EGCS web pages

[Joe Buck]

>     Jason> Yeah, I did that a long time ago because I was afraid that
>     Jason> browsers would munge the patches when people saved them to
>     Jason> files.  I now see the error in my ways, we're going to be
> 
> You did this at my behest, IIRC.  It *is* important that you can
> cut-and-paste patches out of the archives.  Please keep this feature,
> in some shape or form, if at all possible.  (Perhaps `click here for
> plain text version of message' or some such.)

The patch program has code in it to detect whether a patch is present,
and where in the message it starts.

It seems that the ideal solution would be that this patch-detection
code is run on each message to be archived, and if a patch is found,
a plain-text copy of the message is saved, and in the HTML page a
link is made to the plain-text message.  We might chop the patch
out of the HTML message as well, or else it could just be HTMLified
as usual (meaning the <,>,and & characters are replaced by the 
< > & strings.

There are other possibilities, like a CGI program that produces
a plain-text file which is the original message: chop out the hypermail
links, and inverse-translate the <, > and &amp strings.