views:

230

answers:

2

Hi I need your help if you are an expert in MOSS.

I have a system that needs to upload documents into a MOSS document library.

I decided that the easiest approach for a phase 1 system would simply be to map a network path to a MOSS Document library.

The whole thing seems too easy. After that its a straight copy using System.IO.

What I would like to know, is this method reliable enough to be used in a production system?

Speculation would be great, but if you have real experience with working with MOSS in this way, your answer would mean a lot.

Thanks.

+1  A: 

So long as you do the proper error checking around the copy, its fine - if you bear in mind the standard caveats with SharePoint document libraries and file naming conventions.

SharePoint does not allow some characters in file names which NTFS and FAT do - these will cause an error when you try to copy them to the DL, regardless of how you do that, so you will need to sanitise your filenames beforehand.

The only downside to using a network path to the webdav interface of SharePoint is that if you stress it too much (a large copy of a lot of files), you can easily overwhelm it and it will cause the network share to become unavailable for a period of time. If you are talking about a few files every now and then, several an hour for example, it should be fine.

Moo
I think max load should be around 2000 files at once copy.
JL
A: 

You are better off reading the files off the network path and then using the Object Model API or the web services to upload it.

You can use timer jobs which can be scheduled to run at a convenient time. The timer job can read it's configuration settings from an XML file.

This system would be easier to maintain, and troubleshoot, when compared to a straight copy using System.IO.

ashwnacharya