Hi Pappetrou,
Here is the global picture:
1) The lexer:
The lexer is in charge of consuming multi-line comments. So, in your case, you end up with a regular expression which looks similar to "(?s)/\\*.*?\\*/"
Note the (?s) DOT_ALL flag which enables the dot character '.' to consume any characters, including new lines.
2) The CommentAnalyzer
Currently, the goal of the comment analyzer is two-fold:
a) To extract the comment tags from the comment itself, to get the content. In your case, you simply have to remove the leading "/*" and trailing "*/" to handle your multi-line comment syntax, i.e. return comment.substring(2, comment.length() - 2)
b) Determine whether or not a single line of comment is blank or not. This function is later going to be used by the SSLR-Squid-Bridge CommentsVisitor, see part 3
3) SSLR-Squid-Bridge's CommentsVisitor
This class will enable you to collect basic metrics regarding your comments.
It will start by removing the comments tags, using the CommentsAnaylzer, and then split your multi-line comment into separate lines.
Afterwards, on each line, it will call the isBlank() method with the line's content to determine which metric (blank, comments, etc..) to feed.
Et voila!
Let me know if this helps
--
Dinesh Bolkensteyn
On Fri, Jun 29, 2012 at 2:39 PM, Papapetrou P.Patroklos
<[hidden email]> wrote:
Hi guys
I'm wondering if and how I can use SSLR CommentAnalyzer to handle multilines comment like the following
/* blah blah
blih bloh
blah blah */
As far as I understand CommentAnalyzer assumes that a comment is only one code line... right? or am I missing something (again) :) ?
Regards
Patroklos
Locked
