Re: LYNX-DEV Lynx 2-8 Problems on HP-UX 10.20

[T.E.Dickey]

> I have been running lynx2-7 on an HP-UX 9.05 system. I mainly use 
> "expect" software to parse http documents from lynx 2-7 with no problem. 
> The TERM is hpterm. 
>  
> I have just installed lynx 2-8 on an HP-UX 10.20 system. The parsing 
> using "expect" is not working. When I dump the datastreams from lynx I 
> notice a lot of escape sequences embedded within the text that are not 
> there in my 2-7/UX 9.05 system ( I bring up the windows side-by-side ). 
> For example, the http document would have "Username" in it. When you 
> look at the lynx 2-8 datastream it has something like : U\x1b&dBsername 
(that's "U" followed by <reverse video> followed by "sername").
What does the original html look like?

You can always suppress all of the attributes by setting up a terminal
description that doesn't have any (bold, reverse, underline, standout).
Lynx will run (but won't be very interesting to look at).  Your 'expect'
script still has to know how to filter out cursor addressing.

> in it.  These type of  escape sequences are prevalent throughout the 
> document. I try vt100 and still get stuff interspersed in text fields. 
well, at least we're consistent.
  
> How do I get just the text for a field without all this crud? The 2-7 
> version I am using is a precompiled version I pulled off the web site. I 
> have tried both the precompiled 2-8 version and compiled my own version 
> and I still get this stuff. Somebody please give guidance. 
>  
> ************************************************ 
> Paul  V. Kalmbach 
> SBC Services, Inc. 
> Room 3139 
> San Diego, CA 92121 
>  
> e-mail: address@hidden 
> phone: 619-886-4363 
> 


-- 
Thomas E. Dickey
address@hidden
http://www.clark.net/pub/dickey