Interesting puzzles, and I don't know the answer.
On a practical level though, the comments are meant to be read by
humans, so I would want to make sure that in my code it visually clear
what part is a comment and what not. A habbit like that should avoid
any ambiguities. At least, I cannot remember an instance where I have
been bitten by these things.
-- Maarten
--- Sergiy Radyakin <[email protected]> wrote:
> Thank you Maarten,
>
> this is exactly the point: does the end-of-the-line as it is
> understood by #delimit agree with the end of the line as it is
> understood by comments?
>
> I have 2 possible interpretations of "line".
> 1. Line in the text file (as it is understood by a text editor,
> showing current line number and cursor position in the line)
> 2. Line of code (which can occupy more than one text-line if the
> delimiter is set to ";")
>
>
> Here are my doubts:
>
> A. If the manual really means "line" as defined by 1, than the
> following code must yell "Hello one" and "Hello two", but it does
> not:
>
> #delimit ;
> *include "something"
> // comment
> di "Hello one";
> di "Hello two";
> // end of file
>
> I think it must yell two times, because the commands are unaffected
> by
> commentaries in the other text lines if definition 1 is true.
>
> B. If the manual really means "line" as defined by 2, than the
> following code must yell only "Hello two", but it does not:
>
> #delimit ;
> // comment
> di "Hello one";
> di "Hello two";
> // end of file
>
> I think it must yell only once, since everything from "//" till ";"
> must be a comment, and only di "Hello two"; must be executed.
>
> It seems that "*" and "//" are somehow different. I have an external
> program, that modifies .ado files to it's needs, and it get's
> confused
> in these situations. So any advice would be very welcomed.
>
> I am also experiencing difficulties with understanding priorities
> among the commentaries.
>
> Consider the following code:
>
> #delimit ;
> /*;
> /*;
> */;
> di "test";
> */;
> di "test2";
> // end of file
>
> If we run this code, only "test2" is output, because /* */ comments
> may be "shelled", so that if the comment is opened several times with
> "/*" Stata will require exactly as many "*/" to cancel the comment.
>
> It seems that "//" has priority and cancels "/*" and "*/":
>
> #delimit ;
> // /*;
> /*;
> */;
> di "test";
> // */;
> di "test2";
> // end of file
>
> (both "test" and "test2" are written).
>
> However, not always???
> #delimit ;
> // /*;
> /*;
> // */;
> di "test";
> // */;
> di "test2";
> // end of file
>
> (If the above logic is true, than the comment is open in line 3 spans
> until the end of the file)
>
> In general, has anybody done parsing of a Stata .do / .ado file and
> could share a boolean function (in any language), which can answer
> whether a symbol (i,j) in the file is a part of a comment or not? (i
> -
> is the line number, j - position in line). If this is at all
> possible?
>
> Thank you, Sergiy Radyakin
>
> PS: I am working with Stata for Windows 9.2 if the version matters.
>
>
> On 5/1/08, Maarten buis <[email protected]> wrote:
> > --- Sergiy Radyakin <[email protected]> wrote:
> > > I have a question regarding the use of comments in Stata
> programs.
> > >
> > > The two possibilities to comment out a line of code are to put a
> "*"
> > > or "//" in the beginning of a line.
> > >
> > > The manual (e.g. online here:
> http://www.stata.com/help.cgi?comments)
> > > says:
> > >
> > > " * indicates that the line is to be ignored."
> > > and
> > > "// indicates that the rest of the line is to be ignored."
> > >
> > > Both definitions are using the concept of a "line". I would like
> to
> > > double-check, what is really meant by "line" in this context?
> >
> > that is determined by the -#delimit- command, see: -help #delimit-,
> but
> > by default a line ends with a carriage return.
> >
> > Hope this helps,
> > Maarten
> >
> > -----------------------------------------
> > Maarten L. Buis
> > Department of Social Research Methodology
> > Vrije Universiteit Amsterdam
> > Boelelaan 1081
> > 1081 HV Amsterdam
> > The Netherlands
> >
> > visiting address:
> > Buitenveldertselaan 3 (Metropolitan), room Z434
> >
> > +31 20 5986715
> >
> > http://home.fsw.vu.nl/m.buis/
> > -----------------------------------------
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
Yahoo! For Good. Give and get cool things for free, reduce waste and help our planet. Plus find hidden Yahoo! treasure
http://green.yahoo.com/uk/earth-day/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
