This is the complete rewrite of template parser. It provides following improvements and changes:
Added separate parser functions for different tasks
Now the cursor moves in chunks by using simpler regular expressions
Empty start & end tags are now supported (<></>)
Text in between an opening and a closing tag is captured and can be accessed as ._innerText. But the text must not contain < char.
Parser functions now works in a complex sequence as middleware and on each cycle cursor position is checked (that approach prevents many problems might occur when the template definition is complex)
Attributes in nameless tags are no longer supported and the test case related to that has been commented out.
Main parser function has own output format and there a formatter function to produce required output format for the code generation.
The template syntax is checked while parsing operation and a custom Error is thrown when there is a syntax/format related problem. The error also shows where the parsing error occurred.
While formatting the code it also checks if there is a closing tag for each opening tag at the same level. And it throws an error if there is a mismatch.
Attribute values can be defined by using both single or double quotes.
Attribute values can contain any character except the used delimiter (if attribute defined like this x='example', the value cannot contain a single quote char ' or vice versa)
Attribute names can contain characters defined in this regex: [A-Za-z0-9:.\-_@]
Tag names can contain only following chars: [a-zA-Z0-9_\-.] (nameless tags is a special case)
This is the complete rewrite of template parser. It provides following improvements and changes:
<></>)._innerText. But the text must not contain<char.Erroris thrown when there is a syntax/format related problem. The error also shows where the parsing error occurred.x='example', the value cannot contain a single quote char'or vice versa)[A-Za-z0-9:.\-_@][a-zA-Z0-9_\-.](nameless tags is a special case)