#hypothes.is

/

      • GitHub185 joined the channel
      • NOTICE: [h] aron deleted sentry at f3a0678: https://github.com/hypothesis/h/commit/f3a0678
      • GitHub185 has left the channel
      • nickstenn
        aron_: alright, no worries
      • i'll push it up to stage and see how it goes
      • yay, random angular issues that don't throw any errors
      • Treora___ has quit
      • aron_
        yay!
      • GitHub44 joined the channel
      • NOTICE: [h] aron pushed 5 new commits to syncbridge: https://github.com/hypothesis/h/compare/1f481740d54a...c537f88be709
      • NOTICE: h/syncbridge 22212f9 Aron Carroll: Swap AnnotationUISync methods for toggling annotations...
      • NOTICE: h/syncbridge 40384bf Aron Carroll: Add additional helper methods to annotatorUI service...
      • NOTICE: h/syncbridge 87ceb4a Aron Carroll: Always modify the annotatorUI state...
      • GitHub44 has left the channel
      • nickstenn you resolve the issue?
      • nickstenn
        i'm not entirely sure
      • because I can't replicate it
      • every so often when I click "edit" on an annotation the edit form is mangled
      • it's ok though
      • because I have another issue that I can replicate
      • the expand/collapse plus/minus icon doesn't work
      • in the sidebar only, apparently
      • ah
      • no, i'm not sure what's going wrong there
      • there's a whole host of issues, both behavioural and display
      • such as the fact that "n replies ... reply, share, edit, delete" is highlighted on hover over the right-hand part of the annotation
      • but not when you hover over "n replies"
      • ahhhh
      • this is the good old digest bullshit again
      • the plus/minus button does work if you then mouseover something else
      • aron_
        the n-replies hover state is an issue with the n replies being outside of the annotation
      • it's in the thread, which is the element that wraps ti
      • but then negatively positioned to look like it's inside it obvs
      • nickstenn
        yeah, i know, I've looked at that one befor
      • *before
      • aron_
        what do you mean by mangled by the way?
      • shepazu_away is now known as shepazu
      • nickstenn
        aron_: i'll see if I can screencap it
      • aron_
        Does that have any math in it?
      • nickstenn
        nope
      • here's the rendered state at that point: https://gist.github.com/nickstenning/59fc7525d3...
      • i can't give you instructions on how to replicate that other than try editing an annotation a few times...
      • aron_
        Ah it's an edit
      • Something is whack with the permissions on that branch, I cant edit annotations I've created
      • Oh, I can but I have to trigger a digest :(
      • dwhly
        aron_: just got your reply
      • ujvari
        aron_, nickstenning: Are you trying to edit an annotation in a dokku instance?
      • aron_
        ujvari yes
      • nickstenn
        ujvari: it's stage.hypothes.is
      • ujvari
        hmm
      • aron_
        dwhly: sorry it wasn't very interesting
      • nickstenn
        aron_: one notable feature of the above screenshot is that the annotation in question is collapsed
      • so all of this might be related
      • aron_
        nickstenn oh haha
      • ujvari
        on dokku we have a known issue that if you create an annotation with a dokku instance you cannot edit it using an another dokku instance
      • shepazu
        how's the de-forjing going?
      • er...
      • de-dorking
      • lol
      • merging
      • nickstenn
        i mean, it wasn't collapsed when i clicked "edit," otherwise there wouldn't have been an "edit" to click
      • dwhly
        Hey shep
      • nickstenn
        shepazu: good thanks -- we now don't have our own fork of Annotator any more
      • which is great :)
      • aron_
        nickstenn right, and I don't think you can collapse the editor
      • ujvari
        Oh I see
      • shepazu
        yay
      • nickstenn
        shepazu: we do still have some monkeypatches and plenty still to do, but we're in a much better position than we were for upstream contributions
      • aron_: ahhh i think i can replicate this, and I think it is all related
      • click just a pixel or two to the left of the "edit" icon, and then click on the edit icon
      • and I would guess it's that the first click triggers a collapse event which doesn't happen because of the digest bug
      • which in turn wouldn't happen if the "n replies" link weren't positioned in such a stupid way
      • if you could let me know if you see the same thing I'll take a look at fixing it after lunch
      • aron_
        k
      • csillag1 has quit
      • kfogel joined the channel
      • GitHub163 joined the channel
      • NOTICE: [h] aron pushed 3 new commits to syncbridge: https://github.com/hypothesis/h/compare/c537f88be709...86f714d8d312
      • NOTICE: h/syncbridge 6bb4a09 Aron Carroll: Rename CrossFrameBridge to Bridge
      • NOTICE: h/syncbridge 428ca6f Aron Carroll: Rename CrossFrameDiscovery to Discovery
      • NOTICE: h/syncbridge 86f714d Aron Carroll: Rename Plugin.Bridge to Plugin.CrossFrame...
      • GitHub163 has left the channel
      • nickstenn: yup I can replicate that
      • nickstenn
        aron_: what was the test username you created for checking fabian's problem
      • i want to compare his and your user accounts
      • aron_
        aron.test
      • nickstenn
        cheers
      • aron_: well that's weird
      • I can't see what's happening to fabian either, but I can replicate it with his account
      • GitHub50 joined the channel
      • NOTICE: [h] gergely-ujvari deleted analyse_uri_field at f137b74: https://github.com/hypothesis/h/commit/f137b74
      • GitHub50 has left the channel
      • aron_
        multiple reset tokens?
      • GitHub52 joined the channel
      • NOTICE: [h] gergely-ujvari deleted 1796-use-search-api-for-search at f43c009: https://github.com/hypothesis/h/commit/f43c009
      • GitHub52 has left the channel
      • Symon|away is now known as Symon
      • GitHub153 joined the channel
      • NOTICE: [h] gergely-ujvari created 1537-analyse-uri-field (+1 new commit): https://github.com/hypothesis/h/commit/3b1425821b08
      • NOTICE: h/1537-analyse-uri-field 3b14258 Gergely Ujvari: Change the search_analyzer for uri...
      • GitHub153 has left the channel
      • travis-ci joined the channel
      • travis-ci
        hypothesis/h#4643 (1537-analyse-uri-field - 3b14258 : Gergely Ujvari): The build passed.
      • travis-ci has left the channel
      • shepazu has quit
      • shepazu joined the channel
      • nickstenn
        pushing a debug branch to staging FYI
      • Mitar joined the channel
      • csillag1 joined the channel
      • kfogel has quit
      • shepazu is now known as shepazu_away
      • shepazu_away is now known as shepazu
      • ujvari
        nickstenn: About the username
      • you should check this part of the code: https://github.com/hypothesis/h/blob/master/h/a...
      • tilgovi joined the channel
      • tilgovi is back in CA and very happy about it
      • nickstenn
        ujvari: yep! thanks.
      • tilgovi: you around?
      • tilgovi
        yeah
      • nickstenn: pong
      • nickstenn
        so, we apparently normalise usernames with dots in them on lookup
      • tilgovi
        yes
      • nickstenn
        but we don't (or at least didn't) do that when they were created
      • which is a bit of a bummer
      • to say the least
      • tilgovi
        I don't fully understand what you mean, but I intuitively understand and was already guessing it was related to the dots
      • I haven't looked at the code to understand why it's a problem
      • nickstenn
        it's a problem if we have two users in our database called 'foobar' and 'foo.bar'
      • tilgovi
        I would have thought it didn't matter which way it was in the DB because I thought the lookup was agnostic to dots.
      • that _would_ be a problem indeed!
      • nickstenn
        which is the case
      • tilgovi
        Ugh
      • nickstenn
        which is why i was going to ask
      • what was the motivation for attempting to normalise dots in the first place?
      • tilgovi
        I don't know how that could be possible given that the registration should do a lookup to determine if it's in use already, and the lookup ignores the dots
      • Unless it's an old registration
      • nickstenn
        well I suspect once upon a time it didn't
      • tilgovi
        that was done via the PHP on the homepage
      • nickstenn
        yes, it looks like it could well be
      • tilgovi
        well, damn :)
      • nickstenn
        but we might well have plenty of those
      • tilgovi
        we might well have no conflicts other than this person
      • we should check
      • in which case we can just fix it by hand and move on
      • I'm assuming this person is the same person using both usernames?
      • as far as motivation, I don't know. I did it on a whim.
      • nickstenn
        tilgovi: i'll have a look