Touhou-Project.com
patreon
Filler up
🕑 Added 2018-04-15 20:12:42 +0000 UTC I had hoped to have the latest additions I’ve been working on all sorted out by now but a combination of lack of time, project scope and testing has meant that things are coming along a little slower than I’d like. So I don’t really want to give any sort of ETA just yet or really talk about that in detail (in case I end up cutting some of the stuff I’m working on) but we are more or less heading towards completion. It’ll probably just take 3-4 entire days’ worth of work time for me to get it all sorted out. Since I rarely have entire days where I can just sit and code, I’ve got to do it instead in smaller chunks of time across a lot more days.But enough about that! I’ll use the opportunity to talk about some of the general goals I have for THP in the near and medium term. You can divide them up into roughly a few (non-exhaustive) categories:
1) Changes in the User Interface/Experience
Most of this translates into reworking the various scripts that run on any of the site’s pages. You may not realize it, but a lot of things run on the user-side. Things like expanding images, hiding threads, random banners, managing watched threads, checking if the client is a mobile device and serving the mobile CSS etc. All that code needs an overhaul due to outdated libraries that they depend on and, in some cases, outright inefficient code. There’s all sorts of modern features that other chans have that would be also nice to have. I won’t get too specific over these since I’m still seeing what would be realistic to have but I’m sure you can imagine some examples.
There is some interdependence on other bits of code so it’s not as straightforward as I’d like to change but a nice side effect is that we’d be closer to adhering to CSP by getting rid of a lot of the inline scripts that exist on every page.
There’s also few tweaks that need to be made to CSS as well in some parts to fix lack of consistency in several edge cases.
2) Overhauling archived threads
I think that the storylist needs to be redone in presentation and how it hooks into the rest of the site. Easier said than done, however. The actual mechanism for archiving threads on the backend needs to be updated and automated (currently a script is manually run to strip it of certain things and add other bits and bobs). This has to then be followed by a way to either automatically add those threads to a list of either existing stories or create new entries and what can’t be automated should be handled by the community or writers themselves (descriptions, tags, etc). So you need an interface and backend that’s idiot-proof in order to keep things swifter and less dependent on action by myself or Kapow.
And, well, obviously the storylist also needs to be changed in order to accommodate all that. So that page and how it communicates with everything else needs to be worked on. I’ve simplified the objectives but trust me when I say that there’s a lot of caveats and design decisions that need to be sorted out along the way. A lot of testing will have to be done as well before something like this is live.
3) Things generally invisible to the user
This is probably the messiest to categorize. In no particular order here’s a few things that need doing (and why):
- Overhauling the front page’s code. A lot of its calls are inefficient even accounting for caching on the PHP server’s part. There’s no need to fetch certain data from the database every time the page is reloaded.
- Overhauling the html of thread and board pages. It’s honestly just ghastly to try to mess around with elements which is sometimes needed when you want to, say, have scripts play around with presentation in some cases. The Kusaba X people weren’t very forward-thinking in this regard (in-line scripts and styles abound as well).
- Fixing bugs especially on the administration side of things. You couldn’t know but currently one of the functions when the boards are rebuilt is broken under certain circumstances. I broke that when doing another thing a few months ago, just haven’t gotten around it since it’s not high priority. There’s a lot of superfluous or small things that need similar fixes or overhauls.
- Reworking or optimizing a lot of PHP. There’s even ancient comments in some places talking about the need to refactor code into functions for both clarity of reading and speed. The query to the database when it builds threads/boards is the biggest example of this. It’s several hundred lines of pure spaghetti code.
- More swapping out of old libraries for modern equivalents. The biggest thing that should be changed is the template engine but I probably won’t do that, instead I’d rather do a lot of that when building new board software from scratch. So just the smaller things for the short and medium-term.
So yes, those are the three general areas that I plan to work on in the coming weeks and months. I’ve already done things in all three areas in the past and talked about them in previous posts. I often have to read up a lot of documentation and learn new things to do the changes I’m talking about which is why it hasn’t been as swift as some of you have hoped. Work on individual parts like the mystery box recently updated threads will continue as it’s very much part of those categories as well. Sometimes it’s not worth starting work on one item before I get a few other things done first so this is also a matter of setting realistic goals and priorities according to available time, resources and skill.
I’d like to conclude this filler episode by telling you that you’re always welcome to suggest improvements you’d like to see for the site. Just post on Patreon, the site itself or tell me on an instant messaging client. Whatever is most convenient. I can’t promise I can do most requests but knowing what the community wants is certainly valuable knowledge and helps me plan ahead.
Until next time, take it easy.
