algorithmic modeling for Rhino
Today I have noticed a strange behaviour of the Read File Node.
I am reading a .csv file of 58kb, which has close to 3000 lines of numbers. So nothing special.
If I read the file from a network share (via SMB from a Mac Server), then the Read File node takes forever (a few minutes) to read the file. Grasshopper is unresponsive in that time.
If I read the very same file from a local harddrive (its synced also via Dropbox), then the time to read the file goes down to a normal level of 28ms.
Files from the network share open normally and very fast otherwise (we have Gigabit Ethernet and a very fast Raid setup on the server). The problem is clearly not the setup per se.
Anyone come across this bug and/or have a solution (other than copying files locally all the time).
1. Loading from network share:
2. loading from local harddrive:
It's probably the mechanism that keeps track to see if the file changes. It communicates a lot with the drive and seems to retard other IO streams. Are you running the latest version of GH?
Ah alright, yeah that could be a reason. I just did a little test. If I edit the file I get a red warning that the process cant access the file because it is in use, but upon saving and closing the editor it doesn't revert to a non-red state. But ok, that is understandable - I guess once it goes red it stops checking continuosly.
Still strange though that it would take so long when loading the file. Its probably stuck in some sort of loop. Technically accessing the file to see if it has changed should still be possible via the network. I'm sure there is a way to fix it and I guess a lot of people in offices have some sort of file server.
Maybe an update pin, which can be set to automatic. It doesn't always make sense to check for changes constantly?!
Or it could save a local temporary version!?
I am running 0075.
There's less intrusive ways for me to check changes to files. I've updated code in a few places, but I think this component still uses stupid code.
You can use the attached VB script component to read all the lines in a file. It shouldn't suffer from the same delays caused by attempted aynchronisation.
Hello again David.
As a follow-up I have noticed that the Image Sampler REALLY doesn't like images from a network location. It just does stuff forever and I basically have to force quit Rhino and even that doesn't work sometimes, forcing me to restart the computer. :/
Any Ideas about that one?