[SVN] Revision 784 (HTML)
mats at imediatec.co.uk
Mon May 2 10:44:22 UTC 2005
On 2 May 2005, at 09:43, Jeroen van der Ham wrote:
> On 01-05-2005 13:06, Mats Persson wrote:
>> I may have misunderstood the value of the firstLineMatch thing,
>> but I don't think it will work within the scope of HTML as well as
>> it will do in .sh files.
> Yes it will work, perfectly, as long as people write proper HTML...
> A proper HTML page has at least a line like the following as it's
> first line:
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
> Or even better:
> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
I was uncertain of the extent of sophistication of the regex within
firstLineMatch that could handle all the various implementations of
HTML/XHTML starts, in comparison with for instance a bash .sh script.
Here are some points to consider:
1. in a .sh file we HAVE TO (as far as I know) start with a shebang.
2. in valid HTML we should start with <!DOCTYPE...>, but it can be
and is often left out.
3. in valid XHTML we should start with an XML declaration <?xml
version="1.0" encoding="utf-8"?>, but since Win IE has problems with
that start it's left out most of the time.
4. in my normal work I work with smaller snippets of (X)HTML where
the <!Doctype is not available, then firstLineMatch might cause more
problems than it's worth.
I don't know and as I said I might have misunderstood the full
meanings/values of the firstLineMatch, and perhaps someone could set
me right on it, if I'm wrong.
> So there are the ingredients for distinguishing between different
> types of HTML files on the first lines.
> It might have to be extended to grab the first two, but the first
> one should suffice already.
I am toying with the concepts of having a normal regexp type
structure that checks for specific <!DocTypes and then includes
"full" syntaxes accordingly much like we do today with PHP, CSS & JS,
so that we would have something like this structure wise:
|-> if DocType === XHTML 1.1
| # includes XHTML 1.1 specific syntax in separate file
|-> if DocType === XHTML 1.0
| # includes XHTML 1.0 specific syntax in separate file
|-> if DocType === HTML 4
| # includes HTML 4 specific syntax in separate file
|-> if DocType === ??? or missing
| # includes basic HTML syntax
The above is just an idea that may or may not work, be useful or
whatever. It is too complicated for me to get my head around at this
point in time. I actually think that the (X)HTML language is far
more complex in structure and possible variations etc. than say CSS,
JS or PHP.
On top of all this, we have to somehow at some point in time be
prepared to easily enable code_completion based on context. There's
lots of work to do in the language over the coming months.
"TextMate, coding with an incredible sense of joy and ease"
- www.macromates.com -
More information about the textmate-dev