| 1 | =head1 NAME |
| 2 | |
| 3 | perlintern - autogenerated documentation of purely B<internal> |
| 4 | Perl functions |
| 5 | |
| 6 | =head1 DESCRIPTION |
| 7 | |
| 8 | This file is the autogenerated documentation of functions in the |
| 9 | Perl interpreter that are documented using Perl's internal documentation |
| 10 | format but are not marked as part of the Perl API. In other words, |
| 11 | B<they are not for use in extensions>! |
| 12 | |
| 13 | |
| 14 | =head1 Global Variables |
| 15 | |
| 16 | =over 8 |
| 17 | |
| 18 | =item PL_DBsingle |
| 19 | |
| 20 | When Perl is run in debugging mode, with the B<-d> switch, this SV is a |
| 21 | boolean which indicates whether subs are being single-stepped. |
| 22 | Single-stepping is automatically turned on after every step. This is the C |
| 23 | variable which corresponds to Perl's $DB::single variable. See |
| 24 | C<PL_DBsub>. |
| 25 | |
| 26 | SV * PL_DBsingle |
| 27 | |
| 28 | =for hackers |
| 29 | Found in file intrpvar.h |
| 30 | |
| 31 | =item PL_DBsub |
| 32 | |
| 33 | When Perl is run in debugging mode, with the B<-d> switch, this GV contains |
| 34 | the SV which holds the name of the sub being debugged. This is the C |
| 35 | variable which corresponds to Perl's $DB::sub variable. See |
| 36 | C<PL_DBsingle>. |
| 37 | |
| 38 | GV * PL_DBsub |
| 39 | |
| 40 | =for hackers |
| 41 | Found in file intrpvar.h |
| 42 | |
| 43 | =item PL_DBtrace |
| 44 | |
| 45 | Trace variable used when Perl is run in debugging mode, with the B<-d> |
| 46 | switch. This is the C variable which corresponds to Perl's $DB::trace |
| 47 | variable. See C<PL_DBsingle>. |
| 48 | |
| 49 | SV * PL_DBtrace |
| 50 | |
| 51 | =for hackers |
| 52 | Found in file intrpvar.h |
| 53 | |
| 54 | =item PL_dowarn |
| 55 | |
| 56 | The C variable which corresponds to Perl's $^W warning variable. |
| 57 | |
| 58 | bool PL_dowarn |
| 59 | |
| 60 | =for hackers |
| 61 | Found in file intrpvar.h |
| 62 | |
| 63 | =item PL_last_in_gv |
| 64 | |
| 65 | The GV which was last used for a filehandle input operation. (C<< <FH> >>) |
| 66 | |
| 67 | GV* PL_last_in_gv |
| 68 | |
| 69 | =for hackers |
| 70 | Found in file thrdvar.h |
| 71 | |
| 72 | =item PL_ofs_sv |
| 73 | |
| 74 | The output field separator - C<$,> in Perl space. |
| 75 | |
| 76 | SV* PL_ofs_sv |
| 77 | |
| 78 | =for hackers |
| 79 | Found in file thrdvar.h |
| 80 | |
| 81 | =item PL_rs |
| 82 | |
| 83 | The input record separator - C<$/> in Perl space. |
| 84 | |
| 85 | SV* PL_rs |
| 86 | |
| 87 | =for hackers |
| 88 | Found in file thrdvar.h |
| 89 | |
| 90 | |
| 91 | =back |
| 92 | |
| 93 | =head1 GV Functions |
| 94 | |
| 95 | =over 8 |
| 96 | |
| 97 | =item is_gv_magical |
| 98 | |
| 99 | Returns C<TRUE> if given the name of a magical GV. |
| 100 | |
| 101 | Currently only useful internally when determining if a GV should be |
| 102 | created even in rvalue contexts. |
| 103 | |
| 104 | C<flags> is not used at present but available for future extension to |
| 105 | allow selecting particular classes of magical variable. |
| 106 | |
| 107 | bool is_gv_magical(char *name, STRLEN len, U32 flags) |
| 108 | |
| 109 | =for hackers |
| 110 | Found in file gv.c |
| 111 | |
| 112 | |
| 113 | =back |
| 114 | |
| 115 | =head1 IO Functions |
| 116 | |
| 117 | =over 8 |
| 118 | |
| 119 | =item start_glob |
| 120 | |
| 121 | Function called by C<do_readline> to spawn a glob (or do the glob inside |
| 122 | perl on VMS). This code used to be inline, but now perl uses C<File::Glob> |
| 123 | this glob starter is only used by miniperl during the build process. |
| 124 | Moving it away shrinks pp_hot.c; shrinking pp_hot.c helps speed perl up. |
| 125 | |
| 126 | PerlIO* start_glob(SV* pattern, IO *io) |
| 127 | |
| 128 | =for hackers |
| 129 | Found in file doio.c |
| 130 | |
| 131 | |
| 132 | =back |
| 133 | |
| 134 | =head1 Pad Data Structures |
| 135 | |
| 136 | =over 8 |
| 137 | |
| 138 | =item CvPADLIST |
| 139 | |
| 140 | CV's can have CvPADLIST(cv) set to point to an AV. |
| 141 | |
| 142 | For these purposes "forms" are a kind-of CV, eval""s are too (except they're |
| 143 | not callable at will and are always thrown away after the eval"" is done |
| 144 | executing). |
| 145 | |
| 146 | XSUBs don't have CvPADLIST set - dXSTARG fetches values from PL_curpad, |
| 147 | but that is really the callers pad (a slot of which is allocated by |
| 148 | every entersub). |
| 149 | |
| 150 | The CvPADLIST AV has does not have AvREAL set, so REFCNT of component items |
| 151 | is managed "manual" (mostly in op.c) rather than normal av.c rules. |
| 152 | The items in the AV are not SVs as for a normal AV, but other AVs: |
| 153 | |
| 154 | 0'th Entry of the CvPADLIST is an AV which represents the "names" or rather |
| 155 | the "static type information" for lexicals. |
| 156 | |
| 157 | The CvDEPTH'th entry of CvPADLIST AV is an AV which is the stack frame at that |
| 158 | depth of recursion into the CV. |
| 159 | The 0'th slot of a frame AV is an AV which is @_. |
| 160 | other entries are storage for variables and op targets. |
| 161 | |
| 162 | During compilation: |
| 163 | C<PL_comppad_name> is set the the the names AV. |
| 164 | C<PL_comppad> is set the the frame AV for the frame CvDEPTH == 1. |
| 165 | C<PL_curpad> is set the body of the frame AV (i.e. AvARRAY(PL_comppad)). |
| 166 | |
| 167 | Itterating over the names AV itterates over all possible pad |
| 168 | items. Pad slots that are SVs_PADTMP (targets/GVs/constants) end up having |
| 169 | &PL_sv_undef "names" (see pad_alloc()). |
| 170 | |
| 171 | Only my/our variable (SVs_PADMY/SVs_PADOUR) slots get valid names. |
| 172 | The rest are op targets/GVs/constants which are statically allocated |
| 173 | or resolved at compile time. These don't have names by which they |
| 174 | can be looked up from Perl code at run time through eval"" like |
| 175 | my/our variables can be. Since they can't be looked up by "name" |
| 176 | but only by their index allocated at compile time (which is usually |
| 177 | in PL_op->op_targ), wasting a name SV for them doesn't make sense. |
| 178 | |
| 179 | The SVs in the names AV have their PV being the name of the variable. |
| 180 | NV+1..IV inclusive is a range of cop_seq numbers for which the name is valid. |
| 181 | For typed lexicals name SV is SVt_PVMG and SvSTASH points at the type. |
| 182 | |
| 183 | If SvFAKE is set on the name SV then slot in the frame AVs are |
| 184 | a REFCNT'ed references to a lexical from "outside". |
| 185 | |
| 186 | If the 'name' is '&' the the corresponding entry in frame AV |
| 187 | is a CV representing a possible closure. |
| 188 | (SvFAKE and name of '&' is not a meaningful combination currently but could |
| 189 | become so if C<my sub foo {}> is implemented.) |
| 190 | |
| 191 | AV * CvPADLIST(CV *cv) |
| 192 | |
| 193 | =for hackers |
| 194 | Found in file cv.h |
| 195 | |
| 196 | |
| 197 | =back |
| 198 | |
| 199 | =head1 Stack Manipulation Macros |
| 200 | |
| 201 | =over 8 |
| 202 | |
| 203 | =item djSP |
| 204 | |
| 205 | Declare Just C<SP>. This is actually identical to C<dSP>, and declares |
| 206 | a local copy of perl's stack pointer, available via the C<SP> macro. |
| 207 | See C<SP>. (Available for backward source code compatibility with the |
| 208 | old (Perl 5.005) thread model.) |
| 209 | |
| 210 | djSP; |
| 211 | |
| 212 | =for hackers |
| 213 | Found in file pp.h |
| 214 | |
| 215 | =item LVRET |
| 216 | |
| 217 | True if this op will be the return value of an lvalue subroutine |
| 218 | |
| 219 | =for hackers |
| 220 | Found in file pp.h |
| 221 | |
| 222 | |
| 223 | =back |
| 224 | |
| 225 | =head1 SV Manipulation Functions |
| 226 | |
| 227 | =over 8 |
| 228 | |
| 229 | =item report_uninit |
| 230 | |
| 231 | Print appropriate "Use of uninitialized variable" warning |
| 232 | |
| 233 | void report_uninit() |
| 234 | |
| 235 | =for hackers |
| 236 | Found in file sv.c |
| 237 | |
| 238 | =item sv_add_arena |
| 239 | |
| 240 | Given a chunk of memory, link it to the head of the list of arenas, |
| 241 | and split it into a list of free SVs. |
| 242 | |
| 243 | void sv_add_arena(char* ptr, U32 size, U32 flags) |
| 244 | |
| 245 | =for hackers |
| 246 | Found in file sv.c |
| 247 | |
| 248 | =item sv_clean_all |
| 249 | |
| 250 | Decrement the refcnt of each remaining SV, possibly triggering a |
| 251 | cleanup. This function may have to be called multiple times to free |
| 252 | SVs which are in complex self-referential hierarchies. |
| 253 | |
| 254 | I32 sv_clean_all() |
| 255 | |
| 256 | =for hackers |
| 257 | Found in file sv.c |
| 258 | |
| 259 | =item sv_clean_objs |
| 260 | |
| 261 | Attempt to destroy all objects not yet freed |
| 262 | |
| 263 | void sv_clean_objs() |
| 264 | |
| 265 | =for hackers |
| 266 | Found in file sv.c |
| 267 | |
| 268 | =item sv_free_arenas |
| 269 | |
| 270 | Deallocate the memory used by all arenas. Note that all the individual SV |
| 271 | heads and bodies within the arenas must already have been freed. |
| 272 | |
| 273 | void sv_free_arenas() |
| 274 | |
| 275 | =for hackers |
| 276 | Found in file sv.c |
| 277 | |
| 278 | |
| 279 | =back |
| 280 | |
| 281 | =head1 AUTHORS |
| 282 | |
| 283 | The autodocumentation system was originally added to the Perl core by |
| 284 | Benjamin Stuhl. Documentation is by whoever was kind enough to |
| 285 | document their functions. |
| 286 | |
| 287 | =head1 SEE ALSO |
| 288 | |
| 289 | perlguts(1), perlapi(1) |
| 290 | |