Downloading a File from a web page?
Eric Pettifor
>This may be a really dumb question- But I can't get a file to download
>from one of my pages without my broswer (Netscape) just dumping the hex
>into my window. How I let Netscape know that this file is to be downloaded
>& opened with a helper, and not opened as text? Should I be using a
>special Tag?
Nope. It has to do with definitions on the server side. You can
contact your ISP and see what they can do about it, or on your side
two things you might try are zipping the file, since that's a common
one the server is likely set up to recognize, or putting a note next
to the link instructing the user to save the file to disk (shift click
in Netscape and Mosaic).
Test it out yourself first. I found that Netscape (1.2 and 2.0b1)
corrupted it such that the resulting file crashed the programs the
files were subsequently loaded into (Word .doc in one case, Fractal
Design Painter .rif in the other). The solution was to add the
extension .bin (i.e. test.doc.bin), in which case you might want to
instruct the Netscape user to 'save as' the file name with actual
extension. In my test cases the problem was only with Netscape. NCSA
Mosaic saved to disk just fine, and as well I configured it to
automatically open Word even with the .bin extension, but that would
be a little much to ask the casual visitor to your page.
Good luck with it.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Eric Pettifor ~~~~~~~~~~~~~~~~~~~~~~~~~~~
er...@wimsey.com ~~~~~~~~~~~~~~~~~~~~~~~~
http://www.wimsey.com/~ericp/giganto.html
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Walt Carroll
upload .exe and .zip files to my Web Page so that my students can
download them (without asking students to FTP [ah, hhm] somewhere to
fetch the files).
Let's not get this thread get lost. It's important.
Walt Carroll
http://www1.mhv.net/~omi/courses.htm
Peter Hahndorf
> But I can't get a file to download
>from one of my pages without my broswer (Netscape) just dumping the hex
>into my window. How I let Netscape know that this file is to be downloaded
>& opened with a helper, and not opened as text?
for zip-files in Netscape, goto to Options - Preferences - Helper Apps
create a new type using:
Mine Type: application
Subtype: x-zip-compressed
File Extensions: zip
Action: Save To Disk (you can also choose a your favourite ZIP-shell
to be started)
on your page you need a simple link to the zip-file like:
<A HREF="download/test.zip">Download the file</A>
Pete
--
mails over 16Kb to: pe...@zfn.uni-bremen.de
http://www.uni-bremen.de/~pete/
Alex Kremer
>This may be a really dumb question- But I can't get a file to download
>from one of my pages without my broswer (Netscape) just dumping the hex
>into my window. How I let Netscape know that this file is to be downloaded
>& opened with a helper, and not opened as text? Should I be using a
>special Tag?
Really nothing easier than that. Go into Netscape/Preferences (Or
Options/General if 2.0), click on the helper apps tab, define .exe as
SAVE TO DISK, and you're all set...You can do this in almost any
browser, just tell the people that want to download the stuff they
have to do the above on the Web Page.
Alex Kremer
a...@raven.cybercomm.net
Reinier Post
>Good question that should be persued. I would _love_ to be able to
>upload .exe and .zip files to my Web Page so that my students can
>download them (without asking students to FTP [ah, hhm] somewhere to
>fetch the files).
Uploading files to a WWW server is not supported by many browsers.
If you can put the files on your WWW space, or in some space on an
anonymous FTP server, they can use their WWW browser to get them.
Whether or not the files are in HTML is *irrelevant*.
>Let's not get this thread get lost. It's important.
The 'how do I download files' question recurs about five times a week in
this newsgroup. As this newsgroup is about HTML, and downloading files is
completely unrelated to HTML (it's a browser option), this is causing some
annoyance. Sometimes I wish Netscape would add these three tags to their HTML:
<DOWNLOAD FILE="nicepic.gif">
<BACK>
<COUNTER>
then we would probably get rid of the three most annoying FAQs on this
newsgroup. (Instead, we would have more fuel for HTML wars. I prefer the
HTML wars; they're more fun to read.)
Speaking only for myself of course.
Being lazy, I'll just repost my previous posting on this matter. I know,
it's much too long. Hope you find it useful anyway.
Best regards,
Reinier
| From reinpost Wed Sep 27 20:19:01 MET 1995
| Newsgroups: comp.infosystems.www.authoring.html
| Subject: Re: How to link to a downloadable file?
| References: <tkidd.810521045@hubcap> <42q025$o...@news.cais.com> <42vmg9$9...@grid.direct.ca>
| Followup-To: comp.infosystems.www.authoring.misc
|
| In article <42vmg9$9...@grid.direct.ca>,
| >tk...@hubcap.clemson.edu (Travis Kidd) wrote:
| >
| >>I'm trying to make a link to a file that should be downloaded.
| >>I've tried several things in vain.
| >>Any help would be appreciated.
| >>-Travis
|
| *Every* page you view is a file that is downloaded. Just make a link to
| the file and the user will get it.
|
| The way the browser handles the file will depend on how your server
| is configured, and on how the user's browser is configured. HTTP servers
| can explicitly indicate the type of file being sent (HTML, plain text,
| GIF, PostScript, binary DOS executable, etc.) by providing a so-called MIME
| type. Usually, it automatically deduces this type from the filename
| extension; this is configurable but for most of the well-known file types
| you don't need to bother.
|
| Web clients use this MIME type to determine what to do with the received
| document. This is again configurable. Most browsers display HTML, plain
| text, and some of the image formats; they are configured to start up
| external applications on a number of other types; and for all the rest,
| they ask the user to provide a filename, and save the document to the local
| disk under that name.
|
| However, not all documents are served with a MIME type.
|
| As explained, HTTP servers *do* provide MIME types for their documents,
| but Web browsers can also visit documents from FTP servers, Gopher servers,
| the local filesystem, and some other sources. Such documents tend *not* to
| have explicit MIME types, so the browser needs its own heuristics to guess
| a MIME type in cases where none is provided. This is similar to what an HTTP
| server does, for usually the MIME type is derived from the filename
| extension. And it is again configurable for most browsers.
|
| For example, I have recently bought a PC with a sound card. I can now
| record and play musical scores in MIDI format. As it turns out, there are
| many MIDI files on the Web: some entangled in a Web of HTML pages, some in
| plain directory structures, usually served from an FTP server. My local
| collection of MIDI files is growing.
|
| My PC is connected to the Web over a SLIP connection.
| I soon found the Classical Music Archive (
|
| http://www.ima.net/~prs/midi.html
|
| ) and of course I wanted my Web browser to fire up one of my players on
| the MIDI files exhibited there. It took some trouble (not too much) to
| find out the MIME type provided with these files (it's 'audio/midi').
| I configured my WWW browser to start up my favourite MIDI player on that
| MIME type.
|
| However, I also use my Web browser as a file browser for local files.
| In addition I found a large collection of MIDI files on an ftp site
| (ftp.cs.ruu.nl). My local files and the files in the FTP archive have
| the extension .mid; so I configured my browser to associate with .mid
| the MIME type audio/midi.
|
| Soon, I decided I wanted a two-way connection to the Web; I installed my
| own HTTP server. Whenever I'm dialing in, the rest of the world is able
| to point its WWW browser to my local machine, fetch my homebrew MIDI files,
| and play them on their synthesizer. In order to make this work, I had
| to configure my HTTP server to serve all *.mid-files with MIME type
| audio/midi.
|
| (I don't really dial in that often. It's mainly for my own benefit: the
| server provides an easy way for me to upload files to the dialin host.)
|
| Now to get back to your original question: you want people with WWW browsers
| to be able to download files. If all you mean is, the browser should fetch
| the files like it fetches HTML pages, all you have to do is put them on the
| Web server and make links to them. If you mean, the browser must display
| a 'download file' dialog box upon loading one of these files, my question is:
| why would you wish to decide how the browser handles the documents it
| receives? Put type information in the MIME type, and let the user decide!
| All browsers have options to download documents to disk, even for documents
| they know how to handle.
|
| comp.infosystems.www.authoring.html really is the wrong newsgroup for this
| problem, because, as I hope my lengthy (and pedantic) explanation tells you,
| the problem is 100% unrelated to HTML. Followups set to c.i.q.a.misc.
|
| Tor Rognmo <tro...@direct.ca> responded:
|
| >If the server you are using is set up for FTP, the safest is to store
| >your file there and point to the ftp site in your link.
|
| [useful explanation deleted]
|
| This is beside the point. Whether the document is being served from an
| HTTP server ('a WWW server') or an FTP server has little to do with the
| original problem. A Web of HTML pages can be served from an FTP server;
| a bunch of binary files can be served from an HTTP server.
|
| To 'download' is purely a browser action, related to the document type.
| The choice between using an HTTP or FTP server is unrelated to the types
| of documents you want to serve. Both have their disadvantages: FTP servers
| don't offer explicit indication of content types like HTTP servers do, and
| are more inflexible in other ways; HTTP servers on the other hand cannot
| be used with those nifty FTP client programs that people seem to fancy.
|
| >protocol recognizes virtually any file format, and always prompts for
| >information before saving through your browser.
| >Storing the file on you web site is tricky unless it is in one of the
| >few formats that is supported by browsers - e.g. text, Acrobat pdf or
| >a gif image.
|
| Tricky in that, for documents with uncommon types (like MIDI), the server
| must be configured to serve the correct MIME type. But even if it's not,
| all WWW browsers I know will behave like you want them to, offering a
| download dialog.
|
| >Examples of files that normally will NOT download correctly unless
| >they are retrieved via ftp, are program files with the extension .exe
| >and compressed files with extensions like .zip.
|
| This is correct, but it means that either the WWW browser or its
| configuration is broken. In order to make your downloadable files (like
| compressed ones) work with such browsers, you may wish to hack something
| at the server side (ie. provide a different MIME type) but it should never
| be necessary to switch to FTP purely for this reason. In any case, it's
| a hack you shouldn't be needing to make.
|
| >These will be
| >retrieved as text files, and unless you're interested in the source
| >code for the file that's of little use.
|
| The user should reload the file with the 'download to disk' option enabled.
|
| --
| Reinier Post rein...@win.tue.nl
sEiKo
anything but a .GIF/.JPG or an .HTML.
--
________________________________________________________________
/ /|
+---------------------------------------------------------------+ |
| - Shawn C. Reed - | |
| | |
| S H A W N R E E D & C O M P A N Y | |
| Personal Computer Consultants | |
| | |
+---------------------------------------------------------------+ |
| Phone: (208) 234-3532 | |
| *** | |
| email: reed...@ba.isu.edu | |
| | |
| WWW Page Coming Soon! | /
+---------------------------------------------------------------+
Alex Kremer
>What is the URL you are using to access the file? It should be an FTP link if it is
>anything but a .GIF/.JPG or an .HTML.
Never heard of that one before...Put a ZIP in one of your WWW Dirs,
and then download it...No FTP Link there..
Alex Kremer
a...@raven.cybercomm.net