| --- Log opened Tue Mar 23 00:00:19 2021 |
| 00:29 | <@Reiv> | Help me with some basic math? |
| 00:30 | <@Reiv> | If I have something worth $400, and it is now $500, it's gone up in value by 25%. If $500 were to drop to $400, that's a 20% drop. |
| 00:30 | <@Reiv> | I cannot remember the actual formulas to work out the latter beyond the intuitive math >_> |
| 00:31 | < ErikMesoy> | change value divided by starting value |
| 00:31 | < ErikMesoy> | 100/500 = 0.2 = 20% |
| 00:33 | <@Reiv> | right, thank you |
| 00:59 | <&[R]> | Have a system that won't power on, mobo is getting power for RGB lights, but when I close the power-on circuit nothing happens, no fan spin, power supply fan doesn't ramp up, just... nothing. |
| 00:59 | <&[R]> | Any ideas what to do? |
| 01:08 | <~Vornicus> | F |
| 01:15 | < Mahal> | replace PSU, replace motherboard |
| 01:25 | <@Reiv> | Replace PSU. If nothing, salvage hard drives and consider state of modern case technology vs current for building the new machine. >_> |
| 02:24 | | abudhabi [abudhabi@Nightstar-75a9fn.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds] |
| 02:25 | | abudhabi [abudhabi@Nightstar-ivpm6u.adsl.tpnet.pl] has joined #code |
| 02:34 | | Degi [Degi@Nightstar-454h01.pool.telefonica.de] has quit [Operation timed out] |
| 02:37 | | Degi [Degi@Nightstar-h8g749.pool.telefonica.de] has joined #code |
| 04:16 | <@celticminstrel> | I think Spotlight is broken… :/ |
| 04:16 | | celticminstrel [celticminst@Nightstar-rh8g14.dsl.bell.ca] has quit [Client exited] |
| 04:27 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has quit [[NS] Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] |
| 04:28 | | VirusJTG [VirusJTG@Nightstar-42s.jso.104.208.IP] has quit [Connection closed] |
| 04:29 | | VirusJTG [VirusJTG@Nightstar-42s.jso.104.208.IP] has joined #code |
| 04:29 | | mode/#code [+ao VirusJTG VirusJTG] by ChanServ |
| 05:32 | | celticminstrel [celticminst@Nightstar-rh8g14.dsl.bell.ca] has joined #code |
| 05:32 | | mode/#code [+o celticminstrel] by ChanServ |
| 05:32 | <@celticminstrel> | I think Spotlight is broken… :/ |
| 05:36 | <@celticminstrel> | It uses 100% CPU and several GB of memory until the system runs out of swap space, then, just before the system is forced to suspend other programs it backs off (or maybe the system terminates it, I’m not sure) and then it all starts again. |
| 05:36 | | himi [sjjf@Nightstar-1drtbs.anu.edu.au] has quit [Ping timeout: 121 seconds] |
| 05:38 | <@celticminstrel> | Maybe it’ll eventually quiet down, I dunno… |
| 05:39 | <@celticminstrel> | I can’t imagine it being good for the SSD if it doesn’t tho… |
| 05:48 | <&McMartin> | Frustrating progress on my ST nonsense. https://bumbershootsoft.wordpress.com/2021/03/23/line-a-flood-fills/ |
| 06:03 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has joined #code |
| 06:03 | | mode/#code [+o Reiv] by ChanServ |
| 06:33 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has quit [[NS] Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] |
| 07:06 | <@celticminstrel> | Oh? It actually did finally calm down… |
| 07:06 | <@celticminstrel> | Still seems broken for it to do this in the first place tho… |
| 08:30 | <@sshine> | in PHP it makes sense to write '$foo[$bar] ?? null' |
| 08:31 | <@sshine> | because the null-coalescing operator also catches array index error exceptions :-D |
| 08:45 | < catalyst> | things in php make sense?! |
| 08:54 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has joined #code |
| 08:54 | | mode/#code [+o himi] by ChanServ |
| 09:12 | <@sshine> | catalyst, I should say that that piece of code actually does something that isn't equivalent to $foo[$bar] |
| 09:14 | | Kindamoody is now known as Kindamoody|afk |
| 09:35 | <@sshine> | what do you tell someone who wants to name a class IntegrationConnectionTypeToDataConnectionImplementationMapping? |
| 09:35 | <@sshine> | I said "this is way too long!" and he said "OK, how about IntegrationConnectionTypeToDataConnectionClassNameMapping?" |
| 09:35 | <@sshine> | which is... shorter! |
| 09:36 | <@sshine> | MaybeI'llSpeakLikeThisToProveAPointFactoryConnectionMapper |
| 09:36 | <~Vornicus> | i mean it does do what it says on the tin |
| 09:36 | <~Vornicus> | probably |
| 09:36 | <~Vornicus> | if you're bitching about the name being long |
| 09:37 | <@TheWatcher> | That some pure, distilled Java API there. |
| 09:38 | <@TheWatcher> | All you need now is a singleton factory to create IntegrationConnectionTypeToDataConnectionImplementationMapping so it can be called IntegrationConnectionTypeToDataConnectionImplementationMappingFactorySingleton |
| 09:38 | <@TheWatcher> | Possilby as an interface that needs implementation. |
| 09:40 | <@sshine> | it certainly is more descriptive than 'Blerb'. |
| 12:15 | <@gnolam> | https://twitter.com/j_opdenakker/status/1374075366625705984 |
| 12:19 | < catalyst> | things that don't need to be classes: everything |
| 12:20 | | * catalyst runs away to Rust |
| 12:52 | | catalyst [catalyst@Nightstar-ejd4sd.cable.virginm.net] has quit [Ping timeout: 121 seconds] |
| 12:53 | | catalyst [catalyst@Nightstar-761cgj.dab.02.net] has joined #code |
| 12:57 | | Alek [Alek@Nightstar-kaiguq.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 13:00 | <@celticminstrel> | Meanwhile, UnrealHeaderTool complains if a class name is more than 32 characters long. |
| 13:01 | | Alek [Alek@Nightstar-kaiguq.il.comcast.net] has joined #code |
| 13:01 | <&[R]> | Like when the bank complains your password is more than 8 characters long |
| 13:01 | <@celticminstrel> | Not quite the same… |
| 13:19 | <~Vornicus> | Big talk for something that's already 16 characters long |
| 13:23 | <@celticminstrel> | Heh |
| 13:26 | | catalyst [catalyst@Nightstar-761cgj.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 13:27 | | catalyst [catalyst@Nightstar-ejd4sd.cable.virginm.net] has joined #code |
| 14:29 | | Kindamoody|afk is now known as Kindamoody |
| 14:43 | | * Vornicus grumps at complicated unit tests that end up giving "true != false" as the result |
| 14:44 | < catalyst> | handy |
| 14:44 | < catalyst> | that's really letting you know the details that is |
| 14:46 | <@TheWatcher> | I mean, it's accurate. Just about as useful as a concrete parachute. |
| --- Log closed Tue Mar 23 15:36:22 2021 |
| --- Log opened Tue Mar 23 15:36:28 2021 |
| 15:36 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 15:36 | | Irssi: #code: Total of 27 nicks [17 ops, 0 halfops, 0 voices, 10 normal] |
| 15:36 | | mode/#code [+o TheWatcher] by ChanServ |
| 15:36 | | Irssi: Join to #code was synced in 16 secs |
| 16:06 | <@sshine> | TheWatcher, if (complicated expression) { assert (true != false); }? |
| 16:07 | <@sshine> | catalyst, haha yeah, totally not a fan of classes. I think that when using classes as first-class modules, that can be fine. most programming languages don't yet have excellent module systems. |
| 16:08 | <~Vornicus> | in this particular case it's "the actual test case is a huge array, and if any of the entries isn't the same, return false" |
| 16:10 | <&[R]> | What languages actually have good module systems? |
| 16:10 | <@sshine> | [R], 1ML? ;) https://people.mpi-sws.org/~rossberg/1ml/ |
| 16:10 | <~Vornicus> | ∅ |
| 16:11 | <&ToxicFrog> | Here we have quite a lot of work put into test libraries to handle stuff like that, so it would be something like: |
| 16:11 | <@sshine> | it's a PoC rework of Standard ML, haha. |
| 16:11 | <&ToxicFrog> | ASSERT_THAT(test_array, EqualsContainer(expected_array)) |
| 16:11 | <&ToxicFrog> | and the error would be something like "contents of test_array do not match contents of expected_array: test_array has the following [missing/extra] elements: ..." |
| 16:11 | <&ToxicFrog> | But this is possible because we have an entire team that just writes test infrastructure for everyone else to use |
| 16:11 | <@sshine> | Vornicus, I recall climbing this hill a couple of times. it somehow never ends up as nicely as when using a modern test library that does colorized diffs and cutting to the part that differs if the structure is big. |
| 16:12 | <~Vornicus> | in my particular case it'd be nice if it gave ... ... uh ... how do I put this. |
| 16:14 | <@TheWatcher> | A million dollars?~ |
| 16:14 | <~Vornicus> | the test inputs are a mesh, and the function calculates the area of a thing on the mesh; the case is the area of all such things on the mesh. ...which would be *much more helpful* if the test went "this entry is wrong, here is the data this entry should be working from" where that data is a much much smaller portion of the mesh |
| 16:15 | <@sshine> | [R], but if we disregard purely researchy languages and only consider something used in production somewhere, OCaml has the best module system that I know. Jane Street re-did the standard library using higher-order modules: https://ocaml.janestreet.com/ocaml-core/v0.12/doc/base/index.html -- e.g. you have a Set module that can take another module as parameter that contains the element type and its ordering |
| 16:15 | <@sshine> | function. |
| 16:15 | <~Vornicus> | so I can 1. see what the right answer is and 2. sit down with Excel and see if I can calculate the right answer and 3. sit down with the debugger and see if I am seeing the data I need it to |
| 16:16 | <@sshine> | Vornicus, right. |
| 16:17 | <@sshine> | Vornicus, so I meant to display the right and the wrong answer, but use ANSI colors to highlight the difference, like in GitHub diffs. |
| 16:18 | <~Vornicus> | yeah that would help a *little* but in this case it's "every entry is wrong because you did not math correctly" which also isn't helpful at all |
| 16:18 | <@sshine> | oh, right :) |
| 16:25 | <@sshine> | [R], https://dev.realworldocaml.org/maps-and-hashtables.html#scrollNav-1-3 -- Map.of_alist_exn (module String) ["foo",1; "bar",2; "snoo",0];; disregarding the syntax and naming in OCaml, here is a Map module that exposes a type Map.t that takes as parameter the 'key and 'value, but also a third type tying this kind of map to the module that is responsible for comparing the 'key. so you have a generic |
| 16:25 | <@sshine> | dictionary type that won't let you accidentally mix a case-sensitive and a case-insensitive dictionary. |
| 16:26 | <@sshine> | I think Python's modules are also very neat because imported modules get mixed in with the values. but maybe they're a little unpredictable, too. |
| 16:50 | | Emmy [Emmy@Nightstar-l49opt.fixed.kpn.net] has joined #code |
| 17:20 | <@sshine> | question: at an old job we had some gettext functions besides __() that provided a number of variations like __f('You are number %d.', n), __x('Agreement Number:', 'Derived from Danish "Aftalenummer"'). can anyone tell me if those were added locally on top of Gettext's __(), or are there standard extensions to __()? I can't seem to find any references to functions beyond __(). |
| 17:23 | < Yossarian> | Anyone her work with graphviz a bunch? How do you get digraph to draw left to right instead of top-to-bottom? I've tried rankdir=LR but maybe like, my diagram needs sub diagrams? |
| 17:25 | < Yossarian> | I'm trying to use graphviz to draw basic design diagrams and this particular digram would be drawn better left to right. |
| 17:29 | < Yossarian> | seems like `neato` supports more fine-tune parameters |
| 17:30 | <@sshine> | Yossarian, like this? https://tinyurl.com/hhcavpkk |
| 17:31 | <@sshine> | and yeah, could be that sub-graphs are the way to go if you want some part horizontal and then fit that part into a bigger graph that isn't only horizontal. |
| 17:31 | <@sshine> | (also, I haven't touched graphviz in 10+ years, so I may be blatantly misunderstanding what you're doing.) |
| 17:31 | < Yossarian> | yeah like that |
| 17:32 | < Yossarian> | https://stackoverflow.com/questions/1554635/graphviz-how-to-have-a-subgraph-be-left-to-right-when-main-graph-is-top-to-bot seems to have some answers as well |
| 17:32 | < Yossarian> | sshine: and then inserting another graph or element under "start" ? |
| 17:33 | < Yossarian> | oh |
| 17:35 | < Yossarian> | https://tinyurl.com/graphviz2 |
| 17:35 | <@sshine> | ohhh. |
| 17:36 | < Yossarian> | however, the rendering is backwards in this case, you want to tell it to draw the bottom first... |
| 17:36 | <@sshine> | yeah, having sub-graphs with rankdir="LR" inside a graph with rankdir="TB" didn't work. |
| 17:36 | < Yossarian> | cut and paste start2's charting to the top |
| 17:36 | <@sshine> | yeah, loses control over rendering |
| 17:37 | < Yossarian> | the rendering in this case could be described as FIFO |
| 17:37 | < Yossarian> | but I know there is a way to make subgraphs which might let me reverse that? |
| 17:38 | <@sshine> | https://tinyurl.com/35eahsrs |
| 17:38 | < Yossarian> | so the way it is described in the file is not FIFO but rather ah |
| 17:38 | <@sshine> | when I change the order of those sub-graphs in the source, they seem to change order in the rendering. not sure if that's what you were thinking about. |
| 17:39 | < Yossarian> | but G2 is drawn last, still FIFO ordering |
| 17:39 | <@sshine> | do you mean that it shows up in the source code first? |
| 17:40 | < Yossarian> | yeah, now just the minor (very minor point) of making it draw G1 first |
| 17:40 | < Yossarian> | this way it is easier to read and go back to and modify |
| 17:40 | < Yossarian> | the way I figure graphviz is that if I want to make simple design diagrams then I don't have to rely on some GUI tool to do such a thing |
| 17:42 | < Yossarian> | but dot's raw output if you just used 'dot graph.dot' will draw whatever comes first, there must be a way to order around that for ease of managing the file itself. Like I said, it is a minor point but still |
| 17:43 | < Yossarian> | I believe there is a way to order 'em via rank |
| 17:45 | < Yossarian> | like |
| 17:46 | <@sshine> | I seem to recall that 'dot' gives you less control of the output, but nicer input syntax, and that Tikz provides you with very exact rendering options, but only afaik for LaTeX, and complicated input syntax. |
| 17:46 | <@sshine> | I never dug into how to make 'dot' work beyond the basics |
| 17:47 | < Yossarian> | https://tinyurl.com/rankingGraph |
| 17:47 | <@sshine> | ahhh, so by specifying the vertices in the right order before connecting them does it? |
| 17:48 | <@sshine> | neat :) |
| 17:48 | <@sshine> | but does it also affect rendering order? |
| 17:48 | < Yossarian> | I still don't know what that means though. |
| 17:48 | < Yossarian> | mmmm, not sure |
| 17:48 | <@sshine> | what 'p a' means? |
| 17:48 | <@sshine> | it just means "let there be vertices 'p' and 'a'" without specifying edges. |
| 17:49 | < Yossarian> | yeah, looks like if you change Format to plain you get a good idea without the verbosity of 'raw output' |
| 17:49 | < Yossarian> | I tried ranking the subgraphs but it draws G1 and G2 vertices instead |
| 17:51 | <@sshine> | and yeah, the raw SVG output, at least, renders the first one of those first, even though it visually gets rendered last. |
| 17:52 | < Yossarian> | I was wondering if there was a nicer way to issue rank of what gets drawn first |
| 17:52 | < Yossarian> | graphviz's documentation is actually quite... hefty. Long. |
| 17:52 | < Yossarian> | Because you don't just have dot as a renderer, you have neato and all the other things. |
| 17:54 | < Yossarian> | But my understanding too is graphviz is used to render visual outputs of lots of things, not just making LateX documents but like... for data stuff people use graphviz apparently |
| 17:55 | < Yossarian> | https://stackoverflow.com/questions/46743365/graphviz-subgraphs-order |
| 17:56 | < Yossarian> | { 1 2 } -- b -- { 10 11 }; |
| 17:58 | < Yossarian> | interesting but still in that stackoverflow, the last entry does have groupings near the top |
| 17:59 | < Yossarian> | if I interpret the node numbers the way I have for setting p a, then from going left to right, the right most ranked thing (written in this fashion) will be drawn first |
| 18:00 | < Yossarian> | and then if you try to attach say in our example we've been tinyurl-ing back and forth, attach at the end b -> q |
| 18:01 | < Yossarian> | the ordering of `p a` actually does the reverse of what we want, remove `p a` and it will draw a-> b -> c subgraph first |
| 18:02 | < Yossarian> | or if you wanted to attach the middle node of each subgraph together, p subgraph gets offset |
| 18:03 | < Yossarian> | if you select a "head" element though, that doesn't happen |
| 18:04 | < Yossarian> | hmmm, aren't there other types of lines to use to draw besides -> ? |
| 18:06 | < Yossarian> | but sshine you got the basic LR format going for me, so thank you. This online tool is useful. I might have to write a script for my editor that upon any changes or saving of the file that it executes renderer var $file -T x11 |
| 18:19 | < Yossarian> | sshine: thanks for your preliminary work on getting rankdir=LR; working, now I need to read the graphviz documentation on ranking subgraph drawing order as well as some other things. |
| 18:20 | < Yossarian> | If one were to understand graphviz very well and use ctags or exuberant ctags and use a script to render the relationship between functions, classes, etc in the ctag file to .dot file, that'd be pretty wicked |
| 18:21 | < Yossarian> | HTML is an output, I wonder if you could have element give you a command upon clicking on it and bring you to the file in which it is contained but in terms of HTML that seems unlikely |
| 18:21 | < Yossarian> | the kind of design document stuff I want to draw is very simple stuff so I don't have to draw it by hand on paper and take a picture or scan it. |
| 18:22 | < Yossarian> | So, again, thanks. If anyone else here has more knowledge of graphviz and such please lemme know. Seems like a pretty useful tool to get to know. |
| 18:25 | <@sshine> | Yossarian, I think you can get other types of lines by annotating them, yes |
| 18:28 | < Yossarian> | but the FIFO drawing of subgraphs can reverse itself if you link elements between the two graphs |
| 19:03 | < Yossarian> | https://graphs.grevian.org/reference |
| 19:26 | < Yossarian> | sshine: if you were curious, here is an example of how {rank= } works: https://tinyurl.com/digraphRank |
| 19:27 | < Yossarian> | almost does draw LR but what I can't see is how to modify lines to be straight |
| 19:28 | < Yossarian> | I also wonder if rank can be applied to names of subgraphs |
| 19:35 | < Yossarian> | https://tinyurl.com/RankLRcomedout so if you do the ranks if you want the two ranked graphs connected in some way, it actually does follow a LR ranking, uncomment LR and you will get default TB ordering. |
| 19:45 | < Yossarian> | Other than that, I wonder how to make edges straight. I suppose I'll have to read the documentation. |
| 20:32 | < Yossarian> | https://gitlab.com/graphviz/graphviz/-/issues/1592 did I fix this guy's issue? https://tinyurl.com/graphvizGitlabFix |
| 20:49 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 23:00 | | himi [sjjf@Nightstar-1drtbs.anu.edu.au] has joined #code |
| 23:00 | | mode/#code [+o himi] by ChanServ |
| 23:12 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: WeeChat 2.9] |
| 23:13 | | Kindamoody is now known as Kindamoody[zZz] |
| 23:21 | | Emmy [Emmy@Nightstar-l49opt.fixed.kpn.net] has quit [Ping timeout: 121 seconds] |
| 23:21 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 23:31 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 23:31 | | mode/#code [+ao ToxicFrog ToxicFrog] by ChanServ |
| --- Log closed Wed Mar 24 00:00:20 2021 |