00:00:29 <kmc> oh, well, anything that can't be called outside the compilation unit can be optimized arbitrarily
00:00:50 <kmc> for example you wouldn't need to save caller-save registers around a call to a function that happens to not use those registers
00:00:58 <kmc> (that's assuming the function isn't inlined anyway)
00:04:52 -!- Lord_of_Life_ has joined.
00:07:20 -!- Lord_of_Life has quit (Ping timeout: 272 seconds).
00:07:20 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
00:07:23 <b_jonas> `ehlist http://eheroes.smackjeeves.com/comics/2750792/how-to-defeat-a-hero/
00:07:24 <HackEso> ehlist http://eheroes.smackjeeves.com/comics/2750792/how-to-defeat-a-hero/: b_jonas
00:10:03 <shachaf> kmc: I mean putting things like clobbered registers and maximum stack size in the "type" of a function.
00:10:31 <shachaf> Also do C compilers actually change the calling convention for static non-inline functions?
00:10:54 <kmc> they might
00:11:14 <kmc> in Rust the calling convettion is part of the type
00:11:23 <kmc> you can't pass a fn as an extern "C" fn
00:11:36 <shachaf> That makes sense. But there are a few fixed conventions.
00:13:02 <kmc> how do you mean
00:13:13 <kmc> are you saying you should be able to define your own conventions?
00:16:51 <shachaf> I'm saying you can have a convention per function depending on how many registers it needs and so on.
00:17:20 -!- b_jonas has quit (Quit: leaving).
00:19:54 <shachaf> These seem like things you can do in general. If you have a DAG of functions it's easy, and if you have a cycle I guess you can just go with the conservative option (unlimited stack, standard calling convention).
00:20:27 <esowiki> [[Farm]] https://esolangs.org/w/index.php?diff=60005&oldid=59958 * BMO * (+4094)
00:20:51 -!- Remavas has quit (Read error: Connection reset by peer).
00:21:36 -!- Remavas has joined.
00:21:39 -!- Remavas has changed nick to Remavas[AFK].
00:54:03 -!- copumpkin has joined.
01:26:32 -!- orbitaldecay has joined.
02:22:33 -!- sebbu has quit (Ping timeout: 245 seconds).
02:28:14 -!- sebbu has joined.
02:47:33 <esowiki> [[User:Orby]] M https://esolangs.org/w/index.php?diff=60006&oldid=59939 * Orby * (-30)
03:00:28 -!- Remavas has joined.
03:02:33 -!- Remavas[AFK] has quit (Ping timeout: 245 seconds).
03:32:38 -!- oerjan has joined.
03:47:11 <HackEso> Saneb is too normal to invent things, but he likes to watch SV and play Vorld of Varcraft.
03:51:41 -!- FreeFull has quit.
04:01:40 <oerjan> <b_jonas> shachaf: also mediawiki doesn't allow us to put pluses in the language name <-- . o O ( i'm starting to worry about b_jonas - this and claiming helium in zeppelin's explode... )
04:01:47 -!- Remavas has quit (Ping timeout: 244 seconds).
04:02:03 <oerjan> also about my own gramar and spling
04:03:23 <shachaf> Hmm, there are definitely pluses in language names on the wiki.
04:08:13 <oerjan> that's why i'm worried hth
04:57:12 <shachaf> maybe "that was actually clear from the context" or something along those lines
05:22:24 -!- arseniiv has joined.
05:50:26 <oerjan> huh they've rewritten the dialogue of the third last girl genius comic. seems like they remembered tarvek had already met the castle fragment
05:55:26 <oerjan> oh there's a comment edit below
06:02:33 -!- tromp has quit (Remote host closed the connection).
06:11:03 <oerjan> the second monster statue from the right looks a bit like the one over mechanicsburg
06:17:51 <oerjan> (http://www.girlgeniusonline.com/comic.php?date=20140611 for that one)
06:37:09 <kmc> shachaf: should there be more cat emojis in the Unicode, y/y
06:37:57 <shachaf> kmc: will you adopt a cat and/or a cat emoji
06:41:29 -!- tromp has joined.
06:47:17 -!- tromp has quit (Ping timeout: 255 seconds).
07:09:48 -!- AnotherTest has joined.
07:14:25 -!- AnotherTest has quit (Ping timeout: 250 seconds).
07:46:25 -!- oerjan has quit (Quit: Nite).
08:29:02 -!- tromp has joined.
08:31:01 -!- tromp has quit (Remote host closed the connection).
08:31:17 -!- tromp has joined.
08:40:39 <orin> I won't be adding anymore emoji to my font since windows refuses to display them in my font
08:41:15 <orin> I will hoever, be getting back to adding more characters regularly
08:57:01 <orin> also remember how killing pulseaudio was common solution for ubuntu problems a few years ago?
08:57:22 <orin> well today I ahve a problem with my mouse freezing up on windows 10
08:57:44 <orin> google says the solution is to kill the Realtek Audio Manager
08:58:16 <myname> at uni we have an issue with debian where the cursor sometimes is invisible for a few minuts after login
09:17:49 -!- tromp has quit (Remote host closed the connection).
09:36:48 -!- tromp has joined.
09:37:06 -!- AnotherTest has joined.
09:58:51 -!- orbitaldecay has quit (Ping timeout: 256 seconds).
10:20:59 <orin> good to see that bdf2ttf.c is still working
10:21:14 <orin> http://orenwatson.be/bdf2ttf.htm
11:04:21 -!- xkapastel has joined.
11:06:21 <esowiki> [[Bitch]] https://esolangs.org/w/index.php?diff=60007&oldid=60002 * Helen * (-257) Updated impossible popular problems entry
11:14:31 <esowiki> [[Bitch]] https://esolangs.org/w/index.php?diff=60008&oldid=60007 * Helen * (+1005) Added a section for common algorithms
11:27:28 -!- MDude has quit (Ping timeout: 246 seconds).
11:32:15 -!- MDude has joined.
11:37:20 <orin> ᕫᕬᕭᕮᕯᕰᕱᕲᕳᕴᕵᕶᕷᕸᕹᕺᕻᕼᕽ᷂᷊᷀᷁᷃᷄᷅᷆᷇᷈᷉⨌⨍⨎⨏⨐㌀㌁兩硝硫硬碁碑磨礁礎秀租秩称稚稲稼稽稿穂穏穫突窃窒窟窮窯竜端符筒箇箋箸範篤簿籍籠粋粒粗粘粛粧糧糾紋紛紡索紫累紳紹絞絡継維綱網綻緊緒締緩緯緻縁縄縛縝繁繊繕繭繰罰罵罷羞羨翁翻翼耐耗聘聵肌肖肘肝股肢肩肪肯胆胎胞胴脂脅脇脊脚脱腎
11:38:46 <Taneb> Looks like you're testing a font there
11:39:12 <orin> Taneb: yah these are the new characters in the newest version of my font
11:40:10 <orin> I got back into it yesterday and have been adding characters again
11:47:37 <orin> I refactored my ttf parser
11:48:23 <orin> http://orenwatson.be/ttfinvread.htm
11:49:35 <orin> a and wrote a program to compare the character inventories of two files
11:49:39 <orin> http://orenwatson.be/invcmp.htm
11:51:03 <orin> which allowed me to more easily get a list of which characters exist in a new ttf as opposed to the old one
12:05:53 -!- Lord_of_Life has quit (Ping timeout: 255 seconds).
12:11:54 -!- Lord_of_Life has joined.
12:31:08 <esowiki> [[Assembly language]] https://esolangs.org/w/index.php?diff=60009&oldid=59992 * A * (+458)
12:32:01 <esowiki> [[Assembly language]] https://esolangs.org/w/index.php?diff=60010&oldid=60009 * A * (-18) /* Examples */
12:44:20 <esowiki> [[Bitch]] https://esolangs.org/w/index.php?diff=60011&oldid=60008 * Helen * (+31) /* Continuation on the above by User:Helen */ Corrected myself on my statement about a subtraction algorithm
13:00:10 -!- Meme19 has joined.
13:09:15 -!- Meme19 has quit (Ping timeout: 256 seconds).
13:10:24 <esowiki> [[Assembly language]] https://esolangs.org/w/index.php?diff=60012&oldid=60010 * A * (-63) /* Examples */
13:10:49 <esowiki> [[Assembly language]] https://esolangs.org/w/index.php?diff=60013&oldid=60012 * A * (+29) /* Machine code interpreter */
13:15:23 <esowiki> [[Bitch]] https://esolangs.org/w/index.php?diff=60014&oldid=60011 * A * (-96) Division would be unneccesary, since subtraction is not yet implemented.
13:15:50 <esowiki> [[Bitch]] https://esolangs.org/w/index.php?diff=60015&oldid=60014 * A * (-1) /* Continuation on the above by User:Helen */
13:25:30 -!- danieljabailey has quit (Ping timeout: 250 seconds).
14:37:44 <orin> hmm is there an efficient way to verify that there are no characters that look exaclty the same?
14:40:31 <orin> if I loaded all chars into memry at same time it would be uhh
14:41:35 <Taneb> Sort them graphically in some way?
14:42:55 <orin> the characters are represented as 18x16 bitmaps
14:43:03 <Taneb> There are definitely sorting algorithms designed so you don't load everything at once
14:43:45 <orin> so I could just output those as strigs somehow
14:43:58 <orin> then use gnu sort?
14:44:51 <orin> 18 divides 6 evenly so each pixel row could be 3 base64 digits
14:45:29 <orin> does gnu sort use a mergesort on large files?
14:47:38 <orin> hmm, never mind the data isn't *that* large, only 2 megabytes
14:47:47 <orin> I'll use qsort
14:48:19 <Taneb> > let n = 2 * 2^20 in n * log n
14:48:58 <orin> might take a bit
14:49:42 <myname> you can do integer sorting in n * sqrt(log log n)
14:50:24 <Taneb> This is exactly what something like radix sort is good for
14:53:23 -!- AnotherTest has quit (Ping timeout: 245 seconds).
14:53:28 <orin> ah yeah. If I keep splitting the dataset based on each bit then when I'm at the last one, I'll have verified whether there are any repeats
14:54:18 <lambdabot> 4973232364097866421553822481468208401004561507973477174404639768931594970125...
14:54:34 <Taneb> Well, there's certainly a lot of possible characters...
14:57:36 <orin> yeah but most of those possibilities look like white noise
14:58:15 <orin> and one of the goals of my font is to as much as possible keep characters looking distinct from each other
14:59:47 <orin> so what I'll do is make two buffers
15:00:00 <orin> each containing pointers to bitmaps
15:00:24 <orin> and in a loop split the buffer by one bit
15:00:54 <myname> you could also make a giant hashmap with a reasonable hash function for each character
15:01:24 <orin> hmmm or maybe double buffering is overkill
15:02:01 <orin> how many recursion levels can tpyical C environment handle
15:03:21 <orin> yeah I think I'll do the loop instead even if I have to be clever
15:05:02 <int-e> > length ['\0'..] * 18*2 -- bytes
15:06:24 <int-e> I'm sure orin has fewer glyphs than that. :)
15:06:35 <int-e> So sorting them all to find duplicates shouldn't be a big deal, really.
15:06:54 <orin> Currently I've got 21130
15:08:57 -!- sleepnap has joined.
15:09:06 <int-e> Ah, catching up is hard. "<orin> I'll use qsort"
15:09:52 <Taneb> int-e: it's fun to overthink things
15:10:20 <int-e> . o O ( overthinking, what does that even mean? )
15:10:47 <int-e> what's more problematic is that there will be duplicates
15:11:26 <orin> int-e: that's what I'm trying to find
15:11:41 <HackEso> [U+006F LATIN SMALL LETTER O] [U+043E CYRILLIC SMALL LETTER O] [U+03BF GREEK SMALL LETTER OMICRON]
15:11:42 <orin> if there are duplicates I want to knwo what they are
15:11:56 <orin> int-e: thsoe all look different in my font
15:12:18 <int-e> that seems awkward. :)
15:12:34 <orin> cyrillic о is more squarish and greek ο is shorter
15:13:32 <orin> still different.
15:13:47 <orin> I'm more worried about the math symbols and such
15:15:43 <orin> em space is shown with little letters saing "EMSP"
15:16:25 <orin> although some environments stubbornly refuse to show any glyph for space characters
15:16:44 <orin> ok I don't even
15:16:52 <HackEso> U+26D0 CAR SLIDING \ UTF-8: e2 9b 90 UTF-16BE: 26d0 Decimal: ⛐ \ ⛐ \ Category: So (Symbol, Other) \ Bidi: ON (Other Neutrals)
15:17:05 <orin> I didn't draw that one yet
15:17:37 <orin> http://orenwatson.be/fontdemo.htm
15:18:51 <myname> what about invisible plus vs invisible minus?
15:19:32 <int-e> myname: dym times, rather than minus?
15:21:12 <orin> I need to update my cjkwidth function, some charactrs are being whown the wrong width
15:22:13 <orin> and some of these need to be redrawn becuase they became emoji and thus became wide
15:23:24 <orin> hmm or maybe it's better to leave them... I'll think about it
15:24:16 <myname> there is a 69 emoji? that will be so misused
15:24:58 <orin> myname: it's the Cancer astrological sign lol
15:25:12 <myname> what's the difference :D
15:25:29 <orin> I agree emojis have become a cancer
15:26:50 -!- AnotherTest has joined.
15:27:09 <int-e> myname: there are emoji modifiers as well
15:27:54 <orin> yeah i've yet to find a terminal, or terminal-oriented program, that handles those things correctly or at all really
15:28:27 <int-e> `` unidecode $(unicode 1F9B0)
15:28:28 <HackEso> [U+1F9B0 - No such unicode character name in database]
15:29:06 <int-e> Unicode Character 'EMOJI COMPONENT RED HAIR' (U+1F9B0)
15:29:52 <int-e> I'm secretly waiting for the emoji part of Unicode to become TC.
15:32:54 <orin> my font includes some emoji but browsers and even some terminals simply refuse to display them in my font
15:33:34 <orin> so I'm focusing on non-emojis
15:34:47 <orin> emojis can be handled by some other lunatic, this lunatic is working on pixelating every possible mathematic operator and all the most common kanji
16:24:12 <orin> wow google is pushing ahead with crippling ad blockers
16:24:28 <Taneb> Of course they are, ads are their main source of income
16:25:15 <orin> they gave a press release that superficially appears to back down but actually doesn't
16:30:40 <orin> idiot newspapers are reporting that google won't block ad blockers
16:31:18 <orin> "Another clarification is that the webRequest API is not going to be fully removed as part of Manifest V3. In particular, there are currently no planned changes to the observational capabilities of webRequest (i.e., anything that does not modify the request)."
16:32:00 <orin> ^ meanwhile, in reality, the post confirms that ublock and other ad blockers will no longer be allowed to use the existing API to filter content
16:32:22 <orin> only "observe" it
16:32:36 <int-e> "would have blocked 10/11 requests"
16:33:00 <orin> https://groups.google.com/a/chromium.org/forum/#!topic/chromium-extensions/WcZ42Iqon_M
16:33:12 <orin> here's the full post
16:33:28 <int-e> https://www.theregister.co.uk/2019/02/20/google_nest_secret_microphone/ is nicely written
16:33:41 <int-e> (unrelated, except it's also about Google)
16:34:01 <orin> I'm glad I never switched to chrome in the first place
16:35:04 <int-e> "Dynamic Rule Support: We agree that this is valuable in creating sophisticated content blocking extensions, and will be adding support for declarative rules that can be added or removed at runtime to the declarativeNetRequest API." ... uh, right, that's not what we mean by dynamic blocking.
16:36:12 <orin> hehehehe such weasel. much handwave. wow
16:40:02 -!- Hydra_ has joined.
16:42:46 <orin> I mean, they are aware that people can tell whether ads are being blocked?
16:43:24 <orin> I don't see how this won't lead to a large exodus of adblock users to other browsers
16:43:26 -!- Hydra_ has quit (Client Quit).
16:44:42 <myname> well, as long as they won't sync with your bookmarks and passwords, that may be hard
16:44:57 -!- Erika has joined.
16:45:46 <myname> multiple devices as well
16:45:49 <int-e> (though there the trend seems to be not to have a website but a custom app)
16:45:51 <orin> int-e: I use firefox on mobile
16:45:58 <myname> and google did forbid chromium forks to sync
16:46:00 <orin> I don't use any website-apps
16:46:14 <myname> mobile browsers all suck
16:46:21 <myname> but the duet mode in chrome dev is nice
16:46:26 <orin> firefox munges chromium bookmarks
16:46:30 <orin> not sure about passwords
16:47:09 <int-e> of course the timing of these changes is also interesting
16:47:33 <int-e> this is happening shortly after MS announced they're discontinuing Edge... https://www.theverge.com/2018/12/4/18125238/microsoft-chrome-browser-windows-10-edge-chromium
16:47:42 <orin> if you have a website and an app I always use the website. having so few apps means my phone battery lasts much longer than my dad's
16:49:38 <myname> that depends on what you do imho
16:52:47 <orin> myname: well my dad has all these apps he rarely uses constantly loading new messeges and content whereas my phone only does this when I visit a website
16:52:56 -!- moei has joined.
16:53:32 <orin> my phone therefore does less work when I'm not using it
16:53:52 <orin> and as a result only needs charging every 4 or 5 days
16:54:06 <myname> yeah, that's crap. but there are services where an app does provide additional features
16:54:37 <myname> like, i will always take the youtube app over the website
16:54:55 <orin> you can't block ads in the youtube app
16:55:01 <orin> so it has less features
16:55:22 <orin> and wastes data
16:55:43 <int-e> "you can't block ads in the youtube app" is the main reason for all those apps, I'm afraid
16:56:02 <orin> yup. firefox mobile with ublock origin
16:58:27 <orin> you can watch youtube on mibile with no ads, and look at another website while it plays
16:58:43 <orin> all features that the youtube app deliberately does not have
16:59:19 -!- Erika has quit (Ping timeout: 256 seconds).
16:59:33 <orin> billionaires hate him! this 1 cool trick
17:03:48 <myname> orin: maybe _you_ can't
17:04:49 <myname> and you can watch a video while doing whatever you want
17:04:50 <orin> myname: oh are you paying for redtube, er, youtube red, er, youtube premium?
17:05:16 <myname> orin: http://vanced.app
17:05:53 <orin> eh, that's an extra app
17:06:11 <orin> I don't go for these extra apps
17:06:54 <myname> at least the rooted way just replaces it
17:07:36 <orin> just use firefox
17:08:24 <myname> watchlist handling is horrible in browser
17:11:00 <orin> you could just hit "request desktop site" if your phone has high DPI
17:12:40 <myname> the website lacks features the app has
17:14:33 <orin> I don't really go for watchlists as much as super long videos
17:14:44 <orin> https://www.youtube.com/watch?v=DX2-yTicXs0 liek this
17:17:54 <orin> because I hate the slight lag betweeen longs in playlists
17:19:14 <myname> i rarely use youtube for music
17:19:53 <orin> I rarely use youtube for anything except music
17:26:45 -!- danieljabailey has joined.
17:59:43 <esowiki> [[BitBounce]] N https://esolangs.org/w/index.php?oldid=60016 * Hakerh400 * (+12154) Publish BitBounce
18:00:45 <esowiki> [[Language list]] https://esolangs.org/w/index.php?diff=60017&oldid=59915 * Hakerh400 * (+16) Add BitBounce to the list
18:29:46 <orin> lol nike shoes "bricked" by android app
18:30:00 <orin> intenet of shit
18:32:55 -!- Phantom_Hoover has joined.
18:38:50 <orin> weird fact: 0x64 == 0144 == 100
18:39:27 <orin> this is weird because all of those are squares in decimal
19:03:53 <orin> U-FF60 to U-FF9F can augment ascii to form a base-128 system. this has the advantage of looking like the matrix raining code
19:05:55 <orin> therefore the computers in the matrix must have used 7-bit groupings
19:18:56 <orin> for example, in this version fo base 128, the number ｲｪpK is 57681234 in dec
19:32:27 -!- danieljabailey has quit (Ping timeout: 240 seconds).
20:11:00 -!- b_jonas has joined.
20:29:48 <kmc> half-width kana?
20:30:14 -!- tswett has joined.
20:31:55 <tswett> I'm starting to get kinda desperate for work. If anyone's got stuff that needs doing, let me know.
20:35:52 -!- FreeFull has joined.
20:43:11 <b_jonas> shachaf: gcc allows a little bit of control over function calling conventions, but not as much as you're asking for. https://gcc.gnu.org/onlinedocs/gcc-8.2.0/gcc/x86-Function-Attributes.html#x86-Function-Attributes
20:45:25 -!- tswett has quit (Ping timeout: 256 seconds).
20:48:00 <shachaf> b_jonas: I don't think it's very practical in a language like C or C++
20:48:25 <esowiki> [[Pizza Delivery]] https://esolangs.org/w/index.php?diff=60018&oldid=59949 * Cortex * (+725)
20:59:34 <esowiki> [[Pizza Delivery]] https://esolangs.org/w/index.php?diff=60019&oldid=60018 * Cortex * (+21)
21:21:11 <b_jonas> `bobadventureslist http://bobadventures.comicgenesis.com/d/20190219.html
21:21:12 <HackEso> bobadventureslist http://bobadventures.comicgenesis.com/d/20190219.html: b_jonas
22:04:14 -!- Remavas has joined.
22:10:31 <lambdabot> LHBP 202200Z VRB01KT CAVOK M02/M02 Q1025 NOSIG
22:11:35 -!- arseniiv has quit (Ping timeout: 255 seconds).
22:24:04 -!- xkapastel has quit (Quit: Connection closed for inactivity).
22:25:08 <esowiki> [[BitBounce]] M https://esolangs.org/w/index.php?diff=60020&oldid=60016 * Hakerh400 * (+0) fixed typo
22:25:08 -!- Essadon has joined.
22:25:23 -!- Essadon has quit (Max SendQ exceeded).
22:38:18 -!- xkapastel has joined.
23:20:39 -!- sleepnap has left.
23:33:56 -!- AnotherTest has quit (Ping timeout: 255 seconds).
23:39:26 <shachaf> int-e: imo what if snprintf was a coroutine that you could resume with a new buffer if it runs out of space?
23:40:25 <kmc> cocoa coated co-cones
23:51:07 -!- Remavas has changed nick to Remavas[AFK].
23:54:16 -!- tromp has quit.
23:56:09 -!- Remavas[AFK] has changed nick to Remavas.