On Fri, 31 Mar 2000 11:52:21 +0800, Cristian Burneci wrote:
> I wish to ask rhe following question: What is Arachne's behaviour
> when downloading files of different types than the ones listed in mime.cfg.
> While surfing today, I have reached a certain site (which I've forgotten to
> add to hotlist). There were links to some files to download, mainly ".sch"
> (Orcad schematics) and ".lib" (Orcad schematic libraries). The links had tis
> format (generic): "http://www.someprovider.com/somedirs/afile.sch (or
> afile.lib)". Instead of downloading the files to cache and asking me what to
> do with them, Arachne tried to display them as if they were HTML files. I
> waited patiently until all the garbage finished appearing on the screen,
> then quitted Arachne and searched the cache for the files.
What happens with a file as it is being downloaded is stricly controled
by the "content-type" which is sent by the server commputer you are
downloading from.
If the content-type is "text/html", your brouwser will attempt to
decode it as html.
If it "text/plain", the browser will disply it as text.
If it is "image/jpeg", the browser will attempt to decompress and display
the .JPG image.
This holds-true for all file types.... image/gif, application/zipped,
application/octet-stream, etc...etc...etc...
> Also I tried for a couple of times to download the software for
> ELNEC's 51&AVR flash programmer (it can be found at http://www.elnec.com, and
> it consists of an autoexpandable LZH archive named pg4uarc.exe). Every time I
> try, Arachne downloads about 50% of the file, then, strangely enough, tries
> to display it!.
While the file is downloading, press ESC or left-click the "X" button
to abort the D/L. Next, press the = key to view that file's .HTT file.
I think you'll see that the server computer sending the file to you is
incorrectly using "content-type: text/plain"
(it _should_ be using "content-type: application/octet-stream")
Now, you might ask: Why do the other browsers D/L this file just fine?
The answer is simple: Those "other browsers" are bloated-up with code
to over-ride the "content-type" and look inside of the file itself.
In this way, they identify the file type regardless of what the server
_says_ it is.
> Is there a key combination or other feature, allowing the user to
> simply download the target of the link which has the mouse pointer over it?
Agian, this is another piece of "bloated code" contained in those other
browsers. (right-click the link to pop-up a menu of choices for what to
do with the file).
If you have this kind of problem on a particular server. The solution
is simple.
1) contact the webmaster of the site in question and ask them to fix the
"broken content-type" sections of thier setup.
2) while you are waiting for h**l to freeze-over, fix it yourself on
your end
a) Alt+E out of Arachne
b) rename mime.cfg to mime.cf_
c) rename mime.brk to mime.cfg
(mime.brk is a file you create which has just this one line....
text/html HTM
[This will allow you to view the page with the link after re-entering
Arachne])
d) "exit" the DosShell back into Arachne.
e) download the file and it will be saved to disk
(without any mime.cfg refferences to use, Arachne will have no option
but to save the file to disk)
When you are finished at the "broken site", restore your mime.cfg file
and continue to surf at sites which are not broken.
--
Glenn McCorkle [EMAIL PROTECTED] North Jackson, Ohio, USA
DOS prog. for QV cameras http://www.angelfire.com/id/glenndoom/qvplay.html
Other stuff http:[EMAIL PROTECTED]/
Arachne, The Web Browser for DOS
Open the 'DOOR' to the WWW. Keep the 'windows' closed.
http://arachne.browser.org/ http://arachne.cz/
