@(#)README	1.1 93/06/16	- ANSWERBOOK UTILITIES.

dbgen and dbascii are two complementary Answerbook utility programs.

dbascii is used to decompile a Answerbook DBM file into an ascii format
that describes the document structure of a given (Answer) book. The
file can be edited to change the layout of the book but, more importantly,
additional books can be inserted at some place in the hierarchy.
See ascii.xtoc.sample for an example how to re-integrate the X Window
books that went missing between release 1.3 and 1.4 of Sun's Answerbook
into Answerbook 1.4. (in addition to this, the bookinfo file should also
be updated, so that navigators and viewers can look for the actual Postscript
docs in the proper directory, see bookinfo.sample for this).

dbgen undertakes the opposite job of generating the DBM files from an
ascii representation of the document structure. It takes as arguments
a book-identifier which goes into the DBM file and the basename of the
desired DBM files (basename gets augmented by .dir and .pag suffixes).

So
	dbgen SUNWab_1_4a SUNWab_1_4a < ascii.xtoc.new

would generate the files `SUNWab_1_4a.dir' and `SUNWab_1_4a.pag', and
running `navigator SUNWab_1_4a' would enable you to view the documents
added to the new `ascii.xtoc.new' file.

When making changes or additions to an "ascii.xtoc" file, be sure to keep
the <some_value> part of `id = <some_value>' lines unique at all times,
since these are used as DBM keys when accessing navigation information.

NOTES
Both dbgen and dbascii are descendents of some initial hackery with the
AB index format, and never fully recovered from that. `dbgen' is based
on a simple recursive descent parser and got a bit dented over time.
YACC would probably be better.

-pk

