<gsuberland>
whitequark[cis]: there seems to be an issue with the dead link detection in the github actions where it sees line range (#Lx-y) anchors on links as dead links. presumably because it's handled by GH's JS rather than static.
<whitequark[cis]>
gsuberland: please look at conf.py, there's an exclusion for that particular reason
<whitequark[cis]>
I mean, you can add an exclusion for line ranges there
<gsuberland>
ah nice, I'll take a look, thanks
<gsuberland>
whitequark[cis]: d'you want that as a separate PR, or should I just include the commit in my existing PR?
<gsuberland>
(the config change, that is)
<whitequark[cis]>
gsuberland: same commit, before the docs modification
<whitequark[cis]>
thanks!
<gsuberland>
sorry, not quite sure what you mean by same commit? also just realised including it in my PR wouldn't work 'cos then the fix for the checks would be in the PR with checks failing
<gsuberland>
I don't do a lot of stuff with GH actions so maybe I'm misunderstanding how it works there
<whitequark[cis]>
errrr
<whitequark[cis]>
* gsuberland: same PR, commit before the docs modification
<whitequark[cis]>
fixed
<whitequark[cis]>
gsuberland: how so?
<gsuberland>
ok this is embarassing but I did not actually know you can re-order commits lol
<whitequark[cis]>
make a backup of the branch first
<gsuberland>
ohhhh ok so basically just yeet the branch contents and redo the things
<gsuberland>
gotcha
<gsuberland>
can you tell I basically never use git collaboratively? :P
<tpw_rules>
(you can get the branch back with the reflog)
<tpw_rules>
rebasing is so awesome though
<whitequark[cis]>
<gsuberland> "ohhhh ok so basically just..." <- i *guess*
<gsuberland>
whitequark[cis]: is that at my description of it or are there more steps?
<whitequark[cis]>
you can just do that if you don't want to figure out how rebase works
<gsuberland>
oh, right, I was referring to rebase *as* that, so I guess I misunderstood how it works
<gsuberland>
so yeah I'll figure out rebase ^_^
<whitequark[cis]>
rebase lets you reorder commits
<gsuberland>
ahhh
<whitequark[cis]>
ok I see where the confusion came from :)
<gsuberland>
:D
<ewenmcneill[m]>
In particular "git rebase -i" will do an *interactive* rebase, which will drop you into an editor that lets you reorder the commit lines.
<ewenmcneill[m]>
But I'd echo the "make a back up" suggestion, as it's possible to confuse things if you've not done it before.
<gsuberland>
now I am glad I changed my git editor away from vi
<gsuberland>
and yeah I'll back it up
<gsuberland>
oh right I just realised why I was confused, I was mentally aliasing rebase as a kind of reset / forced checkout, but obviously it's not that
<gsuberland>
so yes this makes a lot more sense
<ewenmcneill[m]>
git rebase is about 5 things in a trenchcoat.
<ewenmcneill[m]>
Depending on the arguments you give it.
<gsuberland>
like so many things in git :D
<Darius>
if only they cared half as much about the UI/UX as the backend performance
<omnitechnomancer>
aside from being another way to fast forward a branch I believe all the other things rebase does are some flavour of putting some set of commits ontop of another starting point
<whitequark[cis]>
hence, rebase
<omnitechnomancer>
indeed
notgull has joined #glasgow
cr1901_ has joined #glasgow
cr1901 has quit [Read error: Connection reset by peer]
ar-jan has joined #glasgow
ar-jan has quit [Remote host closed the connection]
ar-jan has joined #glasgow
bvernoux has joined #glasgow
Eli2 has joined #glasgow
Eli2_ has quit [Ping timeout: 240 seconds]
Lord_Nightmare has quit [Ping timeout: 255 seconds]
yuriks has quit [Ping timeout: 256 seconds]
yuriks has joined #glasgow
notgull has quit [Ping timeout: 268 seconds]
redstarcomrade has quit [Read error: Connection reset by peer]
FFY00 has quit [Read error: Connection reset by peer]
notgull has joined #glasgow
FFY00 has joined #glasgow
Chips4MakersakaS has joined #glasgow
<Chips4MakersakaS>
<ewenmcneill[m]> "But I'd echo the "make a back up..." <- What I do most of the time is tag the current branch with a branch named 'old'. After the branch is rebased and reordered I then do a `git diff old`. I've grown this habit as I saw that conflict resolution during reordering often introduced unwanted code changes.
<vegard_e[m]>
I love interactive rebase so much, it's probably my favorite git feature
<galibert[m]>
and if you forgot to tag you still have the commit ref in the reflog
<omnitechnomancer>
I enjoy --autosquash
<vegard_e[m]>
ooh, that's new to me
<omnitechnomancer>
If you make a commit with --fixup pointing at an earlier commit then --autosquash on the interactive rebase will move that commit as a fixup after the referenced commit
<vegard_e[m]>
yeah, looks perfect, gonna start using that
<omnitechnomancer>
It takes some of hte busywork out of targeting small fixups to commits
notgull has quit [Ping timeout: 268 seconds]
notgull has joined #glasgow
alen6060[m] has joined #glasgow
<alen6060[m]>
Finaly mu Glasgow. TNX esden (@_discord_269693955338141697:catircservices.org) and team
<whitequark[cis]>
do you see how writes with zero latency have a latency of 1?
<gsuberland>
it's an engineering zero, where 0~=1
<gsuberland>
just like pi=3
<gsuberland>
hyperram support is cool to see though. can see that being extremely useful for buffering state history in all sorts of applications
<esden[m]>
Yeah, I will be manufacturing the RAM-Paks for anyone to get, when I dig myself out of this hole 😛
<gsuberland>
^_^
<esden[m]>
I might assemble some more prototypes with the materials I have here so that more people can play with it. If someone is eager to collaborate with Catherine on support for it let me know.
<Wanda[cis]>
I could use one
<whitequark[cis]>
^ seconding
<whitequark[cis]>
also, nanographs should get one, for microscope crime reasons
<whitequark[cis]>
the collab with nanographs actually has been very useful for development of Amaranth itself, because that is the first time I apply streams on a large scale and I'm learning quite a bit of stuff I'll need to put into the docs
<whitequark[cis]>
me & Wanda had to rewrite the HyperRAM PHY like four times before it started, like, working
<whitequark[cis]>
the protocol is pretty batshit
<Attie[m]>
🙈
<esden[m]>
Ok I will see how many materials I have and I will try to put some together. /me pulls out the RAM-Pak project bin onto the EE bench. 😄 Hopefully some time next week. Then I can send some care packages out. (I will likely hand deliver to nanographs the next time I drive up to PDX)
<esden[cis]>
Wanda: I will ping you for your address when I have the goods. :D
<esden[cis]>
Catherine: re protocol, that is what I heard... I used tnt's implementation, and I heard him grumble about it.
dne has quit [Read error: Connection reset by peer]
dne has joined #glasgow
<rcombs>
so, glasgow acquired; if I'm going to build a retro-game TAS playback engine as Baby's First Applet, what category should it go in? interface? control? a new category for this kind of thing?
<whitequark[cis]>
no idea
<whitequark[cis]>
you don't have to categorize it right away, build it first :D
<rcombs>
the basic concept is "you have a buffer of input frames, either sent in over USB at runtime or compiled in at build-time; the console will poll for input over some sort of Weird Serial protocol; you need to respond with a frame of input very quickly, and it's very important that your input buffer never starves"
<whitequark[cis]>
it's basically a reverse logic analyzer, yes
<rcombs>
heh
<SnoopJ>
👀
<jn>
a logic synthesizer
<whitequark[cis]>
i... i .... I Guess
<tpw_rules>
rcombs: hi, what console?
<rcombs>
starting with nintendo 64, since it's what I'm most familiar with
<SnoopJ>
rhymes with your flipper zero build I'd assume?
<SnoopJ>
(which was very cool btw)
<tpw_rules>
ah. i had ideas to do one with nes and snes but kinda hadn't bothered yet since glasgows are still so rare
<rcombs>
SnoopJ: one of these days I need to get that to work reliably
<tpw_rules>
are out of tree applets still not recommended
<whitequark[cis]>
there is no stable API
<SnoopJ>
rcombs, I may have misunderstood how reliable it was from your post before last SGDQ. It would be very cool to see the same thing done with glasgow though
<whitequark[cis]>
there will not be stable API until our staffing will get like 100x the level it has now
<whitequark[cis]>
(the current level is "an hour or two a month, when i find time")
<rcombs>
SnoopJ: it _works_, and I used it for some stuff at SGDQ, but it occasionally crashes for mysterious reasons that I haven't tracked down yet
<tpw_rules>
ok that's fair
<galibert[m]>
you use the n64 scanning protocol to count the frames?
<galibert[m]>
making that reliable must be.. complicated
<rcombs>
galibert[m]: not sure what you mean by scanning protocol?
<duskwuff[m]>
* tbh, both genesis controller types
bvernoux has quit [Read error: Connection reset by peer]
<omnitechnomancer>
The master system does not have any output to the controller for the standard pad (I believe the Japanese region consoles (Mk III and Master System) may even lack the ability to do output on any of the controller pins at all)