@Section
   @Title { Organizing large documents }
   @Tag { organizing }
@Begin
@PP
It is not a good plan to store a large document in a single large
organizing.large @Index { organizing large documents }
file.  It takes too long to find things in it, and if some catastrophe
occurs, you lose the lot.  Lout encourages you to break documents into
pieces by its willingness to read a sequence of files
({@Code "lout file1 file2 ..."}).  For large documents, the following
plan is recommended.
@PP
Suppose you are making a book whose third chapter contains sections on
banksias, grevilleas, acacias, and eucalypts.  Place each section, from
@Code "@Section" to {@Code "@End @Section"}, in a separate file, making
four files called, say, {@Code banksias}, {@Code grevilleas},
{@Code acacias}, and {@Code eucalypts}.  Then make a single file for the
chapter as a whole whose contents are as follows:
@ID @OneRow @Code {
"@Chapter"
"    @Title { Australian Native Plants }"
"    @Tag { natives }"
"@Begin"
"Australian native plants provide a distinctive identity to the garden.  Although"
"less colourful than their European alternatives, some banksias and grevilleas do"
"flower strongly, and of course the acacias (wattles) are unsurpassable in late winter."
"@BeginSections"
"@Include { banksias }"
"@Include { grevilleas }"
"@Include { acacias }"
"@Include { eucalypts }"
"@EndSections"
"@End @Chapter"
}
The @Code "@Include" symbol causes Lout to read the file whose name follows
include. @Index @Code "@Include"
it between braces, just as though the contents of that file had been included
at that point.
@PP
With this arrangement you can easily rearrange the order of the
sections:  just swap their @Code "@Include" lines.  You should be using
Lout's automatic cross referencing features (Section {@NumberOf cross}),
so you don't have to worry about keeping cross references up to date.  You
can also temporarily delete a section by placing a @Code "#" character at
the start of its line:
@ID @Code "# @Include { acacias }"
This works because @Code "#" is the @I { comment character }:  Lout will
comment. @Index { comments }
ignore this character (unless enclosed in double quotes) and everything
following it up to the end of the line.  You can even temporarily delete
every section except the one you are working on at the moment, using
these comments.
@PP
Suppose now that this chapter file is called @Code { natives }, and you
have others called @Code { preface }, @Code { flowers }, etc.  Then you
can make one file (call it @Code { garden }) for the whole book like this:
@ID @OneRow @Code {
"@SysInclude { book }"
"@Book"
"    @Title { The Australian Garden }"
"    @Author { Martha S. Vineyard }"
"//"
"@Include { preface }"
"@Include { flowers }"
"@Include { shrubs }"
"@Include { natives }"
"@Include { trees }"
}
You can play the same tricks here:  swap chapters around, or temporarily
delete one or more with a {@Code "#"}.  When a chapter is finished
you can temporarily delete it to save formatting time and paper, and
bring it back at the end.  To format the book, use
@Code "lout garden > out.ps"
in Unix.  Lout will read each @Code "@Include" file as it comes to it,
and if it finds an @Code "@Include" of a section while reading a chapter
file, it will read the section too.
@PP
If the order of your chapters is fairly stable, it might be advantageous
to use the @Code "@BypassNumber" option of @Code "@Chapter" (described
in Appendix {@NumberOf bypass}) to fix the numbers of all your chapters,
so that you get correct chapter numbers even when formatting one
chapter at a time.
@PP
If you decide to store chapters in separate Unix directories, make sure
that any @Code "/" characters in the file names are enclosed in double
quotes:
@ID @Code "@Include { \"natives.dir/acacias\" }"
Be careful not to give the directory the same name as your chapter
file.  You might also find it useful to construct your book
@I { top-down }, as computer scientists call it, laying out all the
chapters and sections as empty skeletons and filling their contents in
later.
@PP
When some part of your document has to be repeated, one way to avoid
entering it twice is to place that part in a separate file and use
@Code "@Include" in two places to include it twice.  This works, but
there are two caveats.  First, it may be better to use a definition
(Section {@NumberOf definitions}), since that gives you an object
with a name, which you can use with confidence anywhere at all.
Second, Lout treats @Code "@Include" in a peculiar way when it
appears in the setup part of a document (in definitions, databases,
and the @Code "@Use" clauses that appear within setup files):  it
reads the file as usual the first time, and silently skips it the
other times.  This is done to simplify the handling of files of
definitions that depend on other files of definitions, as described
in the Expert's Guide @Cite { $kingston1995lout.expert }.
@End @Section