← Index
NYTProf Performance Profile   « line view »
For svc/members/upsert
  Run on Tue Jan 13 11:50:22 2015
Reported on Tue Jan 13 12:09:49 2015

Filename/usr/share/perl/5.20/_charnames.pm
StatementsExecuted 56 statements in 3.67ms
Subroutines
Calls P F Exclusive
Time		 Inclusive
Time		 Subroutine
1111.87ms2.09ms_charnames::::BEGIN@14_charnames::BEGIN@14
11125µs139µs_charnames::::import_charnames::import
11115µs15µs_charnames::::CORE:regcomp_charnames::CORE:regcomp (opcode)
11113µs28µs_charnames::::BEGIN@165_charnames::BEGIN@165
11112µs23µs_charnames::::BEGIN@7_charnames::BEGIN@7
1117µs11µs_charnames::::BEGIN@8_charnames::BEGIN@8
3316µs6µs_charnames::::CORE:qr_charnames::CORE:qr (opcode)
1116µs6µs_charnames::::BEGIN@9_charnames::BEGIN@9
1114µs4µs_charnames::::BEGIN@11_charnames::BEGIN@11
4114µs4µs_charnames::::CORE:pack_charnames::CORE:pack (opcode)
1114µs4µs_charnames::::BEGIN@177_charnames::BEGIN@177
1113µs3µs_charnames::::BEGIN@13_charnames::BEGIN@13
0000s0s_charnames::::alias_charnames::alias
0000s0s_charnames::::alias_file_charnames::alias_file
0000s0s_charnames::::carp_charnames::carp
0000s0s_charnames::::charnames_charnames::charnames
0000s0s_charnames::::croak_charnames::croak
0000s0s_charnames::::lookup_name_charnames::lookup_name
0000s0s_charnames::::not_legal_use_bytes_msg_charnames::not_legal_use_bytes_msg
0000s0s_charnames::::viacode_charnames::viacode
Call graph for these subroutines as a Graphviz dot language file.
Line State
ments		 Time
on line		 Calls Time
in subs		 Code
1# !!!!!!! INTERNAL PERL USE ONLY !!!!!!!
2# This helper module is for internal use by core Perl only. This module is
3# subject to change or removal at any time without notice. Don't use it
4# directly. Use the public <charnames> module instead.
5
6package _charnames;
7223µs235µs
# spent 23µs (12+11) within _charnames::BEGIN@7 which was called: # once (12µs+11µs) by charnames::BEGIN@6 at line 7
use strict;
# spent 23µs making 1 call to _charnames::BEGIN@7 # spent 11µs making 1 call to strict::import
8219µs215µs
# spent 11µs (7+4) within _charnames::BEGIN@8 which was called: # once (7µs+4µs) by charnames::BEGIN@6 at line 8
use warnings;
# spent 11µs making 1 call to _charnames::BEGIN@8 # spent 4µs making 1 call to warnings::import
9230µs16µs
# spent 6µs within _charnames::BEGIN@9 which was called: # once (6µs+0s) by charnames::BEGIN@6 at line 9
use File::Spec;
# spent 6µs making 1 call to _charnames::BEGIN@9
101500nsour $VERSION = '1.39';
11220µs14µs
# spent 4µs within _charnames::BEGIN@11 which was called: # once (4µs+0s) by charnames::BEGIN@6 at line 11
use unicore::Name; # mktables-generated algorithmically-defined names
# spent 4µs making 1 call to _charnames::BEGIN@11
12
13217µs13µs
# spent 3µs within _charnames::BEGIN@13 which was called: # once (3µs+0s) by charnames::BEGIN@6 at line 13
use bytes (); # for $bytes::hint_bits
# spent 3µs making 1 call to _charnames::BEGIN@13
142945µs22.13ms
# spent 2.09ms (1.87+223µs) within _charnames::BEGIN@14 which was called: # once (1.87ms+223µs) by charnames::BEGIN@6 at line 14
use re "/aa"; # Everything in here should be ASCII
# spent 2.09ms making 1 call to _charnames::BEGIN@14 # spent 37µs making 1 call to re::import
15
1611µs$Carp::Internal{ (__PACKAGE__) } = 1;
17
18# Translate between Unicode character names and their code points. This is a
19# submodule of package <charnames>, used to allow \N{...} to be autoloaded,
20# but it was decided not to autoload the various functions in charnames; the
21# splitting allows this behavior.
22#
23# The official names with their code points are stored in a table in
24# lib/unicore/Name.pl which is read in as a large string (almost 3/4 Mb in
25# Unicode 6.0). Each code point/name combination is separated by a \n in the
26# string. (Some of the CJK and the Hangul syllable names are determined
27# instead algorithmically via subroutines stored instead in
28# lib/unicore/Name.pm). Because of the large size of this table, it isn't
29# converted into hashes for faster lookup.
30#
31# But, user defined aliases are stored in their own hashes, as are Perl
32# extensions to the official names. These are checked first before looking at
33# the official table.
34#
35# Basically, the table is grepped for the input code point (viacode()) or
36# name (the other functions), and the corresponding value on the same line is
37# returned. The grepping is done by turning the input into a regular
38# expression. Thus, the same table does double duty, used by both name and
39# code point lookup. (If we were to have hashes, we would need two, one for
40# each lookup direction.)
41#
42# For loose name matching, the logical thing would be to have a table
43# with all the ignorable characters squeezed out, and then grep it with the
44# similiarly-squeezed input name. (And this is in fact how the lookups are
45# done with the small Perl extension hashes.) But since we need to be able to
46# go from code point to official name, the original table would still need to
47# exist. Due to the large size of the table, it was decided to not read
48# another very large string into memory for a second table. Instead, the
49# regular expression of the input name is modified to have optional spaces and
50# dashes between characters. For example, in strict matching, the regular
51# expression would be:
52# qr/\tDIGIT ONE$/m
53# Under loose matching, the blank would be squeezed out, and the re would be:
54# qr/\tD[- ]?I[- ]?G[- ]?I[- ]?T[- ]?O[- ]?N[- ]?E$/m
55# which matches a blank or dash between any characters in the official table.
56#
57# This is also how script lookup is done. Basically the re looks like
58# qr/ (?:LATIN|GREEK|CYRILLIC) (?:SMALL )?LETTER $name/
59# where $name is the loose or strict regex for the remainder of the name.
60
61# The hashes are stored as utf8 strings. This makes it easier to deal with
62# sequences. I (khw) also tried making Name.pl utf8, but it slowed things
63# down by a factor of 7. I then tried making Name.pl store the ut8
64# equivalents but not calling them utf8. That led to similar speed as leaving
65# it alone, but since that is harder for a human to parse, I left it as-is.
66
67125µs87µsmy %system_aliases = (
# spent 4µs making 4 calls to _charnames::CORE:pack, avg 1µs/call # spent 2µs making 4 calls to utf8::unicode_to_native, avg 625ns/call
68
69 'SINGLE-SHIFT 2' => pack("U", utf8::unicode_to_native(0x8E)),
70 'SINGLE-SHIFT 3' => pack("U", utf8::unicode_to_native(0x8F)),
71 'PRIVATE USE 1' => pack("U", utf8::unicode_to_native(0x91)),
72 'PRIVATE USE 2' => pack("U", utf8::unicode_to_native(0x92)),
73);
74
75# These are the aliases above that differ under :loose and :full matching
76# because the :full versions have blanks or hyphens in them.
77#my %loose_system_aliases = (
78#);
79
80#my %deprecated_aliases;
81#$deprecated_aliases{'BELL'} = pack("U", utf8::unicode_to_native(0x07)) if $^V lt v5.17.0;
82
83#my %loose_deprecated_aliases = (
84#);
85
86# These are special cased in :loose matching, differing only in a medial
87# hyphen
881300ns12µsmy $HANGUL_JUNGSEONG_O_E_utf8 = pack("U", 0x1180);
# spent 2µs making 1 call to main::CORE:pack
891100ns1600nsmy $HANGUL_JUNGSEONG_OE_utf8 = pack("U", 0x116C);
# spent 600ns making 1 call to main::CORE:pack
90
91
921100nsmy $txt; # The table of official character names
93
94my %full_names_cache; # Holds already-looked-up names, so don't have to
95# re-look them up again. The previous versions of charnames had scoping
96# bugs. For example if we use script A in one scope and find and cache
97# what Z resolves to, we can't use that cache in a different scope that
98# uses script B instead of A, as Z might be an entirely different letter
99# there; or there might be different aliases in effect in different
100# scopes, or :short may be in effect or not effect in different scopes,
101# or various combinations thereof. This was solved in this version
102# mostly by moving things to %^H. But some things couldn't be moved
103# there. One of them was the cache of runtime looked-up names, in part
104# because %^H is read-only at runtime. I (khw) don't know why the cache
105# was run-time only in the previous versions: perhaps oversight; perhaps
106# that compile time looking doesn't happen in a loop so didn't think it
107# was worthwhile; perhaps not wanting to make the cache too large. But
108# I decided to make it compile time as well; this could easily be
109# changed.
110# Anyway, this hash is not scoped, and is added to at runtime. It
111# doesn't have scoping problems because the data in it is restricted to
112# official names, which are always invariant, and we only set it and
113# look at it at during :full lookups, so is unaffected by any other
114# scoped options. I put this in to maintain parity with the older
115# version. If desired, a %short_names cache could also be made, as well
116# as one for each script, say in %script_names_cache, with each key
117# being a hash for a script named in a 'use charnames' statement. I
118# decided not to do that for now, just because it's added complication,
119# and because I'm just trying to maintain parity, not extend it.
120
121# Like %full_names_cache, but for use when :loose is in effect. There needs
122# to be two caches because :loose may not be in effect for a scope, and a
123# loose name could inappropriately be returned when only exact matching is
124# called for.
125my %loose_names_cache;
126
127# Designed so that test decimal first, and then hex. Leading zeros
128# imply non-decimal, as do non-[0-9]
12918µs14µsmy $decimal_qr = qr/^[1-9]\d*$/;
# spent 4µs making 1 call to _charnames::CORE:qr
130
131# Returns the hex number in $1.
13213µs1900nsmy $hex_qr = qr/^(?:[Uu]\+|0[xX])?([[:xdigit:]]+)$/;
# spent 900ns making 1 call to _charnames::CORE:qr
133
134sub croak
135{
136 require Carp; goto &Carp::croak;
137} # croak
138
139sub carp
140{
141 require Carp; goto &Carp::carp;
142} # carp
143
144sub alias (@) # Set up a single alias
145{
146 my @errors;
147
148 my $alias = ref $_[0] ? $_[0] : { @_ };
149 foreach my $name (sort keys %$alias) { # Sort only because it helps having
150 # deterministic output for
151 # t/lib/charnames/alias
152 my $value = $alias->{$name};
153 next unless defined $value; # Omit if screwed up.
154
155 # Is slightly slower to just after this statement see if it is
156 # decimal, since we already know it is after having converted from
157 # hex, but makes the code easier to maintain, and is called
158 # infrequently, only at compile-time
159 if ($value !~ $decimal_qr && $value =~ $hex_qr) {
160 my $temp = CORE::hex $1;
161 $temp = utf8::unicode_to_native($temp) if $value =~ /^[Uu]\+/;
162 $value = $temp;
163 }
164 if ($value =~ $decimal_qr) {
165274µs243µs
# spent 28µs (13+15) within _charnames::BEGIN@165 which was called: # once (13µs+15µs) by charnames::BEGIN@6 at line 165
no warnings qw(non_unicode surrogate nonchar); # Allow any of these
# spent 28µs making 1 call to _charnames::BEGIN@165 # spent 15µs making 1 call to warnings::unimport
166 $^H{charnames_ord_aliases}{$name} = pack("U", $value);
167
168 # Use a canonical form.
169 $^H{charnames_inverse_ords}{sprintf("%05X", $value)} = $name;
170 }
171 else {
172 # This regex needs to be sync'd with the code in toke.c that checks
173 # for the same thing
174 if ($name !~ / ^
175 \p{_Perl_Charname_Begin}
176 \p{_Perl_Charname_Continue}*
1772464µs34.62ms
# spent 4µs within _charnames::BEGIN@177 which was called: # once (4µs+0s) by charnames::BEGIN@6 at line 177
$ /x) {
# spent 3.49ms making 1 call to utf8::AUTOLOAD # spent 1.13ms making 1 call to utf8::SWASHNEW # spent 4µs making 1 call to _charnames::BEGIN@177
178
179 push @errors, $name;
180 }
181 else {
182 $^H{charnames_name_aliases}{$name} = $value;
183
184 if (warnings::enabled('deprecated')) {
185 if ($name =~ / ( .* \s ) ( \s* ) $ /x) {
186 carp "Trailing white-space in a charnames alias definition is deprecated; marked by <-- HERE in '$1 <-- HERE " . $2 . "'";
187 }
188
189 # Use '+' instead of '*' in this regex, because any trailing
190 # blanks have already been warned about.
191 if ($name =~ / ( .*? \s{2} ) ( .+ ) /x) {
192 carp "A sequence of multiple spaces in a charnames alias definition is deprecated; marked by <-- HERE in '$1 <-- HERE " . $2 . "'";
193 }
194 }
195 }
196 }
197 }
198
199 # We find and output all errors from this :alias definition, rather than
200 # failing on the first one, so fewer runs are needed to get it to compile
201 if (@errors) {
202 foreach my $name (@errors) {
203 my $ok = "";
204 my $nbsp = chr utf8::unicode_to_native(0xa0);
205 $ok = $1 if $name =~ / ^ ( \p{Alpha} [-\p{XPosixWord} ():$nbsp]* ) /x;
206 my $first_bad = substr($name, length($ok), 1);
207 $name = "Invalid character in charnames alias definition; marked by <-- HERE in '$ok$first_bad<-- HERE " . substr($name, length($ok) + 1) . "'";
208 }
209 croak join "\n", @errors;
210 }
211
212 return;
213} # alias
214
215sub not_legal_use_bytes_msg {
216 my ($name, $utf8) = @_;
217 my $return;
218
219 if (length($utf8) == 1) {
220 $return = sprintf("Character 0x%04x with name '%s' is", ord $utf8, $name);
221 } else {
222 $return = sprintf("String with name '%s' (and ordinals %s) contains character(s)", $name, join(" ", map { sprintf "0x%04X", ord $_ } split(//, $utf8)));
223 }
224 return $return . " above 0xFF with 'use bytes' in effect";
225}
226
227sub alias_file ($) # Reads a file containing alias definitions
228{
229 my ($arg, $file) = @_;
230 if (-f $arg && File::Spec->file_name_is_absolute ($arg)) {
231 $file = $arg;
232 }
23311.97ms21.70ms elsif ($arg =~ m/ ^ \p{_Perl_IDStart} \p{_Perl_IDCont}* $/x) {
# spent 1.70ms making 2 calls to utf8::SWASHNEW, avg 852µs/call
234 $file = "unicore/${arg}_alias.pl";
235 }
236 else {
237 croak "Charnames alias file names can only have identifier characters";
238 }
239 if (my @alias = do $file) {
240 @alias == 1 && !defined $alias[0] and
241 croak "$file cannot be used as alias file for charnames";
242 @alias % 2 and
243 croak "$file did not return a (valid) list of alias pairs";
244 alias (@alias);
245 return (1);
246 }
247 0;
248} # alias_file
249
250# For use when don't import anything. This structure must be kept in
251# sync with the one that import() fills up.
25212µsmy %dummy_H = (
253 charnames_stringified_names => "",
254 charnames_stringified_ords => "",
255 charnames_scripts => "",
256 charnames_full => 1,
257 charnames_loose => 0,
258 charnames_short => 0,
259 );
260
261
262sub lookup_name ($$$) {
263 my ($name, $wants_ord, $runtime) = @_;
264
265 # Lookup the name or sequence $name in the tables. If $wants_ord is false,
266 # returns the string equivalent of $name; if true, returns the ordinal value
267 # instead, but in this case $name must not be a sequence; otherwise undef is
268 # returned and a warning raised. $runtime is 0 if compiletime, otherwise
269 # gives the number of stack frames to go back to get the application caller
270 # info.
271 # If $name is not found, returns undef in runtime with no warning; and in
272 # compiletime, the Unicode replacement character, with a warning.
273
274 # It looks first in the aliases, then in the large table of official Unicode
275 # names.
276
277 my $utf8; # The string result
278 my $save_input;
279
280 if ($runtime) {
281
282 my $hints_ref = (caller($runtime))[10];
283
284 # If we didn't import anything (which happens with 'use charnames ()',
285 # substitute a dummy structure.
286 $hints_ref = \%dummy_H if ! defined $hints_ref
287 || (! defined $hints_ref->{charnames_full}
288 && ! defined $hints_ref->{charnames_loose});
289
290 # At runtime, but currently not at compile time, $^H gets
291 # stringified, so un-stringify back to the original data structures.
292 # These get thrown away by perl before the next invocation
293 # Also fill in the hash with the non-stringified data.
294 # N.B. New fields must be also added to %dummy_H
295
296 %{$^H{charnames_name_aliases}} = split ',',
297 $hints_ref->{charnames_stringified_names};
298 %{$^H{charnames_ord_aliases}} = split ',',
299 $hints_ref->{charnames_stringified_ords};
300 $^H{charnames_scripts} = $hints_ref->{charnames_scripts};
301 $^H{charnames_full} = $hints_ref->{charnames_full};
302 $^H{charnames_loose} = $hints_ref->{charnames_loose};
303 $^H{charnames_short} = $hints_ref->{charnames_short};
304 }
305
306 my $loose = $^H{charnames_loose};
307 my $lookup_name; # Input name suitably modified for grepping for in the
308 # table
309
310 # User alias should be checked first or else can't override ours, and if we
311 # were to add any, could conflict with theirs.
312 if (exists $^H{charnames_ord_aliases}{$name}) {
313 $utf8 = $^H{charnames_ord_aliases}{$name};
314 }
315 elsif (exists $^H{charnames_name_aliases}{$name}) {
316 $name = $^H{charnames_name_aliases}{$name};
317 $save_input = $lookup_name = $name; # Cache the result for any error
318 # message
319 # The aliases are documented to not match loosely, so change loose match
320 # into full.
321 if ($loose) {
322 $loose = 0;
323 $^H{charnames_full} = 1;
324 }
325 }
326 else {
327
328 # Here, not a user alias. That means that loose matching may be in
329 # effect; will have to modify the input name.
330 $lookup_name = $name;
331 if ($loose) {
332 $lookup_name = uc $lookup_name;
333
334 # Squeeze out all underscores
335 $lookup_name =~ s/_//g;
336
337 # Remove all medial hyphens
338 $lookup_name =~ s/ (?<= \S ) - (?= \S )//gx;
339
340 # Squeeze out all spaces
341 $lookup_name =~ s/\s//g;
342 }
343
344 # Here, $lookup_name has been modified as necessary for looking in the
345 # hashes. Check the system alias files next. Most of these aliases are
346 # the same for both strict and loose matching. To save space, the ones
347 # which differ are in their own separate hash, which is checked if loose
348 # matching is selected and the regular match fails. To save time, the
349 # loose hashes could be expanded to include all aliases, and there would
350 # only have to be one check. But if someone specifies :loose, they are
351 # interested in convenience over speed, and the time for this second check
352 # is miniscule compared to the rest of the routine.
353 if (exists $system_aliases{$lookup_name}) {
354 $utf8 = $system_aliases{$lookup_name};
355 }
356 # There are currently no entries in this hash, so don't waste time looking
357 # for them. But the code is retained for the unlikely possibility that
358 # some will be added in the future.
359# elsif ($loose && exists $loose_system_aliases{$lookup_name}) {
360# $utf8 = $loose_system_aliases{$lookup_name};
361# }
362# if (exists $deprecated_aliases{$lookup_name}) {
363# require warnings;
364# warnings::warnif('deprecated',
365# "Unicode character name \"$name\" is deprecated, use \""
366# . viacode(ord $deprecated_aliases{$lookup_name})
367# . "\" instead");
368# $utf8 = $deprecated_aliases{$lookup_name};
369# }
370 # There are currently no entries in this hash, so don't waste time looking
371 # for them. But the code is retained for the unlikely possibility that
372 # some will be added in the future.
373# elsif ($loose && exists $loose_deprecated_aliases{$lookup_name}) {
374# require warnings;
375# warnings::warnif('deprecated',
376# "Unicode character name \"$name\" is deprecated, use \""
377# . viacode(ord $loose_deprecated_aliases{$lookup_name})
378# . "\" instead");
379# $utf8 = $loose_deprecated_aliases{$lookup_name};
380# }
381 }
382
383 my @off; # Offsets into table of pattern match begin and end
384
385 # If haven't found it yet...
386 if (! defined $utf8) {
387
388 # See if has looked this input up earlier.
389 if (! $loose && $^H{charnames_full} && exists $full_names_cache{$name}) {
390 $utf8 = $full_names_cache{$name};
391 }
392 elsif ($loose && exists $loose_names_cache{$name}) {
393 $utf8 = $loose_names_cache{$name};
394 }
395 else { # Here, must do a look-up
396
397 # If full or loose matching succeeded, points to where to cache the
398 # result
399 my $cache_ref;
400
401 ## Suck in the code/name list as a big string.
402 ## Lines look like:
403 ## "00052\tLATIN CAPITAL LETTER R\n"
404 # or
405 # "0052 0303\tLATIN CAPITAL LETTER R WITH TILDE\n"
406 $txt = do "unicore/Name.pl" unless $txt;
407
408 ## @off will hold the index into the code/name string of the start and
409 ## end of the name as we find it.
410
411 ## If :loose, look for a loose match; if :full, look for the name
412 ## exactly
413 # First, see if the name is one which is algorithmically determinable.
414 # The subroutine is included in Name.pl. The table contained in
415 # $txt doesn't contain these. Experiments show that checking
416 # for these before checking for the regular names has no
417 # noticeable impact on performance for the regular names, but
418 # the other way around slows down finding these immensely.
419 # Algorithmically determinables are not placed in the cache because
420 # that uses up memory, and finding these again is fast.
421 if (($loose || $^H{charnames_full})
422 && (defined (my $ord = charnames::name_to_code_point_special($lookup_name, $loose))))
423 {
424 $utf8 = pack("U", $ord);
425 }
426 else {
427
428 # Not algorithmically determinable; look up in the table. The name
429 # will be turned into a regex, so quote any meta characters.
430 $lookup_name = quotemeta $lookup_name;
431
432 if ($loose) {
433
434 # For loose matches, $lookup_name has already squeezed out the
435 # non-essential characters. We have to add in code to make the
436 # squeezed version match the non-squeezed equivalent in the table.
437 # The only remaining hyphens are ones that start or end a word in
438 # the original. They have been quoted in $lookup_name so they look
439 # like "\-". Change all other characters except the backslash
440 # quotes for any metacharacters, and the final character, so that
441 # e.g., COLON gets transformed into: /C[- ]?O[- ]?L[- ]?O[- ]?N/
442 $lookup_name =~ s/ (?! \\ -) # Don't do this to the \- sequence
443 ( [^-\\] ) # Nor the "-" within that sequence,
444 # nor the "\" that quotes metachars,
445 # but otherwise put the char into $1
446 (?=.) # And don't do it for the final char
447 /$1\[- \]?/gx; # And add an optional blank or
448 # '-' after each $1 char
449
450 # Those remaining hyphens were originally at the beginning or end of
451 # a word, so they can match either a blank before or after, but not
452 # both. (Keep in mind that they have been quoted, so are a '\-'
453 # sequence)
454 $lookup_name =~ s/\\ -/(?:- | -)/xg;
455 }
456
457 # Do the lookup in the full table if asked for, and if succeeds
458 # save the offsets and set where to cache the result.
459 if (($loose || $^H{charnames_full}) && $txt =~ /\t$lookup_name$/m) {
460 @off = ($-[0] + 1, $+[0]); # The 1 is for the tab
461 $cache_ref = ($loose) ? \%loose_names_cache : \%full_names_cache;
462 }
463 else {
464
465 # Here, didn't look for, or didn't find the name.
466 # If :short is allowed, see if input is like "greek:Sigma".
467 # Keep in mind that $lookup_name has had the metas quoted.
468 my $scripts_trie = "";
469 my $name_has_uppercase;
470 if (($^H{charnames_short})
471 && $lookup_name =~ /^ (?: \\ \s)* # Quoted space
472 (.+?) # $1 = the script
473 (?: \\ \s)*
474 \\ : # Quoted colon
475 (?: \\ \s)*
476 (.+?) # $2 = the name
477 (?: \\ \s)* $
478 /xs)
479 {
480 # Even in non-loose matching, the script traditionally has been
481 # case insensitive
482 $scripts_trie = "\U$1";
483 $lookup_name = $2;
484
485 # Use original name to find its input casing, but ignore the
486 # script part of that to make the determination.
487 $save_input = $name if ! defined $save_input;
488 $name =~ s/.*?://;
489 $name_has_uppercase = $name =~ /[[:upper:]]/;
490 }
491 else { # Otherwise look in allowed scripts
492 $scripts_trie = $^H{charnames_scripts};
493
494 # Use original name to find its input casing
495 $name_has_uppercase = $name =~ /[[:upper:]]/;
496 }
497
498 my $case = $name_has_uppercase ? "CAPITAL" : "SMALL";
499 return if (! $scripts_trie || $txt !~
500 /\t (?: $scripts_trie ) \ (?:$case\ )? LETTER \ \U$lookup_name $/xm);
501
502 # Here have found the input name in the table.
503 @off = ($-[0] + 1, $+[0]); # The 1 is for the tab
504 }
505
506 # Here, the input name has been found; we haven't set up the output,
507 # but we know where in the string
508 # the name starts. The string is set up so that for single characters
509 # (and not named sequences), the name is preceded immediately by a
510 # tab and 5 hex digits for its code, with a \n before those. Named
511 # sequences won't have the 7th preceding character be a \n.
512 # (Actually, for the very first entry in the table this isn't strictly
513 # true: subtracting 7 will yield -1, and the substr below will
514 # therefore yield the very last character in the table, which should
515 # also be a \n, so the statement works anyway.)
516 if (substr($txt, $off[0] - 7, 1) eq "\n") {
517 $utf8 = pack("U", CORE::hex substr($txt, $off[0] - 6, 5));
518
519 # Handle the single loose matching special case, in which two names
520 # differ only by a single medial hyphen. If the original had a
521 # hyphen (or more) in the right place, then it is that one.
522 $utf8 = $HANGUL_JUNGSEONG_O_E_utf8
523 if $loose
524 && $utf8 eq $HANGUL_JUNGSEONG_OE_utf8
525 && $name =~ m/O \s* - [-\s]* E/ix;
526 # Note that this wouldn't work if there were a 2nd
527 # OE in the name
528 }
529 else {
530
531 # Here, is a named sequence. Need to go looking for the beginning,
532 # which is just after the \n from the previous entry in the table.
533 # The +1 skips past that newline, or, if the rindex() fails, to put
534 # us to an offset of zero.
535 my $charstart = rindex($txt, "\n", $off[0] - 7) + 1;
536 $utf8 = pack("U*", map { CORE::hex }
537 split " ", substr($txt, $charstart, $off[0] - $charstart - 1));
538 }
539 }
540
541 # Cache the input so as to not have to search the large table
542 # again, but only if it came from the one search that we cache.
543 # (Haven't bothered with the pain of sorting out scoping issues for the
544 # scripts searches.)
545 $cache_ref->{$name} = $utf8 if defined $cache_ref;
546 }
547 }
548
549
550 # Here, have the utf8. If the return is to be an ord, must be any single
551 # character.
552 if ($wants_ord) {
553 return ord($utf8) if length $utf8 == 1;
554 }
555 else {
556
557 # Here, wants string output. If utf8 is acceptable, just return what
558 # we've got; otherwise attempt to convert it to non-utf8 and return that.
559 my $in_bytes = ($runtime)
560 ? (caller $runtime)[8] & $bytes::hint_bits
561 : $^H & $bytes::hint_bits;
562 return $utf8 if (! $in_bytes || utf8::downgrade($utf8, 1)) # The 1 arg
563 # means don't die on failure
564 }
565
566 # Here, there is an error: either there are too many characters, or the
567 # result string needs to be non-utf8, and at least one character requires
568 # utf8. Prefer any official name over the input one for the error message.
569 if (@off) {
570 $name = substr($txt, $off[0], $off[1] - $off[0]) if @off;
571 }
572 else {
573 $name = (defined $save_input) ? $save_input : $_[0];
574 }
575
576 if ($wants_ord) {
577 # Only way to get here in this case is if result too long. Message
578 # assumes that our only caller that requires single char result is
579 # vianame.
580 carp "charnames::vianame() doesn't handle named sequences ($name). Use charnames::string_vianame() instead";
581 return;
582 }
583
584 # Only other possible failure here is from use bytes.
585 if ($runtime) {
586 carp not_legal_use_bytes_msg($name, $utf8);
587 return;
588 } else {
589 croak not_legal_use_bytes_msg($name, $utf8);
590 }
591
592} # lookup_name
593
594sub charnames {
595
596 # For \N{...}. Looks up the character name and returns the string
597 # representation of it.
598
599 # The first 0 arg means wants a string returned; the second that we are in
600 # compile time
601 return lookup_name($_[0], 0, 0);
602}
603
604sub import
605
# spent 139µs (25+114) within _charnames::import which was called: # once (25µs+114µs) by charnames::import at line 21 of charnames.pm
{
6061100ns shift; ## ignore class name
607
6081200ns if (not @_) {
609 carp("'use charnames' needs explicit imports list");
610 }
61112µs $^H{charnames} = \&charnames ;
61211µs $^H{charnames_ord_aliases} = {};
6131700ns $^H{charnames_name_aliases} = {};
6141800ns $^H{charnames_inverse_ords} = {};
615 # New fields must be added to %dummy_H, and the code in lookup_name()
616 # that copies fields from the runtime structure
617
618 ##
619 ## fill %h keys with our @_ args.
620 ##
6211700ns my ($promote, %h, @args) = (0);
62211µs while (my $arg = shift) {
6231300ns if ($arg eq ":alias") {
624 @_ or
625 croak ":alias needs an argument in charnames";
626 my $alias = shift;
627 if (ref $alias) {
628 ref $alias eq "HASH" or
629 croak "Only HASH reference supported as argument to :alias";
630 alias ($alias);
631 $promote = 1;
632 next;
633 }
634 if ($alias =~ m{:(\w+)$}) {
635 $1 eq "full" || $1 eq "loose" || $1 eq "short" and
636 croak ":alias cannot use existing pragma :$1 (reversed order?)";
637 alias_file ($1) and $promote = 1;
638 next;
639 }
640 alias_file ($alias) and $promote = 1;
641 next;
642 }
64312µs if (substr($arg, 0, 1) eq ':'
644 and ! ($arg eq ":full" || $arg eq ":short" || $arg eq ":loose"))
645 {
646 warn "unsupported special '$arg' in charnames";
647 next;
648 }
6491500ns push @args, $arg;
650 }
651
6521300ns @args == 0 && $promote and @args = (":full");
65312µs @h{@args} = (1) x @args;
654
655 # Don't leave these undefined as are tested for in lookup_names
65612µs $^H{charnames_full} = delete $h{':full'} || 0;
6571600ns $^H{charnames_loose} = delete $h{':loose'} || 0;
6581500ns $^H{charnames_short} = delete $h{':short'} || 0;
65911µs my @scripts = map { uc quotemeta } keys %h;
660
661 ##
662 ## If utf8? warnings are enabled, and some scripts were given,
663 ## see if at least we can find one letter from each script.
664 ##
66512µs1114µs if (warnings::enabled('utf8') && @scripts) {
# spent 114µs making 1 call to warnings::enabled
666 $txt = do "unicore/Name.pl" unless $txt;
667
668 for my $script (@scripts) {
669 if (not $txt =~ m/\t$script (?:CAPITAL |SMALL )?LETTER /) {
670 warnings::warn('utf8', "No such script: '$script'");
671 $script = quotemeta $script; # Escape it, for use in the re.
672 }
673 }
674 }
675
676 # %^H gets stringified, so serialize it ourselves so can extract the
677 # real data back later.
67813µs $^H{charnames_stringified_ords} = join ",", %{$^H{charnames_ord_aliases}};
67911µs $^H{charnames_stringified_names} = join ",", %{$^H{charnames_name_aliases}};
6801900ns $^H{charnames_stringified_inverse_ords} = join ",", %{$^H{charnames_inverse_ords}};
681
682 # Modify the input script names for loose name matching if that is also
683 # specified, similar to the way the base character name is prepared. They
684 # don't (currently, and hopefully never will) have dashes. These go into a
685 # regex, and have already been uppercased and quotemeta'd. Squeeze out all
686 # input underscores, blanks, and dashes. Then convert so will match a blank
687 # between any characters.
6881100ns if ($^H{charnames_loose}) {
689 for (my $i = 0; $i < @scripts; $i++) {
690 $scripts[$i] =~ s/[_ -]//g;
691 $scripts[$i] =~ s/ ( [^\\] ) (?= . ) /$1\\ ?/gx;
692 }
693 }
694
69514µs $^H{charnames_scripts} = join "|", @scripts; # Stringifiy them as a trie
696} # import
697
698# Cache of already looked-up values. This is set to only contain
699# official values, and user aliases can't override them, so scoping is
700# not an issue.
7011100nsmy %viacode;
702
703513µs42µsmy $no_name_code_points_re = join "|", map { sprintf("%05X",
# spent 2µs making 4 calls to utf8::unicode_to_native, avg 375ns/call
704 utf8::unicode_to_native($_)) }
705 0x80, 0x81, 0x84, 0x99;
706122µs216µs$no_name_code_points_re = qr/$no_name_code_points_re/;
# spent 15µs making 1 call to _charnames::CORE:regcomp # spent 900ns making 1 call to _charnames::CORE:qr
707
708sub viacode {
709
710 # Returns the name of the code point argument
711
712 if (@_ != 1) {
713 carp "charnames::viacode() expects one argument";
714 return;
715 }
716
717 my $arg = shift;
718
719 # This is derived from Unicode::UCD, where it is nearly the same as the
720 # function _getcode(), but here it makes sure that even a hex argument
721 # has the proper number of leading zeros, which is critical in
722 # matching against $txt below
723 # Must check if decimal first; see comments at that definition
724 my $hex;
725 if ($arg =~ $decimal_qr) {
726 $hex = sprintf "%05X", $arg;
727 } elsif ($arg =~ $hex_qr) {
728 $hex = CORE::hex $1;
729 $hex = utf8::unicode_to_native($hex) if $arg =~ /^[Uu]\+/;
730 # Below is the line that differs from the _getcode() source
731 $hex = sprintf "%05X", $hex;
732 } else {
733 carp("unexpected arg \"$arg\" to charnames::viacode()");
734 return;
735 }
736
737 return $viacode{$hex} if exists $viacode{$hex};
738
739 my $return;
740
741 # If the code point is above the max in the table, there's no point
742 # looking through it. Checking the length first is slightly faster
743 if (length($hex) <= 5 || CORE::hex($hex) <= 0x10FFFF) {
744 $txt = do "unicore/Name.pl" unless $txt;
745
746 # See if the name is algorithmically determinable.
747 my $algorithmic = charnames::code_point_to_name_special(CORE::hex $hex);
748 if (defined $algorithmic) {
749 $viacode{$hex} = $algorithmic;
750 return $algorithmic;
751 }
752
753 # Return the official name, if exists. It's unclear to me (khw) at
754 # this juncture if it is better to return a user-defined override, so
755 # leaving it as is for now.
756 if ($txt =~ m/^$hex\t/m) {
757
758 # The name starts with the next character and goes up to the
759 # next new-line. Using capturing parentheses above instead of
760 # @+ more than doubles the execution time in Perl 5.13
761 $return = substr($txt, $+[0], index($txt, "\n", $+[0]) - $+[0]);
762
763 # If not one of these 4 code points, return what we've found.
764 if ($hex !~ / ^ $no_name_code_points_re $ /x) {
765 $viacode{$hex} = $return;
766 return $return;
767 }
768
769 # For backwards compatibility, we don't return the official name of
770 # the 4 code points if there are user-defined aliases for them -- so
771 # continue looking.
772 }
773 }
774
775 # See if there is a user name for it, before giving up completely.
776 # First get the scoped aliases, give up if have none.
777 my $H_ref = (caller(1))[10];
778 return if ! defined $return
779 && (! defined $H_ref
780 || ! exists $H_ref->{charnames_stringified_inverse_ords});
781
782 my %code_point_aliases;
783 if (defined $H_ref->{charnames_stringified_inverse_ords}) {
784 %code_point_aliases = split ',',
785 $H_ref->{charnames_stringified_inverse_ords};
786 return $code_point_aliases{$hex} if exists $code_point_aliases{$hex};
787 }
788
789 # Here there is no user-defined alias, return any official one.
790 return $return if defined $return;
791
792 if (CORE::hex($hex) > 0x10FFFF
793 && warnings::enabled('non_unicode'))
794 {
795 carp "Unicode characters only allocated up to U+10FFFF (you asked for U+$hex)";
796 }
797 return;
798
799} # viacode
800
80119µs1;
802
803# ex: set ts=8 sts=2 sw=2 et:
 
# spent 4µs within _charnames::CORE:pack which was called 4 times, avg 1µs/call: # 4 times (4µs+0s) by charnames::BEGIN@6 at line 67, avg 1µs/call
sub _charnames::CORE:pack; # opcode
# spent 6µs within _charnames::CORE:qr which was called 3 times, avg 2µs/call: # once (4µs+0s) by charnames::BEGIN@6 at line 129 # once (900ns+0s) by charnames::BEGIN@6 at line 706 # once (900ns+0s) by charnames::BEGIN@6 at line 132
sub _charnames::CORE:qr; # opcode
# spent 15µs within _charnames::CORE:regcomp which was called: # once (15µs+0s) by charnames::BEGIN@6 at line 706
sub _charnames::CORE:regcomp; # opcode
Report produced by the NYTProf 5.06 Perl profiler, developed by Tim Bunce and Adam Kaplan.