[06:39:50] I wonder if the pull from github is going to go smoothly. I haven't checked out from git in a year. [08:50:52] git merge upstream/master [08:50:55] fatal: refusing to merge unrelated histories [08:51:37] They didn't keep history?! [08:55:11] git merge upstream/old-master [08:55:14] Already up-to-date. [09:04:43] [drbean@sac GF]$ git branch --set-upstream-to=remotes/upstream/old-master masterBranch master set up to track remote branch old-master from upstream. [09:04:46] [drbean@sac GF]$ git branch -m master old-master [10:09:38] But what about my github fork of GF at origin? [10:10:48] Wait! I don't have any significant hacks in my fork of GF, I realize. [10:10:58] i forked the GF master about a year ago too and i remember doing a merge upstream without difficulty [10:14:45] The commits I don't want to lose are in my fork of gf-contrib. [10:15:12] freeside: And your merges from upstream continue to go smoothly? [10:26:08] I'm just going to abandon my GF fork and just clone from GrammaticalFramework/GF. [10:28:02] yeah [10:28:29] i just did some cleanup and ended up doing a push -F to my origin [10:28:38] now my master is fully in sync with the upstream. [10:28:55] i will maintain a parallel branch for my edits, i'll stay out of master. [11:01:41] gf REPL usability tip: echo "historyDuplicates: IgnoreConsecutive" >> ~/.haskeline [11:14:36] Is there an option for remebering the search term of the previous command? haskeline forgetting the search term is something I find annoying with the gf shell. [11:17:34] if you mean a history specialized to C-r, i don't think so [11:21:44] r [11:21:46] ahem [11:31:37] so, back to the gerund situation: it turns out that gerunds are defined in the translator, but not in the RGL. [11:31:54] 20170819-11:31:06 mengwong@venice2:~/non-db-src/GF% grep GerundNP **/*.gf [11:32:00] lib/src/translator/Extensions.gf: GerundNP : VP -> NP ; -- publishing the document (by nature definite) [11:32:11] lib/src/translator/ExtensionsSwe.gf: GerundNP vp = { -- infinitive: att dricka öl, att vara glad [11:34:10] well, i should have said, gerunds VP -> N* are defined in the translator [11:34:31] ParseEngAbs.gf defines GerundN : V -> N and GerundAP : V -> AP [11:43:33] I think I saw that too, but for some reason I wrote my own oper. https://github.com/drbean/branson/commit/a6e6ebe1a8752e460260992110ccb24d9a228a21 [11:49:16] nice. [12:43:42] *** Joins: koom (~sirdancea@109.107.211.244) [13:14:04] *** Joins: venicecat (6d6e1376@gateway/web/freenode/ip.109.110.19.118) [16:23:09] *** Quits: venicecat (6d6e1376@gateway/web/freenode/ip.109.110.19.118) (Ping timeout: 260 seconds) [18:57:41] *** Joins: venicecat (6de5c968@gateway/web/freenode/ip.109.229.201.104) [20:35:52] progress. [20:39:48] everything is working! yay! [20:40:31] sighhh what does this error message means? [20:40:34] illegal record type field a [20:41:31] the only place where there's an a mentioned is here [20:41:33] Object = { s : Case => Str ; a = Agreement } ; [20:41:44] maybe you need a colon instead of an equals? [20:46:25] oooh thanks i think you might be right [20:46:50] i'm now getting a different error message that says constant agreement is not found [20:47:18] i'll go define that in param.. but strangely i thought that's not required given i'm importing paradigmsEng [20:50:10] i see Agr defined in ResEng, but i don't see Agreement defined anywhere... [20:51:11] right! [20:51:30] As i suspect, if i define agreement in param, it conflicts with paradigmsEng [20:51:42] ParadigmsEng opens ResEng [20:52:09] i am looking at line 40 of http://www.grammaticalframework.org/lib/doc/browse/#english/ResEng.gf [21:27:05] *** Quits: koom (~sirdancea@109.107.211.244) (Ping timeout: 240 seconds) [21:45:09] *** Quits: venicecat (6de5c968@gateway/web/freenode/ip.109.229.201.104) (Ping timeout: 260 seconds)