Welcome!

By registering with us, you'll be able to discuss, share and private message with other members of our community.

SignUp Now!

copy /g /u "http://...dropbox.com/..." not working

C

Charles G

Apr
1,818
17
c:\copy /g /u "http://dl.dropbox.com/u/14823386/winapp2.ini" "C:\Program Files\CCleaner\DB_winapp2.ini"

The "/u" option does not work; Tried with v8 to 12.1 (latest builds). Does /u work on http copies at all?

I looked at www.dropbox.com and can't tell what type of servers they use...

TCC 12.10.63 Windows XP [Version 5.1.2600]
TCC Build 63 Windows XP Build 2600 Service Pack 3
 
Why then is an error message not given? Also thank you Rex for answering
----- Original Message -----
From: rconn
To: [email protected]
Sent: Thursday, April 28, 2011 08:13 AM
Subject: RE: [Support-t-2806] Re: copy /g /u "http://...dropbox.com/..." not working


Quote:
Originally Posted by Charles G
c:\copy /g /u "http://dl.dropbox.com/u/14823386/winapp2.ini" "C:\Program Files\CCleaner\DB_winapp2.ini"

The "/u" option does not work; Tried with v8 to 12.1 (latest builds). Does /u work on http copies at all?

/U cannot work on HTTP servers, as it is not possible to get the file time on HTTP files.
 
Charles G said:
Why then is an error message not given?
Click to expand...

Because the option processing is about 5 levels removed from the HTTP copy, and there's no way the parser can jump ahead to determine what you're actually trying to do.

You cannot do a DIR on an HTTP directory (or get any file info at all); this is an HTTP standard limitation and there's not anything that TCC can do about it. If you can access the directory via FTP, do that instead.
 
the warning is in the help chm



> -----Original Message-----
> Subject: RE: [Support-t-2806] Re: copy /g /u
> "http://...dropbox.com/..."
> not working
>
>
> Why then is an error message not given? Also thank you Rex
> for answering
> ----- Original Message -----
> From: rconn
> To: [email protected]
> Sent: Thursday, April 28, 2011 08:13 AM
> Subject: RE: [Support-t-2806] Re: copy /g /u
> "http://...dropbox.com/..." not working
>
>
> Quote:
> Originally Posted by Charles G
> c:\copy /g /u
> "http://dl.dropbox.com/u/14823386/winapp2.ini" "C:\Program
> Files\CCleaner\DB_winapp2.ini"
>
> The "/u" option does not work; Tried with v8 to 12.1
> (latest builds). Does /u work on http copies at all?
>
> /U cannot work on HTTP servers, as it is not possible to
> get the file time on HTTP files.
>
>
>
>
Click to expand...
 
rconn said:
You cannot do a DIR on an HTTP directory (or get any file info at all); this is an HTTP standard limitation and there's not anything that TCC can do about it
Click to expand...
Would it not be possible to use the Last-Modified header field, if present?? (I realise that it isn't a required field, and therefore you would have to revert to the current behaviour if it wasn't included, but if it was then it would seem like it should suffice)
 
Steve Pitts said:
Would it not be possible to use the Last-Modified header field, if present?? (I realise that it isn't a required field, and therefore you would have to revert to the current behaviour if it wasn't included, but if it was then it would seem like it should suffice)
Click to expand...

Four problems with that:

1) I'd have to write a custom parser for any command that referenced HTTP, because it would have to be done at a different level than the current "find files" code. This would probably take at least a couple of months.

2) The IPWorks code doesn't support it, so I'd have to write my own HTTP classes. Add another couple of months.

3) Most of the HTTP servers I've looked at don't use it.

4) It cannot work with any wildcards, because I cannot query the directory on an HTTP server. I'd already have to know the exact filename.

And the bonus issue:

5) Unless you have a really large file, it is faster to simply always do a put rather than trying to find the date and doing a compare.
 
Dropbox doesn't allow FTP access - appearently. I can read the file using line[] and compare against the last stored version

The file is to add additional entires to CCleaner, what I use to clean junk files of any computer I work on. It's of the form:

================= BOF
; Application Cleaning files
;
; Version: v1.0.110227
; Notes
; # of entries: 460
...
================= EOF

the 110227 would be of the YYMMDD format.... The rest would be fairly simple
----- Original Message -----
From: rconn
To: [email protected]
Sent: Thursday, April 28, 2011 03:30 PM
Subject: RE: [Support-t-2806] Re: copy /g /u "http://...dropbox.com/..." not working


Quote:
Originally Posted by Charles G
Why then is an error message not given?

Because the option processing is about 5 levels removed from the HTTP copy, and there's no way the parser can jump ahead to determine what you're actually trying to do.

You cannot do a DIR on an HTTP directory (or get any file info at all); this is an HTTP standard limitation and there's not anything that TCC can do about it. If you can access the directory via FTP, do that instead.
 
re: #3: How would I find out what software a particular HTTP server uses? Or if they have a FTP access that may or may not be documented...?
----- Original Message -----
From: rconn
To: [email protected]
Sent: Friday, April 29, 2011 10:21 AM
Subject: RE: [Support-t-2806] Re: copy /g /u "http://...dropbox.com/..." not working


Quote:
Originally Posted by Steve Pitts
Would it not be possible to use the Last-Modified header field, if present?? (I realise that it isn't a required field, and therefore you would have to revert to the current behaviour if it wasn't included, but if it was then it would seem like it should suffice)

Four problems with that:

1) I'd have to write a custom parser for any command that referenced HTTP, because it would have to be done at a different level than the current "find files" code. This would probably take at least a couple of months.

2) The IPWorks code doesn't support it, so I'd have to write my own HTTP classes. Add another couple of months.

3) Most of the HTTP servers I've looked at don't use it.

4) It cannot work with any wildcards, because I cannot query the directory on an HTTP server. I'd already have to know the exact filename.

And the bonus issue:

5) Unless you have a really large file, it is faster to simply always do a put rather than trying to find the date and doing a compare.
 
On Fri, 29 Apr 2011 14:17:14 -0400, Charles G <> wrote:

|re: #3: How would I find out what software a particular HTTP server uses?

If you can packet-sniff, it's usually given in the first HTTP response.
Lucky.syr.edu uses Serv-U (FTP) which has a little HTTP server in it. It's
first HTTP response includes this.

Code: 
0x0030:  2032 3030 2048 5454 5020 4f4b 0d0a 5365  .200.HTTP.OK..Se
0x0040:  7276 6572 3a20 5365 7276 2d55 2f31 302e  rver:.Serv-U/10.
0x0050:  352e 302e 3131 0d0a 4461 7465 3a20 4672  5.0.11..Date:.Fr

Winpcap/windump provide a free, easy, console packet-sniffer. MS's NetMon is
also now readily available; it's GUI and relatively hard to learn.

And the server might identify itself in the HTML itself. Buried in "TYPE
http://lucky.syr.edu" is "Serv-U from RhinoSoft.com".
 
You must log in or register to reply here.
Back
Top
[FOX] Ultimate Translator
Translate