038 | sv_setpvn(dsv,s,len); |
446eaa42 | 3039 | if (SvUTF8(ssv)) |
cb50f42d | 3040 | SvUTF8_on(dsv); |
446eaa42 | 3041 | else |
cb50f42d | 3042 | SvUTF8_off(dsv); |
6050d10e JP |
3043 | } |
3044 | ||
3045 | /* | |
645c22ef DM |
3046 | =for apidoc sv_2pvbyte |
3047 | ||
3048 | Return a pointer to the byte-encoded representation of the SV, and set *lp | |
1e54db1a | 3049 | to its length. May cause the SV to be downgraded from UTF-8 as a |
645c22ef DM |
3050 | side-effect. |
3051 | ||
3052 | Usually accessed via the C<SvPVbyte> macro. | |
3053 | ||
3054 | =cut | |
3055 | */ | |
3056 | ||
7340a771 | 3057 | char * |
5de3775c | 3058 | Perl_sv_2pvbyte(pTHX_ register SV *const sv, STRLEN *const lp) |
7340a771 | 3059 | { |
7918f24d NC |
3060 | PERL_ARGS_ASSERT_SV_2PVBYTE; |
3061 | ||
0875d2fe | 3062 | sv_utf8_downgrade(sv,0); |
97972285 | 3063 | return lp ? SvPV(sv,*lp) : SvPV_nolen(sv); |
7340a771 GS |
3064 | } |
3065 | ||
645c22ef | 3066 | /* |
035cbb0e RGS |
3067 | =for apidoc sv_2pvutf8 |
3068 | ||
3069 | Return a pointer to the UTF-8-encoded representation of the SV, and set *lp | |
3070 | to its length. May cause the SV to be upgraded to UTF-8 as a side-effect. | |
3071 | ||
3072 | Usually accessed via the C<SvPVutf8> macro. | |
3073 | ||
3074 | =cut | |
3075 | */ | |
645c22ef | 3076 | |
7340a771 | 3077 | char * |
7bc54cea | 3078 | Perl_sv_2pvutf8(pTHX_ register SV *const sv, STRLEN *const lp) |
7340a771 | 3079 | { |
7918f24d NC |
3080 | PERL_ARGS_ASSERT_SV_2PVUTF8; |
3081 | ||
035cbb0e RGS |
3082 | sv_utf8_upgrade(sv); |
3083 | return lp ? SvPV(sv,*lp) : SvPV_nolen(sv); | |
7340a771 | 3084 | } |
1c846c1f | 3085 | |
7ee2227d | 3086 | |
645c22ef DM |
3087 | /* |
3088 | =for apidoc sv_2bool | |
3089 | ||
3090 | This function is only called on magical items, and is only used by | |
8cf8f3d1 | 3091 | sv_true() or its macro equivalent. |
645c22ef DM |
3092 | |
3093 | =cut | |
3094 | */ | |
3095 | ||
463ee0b2 | 3096 | bool |
7bc54cea | 3097 | Perl_sv_2bool(pTHX_ register SV *const sv) |
463ee0b2 | 3098 | { |
97aff369 | 3099 | dVAR; |
7918f24d NC |
3100 | |
3101 | PERL_ARGS_ASSERT_SV_2BOOL; | |
3102 | ||
5b295bef | 3103 | SvGETMAGIC(sv); |
463ee0b2 | 3104 | |
a0d0e21e LW |
3105 | if (!SvOK(sv)) |
3106 | return 0; | |
3107 | if (SvROK(sv)) { | |
fabdb6c0 AL |
3108 | if (SvAMAGIC(sv)) { |
3109 | SV * const tmpsv = AMG_CALLun(sv,bool_); | |
3110 | if (tmpsv && (!SvROK(tmpsv) || (SvRV(tmpsv) != SvRV(sv)))) | |
f2338a2e | 3111 | return cBOOL(SvTRUE(tmpsv)); |
fabdb6c0 AL |
3112 | } |
3113 | return SvRV(sv) != 0; | |
a0d0e21e | 3114 | } |
463ee0b2 | 3115 | if (SvPOKp(sv)) { |
53c1dcc0 AL |
3116 | register XPV* const Xpvtmp = (XPV*)SvANY(sv); |
3117 | if (Xpvtmp && | |
339049b0 | 3118 | (*sv->sv_u.svu_pv > '0' || |
11343788 | 3119 | Xpvtmp->xpv_cur > 1 || |
339049b0 | 3120 | (Xpvtmp->xpv_cur && *sv->sv_u.svu_pv != '0'))) |
463ee0b2 LW |
3121 | return 1; |
3122 | else | |
3123 | return 0; | |
3124 | } | |
3125 | else { | |
3126 | if (SvIOKp(sv)) | |
3127 | return SvIVX(sv) != 0; | |
3128 | else { | |
3129 | if (SvNOKp(sv)) | |
3130 | return SvNVX(sv) != 0.0; | |
180488f8 | 3131 | else { |
f7877b28 | 3132 | if (isGV_with_GP(sv)) |
180488f8 NC |
3133 | return TRUE; |
3134 | else | |
3135 | return FALSE; | |
3136 | } | |
463ee0b2 LW |
3137 | } |
3138 | } | |
79072805 LW |
3139 | } |
3140 | ||
c461cf8f JH |
3141 | /* |
3142 | =for apidoc sv_utf8_upgrade | |
3143 | ||
78ea37eb | 3144 | Converts the PV of an SV to its UTF-8-encoded form. |
645c22ef | 3145 | Forces the SV to string form if it is not already. |
2bbc8d55 | 3146 | Will C<mg_get> on C<sv> if appropriate. |
4411f3b6 | 3147 | Always sets the SvUTF8 flag to avoid future validity checks even |
2bbc8d55 SP |
3148 | if the whole string is the same in UTF-8 as not. |
3149 | Returns the number of bytes in the converted string | |
c461cf8f | 3150 | |
13a6c0e0 JH |
3151 | This is not as a general purpose byte encoding to Unicode interface: |
3152 | use the Encode extension for that. | |
3153 | ||
fe749c9a KW |
3154 | =for apidoc sv_utf8_upgrade_nomg |
3155 | ||
3156 | Like sv_utf8_upgrade, but doesn't do magic on C<sv> | |
3157 | ||
8d6d96c1 HS |
3158 | =for apidoc sv_utf8_upgrade_flags |
3159 | ||
78ea37eb | 3160 | Converts the PV of an SV to its UTF-8-encoded form. |
645c22ef | 3161 | Forces the SV to string form if it is not already. |
8d6d96c1 | 3162 | Always sets the SvUTF8 flag to avoid future validity checks even |
2bbc8d55 SP |
3163 | if all the bytes are invariant in UTF-8. If C<flags> has C<SV_GMAGIC> bit set, |
3164 | will C<mg_get> on C<sv> if appropriate, else not. | |
3165 | Returns the number of bytes in the converted string | |
3166 | C<sv_utf8_upgrade> and | |
8d6d96c1 HS |
3167 | C<sv_utf8_upgrade_nomg> are implemented in terms of this function. |
3168 | ||
13a6c0e0 JH |
3169 | This is not as a general purpose byte encoding to Unicode interface: |
3170 | use the Encode extension for that. | |
3171 | ||
8d6d96c1 | 3172 | =cut |
b3ab6785 KW |
3173 | |
3174 | The grow version is currently not externally documented. It adds a parameter, | |
3175 | extra, which is the number of unused bytes the string of 'sv' is guaranteed to | |
3176 | have free after it upon return. This allows the caller to reserve extra space | |
3177 | that it intends to fill, to avoid extra grows. | |
3178 | ||
3179 | Also externally undocumented for the moment is the flag SV_FORCE_UTF8_UPGRADE, | |
3180 | which can be used to tell this function to not first check to see if there are | |
3181 | any characters that are different in UTF-8 (variant characters) which would | |
3182 | force it to allocate a new string to sv, but to assume there are. Typically | |
3183 | this flag is used by a routine that has already parsed the string to find that | |
3184 | there are such characters, and passes this information on so that the work | |
3185 | doesn't have to be repeated. | |
3186 | ||
3187 | (One might think that the calling routine could pass in the position of the | |
3188 | first such variant, so it wouldn't have to be found again. But that is not the | |
3189 | case, because typically when the caller is likely to use this flag, it won't be | |
3190 | calling this routine unless it finds something that won't fit into a byte. | |
3191 | Otherwise it tries to not upgrade and just use bytes. But some things that | |
3192 | do fit into a byte are variants in utf8, and the caller may not have been | |
3193 | keeping track of these.) | |
3194 | ||
3195 | If the routine itself changes the string, it adds a trailing NUL. Such a NUL | |
3196 | isn't guaranteed due to having other routines do the work in some input cases, | |
3197 | or if the input is already flagged as being in utf8. | |
3198 | ||
3199 | The speed of this could perhaps be improved for many cases if someone wanted to | |
3200 | write a fast function that counts the number of variant characters in a string, | |
3201 | especially if it could return the position of the first one. | |
3202 | ||
8d6d96c1 HS |
3203 | */ |
3204 | ||
3205 | STRLEN | |
b3ab6785 | 3206 | Perl_sv_utf8_upgrade_flags_grow(pTHX_ register SV *const sv, const I32 flags, STRLEN extra) |
8d6d96c1 | 3207 | { |
97aff369 | 3208 | dVAR; |
7918f24d | 3209 | |
b3ab6785 | 3210 | PERL_ARGS_ASSERT_SV_UTF8_UPGRADE_FLAGS_GROW; |
7918f24d | 3211 | |
808c356f RGS |
3212 | if (sv == &PL_sv_undef) |
3213 | return 0; | |
e0e62c2a NIS |
3214 | if (!SvPOK(sv)) { |
3215 | STRLEN len = 0; | |
d52b7888 NC |
3216 | if (SvREADONLY(sv) && (SvPOKp(sv) || SvIOKp(sv) || SvNOKp(sv))) { |
3217 | (void) sv_2pv_flags(sv,&len, flags); | |
b3ab6785 KW |
3218 | if (SvUTF8(sv)) { |
3219 | if (extra) SvGROW(sv, SvCUR(sv) + extra); | |
d52b7888 | 3220 | return len; |
b3ab6785 | 3221 | } |
d52b7888 NC |
3222 | } else { |
3223 | (void) SvPV_force(sv,len); | |
3224 | } | |
e0e62c2a | 3225 | } |
4411f3b6 | 3226 | |
f5cee72b | 3227 | if (SvUTF8(sv)) { |
b3ab6785 | 3228 | if (extra) SvGROW(sv, SvCUR(sv) + extra); |
5fec3b1d | 3229 | return SvCUR(sv); |
f5cee72b | 3230 | } |
5fec3b1d | 3231 | |
765f542d NC |
3232 | if (SvIsCOW(sv)) { |
3233 | sv_force_normal_flags(sv, 0); | |
db42d148 NIS |
3234 | } |
3235 | ||
b3ab6785 | 3236 | if (PL_encoding && !(flags & SV_UTF8_NO_ENCODING)) { |
799ef3cb | 3237 | sv_recode_to_utf8(sv, PL_encoding); |
b3ab6785 KW |
3238 | if (extra) SvGROW(sv, SvCUR(sv) + extra); |
3239 | return SvCUR(sv); | |
3240 | } | |
3241 | ||
4e93345f KW |
3242 | if (SvCUR(sv) == 0) { |
3243 | if (extra) SvGROW(sv, extra); | |
3244 | } else { /* Assume Latin-1/EBCDIC */ | |
c4e7c712 | 3245 | /* This function could be much more efficient if we |
2bbc8d55 | 3246 | * had a FLAG in SVs to signal if there are any variant |
c4e7c712 | 3247 | * chars in the PV. Given that there isn't such a flag |
b3ab6785 KW |
3248 | * make the loop as fast as possible (although there are certainly ways |
3249 | * to speed this up, eg. through vectorization) */ | |
3250 | U8 * s = (U8 *) SvPVX_const(sv); | |
3251 | U8 * e = (U8 *) SvEND(sv); | |
3252 | U8 *t = s; | |
3253 | STRLEN two_byte_count = 0; | |
c4e7c712 | 3254 | |
b3ab6785 KW |
3255 | if (flags & SV_FORCE_UTF8_UPGRADE) goto must_be_utf8; |
3256 | ||
3257 | /* See if really will need to convert to utf8. We mustn't rely on our | |
3258 | * incoming SV being well formed and having a trailing '\0', as certain | |
3259 | * code in pp_formline can send us partially built SVs. */ | |
3260 | ||
c4e7c712 | 3261 | while (t < e) { |
53c1dcc0 | 3262 | const U8 ch = *t++; |
b3ab6785 KW |
3263 | if (NATIVE_IS_INVARIANT(ch)) continue; |
3264 | ||
3265 | t--; /* t already incremented; re-point to first variant */ | |
3266 | two_byte_count = 1; | |
3267 | goto must_be_utf8; | |
c4e7c712 | 3268 | } |
b3ab6785 KW |
3269 | |
3270 | /* utf8 conversion not needed because all are invariants. Mark as | |
3271 | * UTF-8 even if no variant - saves scanning loop */ | |
c4e7c712 | 3272 | SvUTF8_on(sv); |
b3ab6785 KW |
3273 | return SvCUR(sv); |
3274 | ||
3275 | must_be_utf8: | |
3276 | ||
3277 | /* Here, the string should be converted to utf8, either because of an | |
3278 | * input flag (two_byte_count = 0), or because a character that | |
3279 | * requires 2 bytes was found (two_byte_count = 1). t points either to | |
3280 | * the beginning of the string (if we didn't examine anything), or to | |
3281 | * the first variant. In either case, everything from s to t - 1 will | |
3282 | * occupy only 1 byte each on output. | |
3283 | * | |
3284 | * There are two main ways to convert. One is to create a new string | |
3285 | * and go through the input starting from the beginning, appending each | |
3286 | * converted value onto the new string as we go along. It's probably | |
3287 | * best to allocate enough space in the string for the worst possible | |
3288 | * case rather than possibly running out of space and having to | |
3289 | * reallocate and then copy what we've done so far. Since everything | |
3290 | * from s to t - 1 is invariant, the destination can be initialized | |
3291 | * with these using a fast memory copy | |
3292 | * | |
3293 | * The other way is to figure out exactly how big the string should be | |
3294 | * by parsing the entire input. Then you don't have to make it big | |
3295 | * enough to handle the worst possible case, and more importantly, if | |
3296 | * the string you already have is large enough, you don't have to | |
3297 | * allocate a new string, you can copy the last character in the input | |
3298 | * string to the final position(s) that will be occupied by the | |
3299 | * converted string and go backwards, stopping at t, since everything | |
3300 | * before that is invariant. | |
3301 | * | |
3302 | * There are advantages and disadvantages to each method. | |
3303 | * | |
3304 | * In the first method, we can allocate a new string, do the memory | |
3305 | * copy from the s to t - 1, and then proceed through the rest of the | |
3306 | * string byte-by-byte. | |
3307 | * | |
3308 | * In the second method, we proceed through the rest of the input | |
3309 | * string just calculating how big the converted string will be. Then | |
3310 | * there are two cases: | |
3311 | * 1) if the string has enough extra space to handle the converted | |
3312 | * value. We go backwards through the string, converting until we | |
3313 | * get to the position we are at now, and then stop. If this | |
3314 | * position is far enough along in the string, this method is | |
3315 | * faster than the other method. If the memory copy were the same | |
3316 | * speed as the byte-by-byte loop, that position would be about | |
3317 | * half-way, as at the half-way mark, parsing to the end and back | |
3318 | * is one complete string's parse, the same amount as starting | |
3319 | * over and going all the way through. Actually, it would be | |
3320 | * somewhat less than half-way, as it's faster to just count bytes | |
3321 | * than to also copy, and we don't have the overhead of allocating | |
3322 | * a new string, changing the scalar to use it, and freeing the | |
3323 | * existing one. But if the memory copy is fast, the break-even | |
3324 | * point is somewhere after half way. The counting loop could be | |
3325 | * sped up by vectorization, etc, to move the break-even point | |
3326 | * further towards the beginning. | |
3327 | * 2) if the string doesn't have enough space to handle the converted | |
3328 | * value. A new string will have to be allocated, and one might | |
3329 | * as well, given that, start from the beginning doing the first | |
3330 | * method. We've spent extra time parsing the string and in | |
3331 | * exchange all we've gotten is that we know precisely how big to | |
3332 | * make the new one. Perl is more optimized for time than space, | |
3333 | * so this case is a loser. | |
3334 | * So what I've decided to do is not use the 2nd method unless it is | |
3335 | * guaranteed that a new string won't have to be allocated, assuming | |
3336 | * the worst case. I also decided not to put any more conditions on it | |
3337 | * than this, for now. It seems likely that, since the worst case is | |
3338 | * twice as big as the unknown portion of the string (plus 1), we won't | |
3339 | * be guaranteed enough space, causing us to go to the first method, | |
3340 | * unless the string is short, or the first variant character is near | |
3341 | * the end of it. In either of these cases, it seems best to use the | |
3342 | * 2nd method. The only circumstance I can think of where this would | |
3343 | * be really slower is if the string had once had much more data in it | |
3344 | * than it does now, but there is still a substantial amount in it */ | |
3345 | ||
3346 | { | |
3347 | STRLEN invariant_head = t - s; | |
3348 | STRLEN size = invariant_head + (e - t) * 2 + 1 + extra; | |
3349 | if (SvLEN(sv) < size) { | |
3350 | ||
3351 | /* Here, have decided to allocate a new string */ | |
3352 | ||
3353 | U8 *dst; | |
3354 | U8 *d; | |
3355 | ||
3356 | Newx(dst, size, U8); | |
3357 | ||
3358 | /* If no known invariants at the beginning of the input string, | |
3359 | * set so starts from there. Otherwise, can use memory copy to | |
3360 | * get up to where we are now, and then start from here */ | |
3361 | ||
3362 | if (invariant_head <= 0) { | |
3363 | d = dst; | |
3364 | } else { | |
3365 | Copy(s, dst, invariant_head, char); | |
3366 | d = dst + invariant_head; | |
3367 | } | |
3368 | ||
3369 | while (t < e) { | |
3370 | const UV uv = NATIVE8_TO_UNI(*t++); | |
3371 | if (UNI_IS_INVARIANT(uv)) | |
3372 | *d++ = (U8)UNI_TO_NATIVE(uv); | |
3373 | else { | |
3374 | *d++ = (U8)UTF8_EIGHT_BIT_HI(uv); | |
3375 | *d++ = (U8)UTF8_EIGHT_BIT_LO(uv); | |
3376 | } | |
3377 | } | |
3378 | *d = '\0'; | |
3379 | SvPV_free(sv); /* No longer using pre-existing string */ | |
3380 | SvPV_set(sv, (char*)dst); | |
3381 | SvCUR_set(sv, d - dst); | |
3382 | SvLEN_set(sv, size); | |
3383 | } else { | |
3384 | ||
3385 | /* Here, have decided to get the exact size of the string. | |
3386 | * Currently this happens only when we know that there is | |
3387 | * guaranteed enough space to fit the converted string, so | |
3388 | * don't have to worry about growing. If two_byte_count is 0, | |
3389 | * then t points to the first byte of the string which hasn't | |
3390 | * been examined yet. Otherwise two_byte_count is 1, and t | |
3391 | * points to the first byte in the string that will expand to | |
3392 | * two. Depending on this, start examining at t or 1 after t. | |
3393 | * */ | |
3394 | ||
3395 | U8 *d = t + two_byte_count; | |
3396 | ||
3397 | ||
3398 | /* Count up the remaining bytes that expand to two */ | |
3399 | ||
3400 | while (d < e) { | |
3401 | const U8 chr = *d++; | |
3402 | if (! NATIVE_IS_INVARIANT(chr)) two_byte_count++; | |
3403 | } | |
3404 | ||
3405 | /* The string will expand by just the number of bytes that | |
3406 | * occupy two positions. But we are one afterwards because of | |
3407 | * the increment just above. This is the place to put the | |
3408 | * trailing NUL, and to set the length before we decrement */ | |
3409 | ||
3410 | d += two_byte_count; | |
3411 | SvCUR_set(sv, d - s); | |
3412 | *d-- = '\0'; | |
3413 | ||
3414 | ||
3415 | /* Having decremented d, it points to the position to put the | |
3416 | * very last byte of the expanded string. Go backwards through | |
3417 | * the string, copying and expanding as we go, stopping when we | |
3418 | * get to the part that is invariant the rest of the way down */ | |
3419 | ||
3420 | e--; | |
3421 | while (e >= t) { | |
3422 | const U8 ch = NATIVE8_TO_UNI(*e--); | |
3423 | if (UNI_IS_INVARIANT(ch)) { | |
3424 | *d-- = UNI_TO_NATIVE(ch); | |
3425 | } else { | |
3426 | *d-- = (U8)UTF8_EIGHT_BIT_LO(ch); | |
3427 | *d-- = (U8)UTF8_EIGHT_BIT_HI(ch); | |
3428 | } | |
3429 | } | |
3430 | } | |
3431 | } | |
560a288e | 3432 | } |
b3ab6785 KW |
3433 | |
3434 | /* Mark as UTF-8 even if no variant - saves scanning loop */ | |
3435 | SvUTF8_on(sv); | |
4411f3b6 | 3436 | return SvCUR(sv); |
560a288e GS |
3437 | } |
3438 | ||
c461cf8f JH |
3439 | /* |
3440 | =for apidoc sv_utf8_downgrade | |
3441 | ||
78ea37eb | 3442 | Attempts to convert the PV of an SV from characters to bytes. |
2bbc8d55 SP |
3443 | If the PV contains a character that cannot fit |
3444 | in a byte, this conversion will fail; | |
78ea37eb | 3445 | in this case, either returns false or, if C<fail_ok> is not |
c461cf8f JH |
3446 | true, croaks. |
3447 | ||
13a6c0e0 JH |
3448 | This is not as a general purpose Unicode to byte encoding interface: |
3449 | use the Encode extension for that. | |
3450 | ||
c461cf8f JH |
3451 | =cut |
3452 | */ | |
3453 | ||
560a288e | 3454 | bool |
7bc54cea | 3455 | Perl_sv_utf8_downgrade(pTHX_ register SV *const sv, const bool fail_ok) |
560a288e | 3456 | { |
97aff369 | 3457 | dVAR; |
7918f24d NC |
3458 | |
3459 | PERL_ARGS_ASSERT_SV_UTF8_DOWNGRADE; | |
3460 | ||
78ea37eb | 3461 | if (SvPOKp(sv) && SvUTF8(sv)) { |
fa301091 | 3462 | if (SvCUR(sv)) { |
03cfe0ae | 3463 | U8 *s; |
652088fc | 3464 | STRLEN len; |
fa301091 | 3465 | |
765f542d NC |
3466 | if (SvIsCOW(sv)) { |
3467 | sv_force_normal_flags(sv, 0); | |
3468 | } | |
03cfe0ae NIS |
3469 | s = (U8 *) SvPV(sv, len); |
3470 | if (!utf8_to_bytes(s, &len)) { | |
fa301091 JH |
3471 | if (fail_ok) |
3472 | return FALSE; | |
3473 | else { | |
3474 | if (PL_op) | |
3475 | Perl_croak(aTHX_ "Wide character in %s", | |
53e06cf0 | 3476 | OP_DESC(PL_op)); |
fa301091 JH |
3477 | else |
3478 | Perl_croak(aTHX_ "Wide character"); | |
3479 | } | |
4b3603a4 | 3480 | } |
b162af07 | 3481 | SvCUR_set(sv, len); |
67e989fb | 3482 | } |
560a288e | 3483 | } |
ffebcc3e | 3484 | SvUTF8_off(sv); |
560a288e GS |
3485 | return TRUE; |
3486 | } | |
3487 | ||
c461cf8f JH |
3488 | /* |
3489 | =for apidoc sv_utf8_encode | |
3490 | ||
78ea37eb TS |
3491 | Converts the PV of an SV to UTF-8, but then turns the C<SvUTF8> |
3492 | flag off so that it looks like octets again. | |
c461cf8f JH |
3493 | |
3494 | =cut | |
3495 | */ | |
3496 | ||
560a288e | 3497 | void |
7bc54cea | 3498 | Perl_sv_utf8_encode(pTHX_ register SV *const sv) |
560a288e | 3499 | { |
7918f24d NC |
3500 | PERL_ARGS_ASSERT_SV_UTF8_ENCODE; |
3501 | ||
4c94c214 NC |
3502 | if (SvIsCOW(sv)) { |
3503 | sv_force_normal_flags(sv, 0); | |
3504 | } | |
3505 | if (SvREADONLY(sv)) { | |
f1f66076 | 3506 | Perl_croak(aTHX_ "%s", PL_no_modify); |
4c94c214 | 3507 | } |
a5f5288a | 3508 | (void) sv_utf8_upgrade(sv); |
560a288e GS |
3509 | SvUTF8_off(sv); |
3510 | } | |
3511 | ||
4411f3b6 NIS |
3512 | /* |
3513 | =for apidoc sv_utf8_decode | |
3514 | ||
78ea37eb TS |
3515 | If the PV of the SV is an octet sequence in UTF-8 |
3516 | and contains a multiple-byte character, the C<SvUTF8> flag is turned on | |
3517 | so that it looks like a character. If the PV contains only single-byte | |
3518 | characters, the C<SvUTF8> flag stays being off. | |
3519 | Scans PV for validity and returns false if the PV is invalid UTF-8. | |
4411f3b6 NIS |
3520 | |
3521 | =cut | |
3522 | */ | |
3523 | ||
560a288e | 3524 | bool |
7bc54cea | 3525 | Perl_sv_utf8_decode(pTHX_ register SV *const sv) |
560a288e | 3526 | { |
7918f24d NC |
3527 | PERL_ARGS_ASSERT_SV_UTF8_DECODE; |
3528 | ||
78ea37eb | 3529 | if (SvPOKp(sv)) { |
93524f2b NC |
3530 | const U8 *c; |
3531 | const U8 *e; | |
9cbac4c7 | 3532 | |
645c22ef DM |
3533 | /* The octets may have got themselves encoded - get them back as |
3534 | * bytes | |
3535 | */ | |
3536 | if (!sv_utf8_downgrade(sv, TRUE)) | |
560a288e GS |
3537 | return FALSE; |
3538 | ||
3539 | /* it is actually just a matter of turning the utf8 flag on, but | |
3540 | * we want to make sure everything inside is valid utf8 first. | |
3541 | */ | |
93524f2b | 3542 | c = (const U8 *) SvPVX_const(sv); |
63cd0674 | 3543 | if (!is_utf8_string(c, SvCUR(sv)+1)) |
67e989fb | 3544 | return FALSE; |
93524f2b | 3545 | e = (const U8 *) SvEND(sv); |
511c2ff0 | 3546 | while (c < e) { |
b64e5050 | 3547 | const U8 ch = *c++; |
c4d5f83a | 3548 | if (!UTF8_IS_INVARIANT(ch)) { |
67e989fb JH |
3549 | SvUTF8_on(sv); |
3550 | break; | |
3551 | } | |
560a288e | 3552 | } |
560a288e GS |
3553 | } |
3554 | return TRUE; | |
3555 | } | |
3556 | ||
954c1994 GS |
3557 | /* |
3558 | =for apidoc sv_setsv | |
3559 | ||
645c22ef DM |
3560 | Copies the contents of the source SV C<ssv> into the destination SV |
3561 | C<dsv>. The source SV may be destroyed if it is mortal, so don't use this | |
3562 | function if the source SV needs to be reused. Does not handle 'set' magic. | |
3563 | Loosely speaking, it performs a copy-by-value, obliterating any previous | |
3564 | content of the destination. | |
3565 | ||
3566 | You probably want to use one of the assortment of wrappers, such as | |
3567 | C<SvSetSV>, C<SvSetSV_nosteal>, C<SvSetMagicSV> and | |
3568 | C<SvSetMagicSV_nosteal>. | |
3569 | ||
8d6d96c1 HS |
3570 | =for apidoc sv_setsv_flags |
3571 | ||
645c22ef DM |
3572 | Copies the contents of the source SV C<ssv> into the destination SV |
3573 | C<dsv>. The source SV may be destroyed if it is mortal, so don't use this | |
3574 | function if the source SV needs to be reused. Does not handle 'set' magic. | |
3575 | Loosely speaking, it performs a copy-by-value, obliterating any previous | |
3576 | content of the destination. | |
3577 | If the C<flags> parameter has the C<SV_GMAGIC> bit set, will C<mg_get> on | |
5fcdf167 NC |
3578 | C<ssv> if appropriate, else not. If the C<flags> parameter has the |
3579 | C<NOSTEAL> bit set then the buffers of temps will not be stolen. <sv_setsv> | |
3580 | and C<sv_setsv_nomg> are implemented in terms of this function. | |
645c22ef DM |
3581 | |
3582 | You probably want to use one of the assortment of wrappers, such as | |
3583 | C<SvSetSV>, C<SvSetSV_nosteal>, C<SvSetMagicSV> and | |
3584 | C<SvSetMagicSV_nosteal>. | |
3585 | ||
3586 | This is the primary function for copying scalars, and most other | |
3587 | copy-ish functions and macros use this underneath. | |
8d6d96c1 HS |
3588 | |
3589 | =cut | |
3590 | */ | |
3591 | ||
5d0301b7 | 3592 | static void |
7bc54cea | 3593 | S_glob_assign_glob(pTHX_ SV *const dstr, SV *const sstr, const int dtype) |
5d0301b7 | 3594 | { |
70cd14a1 | 3595 | I32 mro_changes = 0; /* 1 = method, 2 = isa */ |
dd69841b | 3596 | |
7918f24d NC |
3597 | PERL_ARGS_ASSERT_GLOB_ASSIGN_GLOB; |
3598 | ||
5d0301b7 NC |
3599 | if (dtype != SVt_PVGV) { |
3600 | const char * const name = GvNAME(sstr); | |
3601 | const STRLEN len = GvNAMELEN(sstr); | |
0d092c36 | 3602 | { |
f7877b28 NC |
3603 | if (dtype >= SVt_PV) { |
3604 | SvPV_free(dstr); | |
3605 | SvPV_set(dstr, 0); | |
3606 | SvLEN_set(dstr, 0); | |
3607 | SvCUR_set(dstr, 0); | |
3608 | } | |
0d092c36 | 3609 | SvUPGRADE(dstr, SVt_PVGV); |
dedf8e73 | 3610 | (void)SvOK_off(dstr); |
2e5b91de NC |
3611 | /* FIXME - why are we doing this, then turning it off and on again |
3612 | below? */ | |
3613 | isGV_with_GP_on(dstr); | |
f7877b28 | 3614 | } |
5d0301b7 NC |
3615 | GvSTASH(dstr) = GvSTASH(sstr); |
3616 | if (GvSTASH(dstr)) | |
daba3364 | 3617 | Perl_sv_add_backref(aTHX_ MUTABLE_SV(GvSTASH(dstr)), dstr); |
159b6efe | 3618 | gv_name_set(MUTABLE_GV(dstr), name, len, GV_ADD); |
5d0301b7 NC |
3619 | SvFAKE_on(dstr); /* can coerce to non-glob */ |
3620 | } | |
3621 | ||
159b6efe | 3622 | if(GvGP(MUTABLE_GV(sstr))) { |
dd69841b BB |
3623 | /* If source has method cache entry, clear it */ |
3624 | if(GvCVGEN(sstr)) { | |
3625 | SvREFCNT_dec(GvCV(sstr)); | |
3626 | GvCV(sstr) = NULL; | |
3627 | GvCVGEN(sstr) = 0; | |
3628 | } | |
3629 | /* If source has a real method, then a method is | |
3630 | going to change */ | |
159b6efe | 3631 | else if(GvCV((const GV *)sstr)) { |
70cd14a1 | 3632 | mro_changes = 1; |
dd69841b BB |
3633 | } |
3634 | } | |
3635 | ||
3636 | /* If dest already had a real method, that's a change as well */ | |
159b6efe | 3637 | if(!mro_changes && GvGP(MUTABLE_GV(dstr)) && GvCVu((const GV *)dstr)) { |
70cd14a1 | 3638 | mro_changes = 1; |
dd69841b BB |
3639 | } |
3640 | ||
159b6efe | 3641 | if(strEQ(GvNAME((const GV *)dstr),"ISA")) |
70cd14a1 CB |
3642 | mro_changes = 2; |
3643 | ||
159b6efe | 3644 | gp_free(MUTABLE_GV(dstr)); |
2e5b91de | 3645 | isGV_with_GP_off(dstr); |
5d0301b7 | 3646 | (void)SvOK_off(dstr); |
2e5b91de | 3647 | isGV_with_GP_on(dstr); |
dedf8e73 | 3648 | GvINTRO_off(dstr); /* one-shot flag */ |
5d0301b7 NC |
3649 | GvGP(dstr) = gp_ref(GvGP(sstr)); |
3650 | if (SvTAINTED(sstr)) | |
3651 | SvTAINT(dstr); | |
3652 | if (GvIMPORTED(dstr) != GVf_IMPORTED | |
3653 | && CopSTASH_ne(PL_curcop, GvSTASH(dstr))) | |
3654 | { | |
3655 | GvIMPORTED_on(dstr); | |
3656 | } | |
3657 | GvMULTI_on(dstr); | |
70cd14a1 CB |
3658 | if(mro_changes == 2) mro_isa_changed_in(GvSTASH(dstr)); |
3659 | else if(mro_changes) mro_method_changed_in(GvSTASH(dstr)); | |
5d0301b7 NC |
3660 | return; |
3661 | } | |
3662 | ||
b8473700 | 3663 | static void |