[04:48:40] Z26779 has stopped working, for over a month it seems. I need to use it to display date values from Wikidata. Any ideas about what changed or how it should be fixed? [04:57:14] It's used in the composition chain of Z35327. (re @u99of9: Z26779 has stopped working, for over a month it seems. I need to use it to display date values from Wikidata. Any ideas about wh...) [09:11:03] I think I’ve fixed it for now, but the fix is not resilient. [09:11:04] The incoming type now begins like this: [09:11:05] {"Z1K1":{"Z1K1":"Z9","Z9K1":"Z4"},"Z4K1":{"Z1K1":{"Z1K1":"Z9","Z9K1":"Z4"},"Z4K1":{"Z9K1":"Z27970"},"Z4K2":… [09:11:07] I don’t know how or why that (presumably) changed ( David ), but it means that there currently needs to be an extra .Z4K1 in the access path to the type reference. This will need to be removed if we revert to the previous structure. (re @u99of9: Z26779 has stopped working, for over a month it seems. I need to use it to display date values from Wikidata. Any ideas about [09:11:07] wh...) [10:19:16] I’ve fixed Z29483 as well but there still seems to be a problem somewhere 🤷‍♂️ (re @Al: I think I’ve fixed it for now, but the fix is not resilient. [10:19:17] The incoming type now begins like this: [10:19:19] {"Z1K1":{"Z1K1":"Z9","Z9K...) [17:01:03] The current blocker seems to be a failure to resolve an embedded argument reference. One example is Z26885. Although this works, Z28505 does not. Replacing the second argument reference with a hardcoded reference works in test, however. The same difficulty seems to occur with Z26936, and Z35343 depends on that, in most cases. Apart from filing the ticket, the only way [17:01:03] forward at [17:01:04] the moment seems to be to compose/code each precision separately (which would help the performance if not our sanity). (re @u99of9: Z26779 has stopped working, for over a month it seems. I need to use it to display date values from Wikidata. Any ideas about wh...) [21:10:34] Thanks for figuring this out. It sounds like a ticket is necessary. (re @Al: The current blocker seems to be a failure to resolve an embedded argument reference. One example is Z26885. Although this works,...) [21:56:09] Yeah, I think it’s two separate tickets. The two issues seem unrelated. I’ll file them tomorrow. (re @u99of9: Thanks for figuring this out. It sounds like a ticket is necessary.)