WebLEXICON v0.1
----------------
Known syntaxes:
DOM/DOM2
HTML/XHTML
SGML
MATHML
JavaScript
XML
--and any similar syntax languages--
Future support:
DocBook
TeX / LaTeX
Global info (applies to all or most of the supported syntaxes):
1. Empty closing tags ("</","<?/") can be used, they will close last opened tag.
Empty opening tags ("<","<?") are not recommended right now.
2. < > are ignored in some cases:
a) inside <!-- --> comment style, they will act as any other character, but inside /* */ they will be skipped.
b) Inside CDATA Marked Sections
CDATA Marked sections start like this: "<![CDATA[" - ID1+ID31 -
3. Closed Start Tags allowed. These are specific to SGML, and they look like this: <tag1<tag2>
You can notice that the first tag is not closed (<tag1><tag2> is correct HTML/XHTML). In this case, the processor
will check each time an "<" is opened that there is no other "<" opened. Otherwise, it will know that the last tag
wasn't closed, and it will act like he found an ">"
NOTE: Closed Smart Tags apply also to "<a></a>" syntax, but no "<" should be contained in the syntax:
correct: <a href="">Google Link</a><next-tag>....
correct: <a href="">Google Link<next-tag>..
incorrect: <a href="">Google<Link<next-tag>..
In last case, "Google<Link" will be interpretted this way:
"Google" = link description (GOOD!)
"<" = new tag opened (BAD!)
"Link" = keyword (BAD!)
4. NET Tags are not yet supported.
A NET Tag looks like this: <name/.../
The HTML equivalent will look like: <name>...</name>
SGML Processing Instructions
-----------------------------
EXAMPLES:
<?>
<?style tt = font courier>
<?page break>
<?experiment> ... <?/experiment>
NOTES: 1. "<" will define a HTML Tag Start, and the "?" will let us know its a processing instruction
2.The code will allow know that "?" is a processing instruction only if before him its an "<",
which means that "< ?" will not be recognized as a processing instruction. To avoid this, the
HTML processor should check if "<" is ID1, "?" is ID 32 or 20.
An ending Processing Instruction Tag will end when "/" is attached after "?"
ie: <?> - opening tag
<?/> - ending tag
3. You dont need to use quoted attribute values as long as you dont use any known keyword.
For example: <?a lang=en> is VALID
<?a lang=html> is INVALID, because HTML is a keyword and will misslead the processor.
LEXICON:
1. <?a lang="en">
ID 1 - HTML Tag Start
ID 32 - Processing Instruction Start
ID 90 - String (keyword)
ID 90* - String (attribute)
ID 35 - Delimiter (=)
ID 34 - Quote
ID 2 - HTML Tag End
HTML / XHTML
-------------
You dont have to double-check your XHTML syntax, because it will be interpreted as HTML otherwise.
For example: <br> is a valid HTML code, while <br /> is a valid XHTML code. If your document is an XHTML
type, <br> will be interpreted as <br />, so you dont have to worry.
<!-- comment here --> is a comment, valid in all document types (even CSS !)
CSS
---
/* comment here */ is a comment, valid in all document types