Re: various Lout requests made by Mark Summerfield

[Mark Summerfield]

On Thu 22-Feb-07, Jeff Kingston wrote:
> Dear Mark,
>
> I've been going through your emails while preparing the next release
> of Lout.  I've sorted everything out except for the weird problem
> with inter-line gaps.  My responses follow.
>
> I will be away for a few days.  When I return I will work on
> the only other major thing to be done for the next release,
> which is incorporating Ludovic's user style stuff.
>
> * import @BasicSetup def @QC right address@hidden pipe{x}{}}
>
>   This does actually call for all occurrences of x in the body of @QC to
>   be replaced by the actual parameter, so when you see all occurrences
>   of x in your program text being replaced by your pipe command, Lout
>   is doing what it should.  In general, it's good for filters to be
>   able to spit out Lout symbols that are subsequently interpreted as
>   such by Lout.
>
>   The fault, then, lies in prg2lout, which should enclose in quotes
>   every scrap of program text that it ever outputs, since it's to be
>   treated verbatim.  However I have not changed prg2lout since the
>   workaround (replace x with a longer name, as you found) is so
>   simple and there may be places where unquoted prg2lout output is
>   useful; I'm not sure.

I use:

import @BasicSetup def @QC right address@hidden pipe{xyzxyzwwwxyz}{}}

Don't think that'll give me name collisions!

> * Ellipsis symbols in Python texts.  I've fixed this by adding a ...
>   token to Python, and ensuring that tokens with dots get enclosed in
>   double quotes.  This may have some effect in some other languages.

Thanks.

> * Style changes in concatenation operator lengths, as in
>
>       A |{{Crystal Base 9p} @Font 4f} B
>
>   I've fixed this.  It may still go wrong in obscure cases, I'm not
>   sure (the code is horrible), but this and most cases should work now.

At the moment I get round it by using some macros based on literal
lengths, but this will be much better.

> * Excessive inter-line gaps when using "lines @Break".  I've looked at
>   this and I agree it looks like a bug.  I've sent you a separate email
>   asking for a complete test that I can unpack and run.

I got around this by using a specific line height:
    def @ConsoleBreak right x {{0.135c address@hidden
    #def @ConsoleBreak right x address@hidden

> * A command which runs a pipe and includes its output verbatim, as in
>   your hypothetical example
>
>      @Verbatim pipe{"my command that does it"}{}
>
>   See the forthcoming @Pipe, @PipeVerbatim, and @PipeRawVerbatim symbols,
>   documented in the Verbatim section of Chapter 2 of the User's Guide.

Thanks!

> * Missing running header above figure.  I believe that Lout is behaving
>   reasonably here, because on odd-numbered pages you have asked for the
>   current section number and title to be displayed (see @RunningOddTop
>   in file book.lout), but there is no current section at the top of
>   that page.  It is true that a section starts lower down, and also
>   true that if the figure was not there Lout would use that section as
>   the basis of the running title.  But I don't think it's unreasonable
>   to not print a section number and title if there is no section at
>   the top of the page.

I accept it isn't unreasonable; but it _is_ a bug. I have to work around
it a few times by moving figures/tables.

>   A good way to work around this would be to make a new section, called
>   "Introduction" or something, out of most of your introductory text.

Hmmm, I'll try it.

[snip]

> * Suggestion for a quartercircle shape for @Diag.  I have not done
>   this since I feel it's a bit specialized.  It's easy to get it
>   by making your own node outline, as you know.

OK. But did you do @ANode to @ZNode (or at least to @HNode?)

> * Suggestion for an arrowhead that looks like -->| in @Diag.  I
>   have done this, in such a way that when two of these arrows meet
>   from opposite directions their bars overstrike, as you suggested.

I look forward to this; doing it manually is v. tiresome!

> * No space follows a TryAfterLine figure that ends up at the top
>   of the page.  I see you have @TopGap { 0c } in one of your
>   setup files, and I suspect this is why you have no space there.
>   Let's keep talking if resetting @TopGap to a more usual value
>   does not fix the problem, or if it causes other problems.

I think there's a more general issue here. If I have @TopGap anything
more than 0c I used to get too much gap above figures & tables. I'll try
again though.

Thanks!

-- 
Mark Summerfield, Qtrac Ltd., www.qtrac.eu