Popping into this thread for the first time to say that sounds very reasonable. I've tried gamesieve on and off but wishlist import and hide lists would be game-changers for me.

Excellent work on the site!

I want to take this opportunity to say thank you, most sincerely! Not so much for the fix - though that, too, of course :) - but mostly for communicating about fixes, and acknowledging when something is a bug, pretty much everywhere on the forum. I've been noticing that a lot more lately, and it's really appreciated that you take the time for it.

Heh, so GOG ended its sale a day early this morning, and then reinstated it, but as I detected the ending, all of the reinstated sale prices count as "more of the same". That basically invalidates my "Improved" discount quality filter, and affects both the "frequency of sales" filter and the secondary sorting by how long ago the previous sale was.

I don't know what's the better course of action here. I could - with some effort - remove the pricepoints from this morning, so the filters are restored to their intended meaning - benefiting visitors who want to use them for the ~20 hours the sale still lasts - but I can also pretend to be an impartial observer and solely stick to recording reality, warts and all? => chose the latter for now, not for any ideological reasons, but simply because I have better things to do with my time than manually cleaning up this particular mess. Have added the case to my todo list for possible automation if it crops up again in the future.

I say leave it as it is. Work done for something like this seems like way too much effort for too little gain.

I finally have a complete and fully functional core loop for importing wishlists (and an unfinished stub for library information) and showing this on the site, and that's kinda really cool. I had some worries about memory overhead and performance, but managed to squash them, and now it's performing beautifully again.

Still loads to do (all of the UI, plus the whole account setup, and a huge number of tiny loose ends), so absolutely no promises on when it'll be available publicly, but it's starting to feel real.

Nice :)
And there was a well-known company out there that used to say the release date for their games was "when it's done", right? :)

Good show!

Stats from the Autumn sale:
4279 games (78% of the total non-free available) discounted, of which 2962 (69%) at all-time-low or better. 189 improved prices.

taking upgraded editions, expansions, goodies and bundles into account:
7939 discounts (77% of the total non-free available), of which 5633 (71%) at all-time-low of better. 384 improved prices.

Numbers will fluctuate a bit once I retrieve the latest releases from the API, and the last of last week's sales end.

Damn, that makes it a great sale indeed.

Yeah, the number of games with improved prices is quite high! => ending sales from last week dropped those particular numbers to 158/333 - *sigh* Still, more than decent.

(Note that number of discounted games in total is very similar to previous big sales, with ~70% of all games/products on sale; I just figured that due to the addition of free modded games, it'd be more useful to report percentage of non-free games, which is where the 77%/78% comes from.)

I really like the way I group expansions e.a. together with the main game on GameSieve. But URGH, this is causing soooo much pain with the wishlists.

Spent effectively the entire day today just figuring out how to make the query - which I was convinced I had already figured out - actually work properly and return the intended games, rather than a collection which looks superficially plausible, but is all wrong.

Oh well, onward...!

Good work on that site wow!
I thought it was an official made by GOG site, not a user made one :o

Yep, I'm aware. :/ (Though the 100% off tends to be only 99% off in Switzerland, where it then costs CHF 0.01) This is a known inconsistency I'm carefully ignoring as not worth my time right now to handle properly. I'm basically treating any discount which is making something free as equivalent to a giveaway, and thus not meaningful to track for determining improvements over regular sale prices. Also, my calculation for how much worse any other discount would be, would always be infinity% (or y'know, crash the page due to division by zero).

It's annoying that this decision now has the unwarranted effect of putting that particular goodie at the very top of the non-grouped page. As far as I know, no other store item regularly gets a 100% discount, so much as it pains me, I'll continue to not handle it any better for the foreseeable future.

When I add full price history visualization (tentatively planned for after lists are all done), what's actually up with an item like that will be a lot more visible.

Thanks. And heh, I was just thinking the other day that I could maybe remove the "Unofficial" bit from the page title one of these days, but sounds like I'd better keep it there. :)