Re: Block comments

[William Poetra Yoga Hadisoeseno]

On 6/1/06, John W. Eaton <address@hidden> wrote:
> On  1-Jun-2006, William Poetra Yoga Hadisoeseno wrote:
>
> | Hi, I've (somewhat) implemented block comments (%{ and %}), but it's
> | still very crude at the moment (look at the diff, it still has
> | debugging features like printf and warning everywhere :p). The current
> | problems are:
> |
> | 1. the block comment start and ends are not stripped. So for example
> | if I have a file a.m:
> |
> | [...]
> |
> | 2. The code looks ugly (lots of duplication of code)
> |
> | 3. I don't know whether some block comments would be missed, or
> | non-block comments would be identified as such, for example:
>
> I think this discussion belongs on the maintainers list.  Please post
> any replies there.
>
> The page
>
> http://www.mathworks.com/access/helpdesk/help/techdoc/matlab_env/edit_d14.html
>
> says that block comments should nest.

Yes, it's a new feature in Matlab 6 IIRC from some user comments on the net.

> I think completely separate code should be used for in-line and block
> comments.  With flex, it might be easiest to handle block comments with
> an exclusive start state.  To use this method, you'll probably need to
> have patterns something like this:
>
>   "%{"  -- begin BLOCK_COMMENT state, set block_comment nesting level
>            to 1, save current start state (saved_start_state = YY_START).
>
>   <BLOCK_COMMENT>"%{"  -- increment block_comment nesting level.
>
>   <BLOCK_COMMENT>"%}"  -- decrement block_comment nesting level.  If
>                           nesting level is 0, restore previous start
>                           state (BEGIN (saved_start_state)).
>
>   <BLOCK_COMMENT>\n    -- increment line counter
>
>   <BLOCK_COMMENT>.     -- all other text in the comment (you might
>                           also be able to match more than one
>                           character at a time with regular expressions
>                           if you are careful -- see the flex manual
>                           for an example of handline C-style comments,
>                           but note that C-style comments don't have to
>                           nest).
>
> In each of these, you'll need to save the comment in the comment
> buffer and keep track of the column counter.

I think we should only match for "^%{$" and "^%}$", because in Matlab,
a block comment starter and ender is only valid if the two characters
%{ or %} are the only characters on the line. So for example this one
doesn't start a block comment:

 myvar = 1; %{ this is not a block comment
 %} and this line should generate a warning

And also,

 myvar = 1;
 %{ not a block comment
 this generates an error here
 %} and a warning here

Another example:

 myvar = 1;
 %{
 i'm in a block comment
 %} this doesn't end the block comment
 myvar2 = 2; this line is still commented
 will generate an error about block comment ended by EOF here

And this is the correct usage of block comments as in Matlab:

 myvar = 1;
 %{
 block comment here
 %}
 myvar2 = 2;

Since I don't have access to Matlab here, I can't verify it
personally. So should we "extend" the syntax of block comments, so
that it doesn't need to appear by itself on a line? If we don't extend
the syntax, we don't have to worry about column numbers (since it
matches only when we begin a line and ends with a newline --> %}\n).

> The only catch here is that there are a few places where comments are
> already matched with regular expressions like "%.*\n", so those would
> have to be modified as well.

OK, I'll take a look. Some functions also deal with '\r'. What is it
for? Is it for compatibility with Matlab scripts with the DOS/Windows
encoding?

> Now that I see it is possible to save and restore start states by
> using YY_START, perhaps the current comment-handling code could also
> be simplified by using an exclusive start state.  So instead of trying
> to handle comments directly in the <MATRIX> start state, we could
> handle comments separately, but be sure to restore the previous start
> state when the end of a comment is reached.
>
> jwe

Uh, could you please briefly explain the start states and the column
numbers? I took a look at the flex info manual, and I haven't really
understood it. About the column numbers (the global variable
current_input_column, right?), yes, I saw those, but some functions
don't update it (and input_line_number, too). Why?

--
William Poetra Yoga Hadisoeseno