On Thu, Aug 27, 2009 at 7:26 AM, <[email protected]> wrote: > I was thinking more in terms of how python would find it more efficient in > retrieving the information, this being: > 1) Storing all the user files in one place and then selecting only the files > with the correct permissions for the user views > 2) Storing the files in the user's folder and then building a list of this > files on the main views, such as /news/index.html > > I guess in the first instance the code will only has to traverse one > directory and build the dictionary of all the files for that specific user > when the user wants their files. i.e /users/user/news/list_of_news_items.html > > Whereas on the second method the code will have to travererse each user's > directory and build a dictionary for all the published documents to display > for the main view i.e /news/index.html
There is no right answer to your question the way you have asked it. Organizing per user will be faster for the per-user search. All files together will be faster for locating all files. Which one do you want to prefer? > The thing that bothers me is that I ma have 10 users or 100,000 users and > really wanted to get an opinion as to which option would scale better, > leaving aside the relational DB approach. I'm no expert on scaling but I can't imagine that any approach based purely on scanning a filesystem is going to scale to 100,000 users. Perhaps you should use a database to make an index of the files? Kent _______________________________________________ Tutor maillist - [email protected] http://mail.python.org/mailman/listinfo/tutor
