[00:04:11] I've done this (re-listed the default as a valid language option in all configurations where it the default is supposed to be used) for English. Others should free to add your languages. I also added a selector implementation Z53026 which will skip the last option if it matches the default, so these long default lists don't have to degrade performance. It certainly [00:04:11] improves the usefulness of https://abstract-data.toolforge.org/ : https://tools-static.wmflabs.org/bridgebot/ec8a822c/file_85215.jpg [00:08:58] For those who want to add their language to the configs that may well want to use the default, here are the ones I added for English: Z39632 Z45394 Z29317 Z46615 Z30790 Z32279 Z42229 Z35676 Z40064 Z25737 Z40803 Z40216 (re @u99of9: I've done this (re-listed the default as a valid language option in all configurations where it the default is supposed to be us...) [00:21:15] Which bring the missing lexemes back into focus as the primary "blockers". This is an interesting list. How did @davidsantamariame 's tool accept all the proper nouns (Sydney, Vegemite, etc) but not accept these (which are also drawn from labels in Z26092 when it does successfully render Australia in English? Is it also requiring that the alternative (non-preferred) [00:21:15] implementation Z37124 (which does use lexemes) also be satiated? Perhaps it should only follow the preferred path for each function? : https://tools-static.wmflabs.org/bridgebot/074552ac/file_85216.jpg [05:52:44] Why do I get a different result at Z53162 (both blanks) than I do in this run (["", "from"]) [05:52:44] https://www.wikifunctions.org/view/en/Z44686?call=%7B%22Z1K1%22%3A%22Z7%22%2C%22Z7K1%22%3A%22Z44686%22%2C%22Z44686K1%22%3A%5B%22Z1%22%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q7873%22%7D%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q9612%22%7D%5D%2C%22Z44686K2%22%3A%7B%22Z [05:52:45] 1K1%22%3A%22Z7%22%2C%22Z7K1%22%3A%22Z873%22%2C%22Z873K1%22%3A%22Z39586%22%2C%22Z873K2%22%3A%5B%22Z1%22%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q140399652%22%7D%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q140399648%22%7D%5D%7D%7D [05:56:31] I've also tested that the map function should send the right pair: https://www.wikifunctions.org/view/en/Z873?call=%7B%22Z1K1%22%3A%22Z7%22%2C%22Z7K1%22%3A%22Z873%22%2C%22Z873K1%22%3A%22Z39586%22%2C%22Z873K2%22%3A%5B%22Z1%22%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q140399652%22%7D%2C%7B%22Z1K1%22%3A%22Z6091%22%2C%22Z6091K1%22%3A%22Q140399648%22%7D%5D%7D [11:18:57] I can’t say why they’re different, but the blank Z11 seems to be a result of a False result from Z13381 in Z39587. A complication is that Z22193 returns the last item in the event of no match, which happens to be the correct result in this particular case. I tweaked the implementation (Z39587) to return a blank Z11 by default, but the test call only succeeds when it is [11:18:57] made i [11:18:58] n test mode. That suggests Python’s == is sometimes failing in switch, which suggests an occasional representation issue (just a guess). (re @u99of9: Why do I get a different result at Z53162 (both blanks) than I do in this run (["", "from"]) https://www.wikifunctions.org/view/...) [11:21:23] Thanks for investigating. Sometimes edit mode works better when the external test call has been cached as failing. So maybe it will fix if we wait. (re @Al: I can’t say why they’re different, but the blank Z11 seems to be a result of a False result from Z13381 in Z39587. A complicatio...) [11:22:17] e.g. this one has now cleared up (re @u99of9: Cached validation calls are still problematic. Z51033 says it failed all tests (validation cacehd from when they did fail), but ...) [11:23:40] Yes, but I got a fresh evaluation when I changed the implementation, and it’s still different in the two modes 🤷‍♂️ (re @u99of9: Thanks for investigating. Sometimes edit mode works better when the external test call has been cached as failing. So maybe it w...) [14:44:06] Hi all, the recording of the last Corner is now available on Commons: https://commons.wikimedia.org/wiki/File:Abstract_Wikipedia_Volunteer_Corner_2026-11.webm [14:44:26] thank you Sannita! [18:17:37] There's a recently-added message: `wikilambda-abstract-preview-provenance-render-date` ("Fragment rendered") [18:17:51] It was as part of https://phabricator.wikimedia.org/T441583 [18:17:57] Is this deployed already? [23:44:49] Thanks. Sorry to dominate the discussion so much! (re @Sannita: Hi all, the recording of the last Corner is now available on Commons: https://commons.wikimedia.org/wiki/File:Abstract_Wikipedia...) [23:45:05] Not at all! (re @u99of9: Thanks. Sorry to dominate the discussion so much!)