Table of Contents
 |
With the transition to the WACS 0.9 release series, it's time to have another look at the core database schemas and see how the current schema is holding up and what attention it needs. A significant feature of this revision is taking a long hard look at those fields that are likely to be needed for full-on commercial use and incorporate them at this time.
This is revision 3 of this document containing those changes that did make it into the new schema settled upon in Wacs 0.8.5, and for which the wacsschema tool introduced in Wacs 0.8.6 should make older database schemas compatible with.
Since it's been a long time since we changed any major schema, it's probably time to uprev the WACS_EXPORT_FORMAT to 2 so that older installations wacsxmlin and co do not fail in nasty ways.
This idea, inspired by the close-up examination sets on EuroNudes, is to have an optional photo of the model standing naked as a general idea of what she looks like, rather than just headshots.
The wacsmodelmanager command has now been updated to incorporate this field and we're now trying out making use of this with images of 200x400 pixels. This appears to be a reasonable size for the most likely uses of this image.
The idea behind this one is to indicate where her pussy lips protrude and if so by how much - this would probably be a three value single character field.
We've added this field to the model manager and created an initial four values for it - they are:
This would be the model's dress size - probably specified in the euro sizes as they have the largest numeric range and should therefore scale with most accuracy to the other names.
This would be for the model's star sign - reasonably easy to calculate if we have her birthdate or this information from elsewhere.
A simple yes/no field as to whether we have the records relating to 18 USC 2257 on file ourselves.
One added major set attribute - this one indicates the nature of the set and in particular the relationship between the model and the camera.
Most of these proposed changes relate to how to navigate amongst the various set records. The issue is basically to better cope with duplicate sets, alternate media sets, crossover sets that are both images and videos and multi-set sequences.
This is the link to the set number that features the same scenario in another media - in an image set, this would link to the video clip; in a view clip, to the corresponding image set. If in future, there are three media types (ie SE's Rene and Linda - Hallway which is a photoset, video clip and DVD track), the video set refers to the DVD, the DVD back to the image set.
This simply links to the next set that is a continuation of this one - likely to be used mainly for movie files, it'll probably also get used by the DFR sets on AMK which continue over two sets.
The code to understand the presence of this has now been incorporated into the wacsmodelpage and is currently under test. At the moment there's no webapp access to updating it.
This simply links to the previous set of which this set is a continuation - basically an other-way pointer for snext. This is expected to be used mostly for movies.
We need more control over Duplicates and multi-part sets than the current system of flagging a set as being a duplicate provides. The idea here is that any free standing or initial set of a movie etc is flagged as primary. In doing a simple overview of sets, you would only select primary sets - continuations and second parts of movies, etc would be hidden. The next level down would consist of continuations of primary sets, ie the second and subsequent parts of movies or similar sets. A third level would include alternative versions such as the multi-part versions of movies if we also have a whole thing file as the primary. There is also the issue of the different branding/same content sets that we're currently handling this way and whether those should be a further level lower or are really not that different from the multi- file versions of things we already have as a single file. A final type could be a "not yet released" set allowing sets to be registered but be of very limited availability until the appointed date. We already have the srelease field for the release date but haven't really used it as we didn't want to add that much complexity to every search we perform. Now this will not be painful or expensive in CPU time.
The long term plan is that these changes would render the D - Duplicate set type obsolete. This is a bit of a misnomer anyway as it still has a genuine type of it's own - ie Solo, Lesbian, Toys, etc. The set manager and unpacking tools in subversion have now been updated to handle the following values of srank and we'll have to see how that works out.
The code is now in wacssetmgr, wacsunpackmgr and updateinfo to make use of this field and allow it to be set. In addition wacsmodelpage can now make pretty extensive use of this information.