bumble.sf.net language and parsing

plain text

text mark-up rules for this site

execute a script in a page

 [** scriptname]

widgets

list all the ".png" files in the "/img/" folder
 [@@filelist /img/icon/*.png]   

The Bumble site uses some bash and sed scripts to generate the html pages from the text documents. certain formats in the text files trigger certain types of html output in the usual style of a "wiki" [->wikipedia] system. Detailed below are some of the text formats and their resulting html. The system described is not a wiki because there is no mechanism for the page visitor to edit the files. It may be asked "why wasnt an existing text (or wiki) markup system used instead of inventing a new one?". Firstly, for capriciousness, and also because the system created allows the text files to be "cleaner", that is, the "markup language" [->wikipedia] is simpler and (hopefully) less intrusive for the editor of the text file. Files of the form "name.txt.html" are being generated by the sed script from the text file "name.txt" (that is, the file "name.txt.html" does not exist as such on the server)

text mark-up rules

  1.  ** heading 
    produces a page section heading as below

    heading

     = page title =
    produces a page heading. Look at the top of the current page to see what this looks like.
     :>
    prevents text on the line being formatted which may be useful for displaying fragments of code ...
     -text-
    makes the text emphasized like this: text
     (-faq-) * a question? an answer (-/faq-)
    produces an faq type of document. actually the tags above must be the first thing on their lines. a star starts a question and a ? ends it. (a table of contents is not produced)
     "term" [->wikipedia] or (wikipedia)
    produces a link to the wikipedia site like this: "term" [->wikipedia]
     "term" (googlemap)
    creates a link to the http://maps.google.com site like this: san francisco
     "term" (google) 
    produces a link to a google search for the term like this: term
     "document;"
    makes a link to one of the documents in the machine/doc directory like this "workspace;"
     file://index.txt or link://index.txt or doc:index.txt
    produces a link to a local file on the server like this: index.txt
     "display" file://index.txt or link://index.txt
    produces a link to a local file on the server like this: display
     "text" www.google.com or "text" http://google.com 
    produce links with the text as display like this: text
     --> and --< 
    prevent a block of text being marked up. These patterns must begin the line (excepting whitespace). No other transformations on the text between the tags will be carried out and html
     tags will surround the block
    in order to assure that the text appears as it is written 
    
Ordered and unordered lists can be marked up in the following way...
   u- one
    - second item in the list which
      spans more that one line
    - three
  
   o- un
    - deux
    - trois
  
  
with a blank line terminating the list. the "dash" character must be the first thing on the line and will appear like this
  1. un
  2. deux
  3. trois
Definition lists are written as below (and also need to be terminated with a blank line)
    D- silence: an absence of noise
     - noise:
        compression waves in air
     -hearing: 
        a way of listening

  
and should appear as below:
silence
an absence of noise
noise
compression waves in air
hearing
a way of listening
nested lists will not render will not render correctly

(castellano )

using the system

If for some strange reason you wish to use the page generation system then you need to put the following files in your cgi-bin directory

file:/script/dir.sed.txt
which makes the menu of sub-folders on the page
file.sed
which makes the menu of files on the page
file:/script/makehtml.txt
which is the sed script which generates the body of text on the page
file:/script/text2html.txt
which is a "common gateway interface" [->wikipedia] cgi bash script which calls the sed scripts and generates the page.
Also you need to put the following mod-rewrite rules in your ".htaccess" file.
     
   RewriteRule ^(.*)\.txt\.html$ /cgi-bin/text2html.cgi?$1
   RewriteRule ^(.*/)$ /cgi-bin/text2html.cgi?$1
   RewriteRule ^(.*doc/.*)\.txt$ /cgi-bin/text2html.cgi?$1
   RewriteRule ^$ /cgi-bin/text2html.cgi?$1
  

Thats all. If you have problems please consult the relevant user manuals.

All this assumes that your are running apache on a "unix" [->wikipedia] style operating system. If not then then you cant really use these scripts without large amounts of annoying installations and configurations.