f90-mode [was Re: How to update major modes?]

[Glenn Morris]

Jared Ahern wrote:

> 1) On initialization expressions, all of the stuff on the RHS of the  
> '=' sign are in the same color

It's always worked like that, and I don't think I'd want to change this.

> 2) DEFAULT when it follows CASE is a keyword, or close enough

That would be easy to change, but it's always been the way it is.

> 3) ... same with IN/OUT/INOUT in INTENT() expressions

Same as previous.

I think 2) and 3) are really things for your .emacs file.

> 4) Syntax coloring may fail across lines continued with '&'

Sure, you can always write the code in such a way as to fool
font-locking. Trying to support statements split over lines would be a
lot of extra work for little gain, IMO.

> 5) TYPE(...) declarations are now hilited (great!), but this is done  
> differently than REAL() and INTEGER(); I'd much prefer this over  
> nothing, so not a huge deal.

You mean in something like:

type(sometype) :: foo

you would prefer only the "type" part to be highlighted, not the whole
"type(sometype)"? Yes, I wondered about this too. It would in some
sense be more consistent with something "integer(kind=foo)" and is
easy to do. But the way it works now is how it was originally supposed
to work, until some typo crept in and broke it at some point.

> 6) INTRINSIC (and NON_INTRINSIC ?) kills the rest of the normal  
> syntax hiliting on a USE statement;

Should be fixed now.

> I don't mean to make demands or anything.

Not at all. It's nice to get some feedback.

> On a side note, in TeX-mode, you can insert '\end{...whatever...}'  
> with C-C C-E, much like typing 'END' and then TAB in F90.  How would  
> I bind the latter to the former's key combo in my .emacs?

I guess you want:

(defun my-f90-mode-hook ()
  (local-set-key "\C-c\C-e" 'f90-match-end))

(add-hook 'f90-mode-hook 'my-f90-mode-hook)