Also the one thing with pickling these files is if any of the files get edited, you should perform some sort of locking since what if two people edit the same file at the same time?
Just a disadvantage of using pickle in a multi-instanced environment such as web servers. But you probably already know this There shouldn't be any performance issues as far as web2py is concerned, just the reading/writing to the file system which you will need to do anyways. You don't necessarily have to store the originals in the uploads folder, that is only a web2py convenience, you could use regular forms and save the files wherever you like (a mounted network share?) -Thadeus On Tue, Jan 26, 2010 at 10:12 AM, mdipierro <mdipie...@cs.depaul.edu> wrote: > I think what matters if how the files are downloaded. > > If the are downloaded via the provided "download" action then they > must be referenced by a database record and the names must follow the > convention "table.field.uuid.extension" (so that web2py can locate the > record and check authorization if any). > > If anybody can download them and they do not need to be referenced by > db, they should go in static. > > In all the other cases you should put them in "private" and create > your own "mydownload" action to retrieve them. > > The "private" folder was supposed to mean "private of the application" > as of "anything this app wants manage without web2py help", not > "private from users". > > Massimo > > > > > On Jan 26, 10:06 am, MikeEllis <michael.f.el...@gmail.com> wrote: >> I have a design question. >> >> I'm part of a team developing a scientific app that will allow users >> to upload seismic data and apply various kinds of processing and >> plotting. It looks as though we'll need to support several different >> raw data formats. The approach we're taking for this is to convert >> the data files after they're uploaded and store them in pickled python >> object files referenced by entries in a db table. All subsequent >> processing would be done with the pickled objects, but users may also >> want to retrieve the raw data in the original format. So my thought >> is that the uploaded data files should go into the uploads folder >> using the usual web2py mechanisms and the derived object files would >> be stored in the private folder. >> >> Is that consistent with the intended purpose of the private folder? >> Are there any (web2py) usage or performance issues that need to be >> taken into account? >> >> Thanks, >> Mike > > -- > You received this message because you are subscribed to the Google Groups > "web2py-users" group. > To post to this group, send email to web...@googlegroups.com. > To unsubscribe from this group, send email to > web2py+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/web2py?hl=en. > > -- You received this message because you are subscribed to the Google Groups "web2py-users" group. To post to this group, send email to web...@googlegroups.com. To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/web2py?hl=en.
