[27174] | 1 | package HTTP::Headers;
|
---|
| 2 |
|
---|
| 3 | use strict;
|
---|
| 4 | use Carp ();
|
---|
| 5 |
|
---|
| 6 | use vars qw($VERSION $TRANSLATE_UNDERSCORE);
|
---|
| 7 | $VERSION = "6.05";
|
---|
| 8 |
|
---|
| 9 | # The $TRANSLATE_UNDERSCORE variable controls whether '_' can be used
|
---|
| 10 | # as a replacement for '-' in header field names.
|
---|
| 11 | $TRANSLATE_UNDERSCORE = 1 unless defined $TRANSLATE_UNDERSCORE;
|
---|
| 12 |
|
---|
| 13 | # "Good Practice" order of HTTP message headers:
|
---|
| 14 | # - General-Headers
|
---|
| 15 | # - Request-Headers
|
---|
| 16 | # - Response-Headers
|
---|
| 17 | # - Entity-Headers
|
---|
| 18 |
|
---|
| 19 | my @general_headers = qw(
|
---|
| 20 | Cache-Control Connection Date Pragma Trailer Transfer-Encoding Upgrade
|
---|
| 21 | Via Warning
|
---|
| 22 | );
|
---|
| 23 |
|
---|
| 24 | my @request_headers = qw(
|
---|
| 25 | Accept Accept-Charset Accept-Encoding Accept-Language
|
---|
| 26 | Authorization Expect From Host
|
---|
| 27 | If-Match If-Modified-Since If-None-Match If-Range If-Unmodified-Since
|
---|
| 28 | Max-Forwards Proxy-Authorization Range Referer TE User-Agent
|
---|
| 29 | );
|
---|
| 30 |
|
---|
| 31 | my @response_headers = qw(
|
---|
| 32 | Accept-Ranges Age ETag Location Proxy-Authenticate Retry-After Server
|
---|
| 33 | Vary WWW-Authenticate
|
---|
| 34 | );
|
---|
| 35 |
|
---|
| 36 | my @entity_headers = qw(
|
---|
| 37 | Allow Content-Encoding Content-Language Content-Length Content-Location
|
---|
| 38 | Content-MD5 Content-Range Content-Type Expires Last-Modified
|
---|
| 39 | );
|
---|
| 40 |
|
---|
| 41 | my %entity_header = map { lc($_) => 1 } @entity_headers;
|
---|
| 42 |
|
---|
| 43 | my @header_order = (
|
---|
| 44 | @general_headers,
|
---|
| 45 | @request_headers,
|
---|
| 46 | @response_headers,
|
---|
| 47 | @entity_headers,
|
---|
| 48 | );
|
---|
| 49 |
|
---|
| 50 | # Make alternative representations of @header_order. This is used
|
---|
| 51 | # for sorting and case matching.
|
---|
| 52 | my %header_order;
|
---|
| 53 | my %standard_case;
|
---|
| 54 |
|
---|
| 55 | {
|
---|
| 56 | my $i = 0;
|
---|
| 57 | for (@header_order) {
|
---|
| 58 | my $lc = lc $_;
|
---|
| 59 | $header_order{$lc} = ++$i;
|
---|
| 60 | $standard_case{$lc} = $_;
|
---|
| 61 | }
|
---|
| 62 | }
|
---|
| 63 |
|
---|
| 64 |
|
---|
| 65 |
|
---|
| 66 | sub new
|
---|
| 67 | {
|
---|
| 68 | my($class) = shift;
|
---|
| 69 | my $self = bless {}, $class;
|
---|
| 70 | $self->header(@_) if @_; # set up initial headers
|
---|
| 71 | $self;
|
---|
| 72 | }
|
---|
| 73 |
|
---|
| 74 |
|
---|
| 75 | sub header
|
---|
| 76 | {
|
---|
| 77 | my $self = shift;
|
---|
| 78 | Carp::croak('Usage: $h->header($field, ...)') unless @_;
|
---|
| 79 | my(@old);
|
---|
| 80 | my %seen;
|
---|
| 81 | while (@_) {
|
---|
| 82 | my $field = shift;
|
---|
| 83 | my $op = @_ ? ($seen{lc($field)}++ ? 'PUSH' : 'SET') : 'GET';
|
---|
| 84 | @old = $self->_header($field, shift, $op);
|
---|
| 85 | }
|
---|
| 86 | return @old if wantarray;
|
---|
| 87 | return $old[0] if @old <= 1;
|
---|
| 88 | join(", ", @old);
|
---|
| 89 | }
|
---|
| 90 |
|
---|
| 91 | sub clear
|
---|
| 92 | {
|
---|
| 93 | my $self = shift;
|
---|
| 94 | %$self = ();
|
---|
| 95 | }
|
---|
| 96 |
|
---|
| 97 |
|
---|
| 98 | sub push_header
|
---|
| 99 | {
|
---|
| 100 | my $self = shift;
|
---|
| 101 | return $self->_header(@_, 'PUSH_H') if @_ == 2;
|
---|
| 102 | while (@_) {
|
---|
| 103 | $self->_header(splice(@_, 0, 2), 'PUSH_H');
|
---|
| 104 | }
|
---|
| 105 | }
|
---|
| 106 |
|
---|
| 107 |
|
---|
| 108 | sub init_header
|
---|
| 109 | {
|
---|
| 110 | Carp::croak('Usage: $h->init_header($field, $val)') if @_ != 3;
|
---|
| 111 | shift->_header(@_, 'INIT');
|
---|
| 112 | }
|
---|
| 113 |
|
---|
| 114 |
|
---|
| 115 | sub remove_header
|
---|
| 116 | {
|
---|
| 117 | my($self, @fields) = @_;
|
---|
| 118 | my $field;
|
---|
| 119 | my @values;
|
---|
| 120 | foreach $field (@fields) {
|
---|
| 121 | $field =~ tr/_/-/ if $field !~ /^:/ && $TRANSLATE_UNDERSCORE;
|
---|
| 122 | my $v = delete $self->{lc $field};
|
---|
| 123 | push(@values, ref($v) eq 'ARRAY' ? @$v : $v) if defined $v;
|
---|
| 124 | }
|
---|
| 125 | return @values;
|
---|
| 126 | }
|
---|
| 127 |
|
---|
| 128 | sub remove_content_headers
|
---|
| 129 | {
|
---|
| 130 | my $self = shift;
|
---|
| 131 | unless (defined(wantarray)) {
|
---|
| 132 | # fast branch that does not create return object
|
---|
| 133 | delete @$self{grep $entity_header{$_} || /^content-/, keys %$self};
|
---|
| 134 | return;
|
---|
| 135 | }
|
---|
| 136 |
|
---|
| 137 | my $c = ref($self)->new;
|
---|
| 138 | for my $f (grep $entity_header{$_} || /^content-/, keys %$self) {
|
---|
| 139 | $c->{$f} = delete $self->{$f};
|
---|
| 140 | }
|
---|
| 141 | if (exists $self->{'::std_case'}) {
|
---|
| 142 | $c->{'::std_case'} = $self->{'::std_case'};
|
---|
| 143 | }
|
---|
| 144 | $c;
|
---|
| 145 | }
|
---|
| 146 |
|
---|
| 147 |
|
---|
| 148 | sub _header
|
---|
| 149 | {
|
---|
| 150 | my($self, $field, $val, $op) = @_;
|
---|
| 151 |
|
---|
| 152 | Carp::croak("Illegal field name '$field'")
|
---|
| 153 | if rindex($field, ':') > 1 || !length($field);
|
---|
| 154 |
|
---|
| 155 | unless ($field =~ /^:/) {
|
---|
| 156 | $field =~ tr/_/-/ if $TRANSLATE_UNDERSCORE;
|
---|
| 157 | my $old = $field;
|
---|
| 158 | $field = lc $field;
|
---|
| 159 | unless($standard_case{$field} || $self->{'::std_case'}{$field}) {
|
---|
| 160 | # generate a %std_case entry for this field
|
---|
| 161 | $old =~ s/\b(\w)/\u$1/g;
|
---|
| 162 | $self->{'::std_case'}{$field} = $old;
|
---|
| 163 | }
|
---|
| 164 | }
|
---|
| 165 |
|
---|
| 166 | $op ||= defined($val) ? 'SET' : 'GET';
|
---|
| 167 | if ($op eq 'PUSH_H') {
|
---|
| 168 | # Like PUSH but where we don't care about the return value
|
---|
| 169 | if (exists $self->{$field}) {
|
---|
| 170 | my $h = $self->{$field};
|
---|
| 171 | if (ref($h) eq 'ARRAY') {
|
---|
| 172 | push(@$h, ref($val) eq "ARRAY" ? @$val : $val);
|
---|
| 173 | }
|
---|
| 174 | else {
|
---|
| 175 | $self->{$field} = [$h, ref($val) eq "ARRAY" ? @$val : $val]
|
---|
| 176 | }
|
---|
| 177 | return;
|
---|
| 178 | }
|
---|
| 179 | $self->{$field} = $val;
|
---|
| 180 | return;
|
---|
| 181 | }
|
---|
| 182 |
|
---|
| 183 | my $h = $self->{$field};
|
---|
| 184 | my @old = ref($h) eq 'ARRAY' ? @$h : (defined($h) ? ($h) : ());
|
---|
| 185 |
|
---|
| 186 | unless ($op eq 'GET' || ($op eq 'INIT' && @old)) {
|
---|
| 187 | if (defined($val)) {
|
---|
| 188 | my @new = ($op eq 'PUSH') ? @old : ();
|
---|
| 189 | if (ref($val) ne 'ARRAY') {
|
---|
| 190 | push(@new, $val);
|
---|
| 191 | }
|
---|
| 192 | else {
|
---|
| 193 | push(@new, @$val);
|
---|
| 194 | }
|
---|
| 195 | $self->{$field} = @new > 1 ? \@new : $new[0];
|
---|
| 196 | }
|
---|
| 197 | elsif ($op ne 'PUSH') {
|
---|
| 198 | delete $self->{$field};
|
---|
| 199 | }
|
---|
| 200 | }
|
---|
| 201 | @old;
|
---|
| 202 | }
|
---|
| 203 |
|
---|
| 204 |
|
---|
| 205 | sub _sorted_field_names
|
---|
| 206 | {
|
---|
| 207 | my $self = shift;
|
---|
| 208 | return [ sort {
|
---|
| 209 | ($header_order{$a} || 999) <=> ($header_order{$b} || 999) ||
|
---|
| 210 | $a cmp $b
|
---|
| 211 | } grep !/^::/, keys %$self ];
|
---|
| 212 | }
|
---|
| 213 |
|
---|
| 214 |
|
---|
| 215 | sub header_field_names {
|
---|
| 216 | my $self = shift;
|
---|
| 217 | return map $standard_case{$_} || $self->{'::std_case'}{$_} || $_, @{ $self->_sorted_field_names },
|
---|
| 218 | if wantarray;
|
---|
| 219 | return grep !/^::/, keys %$self;
|
---|
| 220 | }
|
---|
| 221 |
|
---|
| 222 |
|
---|
| 223 | sub scan
|
---|
| 224 | {
|
---|
| 225 | my($self, $sub) = @_;
|
---|
| 226 | my $key;
|
---|
| 227 | for $key (@{ $self->_sorted_field_names }) {
|
---|
| 228 | my $vals = $self->{$key};
|
---|
| 229 | if (ref($vals) eq 'ARRAY') {
|
---|
| 230 | my $val;
|
---|
| 231 | for $val (@$vals) {
|
---|
| 232 | $sub->($standard_case{$key} || $self->{'::std_case'}{$key} || $key, $val);
|
---|
| 233 | }
|
---|
| 234 | }
|
---|
| 235 | else {
|
---|
| 236 | $sub->($standard_case{$key} || $self->{'::std_case'}{$key} || $key, $vals);
|
---|
| 237 | }
|
---|
| 238 | }
|
---|
| 239 | }
|
---|
| 240 |
|
---|
| 241 |
|
---|
| 242 | sub as_string
|
---|
| 243 | {
|
---|
| 244 | my($self, $endl) = @_;
|
---|
| 245 | $endl = "\n" unless defined $endl;
|
---|
| 246 |
|
---|
| 247 | my @result = ();
|
---|
| 248 | for my $key (@{ $self->_sorted_field_names }) {
|
---|
| 249 | next if index($key, '_') == 0;
|
---|
| 250 | my $vals = $self->{$key};
|
---|
| 251 | if ( ref($vals) eq 'ARRAY' ) {
|
---|
| 252 | for my $val (@$vals) {
|
---|
| 253 | my $field = $standard_case{$key} || $self->{'::std_case'}{$key} || $key;
|
---|
| 254 | $field =~ s/^://;
|
---|
| 255 | if ( index($val, "\n") >= 0 ) {
|
---|
| 256 | $val = _process_newline($val, $endl);
|
---|
| 257 | }
|
---|
| 258 | push @result, $field . ': ' . $val;
|
---|
| 259 | }
|
---|
| 260 | }
|
---|
| 261 | else {
|
---|
| 262 | my $field = $standard_case{$key} || $self->{'::std_case'}{$key} || $key;
|
---|
| 263 | $field =~ s/^://;
|
---|
| 264 | if ( index($vals, "\n") >= 0 ) {
|
---|
| 265 | $vals = _process_newline($vals, $endl);
|
---|
| 266 | }
|
---|
| 267 | push @result, $field . ': ' . $vals;
|
---|
| 268 | }
|
---|
| 269 | }
|
---|
| 270 |
|
---|
| 271 | join($endl, @result, '');
|
---|
| 272 | }
|
---|
| 273 |
|
---|
| 274 | sub _process_newline {
|
---|
| 275 | local $_ = shift;
|
---|
| 276 | my $endl = shift;
|
---|
| 277 | # must handle header values with embedded newlines with care
|
---|
| 278 | s/\s+$//; # trailing newlines and space must go
|
---|
| 279 | s/\n(\x0d?\n)+/\n/g; # no empty lines
|
---|
| 280 | s/\n([^\040\t])/\n $1/g; # initial space for continuation
|
---|
| 281 | s/\n/$endl/g; # substitute with requested line ending
|
---|
| 282 | $_;
|
---|
| 283 | }
|
---|
| 284 |
|
---|
| 285 |
|
---|
| 286 |
|
---|
| 287 | if (eval { require Storable; 1 }) {
|
---|
| 288 | *clone = \&Storable::dclone;
|
---|
| 289 | } else {
|
---|
| 290 | *clone = sub {
|
---|
| 291 | my $self = shift;
|
---|
| 292 | my $clone = HTTP::Headers->new;
|
---|
| 293 | $self->scan(sub { $clone->push_header(@_);} );
|
---|
| 294 | $clone;
|
---|
| 295 | };
|
---|
| 296 | }
|
---|
| 297 |
|
---|
| 298 |
|
---|
| 299 | sub _date_header
|
---|
| 300 | {
|
---|
| 301 | require HTTP::Date;
|
---|
| 302 | my($self, $header, $time) = @_;
|
---|
| 303 | my($old) = $self->_header($header);
|
---|
| 304 | if (defined $time) {
|
---|
| 305 | $self->_header($header, HTTP::Date::time2str($time));
|
---|
| 306 | }
|
---|
| 307 | $old =~ s/;.*// if defined($old);
|
---|
| 308 | HTTP::Date::str2time($old);
|
---|
| 309 | }
|
---|
| 310 |
|
---|
| 311 |
|
---|
| 312 | sub date { shift->_date_header('Date', @_); }
|
---|
| 313 | sub expires { shift->_date_header('Expires', @_); }
|
---|
| 314 | sub if_modified_since { shift->_date_header('If-Modified-Since', @_); }
|
---|
| 315 | sub if_unmodified_since { shift->_date_header('If-Unmodified-Since', @_); }
|
---|
| 316 | sub last_modified { shift->_date_header('Last-Modified', @_); }
|
---|
| 317 |
|
---|
| 318 | # This is used as a private LWP extension. The Client-Date header is
|
---|
| 319 | # added as a timestamp to a response when it has been received.
|
---|
| 320 | sub client_date { shift->_date_header('Client-Date', @_); }
|
---|
| 321 |
|
---|
| 322 | # The retry_after field is dual format (can also be a expressed as
|
---|
| 323 | # number of seconds from now), so we don't provide an easy way to
|
---|
| 324 | # access it until we have know how both these interfaces can be
|
---|
| 325 | # addressed. One possibility is to return a negative value for
|
---|
| 326 | # relative seconds and a positive value for epoch based time values.
|
---|
| 327 | #sub retry_after { shift->_date_header('Retry-After', @_); }
|
---|
| 328 |
|
---|
| 329 | sub content_type {
|
---|
| 330 | my $self = shift;
|
---|
| 331 | my $ct = $self->{'content-type'};
|
---|
| 332 | $self->{'content-type'} = shift if @_;
|
---|
| 333 | $ct = $ct->[0] if ref($ct) eq 'ARRAY';
|
---|
| 334 | return '' unless defined($ct) && length($ct);
|
---|
| 335 | my @ct = split(/;\s*/, $ct, 2);
|
---|
| 336 | for ($ct[0]) {
|
---|
| 337 | s/\s+//g;
|
---|
| 338 | $_ = lc($_);
|
---|
| 339 | }
|
---|
| 340 | wantarray ? @ct : $ct[0];
|
---|
| 341 | }
|
---|
| 342 |
|
---|
| 343 | sub content_type_charset {
|
---|
| 344 | my $self = shift;
|
---|
| 345 | require HTTP::Headers::Util;
|
---|
| 346 | my $h = $self->{'content-type'};
|
---|
| 347 | $h = $h->[0] if ref($h);
|
---|
| 348 | $h = "" unless defined $h;
|
---|
| 349 | my @v = HTTP::Headers::Util::split_header_words($h);
|
---|
| 350 | if (@v) {
|
---|
| 351 | my($ct, undef, %ct_param) = @{$v[0]};
|
---|
| 352 | my $charset = $ct_param{charset};
|
---|
| 353 | if ($ct) {
|
---|
| 354 | $ct = lc($ct);
|
---|
| 355 | $ct =~ s/\s+//;
|
---|
| 356 | }
|
---|
| 357 | if ($charset) {
|
---|
| 358 | $charset = uc($charset);
|
---|
| 359 | $charset =~ s/^\s+//; $charset =~ s/\s+\z//;
|
---|
| 360 | undef($charset) if $charset eq "";
|
---|
| 361 | }
|
---|
| 362 | return $ct, $charset if wantarray;
|
---|
| 363 | return $charset;
|
---|
| 364 | }
|
---|
| 365 | return undef, undef if wantarray;
|
---|
| 366 | return undef;
|
---|
| 367 | }
|
---|
| 368 |
|
---|
| 369 | sub content_is_text {
|
---|
| 370 | my $self = shift;
|
---|
| 371 | return $self->content_type =~ m,^text/,;
|
---|
| 372 | }
|
---|
| 373 |
|
---|
| 374 | sub content_is_html {
|
---|
| 375 | my $self = shift;
|
---|
| 376 | return $self->content_type eq 'text/html' || $self->content_is_xhtml;
|
---|
| 377 | }
|
---|
| 378 |
|
---|
| 379 | sub content_is_xhtml {
|
---|
| 380 | my $ct = shift->content_type;
|
---|
| 381 | return $ct eq "application/xhtml+xml" ||
|
---|
| 382 | $ct eq "application/vnd.wap.xhtml+xml";
|
---|
| 383 | }
|
---|
| 384 |
|
---|
| 385 | sub content_is_xml {
|
---|
| 386 | my $ct = shift->content_type;
|
---|
| 387 | return 1 if $ct eq "text/xml";
|
---|
| 388 | return 1 if $ct eq "application/xml";
|
---|
| 389 | return 1 if $ct =~ /\+xml$/;
|
---|
| 390 | return 0;
|
---|
| 391 | }
|
---|
| 392 |
|
---|
| 393 | sub referer {
|
---|
| 394 | my $self = shift;
|
---|
| 395 | if (@_ && $_[0] =~ /#/) {
|
---|
| 396 | # Strip fragment per RFC 2616, section 14.36.
|
---|
| 397 | my $uri = shift;
|
---|
| 398 | if (ref($uri)) {
|
---|
| 399 | $uri = $uri->clone;
|
---|
| 400 | $uri->fragment(undef);
|
---|
| 401 | }
|
---|
| 402 | else {
|
---|
| 403 | $uri =~ s/\#.*//;
|
---|
| 404 | }
|
---|
| 405 | unshift @_, $uri;
|
---|
| 406 | }
|
---|
| 407 | ($self->_header('Referer', @_))[0];
|
---|
| 408 | }
|
---|
| 409 | *referrer = \&referer; # on tchrist's request
|
---|
| 410 |
|
---|
| 411 | sub title { (shift->_header('Title', @_))[0] }
|
---|
| 412 | sub content_encoding { (shift->_header('Content-Encoding', @_))[0] }
|
---|
| 413 | sub content_language { (shift->_header('Content-Language', @_))[0] }
|
---|
| 414 | sub content_length { (shift->_header('Content-Length', @_))[0] }
|
---|
| 415 |
|
---|
| 416 | sub user_agent { (shift->_header('User-Agent', @_))[0] }
|
---|
| 417 | sub server { (shift->_header('Server', @_))[0] }
|
---|
| 418 |
|
---|
| 419 | sub from { (shift->_header('From', @_))[0] }
|
---|
| 420 | sub warning { (shift->_header('Warning', @_))[0] }
|
---|
| 421 |
|
---|
| 422 | sub www_authenticate { (shift->_header('WWW-Authenticate', @_))[0] }
|
---|
| 423 | sub authorization { (shift->_header('Authorization', @_))[0] }
|
---|
| 424 |
|
---|
| 425 | sub proxy_authenticate { (shift->_header('Proxy-Authenticate', @_))[0] }
|
---|
| 426 | sub proxy_authorization { (shift->_header('Proxy-Authorization', @_))[0] }
|
---|
| 427 |
|
---|
| 428 | sub authorization_basic { shift->_basic_auth("Authorization", @_) }
|
---|
| 429 | sub proxy_authorization_basic { shift->_basic_auth("Proxy-Authorization", @_) }
|
---|
| 430 |
|
---|
| 431 | sub _basic_auth {
|
---|
| 432 | require MIME::Base64;
|
---|
| 433 | my($self, $h, $user, $passwd) = @_;
|
---|
| 434 | my($old) = $self->_header($h);
|
---|
| 435 | if (defined $user) {
|
---|
| 436 | Carp::croak("Basic authorization user name can't contain ':'")
|
---|
| 437 | if $user =~ /:/;
|
---|
| 438 | $passwd = '' unless defined $passwd;
|
---|
| 439 | $self->_header($h => 'Basic ' .
|
---|
| 440 | MIME::Base64::encode("$user:$passwd", ''));
|
---|
| 441 | }
|
---|
| 442 | if (defined $old && $old =~ s/^\s*Basic\s+//) {
|
---|
| 443 | my $val = MIME::Base64::decode($old);
|
---|
| 444 | return $val unless wantarray;
|
---|
| 445 | return split(/:/, $val, 2);
|
---|
| 446 | }
|
---|
| 447 | return;
|
---|
| 448 | }
|
---|
| 449 |
|
---|
| 450 |
|
---|
| 451 | 1;
|
---|
| 452 |
|
---|
| 453 | __END__
|
---|
| 454 |
|
---|
| 455 | =head1 NAME
|
---|
| 456 |
|
---|
| 457 | HTTP::Headers - Class encapsulating HTTP Message headers
|
---|
| 458 |
|
---|
| 459 | =head1 SYNOPSIS
|
---|
| 460 |
|
---|
| 461 | require HTTP::Headers;
|
---|
| 462 | $h = HTTP::Headers->new;
|
---|
| 463 |
|
---|
| 464 | $h->header('Content-Type' => 'text/plain'); # set
|
---|
| 465 | $ct = $h->header('Content-Type'); # get
|
---|
| 466 | $h->remove_header('Content-Type'); # delete
|
---|
| 467 |
|
---|
| 468 | =head1 DESCRIPTION
|
---|
| 469 |
|
---|
| 470 | The C<HTTP::Headers> class encapsulates HTTP-style message headers.
|
---|
| 471 | The headers consist of attribute-value pairs also called fields, which
|
---|
| 472 | may be repeated, and which are printed in a particular order. The
|
---|
| 473 | field names are cases insensitive.
|
---|
| 474 |
|
---|
| 475 | Instances of this class are usually created as member variables of the
|
---|
| 476 | C<HTTP::Request> and C<HTTP::Response> classes, internal to the
|
---|
| 477 | library.
|
---|
| 478 |
|
---|
| 479 | The following methods are available:
|
---|
| 480 |
|
---|
| 481 | =over 4
|
---|
| 482 |
|
---|
| 483 | =item $h = HTTP::Headers->new
|
---|
| 484 |
|
---|
| 485 | Constructs a new C<HTTP::Headers> object. You might pass some initial
|
---|
| 486 | attribute-value pairs as parameters to the constructor. I<E.g.>:
|
---|
| 487 |
|
---|
| 488 | $h = HTTP::Headers->new(
|
---|
| 489 | Date => 'Thu, 03 Feb 1994 00:00:00 GMT',
|
---|
| 490 | Content_Type => 'text/html; version=3.2',
|
---|
| 491 | Content_Base => 'http://www.perl.org/');
|
---|
| 492 |
|
---|
| 493 | The constructor arguments are passed to the C<header> method which is
|
---|
| 494 | described below.
|
---|
| 495 |
|
---|
| 496 | =item $h->clone
|
---|
| 497 |
|
---|
| 498 | Returns a copy of this C<HTTP::Headers> object.
|
---|
| 499 |
|
---|
| 500 | =item $h->header( $field )
|
---|
| 501 |
|
---|
| 502 | =item $h->header( $field => $value )
|
---|
| 503 |
|
---|
| 504 | =item $h->header( $f1 => $v1, $f2 => $v2, ... )
|
---|
| 505 |
|
---|
| 506 | Get or set the value of one or more header fields. The header field
|
---|
| 507 | name ($field) is not case sensitive. To make the life easier for perl
|
---|
| 508 | users who wants to avoid quoting before the => operator, you can use
|
---|
| 509 | '_' as a replacement for '-' in header names.
|
---|
| 510 |
|
---|
| 511 | The header() method accepts multiple ($field => $value) pairs, which
|
---|
| 512 | means that you can update several fields with a single invocation.
|
---|
| 513 |
|
---|
| 514 | The $value argument may be a plain string or a reference to an array
|
---|
| 515 | of strings for a multi-valued field. If the $value is provided as
|
---|
| 516 | C<undef> then the field is removed. If the $value is not given, then
|
---|
| 517 | that header field will remain unchanged.
|
---|
| 518 |
|
---|
| 519 | The old value (or values) of the last of the header fields is returned.
|
---|
| 520 | If no such field exists C<undef> will be returned.
|
---|
| 521 |
|
---|
| 522 | A multi-valued field will be returned as separate values in list
|
---|
| 523 | context and will be concatenated with ", " as separator in scalar
|
---|
| 524 | context. The HTTP spec (RFC 2616) promise that joining multiple
|
---|
| 525 | values in this way will not change the semantic of a header field, but
|
---|
| 526 | in practice there are cases like old-style Netscape cookies (see
|
---|
| 527 | L<HTTP::Cookies>) where "," is used as part of the syntax of a single
|
---|
| 528 | field value.
|
---|
| 529 |
|
---|
| 530 | Examples:
|
---|
| 531 |
|
---|
| 532 | $header->header(MIME_Version => '1.0',
|
---|
| 533 | User_Agent => 'My-Web-Client/0.01');
|
---|
| 534 | $header->header(Accept => "text/html, text/plain, image/*");
|
---|
| 535 | $header->header(Accept => [qw(text/html text/plain image/*)]);
|
---|
| 536 | @accepts = $header->header('Accept'); # get multiple values
|
---|
| 537 | $accepts = $header->header('Accept'); # get values as a single string
|
---|
| 538 |
|
---|
| 539 | =item $h->push_header( $field => $value )
|
---|
| 540 |
|
---|
| 541 | =item $h->push_header( $f1 => $v1, $f2 => $v2, ... )
|
---|
| 542 |
|
---|
| 543 | Add a new field value for the specified header field. Previous values
|
---|
| 544 | for the same field are retained.
|
---|
| 545 |
|
---|
| 546 | As for the header() method, the field name ($field) is not case
|
---|
| 547 | sensitive and '_' can be used as a replacement for '-'.
|
---|
| 548 |
|
---|
| 549 | The $value argument may be a scalar or a reference to a list of
|
---|
| 550 | scalars.
|
---|
| 551 |
|
---|
| 552 | $header->push_header(Accept => 'image/jpeg');
|
---|
| 553 | $header->push_header(Accept => [map "image/$_", qw(gif png tiff)]);
|
---|
| 554 |
|
---|
| 555 | =item $h->init_header( $field => $value )
|
---|
| 556 |
|
---|
| 557 | Set the specified header to the given value, but only if no previous
|
---|
| 558 | value for that field is set.
|
---|
| 559 |
|
---|
| 560 | The header field name ($field) is not case sensitive and '_'
|
---|
| 561 | can be used as a replacement for '-'.
|
---|
| 562 |
|
---|
| 563 | The $value argument may be a scalar or a reference to a list of
|
---|
| 564 | scalars.
|
---|
| 565 |
|
---|
| 566 | =item $h->remove_header( $field, ... )
|
---|
| 567 |
|
---|
| 568 | This function removes the header fields with the specified names.
|
---|
| 569 |
|
---|
| 570 | The header field names ($field) are not case sensitive and '_'
|
---|
| 571 | can be used as a replacement for '-'.
|
---|
| 572 |
|
---|
| 573 | The return value is the values of the fields removed. In scalar
|
---|
| 574 | context the number of fields removed is returned.
|
---|
| 575 |
|
---|
| 576 | Note that if you pass in multiple field names then it is generally not
|
---|
| 577 | possible to tell which of the returned values belonged to which field.
|
---|
| 578 |
|
---|
| 579 | =item $h->remove_content_headers
|
---|
| 580 |
|
---|
| 581 | This will remove all the header fields used to describe the content of
|
---|
| 582 | a message. All header field names prefixed with C<Content-> fall
|
---|
| 583 | into this category, as well as C<Allow>, C<Expires> and
|
---|
| 584 | C<Last-Modified>. RFC 2616 denotes these fields as I<Entity Header
|
---|
| 585 | Fields>.
|
---|
| 586 |
|
---|
| 587 | The return value is a new C<HTTP::Headers> object that contains the
|
---|
| 588 | removed headers only.
|
---|
| 589 |
|
---|
| 590 | =item $h->clear
|
---|
| 591 |
|
---|
| 592 | This will remove all header fields.
|
---|
| 593 |
|
---|
| 594 | =item $h->header_field_names
|
---|
| 595 |
|
---|
| 596 | Returns the list of distinct names for the fields present in the
|
---|
| 597 | header. The field names have case as suggested by HTTP spec, and the
|
---|
| 598 | names are returned in the recommended "Good Practice" order.
|
---|
| 599 |
|
---|
| 600 | In scalar context return the number of distinct field names.
|
---|
| 601 |
|
---|
| 602 | =item $h->scan( \&process_header_field )
|
---|
| 603 |
|
---|
| 604 | Apply a subroutine to each header field in turn. The callback routine
|
---|
| 605 | is called with two parameters; the name of the field and a single
|
---|
| 606 | value (a string). If a header field is multi-valued, then the
|
---|
| 607 | routine is called once for each value. The field name passed to the
|
---|
| 608 | callback routine has case as suggested by HTTP spec, and the headers
|
---|
| 609 | will be visited in the recommended "Good Practice" order.
|
---|
| 610 |
|
---|
| 611 | Any return values of the callback routine are ignored. The loop can
|
---|
| 612 | be broken by raising an exception (C<die>), but the caller of scan()
|
---|
| 613 | would have to trap the exception itself.
|
---|
| 614 |
|
---|
| 615 | =item $h->as_string
|
---|
| 616 |
|
---|
| 617 | =item $h->as_string( $eol )
|
---|
| 618 |
|
---|
| 619 | Return the header fields as a formatted MIME header. Since it
|
---|
| 620 | internally uses the C<scan> method to build the string, the result
|
---|
| 621 | will use case as suggested by HTTP spec, and it will follow
|
---|
| 622 | recommended "Good Practice" of ordering the header fields. Long header
|
---|
| 623 | values are not folded.
|
---|
| 624 |
|
---|
| 625 | The optional $eol parameter specifies the line ending sequence to
|
---|
| 626 | use. The default is "\n". Embedded "\n" characters in header field
|
---|
| 627 | values will be substituted with this line ending sequence.
|
---|
| 628 |
|
---|
| 629 | =back
|
---|
| 630 |
|
---|
| 631 | =head1 CONVENIENCE METHODS
|
---|
| 632 |
|
---|
| 633 | The most frequently used headers can also be accessed through the
|
---|
| 634 | following convenience methods. Most of these methods can both be used to read
|
---|
| 635 | and to set the value of a header. The header value is set if you pass
|
---|
| 636 | an argument to the method. The old header value is always returned.
|
---|
| 637 | If the given header did not exist then C<undef> is returned.
|
---|
| 638 |
|
---|
| 639 | Methods that deal with dates/times always convert their value to system
|
---|
| 640 | time (seconds since Jan 1, 1970) and they also expect this kind of
|
---|
| 641 | value when the header value is set.
|
---|
| 642 |
|
---|
| 643 | =over 4
|
---|
| 644 |
|
---|
| 645 | =item $h->date
|
---|
| 646 |
|
---|
| 647 | This header represents the date and time at which the message was
|
---|
| 648 | originated. I<E.g.>:
|
---|
| 649 |
|
---|
| 650 | $h->date(time); # set current date
|
---|
| 651 |
|
---|
| 652 | =item $h->expires
|
---|
| 653 |
|
---|
| 654 | This header gives the date and time after which the entity should be
|
---|
| 655 | considered stale.
|
---|
| 656 |
|
---|
| 657 | =item $h->if_modified_since
|
---|
| 658 |
|
---|
| 659 | =item $h->if_unmodified_since
|
---|
| 660 |
|
---|
| 661 | These header fields are used to make a request conditional. If the requested
|
---|
| 662 | resource has (or has not) been modified since the time specified in this field,
|
---|
| 663 | then the server will return a C<304 Not Modified> response instead of
|
---|
| 664 | the document itself.
|
---|
| 665 |
|
---|
| 666 | =item $h->last_modified
|
---|
| 667 |
|
---|
| 668 | This header indicates the date and time at which the resource was last
|
---|
| 669 | modified. I<E.g.>:
|
---|
| 670 |
|
---|
| 671 | # check if document is more than 1 hour old
|
---|
| 672 | if (my $last_mod = $h->last_modified) {
|
---|
| 673 | if ($last_mod < time - 60*60) {
|
---|
| 674 | ...
|
---|
| 675 | }
|
---|
| 676 | }
|
---|
| 677 |
|
---|
| 678 | =item $h->content_type
|
---|
| 679 |
|
---|
| 680 | The Content-Type header field indicates the media type of the message
|
---|
| 681 | content. I<E.g.>:
|
---|
| 682 |
|
---|
| 683 | $h->content_type('text/html');
|
---|
| 684 |
|
---|
| 685 | The value returned will be converted to lower case, and potential
|
---|
| 686 | parameters will be chopped off and returned as a separate value if in
|
---|
| 687 | an array context. If there is no such header field, then the empty
|
---|
| 688 | string is returned. This makes it safe to do the following:
|
---|
| 689 |
|
---|
| 690 | if ($h->content_type eq 'text/html') {
|
---|
| 691 | # we enter this place even if the real header value happens to
|
---|
| 692 | # be 'TEXT/HTML; version=3.0'
|
---|
| 693 | ...
|
---|
| 694 | }
|
---|
| 695 |
|
---|
| 696 | =item $h->content_type_charset
|
---|
| 697 |
|
---|
| 698 | Returns the upper-cased charset specified in the Content-Type header. In list
|
---|
| 699 | context return the lower-cased bare content type followed by the upper-cased
|
---|
| 700 | charset. Both values will be C<undef> if not specified in the header.
|
---|
| 701 |
|
---|
| 702 | =item $h->content_is_text
|
---|
| 703 |
|
---|
| 704 | Returns TRUE if the Content-Type header field indicate that the
|
---|
| 705 | content is textual.
|
---|
| 706 |
|
---|
| 707 | =item $h->content_is_html
|
---|
| 708 |
|
---|
| 709 | Returns TRUE if the Content-Type header field indicate that the
|
---|
| 710 | content is some kind of HTML (including XHTML). This method can't be
|
---|
| 711 | used to set Content-Type.
|
---|
| 712 |
|
---|
| 713 | =item $h->content_is_xhtml
|
---|
| 714 |
|
---|
| 715 | Returns TRUE if the Content-Type header field indicate that the
|
---|
| 716 | content is XHTML. This method can't be used to set Content-Type.
|
---|
| 717 |
|
---|
| 718 | =item $h->content_is_xml
|
---|
| 719 |
|
---|
| 720 | Returns TRUE if the Content-Type header field indicate that the
|
---|
| 721 | content is XML. This method can't be used to set Content-Type.
|
---|
| 722 |
|
---|
| 723 | =item $h->content_encoding
|
---|
| 724 |
|
---|
| 725 | The Content-Encoding header field is used as a modifier to the
|
---|
| 726 | media type. When present, its value indicates what additional
|
---|
| 727 | encoding mechanism has been applied to the resource.
|
---|
| 728 |
|
---|
| 729 | =item $h->content_length
|
---|
| 730 |
|
---|
| 731 | A decimal number indicating the size in bytes of the message content.
|
---|
| 732 |
|
---|
| 733 | =item $h->content_language
|
---|
| 734 |
|
---|
| 735 | The natural language(s) of the intended audience for the message
|
---|
| 736 | content. The value is one or more language tags as defined by RFC
|
---|
| 737 | 1766. Eg. "no" for some kind of Norwegian and "en-US" for English the
|
---|
| 738 | way it is written in the US.
|
---|
| 739 |
|
---|
| 740 | =item $h->title
|
---|
| 741 |
|
---|
| 742 | The title of the document. In libwww-perl this header will be
|
---|
| 743 | initialized automatically from the E<lt>TITLE>...E<lt>/TITLE> element
|
---|
| 744 | of HTML documents. I<This header is no longer part of the HTTP
|
---|
| 745 | standard.>
|
---|
| 746 |
|
---|
| 747 | =item $h->user_agent
|
---|
| 748 |
|
---|
| 749 | This header field is used in request messages and contains information
|
---|
| 750 | about the user agent originating the request. I<E.g.>:
|
---|
| 751 |
|
---|
| 752 | $h->user_agent('Mozilla/5.0 (compatible; MSIE 7.0; Windows NT 6.0)');
|
---|
| 753 |
|
---|
| 754 | =item $h->server
|
---|
| 755 |
|
---|
| 756 | The server header field contains information about the software being
|
---|
| 757 | used by the originating server program handling the request.
|
---|
| 758 |
|
---|
| 759 | =item $h->from
|
---|
| 760 |
|
---|
| 761 | This header should contain an Internet e-mail address for the human
|
---|
| 762 | user who controls the requesting user agent. The address should be
|
---|
| 763 | machine-usable, as defined by RFC822. E.g.:
|
---|
| 764 |
|
---|
| 765 | $h->from('King Kong <[email protected]>');
|
---|
| 766 |
|
---|
| 767 | I<This header is no longer part of the HTTP standard.>
|
---|
| 768 |
|
---|
| 769 | =item $h->referer
|
---|
| 770 |
|
---|
| 771 | Used to specify the address (URI) of the document from which the
|
---|
| 772 | requested resource address was obtained.
|
---|
| 773 |
|
---|
| 774 | The "Free On-line Dictionary of Computing" as this to say about the
|
---|
| 775 | word I<referer>:
|
---|
| 776 |
|
---|
| 777 | <World-Wide Web> A misspelling of "referrer" which
|
---|
| 778 | somehow made it into the {HTTP} standard. A given {web
|
---|
| 779 | page}'s referer (sic) is the {URL} of whatever web page
|
---|
| 780 | contains the link that the user followed to the current
|
---|
| 781 | page. Most browsers pass this information as part of a
|
---|
| 782 | request.
|
---|
| 783 |
|
---|
| 784 | (1998-10-19)
|
---|
| 785 |
|
---|
| 786 | By popular demand C<referrer> exists as an alias for this method so you
|
---|
| 787 | can avoid this misspelling in your programs and still send the right
|
---|
| 788 | thing on the wire.
|
---|
| 789 |
|
---|
| 790 | When setting the referrer, this method removes the fragment from the
|
---|
| 791 | given URI if it is present, as mandated by RFC2616. Note that
|
---|
| 792 | the removal does I<not> happen automatically if using the header(),
|
---|
| 793 | push_header() or init_header() methods to set the referrer.
|
---|
| 794 |
|
---|
| 795 | =item $h->www_authenticate
|
---|
| 796 |
|
---|
| 797 | This header must be included as part of a C<401 Unauthorized> response.
|
---|
| 798 | The field value consist of a challenge that indicates the
|
---|
| 799 | authentication scheme and parameters applicable to the requested URI.
|
---|
| 800 |
|
---|
| 801 | =item $h->proxy_authenticate
|
---|
| 802 |
|
---|
| 803 | This header must be included in a C<407 Proxy Authentication Required>
|
---|
| 804 | response.
|
---|
| 805 |
|
---|
| 806 | =item $h->authorization
|
---|
| 807 |
|
---|
| 808 | =item $h->proxy_authorization
|
---|
| 809 |
|
---|
| 810 | A user agent that wishes to authenticate itself with a server or a
|
---|
| 811 | proxy, may do so by including these headers.
|
---|
| 812 |
|
---|
| 813 | =item $h->authorization_basic
|
---|
| 814 |
|
---|
| 815 | This method is used to get or set an authorization header that use the
|
---|
| 816 | "Basic Authentication Scheme". In array context it will return two
|
---|
| 817 | values; the user name and the password. In scalar context it will
|
---|
| 818 | return I<"uname:password"> as a single string value.
|
---|
| 819 |
|
---|
| 820 | When used to set the header value, it expects two arguments. I<E.g.>:
|
---|
| 821 |
|
---|
| 822 | $h->authorization_basic($uname, $password);
|
---|
| 823 |
|
---|
| 824 | The method will croak if the $uname contains a colon ':'.
|
---|
| 825 |
|
---|
| 826 | =item $h->proxy_authorization_basic
|
---|
| 827 |
|
---|
| 828 | Same as authorization_basic() but will set the "Proxy-Authorization"
|
---|
| 829 | header instead.
|
---|
| 830 |
|
---|
| 831 | =back
|
---|
| 832 |
|
---|
| 833 | =head1 NON-CANONICALIZED FIELD NAMES
|
---|
| 834 |
|
---|
| 835 | The header field name spelling is normally canonicalized including the
|
---|
| 836 | '_' to '-' translation. There are some application where this is not
|
---|
| 837 | appropriate. Prefixing field names with ':' allow you to force a
|
---|
| 838 | specific spelling. For example if you really want a header field name
|
---|
| 839 | to show up as C<foo_bar> instead of "Foo-Bar", you might set it like
|
---|
| 840 | this:
|
---|
| 841 |
|
---|
| 842 | $h->header(":foo_bar" => 1);
|
---|
| 843 |
|
---|
| 844 | These field names are returned with the ':' intact for
|
---|
| 845 | $h->header_field_names and the $h->scan callback, but the colons do
|
---|
| 846 | not show in $h->as_string.
|
---|
| 847 |
|
---|
| 848 | =head1 COPYRIGHT
|
---|
| 849 |
|
---|
| 850 | Copyright 1995-2005 Gisle Aas.
|
---|
| 851 |
|
---|
| 852 | This library is free software; you can redistribute it and/or
|
---|
| 853 | modify it under the same terms as Perl itself.
|
---|
| 854 |
|
---|