[Lynx-dev] Handling sites that don't send content-type (was: lynx word bleeding?)

[Luke, Shellworld Support]

On Wed, 2 Feb 2022, Ian Collier wrote:

> Extract from lynx.cfg:
> 
> .h2 PREFERRED_CONTENT_TYPE
> # When doing a GET, lynx expects the server to provide a Content-Type, i.e.,
> # the MIME name which tells it how to present data.  When that is missing
> # lynx uses this value.
> #PREFERRED_CONTENT_TYPE: text/plain

We did not have this section on Shellworld. It must have arrived 
in an update to lynx.cfg that we didn't take for reasons of not scratching 
existing configs.
Lynx.cfg is such a huge config file, that I wish there was an efficient 
way to break it into a lynx.d style arrangement.

Anyway, after including an appropriately modified version of this section, 
and compensating for the below bug, both our old and new versions of lynx 
can handle the URL in question (https://www.radiotapes.com) from the version of 
Apache 
that isn't sending content-type for some reason.

> Note: I believe there shouldn't be a space between the colon and 'text'
> in the lynx.cfg setting - at least it didn't work for me with a space
> but did work without.  This means there is a bug either in the parser
> or in the sample lynx.cfg which comes with Lynx.

I can confirm that bug, whichever it is.

Luke