Pae: The term protocol implies that there is a there is a network based exchange of information.
In the case of 'file://' URLs, your client software ends up making system calls (e.g. open(), read(), close()), no exchange of information across a network. Ah, but what about files on your network drives? you may ask. Same thing goes here. Your client software simply makes open, read, and close system calls. However, your O/S, under the covers, turns these into network exchanges (e.g. NETBIOS, NFS, SMB, Netware, ...) However, your client has no knowledge that these files are hosted elsewhere. As far as it can tell these 'remote' filesystems are hosted locally. Your client is blissfully unaware of any network activity, and of the actual protocol used to retrieve the file. The protocol used, in this case is determined by the O/S. Furthermore, consider that: The http protocol is defined by RFC 2616. The ftp protocol (ftp://) is defined by RFC 0959 / STD009 Can anyone tell me where the file:// protocol is defined? The 'file://' protocol specifier, doesn't actually relate to a known protocol. It is merely a shorthand way of treating local files as URL's. Hope this is helpful. Regards, Tom Drake ----- Original Message ----- From: "Pae Choi" <[EMAIL PROTECTED]> To: "Tomcat Users List" <[EMAIL PROTECTED]>; "Tom Drake" <[EMAIL PROTECTED]> Sent: Monday, January 07, 2002 11:18 AM Subject: Re: Problem with file URL | Hmm.. the communication between the browser(client) and | the file system(server) is not considered in the client-server | communication? An interesting comment. | | | Pae | | | | > There appears to be confusion about the 'file://' protocol. | > | > The "file://" protocol is really not a protocol at all, since there | > are no client-server communications involved. | > This 'false' protocol identifier is simply a 'nice' little feature | > that web-browsers implement (URL/URLConnection also | > implements this), that provides the means of displaying a | > file stored on the client system (or on a drive that is | > mounted on the client system). | > | > The key here is 'client-system'. When you type a "file://" | > URL into your browser, your browser goes straight to | > your filesystem to get the file. It does not contact a web-server | > for this. | > | > If you want to retrieve a 'file' from a web server, you must | > provide a URL that starts with 'http://' (or https://). | > | > Hope this is helpful. | > | > Tom Drake. | > | > ----- Original Message ----- | > From: "Michael Wentzel" <[EMAIL PROTECTED]> | > To: "'Tomcat Users List'" <[EMAIL PROTECTED]> | > Sent: Monday, January 07, 2002 9:44 AM | > Subject: RE: Problem with file URL | > | > | > | BTW, is there a reason this is that important to you? How many users | > | try to access a url using the file protocol instead of http? | > | | > | --- | > | Michael Wentzel | > | Software Developer | > | Software As We Think - http://www.aswethink.com | > | | > | -- | > | To unsubscribe: <mailto:[EMAIL PROTECTED]> | > | For additional commands: <mailto:[EMAIL PROTECTED]> | > | Troubles with the list: <mailto:[EMAIL PROTECTED]> | > | | > | | > | | > | > | > -- | > To unsubscribe: <mailto:[EMAIL PROTECTED]> | > For additional commands: <mailto:[EMAIL PROTECTED]> | > Troubles with the list: <mailto:[EMAIL PROTECTED]> | > | | | -- | To unsubscribe: <mailto:[EMAIL PROTECTED]> | For additional commands: <mailto:[EMAIL PROTECTED]> | Troubles with the list: <mailto:[EMAIL PROTECTED]> | | | -- To unsubscribe: <mailto:[EMAIL PROTECTED]> For additional commands: <mailto:[EMAIL PROTECTED]> Troubles with the list: <mailto:[EMAIL PROTECTED]>
