bumble.sf.net language and parsing

plain text ##! d:/mjb/unixutils/bin/sh.exe # # A script to reformat a text 'documentation' document into # some kind of html # This version makes the section headings into 'heading' elements and makes the # function title or name into an h3 element. This has the disadvantage that unless # there is a style sheet, browsers put quite a lot of space after the # heading elements. However, these heading elements will be usefull if it is # desired to convert the html into some more 'robust' format such as pdf. # By using the 'heading' elements and an html2pdf conversion engine, such as # for example the 'htmldoc' application, it is possible to generate a table of # contents with accurate page numbering for a given output device. This is the # real 'shang-ri-la' of printing as far as I am concerned at the moment. # # By placing a 'for' loop around this command line in a 'sh' shell it should be # possible to batch process plain text documentation files into html and pdf. # There will be those people who are shaking their heads and muttering the words # xml, and sgml and document parsing etc, and really I suppose that I am in # sympathy with those people. But the power of a dodgy plain text format should # not be underestimated. # The lines below seem to crash the 'unxutils' shell but it appears to # run without problems in the cygwin bash shell. # It takes about 20 seconds to run on my MS Win 2000 laptop # It will also crash in the 'cygwin' shell if you point the shell # interpreter to the /unixutils/bin/sh.exe file if [ "$1" = "" ] then echo "usage: txtdoc2html.txt textFileName" exit 1; fi echo "" echo "