1HTML::WikiConverter::DiUasleerctCso(n3t)ributed Perl DocHuTmMeLn:t:aWtiikoinConverter::Dialects(3)
2
6 HTML::WikiConverter::Dialects - How to add a dialect
7
9 # In your dialect module:
10 package HTML::WikiConverter::MySlimWiki;
12 use base 'HTML::WikiConverter';
13 sub rules { {
15 b => { start => '**', end => '**' },
16 i => { start => '//', end => '//' },
17 strong => { alias => 'b' },
18 em => { alias => 'i' },
19 hr => { replace => "\n----\n" }
20 } }
21 # In a nearby piece of code:
23 package main;
25 use Test::More tests => 5;
26 my $wc = new HTML::WikiConverter(
28 dialect => 'MySlimWiki'
29 );
30 is( $wc->html2wiki( '<b>text</b>' ), '**text**', b );
32 is( $wc->html2wiki( '<i>text</i>' ), '//text//', i );
33 is( $wc->html2wiki( '<strong>text</strong>' ), '**text**', 'strong' );
34 is( $wc->html2wiki( '<em>text</em>' ), '//text//', 'em' );
35 is( $wc->html2wiki( '<hr/>' ), '----', 'hr' );
36
38 HTML::WikiConverter (or H::WC, for short) is an HTML to wiki converter.
39 It can convert HTML source into a variety of wiki markups, called wiki
40 "dialects". This manual describes how you to create your own dialect
41 to be plugged into HTML::WikiConverter.
42
44 Each dialect has a separate dialect module containing rules for
45 converting HTML into wiki markup specific for that dialect. Currently,
46 all dialect modules are in the "HTML::WikiConverter::" package space
47 and subclass HTML::WikiConverter. For example, the MediaWiki dialect
48 module is HTML::WikiConverter::MediaWiki, while PhpWiki's is
49 HTML::WikiConverter::PhpWiki. However, dialect modules need not be in
50 the "HTML::WikiConverter::" package space; you may just as easily use
51 "package MyWikiDialect;" and H::WC will Do The Right Thing.
52 From now on, I'll be using the terms "dialect" and "dialect module"
54 interchangeably.
55 Subclassing
57 To interface with H::WC, dialects need to subclass it. This is done
58 like so at the start of the dialect module:
59 package HTML::WikiConverter::MySlimWiki;
61 use base 'HTML::WikiConverter';
62 Conversion rules
64 Dialects guide H::WC's conversion process with a set of rules that
65 define how HTML elements are turned into their wiki counterparts. Each
66 rule corresponds to an HTML tag and there may be any number of rules.
67 Rules are specified in your dialect's "rules()" method, which returns a
68 reference to a hash of rules. Each entry in the hash maps a tag name to
69 a set of subrules, as in:
70 $tag => \%subrules
72 where $tag is the name of the HTML tag (e.g., "b", "em", etc.) and
74 %subrules contains subrules that specify how that tag will be converted
75 when it is encountered in the HTML input.
76 Subrules
78 The following subrules are recognized:
80 start
82 end
83 preserve
85 attributes
86 empty
87 replace
89 alias
90 block
92 line_format
93 line_prefix
94 trim
96 A simple example
98 The following rules could be used for a dialect that uses "*asterisks*"
100 for bold and "_underscores_" for italic text:
101 sub rules {
103 b => { start => '*', end => '*' },
104 i => { start => '_', end => '_' },
105 }
106 Aliases
108 To add "<strong>" and "<em>" as aliases of "<b>" and "<i>", use the
110 "alias" subrule:
111 strong => { alias => 'b' },
113 em => { alias => 'i' },
114 (The "alias" subrule cannot be used with any other subrule.)
116 Blocks
118 Many dialects separate paragraphs and other block-level elements with a
120 blank line. To indicate this, use the "block" subrule:
121 p => { block => 1 },
123 (To better support nested block elements, if a block elements are
125 nested inside each other, blank lines are only added to the outermost
126 element.)
127 Line formatting
129 Many dialects require that the text of an element be contained on a
131 single line of text, or that it cannot contain any newlines, etc. These
132 options can be specified using the "line_format" subrule, which can be
133 assigned the value "single", "multi", or "blocks".
134 If the element must be contained on a single line, then the
136 "line_format" subrule should be "single". If the element can span
137 multiple lines, but there can be no blank lines contained within, then
138 use "multi". If blank lines (which delimit blocks) are allowed, then
139 use "blocks". For example, paragraphs are specified like so in the
140 MediaWiki dialect:
141 p => { block => 1, line_format => 'multi', trim => 'both' },
143 Trimming whitespace
145 The "trim" subrule specifies whether leading or trailing whitespace (or
147 both) should be stripped from the element. To strip leading whitespace
148 only, use "leading"; for trailing whitespace, use "trailing"; for both,
149 use the aptly named "both"; for neither (the default), use "none".
150 Line prefixes
152 Some elements require that each line be prefixed with a particular
154 string. This is specified with the "line_prefix" subrule. For example,
155 preformatted text in MediaWiki is prefixed with a space:
156 pre => { block => 1, line_prefix => ' ' },
158 Replacement
160 In some cases, conversion from HTML to wiki markup is as simple as
162 string replacement. To replace a tag and its contents with a particular
163 string, use the "replace" subrule. For example, in PhpWiki, three
164 percent signs, "%%%", represents a line break, "<br>", hence:
165 br => { replace => '%%%' },
167 (The "replace" subrule cannot be used with any other subrule.)
169 Preserving HTML tags
171 Some dialects allow a subset of HTML in their markup. While H::WC
173 ignores unhandled HTML tags by default (i.e., if H::WC encounters a tag
174 that does not exist in a dialect's rule specification, then the
175 contents of the tag is simply passed through to the wiki markup), you
176 may specify that some be preserved using the "preserve" subrule. For
177 example, to allow "<font>" tag in wiki markup:
178 font => { preserve => 1 },
180 Preserved tags may also specify a list of attributes that may also
182 passthrough from HTML to wiki markup. This is done with the
183 "attributes" subrule:
184 font => { preserve => 1, attributes => [ qw/ style class / ] },
186 (The "attributes" subrule can only be used if the "preserve" subrule is
188 also present.)
189 Some HTML elements have no content (e.g., line breaks, images) and the
191 wiki dialect might require them to be preserved in a more XHTML-
192 friendly way. To indicate that a preserved tag should have no content,
193 use the "empty" subrule. This will cause the element to be replaced
194 with "<tag />" and no end tag. For example, MediaWiki handles line
195 breaks like so:
196 br => {
198 preserve => 1,
199 attributes => [ qw/ id class title style clear / ],
200 empty => 1
201 },
202 This will convert, for example, "<br clear='both'>" into "<br
204 clear='both' />". Without specifying the "empty" subrule, this would be
205 converted into the (probably undesirable) "<br clear='both'></br>".
206 (The "empty" subrule can only be used if the "preserve" subrule is also
208 present.)
209 Rules that depend on attribute values
211 In some circumstances, you might want your dialect's conversion rules
213 to depend on the value of one or more attributes. This can be achieved
214 by producing rules in a conditional manner within "rules()". For
215 example:
216 sub rules {
218 my $self = shift;
219 my %rules = (
221 em => { start => "''", end => "''" },
222 strong => { start => "'''", end => "'''" },
223 );
224 $rules{i} = { preserve => 1 } if $self->preserve_italic;
226 $rules{b} = { preserve => 1 } if $self->preserve_bold;
227 return \%rules;
229 }
230 Dynamic subrules
232 Instead of simple strings, you may use coderefs as values for the
233 "start", "end", "replace", and "line_prefix" subrules. If you do, the
234 code will be called when the subrule is applied, and will be passed
235 three arguments: the current H::WC object, the current HTML::Element
236 node being operated on, and a reference to the hash containing the
237 dialect's subrules associated with elements of that type.
238 For example, MoinMoin handles lists like so:
240 ul => { line_format => 'multi', block => 1, line_prefix => ' ' },
242 li => { start => \&_li_start, trim => 'leading' },
243 ol => { alias => 'ul' },
244 It then defines "_li_start()":
246 sub _li_start {
248 my( $self, $node, $subrules ) = @_;
249 my $bullet = '';
250 $bullet = '*' if $node->parent->tag eq 'ul';
251 $bullet = '1.' if $node->parent->tag eq 'ol';
252 return "\n$bullet ";
253 }
254 This prefixes every unordered list item with "*" and every ordered list
256 item with "1.", which MoinMoin requires. It also puts each list item on
257 its own line and places a space between the prefix and the content of
258 the list item.
259 Subrule validation
261 Certain subrule combinations are not allowed. Hopefully it's intuitive
262 why this is, but in case it's not, prohibited combinations have been
263 mentioned above parenthetically. For example, the "replace" and "alias"
264 subrules cannot be combined with any other subrules, and "attributes"
265 can only be specified alongside "preserve". Invalid subrule
266 combinations will trigger a fatal error when the H::WC object is
267 instantiated.
268 Dialect attributes
270 H::WC's constructor accepts a number of attributes that help determine
271 how conversion takes place. Dialects can alter these attributes or add
272 their own by defining an "attributes()" method, which returns a
273 reference to a hash of attributes. Each entry in the hash maps the
274 attribute's name to an attribute specification, as in:
275 $attr => \%spec
277 where $attr is the name of the attribute and %spec is a
279 Params::Validate specification for the attribute.
280 For example, to add a boolean attribute called "camel_case" which is
282 disabled by default:
283 sub attributes {
285 camel_case => { default => 0 },
286 }
287 Attributes defined liks this are given accessor and mutator methods via
289 Perl's "AUTOLOAD" mechanism, so you can later say:
290 my $ok = $wc->camel_case;
292 $wc->camel_case(0);
293 You may override the default H::WC attributes using this mechanism. For
295 example, while H::WC considers the "base_uri" attribute optional, it is
296 required for the PbWiki dialect. PbWiki can override this default-
297 optional behavior by saying:
298 sub attributes {
300 base_uri => { optional => 0 }
301 }
302 Preprocessing
304 The first step H::WC takes in converting HTML source to wiki markup is
305 to parse the HTML into a syntax tree using HTML::TreeBuilder. It is
306 often useful for dialects to preprocess the tree prior to converting it
307 into wiki markup. Dialects that need to preprocess the tree can define
308 a "preprocess_node" method that will be called on each node of the tree
309 (traversal is done in pre-order). The method receives two arguments,
310 the H::WC object, and the current HTML::Element node being traversed.
311 It may modify the node or decide to ignore it; its return value is
312 discarded.
313 Built-in preprocessors
315 Because they are commonly needed, H::WC automatically carries out two
317 preprocessing steps, regardless of the dialect: 1) relative URIs in
318 images and links are converted to absolute URIs (based upon the
319 "base_uri" parameter), and 2) ignorable text (e.g. between a "</td>"
320 and "<td>") is discarded.
321 H::WC also provides additional preprocessing steps that may be
323 explicitly enabled by dialect modules.
324 strip_aname
326 Removes any anchor elements that do not contain an "href"
327 attribute.
328 caption2para
330 Removes table captions and reinserts them as paragraphs before the
331 table.
332 Dialects may apply these optional preprocessing steps by calling them
334 as methods on the dialect object inside "preprocess_node". For example:
335 sub preprocess_node {
337 my( $self, $node ) = @_;
338 $self->strip_aname($node);
339 $self->caption2para($node);
340 }
341 Postprocessing
343 Once the work of converting HTML is complete, it is sometimes useful to
344 postprocess the resulting wiki markup. Postprocessing can be used to
345 clean up whitespace, fix subtle bugs introduced in the markup during
346 conversion, etc.
347 Dialects that want to postprocess the wiki markup should define a
349 "postprocess_output" method that will be called just before the
350 "html2wiki" method returns to the client. The method will be passed two
351 arguments, the H::WC object and a reference to the wiki markup. The
352 method may modify the wiki markup that the reference points to; its
353 return value is discarded.
354 For example, to replace a series of line breaks with a pair of
356 newlines, a dialect might implement this:
357 sub postprocess_output {
359 my( $self, $outref ) = @_;
360 $$outref =~ s/<br>\s*<br>/\n\n/gs;
361 }
362 (This example assumes that HTML line breaks were replaced with "<br>"
364 in the wiki markup.)
365 Dialect utility methods
367 H::WC defines a set of utility methods that dialect modules may find
368 useful.
369 get_elem_contents
371 my $wiki = $wc->get_elem_contents( $node );
373 Converts the contents of $node into wiki markup and returns the
375 resulting wiki markup.
376 get_wiki_page
378 my $title = $wc->get_wiki_page( $url );
380 Attempts to extract the title of a wiki page from the given URL,
382 returning the title on success, "undef" on failure. If "wiki_uri" is
383 empty, this method always return "undef". See "ATTRIBUTES" in
384 HTML::WikiConverter for details on how the "wiki_uri" attribute is
385 interpreted.
386 is_camel_case
388 my $ok = $wc->is_camel_case( $str );
390 Returns true if $str is in CamelCase, false otherwise. CamelCase-ness
392 is determined using the same rules that Kwiki's formatting module uses.
393 get_attr_str
395 my $attr_str = $wc->get_attr_str( $node, @attrs );
397 Returns a string containing the specified attributes in the given node.
399 The returned string is suitable for insertion into an HTML tag. For
400 example, if $node contains the HTML
401 <style id="ht" class="head" onclick="editPage()">Header</span>
403 and @attrs contains "id" and "class", then "get_attr_str()" will return
405 'id="ht" class="head"'.
406 _attr
408 my $value = $wc->_attr( $name );
410 Returns the value of the named attribute. This is rarely needed since
412 you can access attribute values by treating the attribute name as a
413 method (i.e., "$wc->$name"). This low-level method of accessing
414 attributes is provided for when you need to override an attribute's
415 accessor/mutator method, as in:
416 sub attributes { {
418 my_attr => { default => 1 },
419 } }
420 sub my_attr {
422 my( $wc, $name, $new_value ) = @_;
423 # do something special
424 return $wc->_attr( $name => $new_value );
425 }
426
428 David J. Iberri <diberri@cpan.org>
429
431 Copyright 2006 David J. Iberri, all rights reserved.
432 This program is free software; you can redistribute it and/or modify it
434 under the same terms as Perl itself.
435perl v5.28.0 2008-05-18 HTML::WikiConverter::Dialects(3)