| --- Log opened Sat Jun 10 00:00:52 2017 |
| 00:22 | | Turaiel[Offline] is now known as Turaiel |
| 00:25 | <&McMartin> | Oh dear |
| 00:25 | <&McMartin> | The work slack just discovered the cult of the party parrot |
| 00:26 | | * McMartin reacted with :parrot: |
| 00:54 | | * Vornicus returns to shenzhen i/o, tries to remember how it works, also is apparently dealing with meat |
| 00:55 | <&McMartin> | MEAT PRINTER |
| 00:59 | < Vornicus> | never quite enough instructions to do it right |
| 01:03 | < ToxicFrog> | aka :parrotdrugs: |
| 01:06 | | * ToxicFrog pokes minicom with a stick |
| 01:09 | < Vornicus> | Three short. :( |
| 01:11 | < ToxicFrog> | Boooo it looks like minicom doesn't support local line buffering |
| 01:12 | | * celmin|sleep wonders if there's a logical answer to "min element of an empty list". |
| 01:12 | | celmin|sleep is now known as celticminstrel |
| 01:15 | <&McMartin> | I do not believe so. |
| 01:16 | <&McMartin> | If it's typed, and there's a maximum or positive infinity value, then that will work. |
| 01:16 | <&McMartin> | (By analogy with the way the sum of an empty list is 0 and the product is 1; it's the initial argument to the implied fold operation) |
| 01:17 | <&McMartin> | I guess more generally, you would want it to be the value V such that min(x, V) = x for all valid x. |
| 01:18 | <@celticminstrel> | I guess that means it's impossible if the type is string. |
| 01:18 | <&McMartin> | Yes. |
| 01:18 | <&McMartin> | But "" will work for max! |
| 01:18 | <@celticminstrel> | True. |
| 01:19 | <&McMartin> | Is the string known to be ASCII or UTF-8? |
| 01:19 | <&McMartin> | If so, the string "\xff\xff |
| 01:19 | <&McMartin> | " |
| 01:19 | <&McMartin> | is lexicographically later than all validly encoded sttrings. |
| 01:20 | <&McMartin> | That won't work for any encoding that can freely use every octet, though. |
| 01:20 | <&McMartin> | And it's sneaky hacky bullshit anyway. |
| 01:21 | <@celticminstrel> | o.o |
| 01:21 | < ToxicFrog> | Argh. minicom can't do line buffering. Neither can kermit. Arduino IDE can, but its serial terminal is garbage. |
| 01:22 | < ToxicFrog> | Cutecom can, but it randomly inserts newlines into the output. |
| 01:22 | <&McMartin> | (ASCII only permits byte values of 0x7f or less, so 0xff alone is enough to be later than all valid ASCII strings.) |
| 01:23 | <&McMartin> | (UTF-8 encodes a multibyte code point with one byte larger than 0xBF and then one or more bytes in the 0x80-0xBF range, so two FF bytes in a row is illegal.) |
| 01:24 | < Vornicus> | hooray it worked |
| 01:26 | <&McMartin> | If you can restrict it to "only includes valid unicode code points", then code point FFFF is a noncharacter and thus two FF bytes is also larger than any UTF-16 string in either encoding order. |
| 01:27 | <&McMartin> | For UCS-4 you'd need four FF bytes to handle all endianness/comparison cases. |
| 01:32 | | * Vornicus now works on the keycode lock. |
| 02:24 | | * Vornicus runs out of instructions *again* |
| 02:49 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has quit [Operation timed out] |
| 03:08 | < Vornicus> | oop, it *does* test that edge case :( |
| 03:18 | < Vornicus> | and my reader has one instruction of slack space. Not that I have a plan. |
| 03:59 | | * ToxicFrog bonks head against desk |
| 03:59 | < ToxicFrog> | Just spent twenty minutes trying to figure out why the compiler was emitting OP_CALLWORD,NULL rather than OP_PUSHLITERAL,2 |
| 04:00 | < ToxicFrog> | (intptr_t)OP_CALLWORD is 2. |
| 04:00 | < ToxicFrog> | The *actual* problem is that the OP_PUSHLITERAL opcode never got emitted. |
| 04:00 | <&McMartin> | man, valid zeropage addresses. |
| 04:00 | < ToxicFrog> | McMartin: not in this case! |
| 04:00 | < ToxicFrog> | That's why it was blowing up. |
| 04:01 | <&McMartin> | Oh, that's an enum, right |
| 04:01 | <&McMartin> | I misunderstood the significance of intptr_t there |
| 04:27 | < ToxicFrog> | So here's a fun one |
| 04:27 | < ToxicFrog> | This is valid: |
| 04:28 | < ToxicFrog> | :two 2 const |
| 04:28 | < ToxicFrog> | :double { two * } defn |
| 04:28 | < ToxicFrog> | This is not valid |
| 04:28 | < ToxicFrog> | :two 2 const :double { two * } defn |
| 04:29 | < ToxicFrog> | ...because the definition of `double` tries to resolve `two` at compile time, and if `:two 2 const` hasn't been executed yet, resolution fails. |
| 04:29 | <&McMartin> | ouch |
| 04:31 | < ToxicFrog> | I noticed this because of a separate bug in the lexer that was causing it to eat newlines under some circumstances, effectively concatenating adjacent lines in the test script. |
| 05:48 | <&McMartin> | himi: It occurs to me that based on your discussion earlier about your issues with the RPi... |
| 05:49 | <&McMartin> | ... that what you're really looking for is an updated version of the Iyonix PC. |
| 05:49 | <&McMartin> | (Which was too underpowered on its own to meet your needs, but was built around Intel's XScale ARM chip and the PCI bus) |
| 05:50 | <@himi> | McMartin: yeah, there are better options for what I use it for, but the big advantage of the RPi is that it's really easy to get hold of |
| 05:51 | <&McMartin> | Well, it sounds like what you want is a cheap ARM system that works like a small desktop |
| 05:51 | <&McMartin> | As opposed to media player/control system where the filesystem is read-occasionally, write-even-more-rarely |
| 05:52 | | * himi nods |
| 05:52 | <@himi> | Or more specifically, a small server |
| 05:52 | <&McMartin> | Ouch, yeah. |
| 05:52 | | * McMartin went with an Intel NUC for that. |
| 05:53 | <&McMartin> | Handed it a celeron |
| 05:53 | <@himi> | The RPi works surprisingly well, but the lack of SATA means that I get fairly shitty disk I/O |
| 05:53 | <&McMartin> | It's the one running IRC right now, actually, and I'm SSHed into it from the Pi~ |
| 06:15 | | Turaiel is now known as Turaiel[Offline] |
| 06:25 | <@himi> | ugh |
| 06:26 | <@himi> | TKinter is not masses of fun to work with |
| 06:33 | <&McMartin> | no |
| 06:36 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [[NS] Quit: Kernel upgrade] |
| 06:42 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 06:42 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 06:42 | <@himi> | Sadly I don't know of any other Python GUI toolkit that's basically universally available without requiring additional packages to be installed |
| 06:43 | <&McMartin> | That's quite correct, alas. |
| 06:43 | <&McMartin> | Also, all GUI programming is full of spiders |
| 06:46 | <@himi> | eh, it's not so much that it's full of spiders, it's that it's full of grind |
| 06:48 | <@ErikMesoy> | I have a range(x-1,-1,-1) in my GUI programming, that feels like a spider |
| 06:48 | <@ErikMesoy> | Is there a less spidery way to traverse elements by index last to first? |
| 06:49 | <@himi> | Hunk after hunk of boilerplate with the barest minimum of differences, and then the same again to actually /do/ anything with it |
| 06:49 | <@himi> | ErikMesoy what language? |
| 06:49 | <@ErikMesoy> | Python. |
| 06:49 | <@ErikMesoy> | I'm using wxPython for my toolkit |
| 06:49 | <@himi> | Well, you can do a list.reverse() first |
| 06:49 | <@himi> | That modifies the list in place, though |
| 06:51 | <@ErikMesoy> | Reversing the list won't help as I can't access the elements by name, only by index. And I'm deleting them in this case, which is why I have to iterate backwards: deleting from a list while iterating forwards through it is a great way to foul up. |
| 06:52 | <@himi> | Then reverse and use pop() to pull elements off the list |
| 06:52 | <@himi> | Or rather, just use pop() to pull the last element off the list |
| 06:54 | < Vornicus> | for item in reversed(list) |
| 06:54 | < Vornicus> | oh, you need to deal with indexes |
| 06:54 | < Vornicus> | ...use filter |
| 06:55 | <@himi> | Oh, you need the index while you're iterating? |
| 06:55 | <@himi> | Hm |
| 06:57 | < Vornicus> | and enumerate |
| 06:58 | <@himi> | I'm a little confused as to what the actual data structure you're operating with looks like - if it's just a list of objects then getting access to the objects via pop() or similar should work, but if it's not a list then what is it? |
| 06:59 | < Vornicus> | new_list = filter(function_that_decides_whether_to_keep_it, enumerate(old_list)); ftdwtki accepts index, item and returns True if you want to keep it |
| 07:05 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:06 | <@ErikMesoy> | http://pastebin.starforge.co.uk/121 |
| 07:08 | <@himi> | Oh |
| 07:10 | < Vornicus> | Oh oh, this isn't even that bad |
| 07:10 | < Vornicus> | wait, it is. blah! |
| 07:10 | <@himi> | for i in reversed(list(range(0, x))): |
| 07:11 | < Vornicus> | Ewk, jsut use the negative thing |
| 07:11 | < Vornicus> | But: I'm surprised there's no way to just say "get rid of all the children" |
| 07:11 | <@himi> | Nothing is any more elegant |
| 07:12 | <@himi> | Yeah - I'd check the docs again, because this is the kind of shit that normally has a sensible built in solution |
| 07:12 | < Vornicus> | or, you know: while(..getChildren()): gameobject.contextualbuttons.Hide(0) |
| 07:13 | <@ErikMesoy> | Ooh! |
| 07:13 | <@ErikMesoy> | Thanks. |
| 07:14 | < Vornicus> | But yeah, look for a thing that gets rid of all children instead, I'd say |
| 07:15 | <@ErikMesoy> | I will do that too |
| 07:21 | <@himi> | I'm a little surprised that the output of getChildren() implement __len__() but not __iter__() |
| 07:21 | <@himi> | eh, who knows what's going on in the minds of the wxWidgets devs |
| 07:25 | <@ErikMesoy> | So what is your boilerplate like? |
| 07:33 | <@himi> | http://pastebin.starforge.co.uk/122 |
| 07:33 | <@himi> | Also, you have to pack things in reverse order, or there's some trick I have yet to find that makes it not be so stupid |
| 07:37 | <@himi> | On the plus side, it looks like almost all the widgets and the like can be set up with temporary variables unless you need to do something with it later, so I could probably wrap a good chunk of this stuff without any pain |
| 07:37 | | celticminstrel is now known as celmin|sleep |
| 11:34 | | Kindamoody [Kindamoody@Nightstar-0lgkcs.tbcn.telia.com] has quit [Ping timeout: 121 seconds] |
| 11:36 | | Kindamoody|autojoin [Kindamoody@Nightstar-80id9c.mobileonline.telia.com] has joined #code |
| 11:36 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 12:46 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 16:38 | <@ErikMesoy> | Okay, that's enough improvised similar names, I need a convention for naming Buttons and associated Methods. Any thoughts on the value of naming the former with foo_B, or the latter foo_M, or both? Prefix or postfix? |
| 17:05 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed] |
| 17:12 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code |
| 17:12 | | mode/#code [+ao VirusJTG VirusJTG] by ChanServ |
| 17:52 | | celmin|sleep is now known as celticminstrel |
| 17:57 | | Turaiel[Offline] is now known as Turaiel |
| 19:35 | <&[R]> | ErikMesoy: they share a namespace? |
| 19:36 | <&[R]> | I'd just prefix the name of the form/page/screen the buttons are on if that were the case. Leave the methods alone. Unless you're using proper prefix-notation throughout then prefix both appropriately. |
| 20:04 | | Jessikat [Jessikat@Nightstar-cnb095.dab.02.net] has joined #code |
| 20:31 | <@ErikMesoy> | Yes, same namespace, since I'm creating them at the same time and place in GUI (and I feel I'm missing an integrated function that would set them both up as a unit). After fiddling I have wound up postfixing the buttons and letting the methods use master name. |
| 21:19 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 21:37 | | Kindamoody|autojoin [Kindamoody@Nightstar-80id9c.mobileonline.telia.com] has quit [Ping timeout: 121 seconds] |
| 21:53 | | * TheWatcher ponders Unity, hrms |
| 21:53 | <@TheWatcher> | As near as I can tell, it doesn't really expect you to do runtime scene generation |
| 21:54 | <@TheWatcher> | I wonder if this is going to end up with me fighting the engine to do what I want. |
| 21:57 | | Kindamoody|autojoin [Kindamoody@Nightstar-0lgkcs.tbcn.telia.com] has joined #code |
| 21:57 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 22:06 | | Kindamoody|autojoin is now known as Kindamoody |
| 22:30 | | Jessikat [Jessikat@Nightstar-cnb095.dab.02.net] has quit [[NS] Quit: Bye] |
| 23:19 | | Kindamoody is now known as Kindamoody[zZz] |
| --- Log closed Sun Jun 11 00:00:53 2017 |