Various information about 'sql'. [diff files] Timo Schulz (who I thank again) worked a lot on this, coming to the conclusion that it's not feasible (at least with the current layout of the database) and that right now it's easier and faster to download the diff files, apply them and rebuild the database from scratch. I mostly do agree: not only it's a very complex task, but it must also be very fast to make some sense.
Maybe we can supply a simple (Python?) script to handle the downloading/applying of diff files - right now I'm using a home-made bash script. At the moment, if you need to populate your database as fast as possible, your best option is probably the '-c' argument of the imdbpy2sql.py script, used to create a set of CSV files to be later imported into the database. [info from obsolete files] A recent (and still untested...) improvement of the imdbpy2sql.py: now it ignores information from obsolete files (german-aka-titles, italian-aka-titles, laserdisc, iso-aka-titles and files in the 'contrib' directory) if the movie was never seen elsewhere. [movie titles in the old "Title, The" format] There are still some files containing titles in the old format; if I'm not wrong: complete-crew, distributors, keywords and special-effects-companies. The IMDb staff said that they are aware of this and they will fix the problem, even if it's not a high priority. This can be fixed with the --fix-old-style-titles option of imdbpy2sql.py, which I'll probably activate by default in the SVN version in the next days. Have fun, -- Davide Alberani <davide.alber...@gmail.com> [GPG KeyID: 0x465BFD47] http://erlug.linux.it/~da/ ------------------------------------------------------------------------------ _______________________________________________ Imdbpy-devel mailing list Imdbpy-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/imdbpy-devel
