Writing APE Web Pages (a style guide)

Rules for HTML files:

Try to avoid editors that reformat documents freely. They tend to become unreadable. One should be able to edit all HTML files using 'vi' or 'emacs'. Under html/ there is a template called 'empty.html' which may serve as a start. It contains tha basic layout of the file but no content. Replace title and header, insert your text and copy links from other pages. ALL files are under CVS, so add your files and commit ! Links must always be relative !! Links to other APE pages must not contain the whole http://... address but only ./<filename>

Adding Documents:

Under html/Documentation/ you can freely add other documents. The index pages for these directories are generated on the fly. They use a file called '.description' to output some useful comment on the contents of the file. Entries in the .description file are mandatory ! Check one of the current directories for an example.

Adding new directories:

Under CVS directories are easiest to make in the repository itself. See what CVS/Root says and mkdir there, then 'cvs update -d' locally. If you added the directory under html/Documentation/ then read the paragraph above and also copy AND MODIFY a .htaccess file in the new dir. It is used to redirect the indexing to a CGI script.

Making a new search index:

The APE webpages use a search engine called 'swish++' for searching which is capable of searching not only text- and html-files, but also PDF-, PostScript- and Word-documents. This little tool uses an index file as database for its searches. The index file must be re-generated after new documents have bee added by calling from the root of the APE document tree. Indexing takes some time due to the fact that all non-text files must undergo some 'extraction'-procedure while being indexed. So start it and take a cup of coffee. P.S. There may be some errors while extracting data from PDF files. Ignore them.