Re: [Aspell-user] Spell-checking source code

[Greg Ward]

On 14 October 2004, Christoph Hintermüller said:
> Which version of Aspell are you using:
> < 0.33.X if yo can upgrade to 0.60.X or stop reading this mail
> < 0.60 if you can upgrade to 0.60.X or stop reading this mail

0.50.5 at home (Debian unstable); 0.33.something at work (Red Hat 9).

> >= 0.60.X which language is your code in  ?

I'm a Python bigot myself, but we use Java at work, and this project is
inspired by a couple of co-workers who are orthographically challenged.
So if I had to pick one language to spell-check, it would be Java.
(Once nice thing about Java programmer's tendency for verbosity is that
they tend to use whole words in identifiers, which makes for easier
spell-checking of source code!)

> If not are ther destinct delimiters like " and # or // or % or \n separating 
> the parts to be spelled form the parts not to be spelled ?
> if so try calling

Ah-ha, you just gave me a clue that aspell does *not* do what I want.  I
want to spell-check all text: comments, string literals, *and*
programming identifiers.  I don't care if the error is

  /* this fucntion adds two numbers */

or

  message = "error reeding file";

or

  void getRemaningObjects()

-- I want the spell-checker to catch it.  Which is why my regex to split
mixed-case identifiers like "getRemainingObjects" or "HTTPResponse" into
English words is at the heart of this little hack.

Does aspell 0.60 do *that*?  Or does it only spell-check comments and
string literals?  The latter two are not enough in the face of
programmers who consistently misspell function names.  ;-(

        Greg
-- 
Greg Ward <address@hidden>                         http://www.gerg.ca/
Very few profundities can be expressed in less than 80 characters.