I bet you could use cURL for this, using the FILE protocol. Build your
curl command line with filenames formatted for curl; open it as a
process for read; read that process's output every 500 msecs or so to
get the progress info.
I'm on assignment in NZ right now so I don't have everything at my
fingertips. Josh Mellicker made me aware of curl a few months ago, and
Ken Ray may have made a rev library of it by now. ;-) Also Mark Smith
created a rev curl lib some time ago.
Food for thought...
Phil Davis
On 23/05/10 12:26 PM, Alex Tweedly wrote:
First thought : I thought (according to Dictionary) that "shell" would
wait until the command was complete before returning - dictionary says
The current handler pauses until the shell returns its result. If the
command was successful but did not return anything, the shell
function returns empty.
Second try : why take a checksum, why not just use the file size ?
Getting the checksum may involve copying the file from the remote
server into your laptop to calculate the checksum.
Third try : instead of doing "shell 'cp file ....' " could you do
load url ("file:" & tRemotepathname)
.... check the cachedURLs ....
put url("file:" & tRemotepathname) into url('file:" & tLocalfilename)
-- Alex.
On 22/05/2010 23:35, Sivakatirswami wrote:
Work working on an in-house file manager - RCS for Indesign (since
Adobe did an "end of life" for Version Cue)
It's coming along well, in fact really well.
One challenge is that I'm using lots of these:
put ("mv " & quote & gLocalFilePath & quote &" "& quote &
(gLocalProjectPath& "/"& tShortFileName) & quote) into tShell
get shell (tShell)
put ("cp " & quote & (gLocalProjectPath& "/"& tShortFileName) &
quote &" "& quote & (gServerProjectPath& "/"& tShortFileName) &
quote) into tShell
get shell (tShell)
to more files around and rename them... it works great.
Is there a way to monitor a background shell process like this?
The problem is if you copy from the Big Server on the LAN... to my
Little MacBook Pro... you won't know when the file is completely
copied to the local hard drive before doing:
launch (gLocalProjectPath &"/" & pFileName) with the
uInDesignPath of this stack
Typically RunRev will issue the unix cp command and then immediately
launch inDesign, which "crokes" because the file is incomplete on the
local hard drive.
Now I have some ideas about how to do this: get the checksum some of
the remote file, do a send in 30 ticks (repeatedly) test of the local
file until the local file's checksum matches the remote file and then
launch it.
But before I go after this, I was wondering if anyone had any other
method? Most "cool" would be a progress bar, but I'm not sure RunRev
can monitor a local unix copy file process....or, if it can, how to
do it.
Thanks!
Sivakatirswami
_______________________________________________
use-revolution mailing list
[email protected]
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
