code logs -> 2019 -> Fri, 14 Jun 2019< code.20190613.log - code.20190615.log >
--- Log opened Fri Jun 14 00:00:46 2019
00:55 celmin|away is now known as celticminstrel
01:35 bluefoxx [fuzzylombax@Nightstar-gmbj85.vs.shawcable.net] has quit [[NS] Quit: nice to know you]
02:23 Degi [Degi@Nightstar-vdkf2b.dyn.telefonica.de] has quit [Connection reset by peer]
03:56 Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code
03:56 mode/#code [+qo Vornicus Vornicus] by ChanServ
04:34 macdjord [macdjord@Nightstar-grpbnp.mc.videotron.ca] has joined #code
04:34 mode/#code [+o macdjord] by ChanServ
04:37 macdjord|slep [macdjord@Nightstar-grpbnp.mc.videotron.ca] has quit [Ping timeout: 121 seconds]
04:38 Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed]
04:39 Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code
04:39 mode/#code [+qo Vornicus Vornicus] by ChanServ
04:49 celticminstrel [celticminst@Nightstar-6an2qt.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
07:35 Kindamoody[zZz] is now known as Kindamoody
07:42 Kindamoody is now known as Kindamoody|out
11:14
<~Vornicus>
dangit luaunit why does this traceback not go back far enough
11:14
<~Vornicus>
or rather why does it skip steps
11:16
<~Vornicus>
I know that vorntable.push is getting a nil table, and I know technically it's somewhere in the putPointOnEdge call because the next item in the call stack is addPoint and the line where putPointOnEdge is called, but there are 18 calls to vorntable.push in there, you're going to have to give me a little more to work with.
11:26
<~Vornicus>
and again, this time with vorntable.load which I use 20 times.
11:26
<~Vornicus>
Sigh.
11:42
<~Vornicus>
OKay. REason it's not working is because for some reason my hierarchy face table isn't being completely filled.
11:48
<~Vornicus>
--because I claimed to be using indexes and ended up inserting references instead in one spot..
11:49
<~Vornicus>
Okay, now it's failing in a different spot, let's see
12:29
<~Vornicus>
Aha, typo in the construction protocol, that should have been index 7, day saved.
12:32
<~Vornicus>
GREEN
12:33
<~Vornicus>
fuck yeah
12:34
<@TheWatcher>
And there was much rejoicing!
12:35
<~Vornicus>
All right! TIme to think of a test for the next bit
12:36 * TheWatcher ponders this variable
12:37
<@TheWatcher>
$jaeorbs
12:37
<@TheWatcher>
Well, that's rather old-school.
12:37
<~Vornicus>
did they take 'em?
12:39 * TheWatcher checks
12:40
<@TheWatcher>
Well, it's got a $jaeorbs -> run() down there, but not sure if it's skookem
12:52 Kindamoody|out is now known as Kindamoody
12:55
<&Reiver>
okay, I want to dabble in a teeny bit of programming. To rename files in a directory.
12:55
<&Reiver>
My default would be to bang my head against Python again, get baffled at regexes again, and give up on the whole thing in misery and self-loathing and then do everything by hand.
12:56
<&Reiver>
Do I hear any competing suggestions for methodology in which I actually learn something, and maybe even succeed?
12:56
<~Vornicus>
Define your precise problem here
12:56
<@TheWatcher>
^ what the unsqueakable one says
12:58
<~Vornicus>
Regex is pretty opaque; one of the best ways to learn is to have someone who is very knowledgeable in them explain some relatively simple ones they made themselves.
12:58
<&Reiver>
File.Name.2018.1234p.Text.Foo.Bar.mp4 (or .avi or simply a folder with nothing at all) => File Name (2018).mp4 (or equivalent)
12:59
<&Reiver>
Yeah, I guess I don't really learn that way these days. My years of professional devving mean I mostly learn from taking a working example and then hacking it to a new, similar, example next time. Regex has tended to be resistant to this technique, and its whole paradgim is alien enough I just stare at it blankly and uninstall the entire development stack du jour. (I appreciate this is not a healthy response.)
12:59
<~Vornicus>
Can File.Name.2018 be Mononym.2018, or Really.Long.File.Name.2018
13:00
<@TheWatcher>
But regex are awesome, you can even used them to parse HTML!*
13:00
<@TheWatcher>
(* you can't and shouldn't)
13:01
<~Vornicus>
If so I can name this tune with str.split, str.isdigit, and str.join
13:01
<&Reiver>
Vornicus: Yes
13:02
<@TheWatcher>
Are you guaranteed to have the year after the name?
13:02
<&Reiver>
"Strip all the full stops into spaces up to the first set of numerals. Put those numerals in parenthesis. Then eliminate everything else afterwards, out to the final file type identifier."
13:03
<&Reiver>
Yes. I was planning to use it as the identifier for where it was.
13:03
<&Reiver>
Or rather, where the file name ends and the cruft begins.
13:03
<@TheWatcher>
hokay, yeah
13:04
<~Vornicus>
no need for regex at all, then.
13:04 * TheWatcher sulk~
13:05
<~Vornicus>
oh, you'll also need os.isdir
13:05 celticminstrel [celticminst@Nightstar-6an2qt.dsl.bell.ca] has joined #code
13:05 mode/#code [+o celticminstrel] by ChanServ
13:05
<~Vornicus>
(to determine whether there's an extension or just cruft at the end)
13:10
<~Vornicus>
you can set up a twitch stream and we can yell at your code live, or something
13:16
<@TheWatcher>
XD
13:16
<@TheWatcher>
Not sure that'd be helpful
13:19 * Reiver hides under a rock.
13:19
<&Reiver>
Still, thank you for clarifying that I can skip the regex.
13:19 * TheWatcher vaguely ponders a one-lined regex to do the job, decides that it has some interesting wrinkles and comes to the conclusion - for the purposes of a day where he's supposed to be workin on a gitlab job runner - that it's an ecumenical matter.
13:21
<&Reiver>
what
13:25
<@TheWatcher>
It's from Father Ted, a sitcom. http://preachingandpondering.blogspot.com/2017/05/that-would-be-ecumenical-matter.html but in short 'They land on the phrase "That would be an ecumenical matter", which Ted observes is just enough to imply you are interested or are able to follow some arcane theological discussion'
13:28
<@TheWatcher>
'Just respond with that phrase, he has learned, and others will consider you profound and wise. (And hopefully leave you alone!)'
13:42
<@TheWatcher>
It's actually easy to do except for the \. between the name parts $foo =~ s/^(?<name>(\w+\.)+)(?<date>\d{4})\..*?(?<ext>\.\w+)$/$+{name} \($+{date}\)$+{ext}/; but those damned \.s hrrrrm
13:43
<~Vornicus>
regex is super crazy concise. That's basically all that's good about it: sometimes equivalent code for your regex will be long and annoying so a regex will get it done more neatly and, for those who are familiar with regex, more obviously
13:44
<~Vornicus>
but then when you do something like this and, quite frankly, I can't read that
13:44
<~Vornicus>
and I consider myself not terrible at regex
13:54
<@TheWatcher>
(?<name>(\w+\.)+) = capture at least one group of alphanumerics ending in a \. and store the lot in the 'name' capture. (?<date>\d{4}) = capture four digits, store in the 'date' capture. \..*? = match the next . and do a non-greedy match for any character, (?<ext>\.\w+) = capture the last . and any alphanumerics after, store in the 'ext' capture. And then $+{name}
13:54
<@TheWatcher>
\($+{date}\)$+{ext} pulls the values out of the named capture groups and use them to replace the string.
13:55
<@TheWatcher>
I wouldn't usually bother with named captures, but normal $1 $2 etc refs won't work properly. There's probably a more elegant method to do it that I'm brainfarting on
13:56 Kindamoody is now known as Kindamoody|afk
13:58 celticminstrel is now known as celmin|away
14:02 * Reiver screams and dies.
14:13 Degi [Degi@Nightstar-ksir82.dyn.telefonica.de] has joined #code
14:37
<~Vornicus>
Ok. Test is: place three points in a triangle in a particular way I know is safe - hi, lo, right. This will break up the mesh into four triangles none of which need adjusting, so I should get "this doesn't need adjusting" off of the adjustment checker. Then, add one close enough to the outside of this triangle that it lies inside the circumcircle. that will force a rework of the triangles. That then tests the edge flop.
14:57 Degi [Degi@Nightstar-ksir82.dyn.telefonica.de] has quit [Connection closed]
15:27 gnolam [quassel@Nightstar-0oc.n80.224.185.IP] has joined #code
15:27 mode/#code [+o gnolam] by ChanServ
16:50 gnolam [quassel@Nightstar-0oc.n80.224.185.IP] has quit [[NS] Quit: Interruption]
16:59 gnolam [quassel@Nightstar-0oc.n80.224.185.IP] has joined #code
16:59 mode/#code [+o gnolam] by ChanServ
16:59 Vorntastic [uid293981@Nightstar-6br85t.irccloud.com] has joined #code
16:59 mode/#code [+qo Vorntastic Vorntastic] by ChanServ
17:34 catalyst [Jessikat@Nightstar-5dv16h.cable.virginm.net] has joined #code
17:40
<@gnolam>
Argh.
17:41
<@gnolam>
How did Pascal of all languages ever get introduced as a teaching language?
17:42 gnolam [quassel@Nightstar-0oc.n80.224.185.IP] has quit [[NS] Quit: SJ.]
17:44 Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code
17:53 gnolam [quassel@Nightstar-0m2ors.mobileonline.telia.com] has joined #code
17:53 mode/#code [+o gnolam] by ChanServ
18:43 gnolam [quassel@Nightstar-0m2ors.mobileonline.telia.com] has quit [[NS] Quit: Blergh]
19:16 ErikMesoy [Bruker@Nightstar-hhsrrl.customer.cdi.no] has quit [Connection closed]
19:16 ErikMesoy [Bruker@Nightstar-hhsrrl.customer.cdi.no] has joined #code
20:13 gnolam [lenin@Nightstar-hfrbpd.cust.bahnhof.se] has joined #code
20:13 mode/#code [+o gnolam] by ChanServ
21:10 Kindamoody|afk is now known as Kindamoody
23:03 Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection closed]
23:14
<&McMartin>
gnolam: A lack of viable alternatives at the time; C was not a viable alternative until standardization was de facto complete around the cusp of the 90s.
23:15
<&McMartin>
Modula-2 would not have been a meaningful improvement, and if you wanted an ALGOL that actually produced compilable and runnable code on your university systems, that's literally what Pascal was.
23:16
<&McMartin>
Ease of use and ability to brush off the Dominant Bullshit Of The Age That Got In The Way Of Actually Teaching What The Class Was About, ironically enough, were things it actually delivered
23:16
<&McMartin>
Ultimately, Java did the same 25 years later, which is what ultimately displaced it.
23:17
<@TheWatcher>
The fact that using pascal successfully requires donning a smoking jacket and obtaining a pipe is a secondary concern~
23:17
<&McMartin>
That's a feature, tbh
23:17 * McMartin grew up being taught Pascal in his Serious Programming For Srs classes, was expected to be ahead of the curve due to prior exposure to C, but then teaching switched to C++ in the late 1990s and Java at the turn of the century
23:17
<&McMartin>
And I got to TA the same class twice as a grad student, once in C++ and once in Java
23:17
<&McMartin>
I hate Java with the fire of several hundred suns
23:18
<&McMartin>
But I will say nothing at all unkind about its suitability as a teaching language.
23:18
<&McMartin>
It wasn't built to do that job
23:18
<&McMartin>
But It Did That Job.
23:24
<@gnolam>
Context: for some bizarre reason, Pascal is what Inno Setup uses for its code/scripting bits.
23:25
<&McMartin>
It's old enough that it was probably a reasonable choice at the time, especially if it was a P-Code kind of thing pre-JVM
23:37
<&McMartin>
Delphi also maintains a surprisingly persistent userbase
--- Log closed Sat Jun 15 00:00:47 2019
code logs -> 2019 -> Fri, 14 Jun 2019< code.20190613.log - code.20190615.log >

[ Latest log file ]