Recent changes to bugs

Add "--help" option

Dan Scott — Sun, 23 Sep 2001 02:16:56 -0000

It would be nice to have help available from the
command line with the standard "--help" parameter, e.g.:

./scrollserver.py --help

Usage: scrollserver.py [OPTIONS]

scrollserver.py - Python HTTPD server that provides
access to a scrollkeeper database from a web browser
using the URL http://localhost:8000

-p, --port=PORT
Specify port number on which scrollserver
listens (default 8000)

--help Display this help and exit

--version Display version number and exit

Version no differs in code and package

Dan Scott — Sun, 23 Sep 2001 01:53:48 -0000

The initial release is called "0.4" but when you run
scrollserver it states:

ScrollServer v0.3 -- development version!

Yes, this is minor...

Some files cached in local directory

David Merrill — Sat, 22 Sep 2001 14:41:30 -0000

All cache files should go in /var/cache/scrollserver/
instead.

DocBook stylesheets out of date

David Merrill — Fri, 21 Sep 2001 18:02:53 -0000

Norman Wash has released a newer version of the
DocBook xsl stylesheets, so I need to merge my
enhancements with them.

Document sort is case sensitive

David Merrill — Fri, 21 Sep 2001 04:06:15 -0000

Document list should be insensitive to case.

Post website

David Merrill — Thu, 20 Sep 2001 17:37:57 -0000

Duh, need a website.

Build tarball

David Merrill — Thu, 20 Sep 2001 17:19:27 -0000

Build a tarball for installation.

Display images for DocBook files

David Merrill — Thu, 20 Sep 2001 17:07:26 -0000

Xsltproc converts DocBook XML and SGML source files
into html, but doesn't touch relative links to
embedded graphics. So, when we serve the final output
there are no images.

Fixing this will mean solving the problem of knowing
which document's home directory should be used as the
relative path for these links, which means figuring
out how to pull that information from the
SimpleHTTPServer.

I don't want to create copies of local images, but am
considering adding symlinks in the cache directory to
the files in the document's home directory to simplify
serving them. This may still be a good approach for
local documents. However, it won't work if the
document is being pulled from a remote location via
http. In that case, I will have to actually retrieve
the image file and store a copy locally, in the cache.