I know its been a bit of radio silence from me over the past two months. I've been busy with life but have found some time to work on the rebuild which will be the focus of this post.
First off, all the game and hardware data has been migrated! I know things are probably not 100% perfect as its hard to verify the large amount of data with changing and standardizing it so much. I'm hoping its very very close though and the next piece I'm working on will help me to test that. I'm currently working on migrating my personal collection to the new structure and getting the new collection tables setup in the new database. Since I obviously have familiarity with my own collection I can more easily compare the migration of my old collection records with the new. Once I have more confidence in all of the migration aspects, everyone else's collection will be migrated as well.
Speaking of collections, I have taken this opportunity to implement a newer feature on the database side (it won't be visible of course until the new site is actually running but its good to bring it up now). I've added support for admin configured, console specific tracking columns and user defined custom columns at a folder level. That may seem confusing so here's an example of each:
- Console Specific: In the US, many Turbografx-16 / CD game came with an outerbox that held the jewel case. An admin can configure those systems to have an outerbox quantity column that can be filled in when adding a game to your collection.
- User Specific: Maybe someone is really interested in tracking the games in their collection folder that still have their registration card, they can setup a custom column to add a quantity for registration card.
On the more technical side, a friend and I have been doing some backend work within AWS to get a test site working. We have a test domain registered and configured SSL certificates for https support. We are currently testing ways to host the site as a static webpage, basically meaning we can drop our html files into a location and be able to access them from the test domain, its sort of how the site works now but more modernized. We also believe we have a path forward for the API called a serverless API. Its kind of a recent buzzword but all it really means is we can setup an API without having to manage any infrastructure but with a lot of extra benefits we don't have today. We can more easily scale parts of the API that are used more for instance. It also lets us only pay for the requests that are made. Hopefully this will be cheaper than a flat hosting rate.
Oh and one last thing. I was able to make two small improvements to the current site. First, and more visible to everyone, comments you add to entries in your collection are now visible when you do a csv export. CSV export limits are now much higher (before it was capped at 10000 records, now its 100000). The second, SQL connections are now terminated much sooner. This won't mean anything to most people but if you noticed the site randomly going down, this was because the connections were being exhausted and not cleaned up. This should lead to greater uptime for the site overall.
