IMPORTANT: This project has been moved to https://bitbucket.org/byuhbll/filmmatch.

Film Match

INTRO

Film Match is an application written by Charles Draper of the Harold B. Lee Library at Brigham Young University to crowdsource the matching of locally cataloged films to their corresponding films in IMDb. Given a table of catalog record IDs and titles, Film Match attempts to find that film in IMDb. Films that are matched are presented to the user for human verification. Once all films have been processed, the IMDb ID's can be used in conjunction with themoviedb.org to pull down film poster images.

SETUP

Experience managing a LAMP stack is required.

Film Match was developed to run on a LAMP stack (Linux, Apache, MySQL, and PHP). Other software bundles are possible, but may require tweaks to the code.

1. Download Film Match from https://bitbucket.org/byuhbll/filmmatch which I'm assuming you have already done.

2. Copy or link the contents of filmmatch to where your web application will be executed. This will vary depending on the host. I will refer to this location as the APPDIR. APPDIR/htdocs is the document root.

   Note: the following packages need to be installed on the host machine a. php5-curl b. php5-mysql

3. Set up a MySQL database with a user that has SELECT, INSERT, and UPDATE privileges on that database.

4. Copy APPDIR/config.inc.sample to APPDIR/config.inc and configure config.inc to your environment. Additional instructions are found with each setting in this file.

5. Initialize the database by executing the sql script found in APPDIR/init/init.sql.

6. Populate the Record table in your database with films from your catalog. You only need to populate the catId and title fields. You can also optionally populate the notes field with any other additional information you like, such as a call number.

   a. Record->catId: The catalog record ID for the film b. Record->title: The film title c. Record->notes: Anything extra you want to add to the entry

   Example

   catID title notes

   ---

   3160949 The emperor's new groove DVDC 3138 3152597 Ever after a Cinderella story DVDC 3102 3193309 Fantasia 2000 DVD 1042

   All other fields and tables are populated by filmmatch.

RUN

The application is now ready for use. Direct people to the uri of your application (ie, http://your.server.com/) to start matching.

Note: As long as Record->status is NULL for a given film, the application will attempt to find the next best IMDb match.

RESULTS

Results from matching are placed in the Result table. You will want to pay attention to those records with a status of 'yes' in this table. Using the imdbId of such records, enrich the corresponding catalog record with that imdbId. For marc you could add an entry to 035/a with (IMDb)tt00112233. You can also view just the yes results by going to http://your.server.com/results.php?status=yes .

The History table is a log of all actions taken by users. Result carries the final say as to whether catalog records match or not with a particular imdbId. Keep in mind that films are always being added to themoviedb.org and IMDb so you will want to keep the Result table intact for future rounds of matching as the Result table carries all the "no match" decisions and you won't want to continually re-ask those same questions.