#jsmess 2017-07-12,Wed

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)

WhoWhatWhen
JohnTalendb48x: Like you said the functions as they stand will just be indented 'relative to the function keyword that began them'. So this one change that aspect at all. And it won't cover the attrocities that preceed and follow it.
s/one/won't
[00:00]
db48xheh
which atrocities are those?
[00:00]
JohnTalenI'll be happy to work on emulators providing you with an interface to them or rewriting emularity to then work in the newly written emularity.
absolutely little to no code reuse. Only having the facing html interface consistent and clean. Functions that serve only as empty repositories to then be modded throughout without any rhyme or reason why these dead structs exist without code relations.
the list goes on.
Who wrote that anyway? And why do they want to abandon it?
[00:01]
db48xwhich functions are only "dead repositories" to be modded?
s/dead/empty
[00:04]
are you referring to things like BaseLoader.emulatorJS, which just return trivial objects? [00:16]
JohnTaleni looked at it over 1 1/2 weeks ago. Hang on. [00:17]
baithe only thing crazy about db48x is his beard. and right now it's not even the craziest beard on the team [00:18]
JohnTalenThere are instances of structures that could be inheritenting code and that simple do nothing but store. And these are themselves are treated like functions. It's maddening. [00:18]
db48xyea, I think you're talking about BaseLoader/DosBoxLoader/MAMELoader/SAELoader, but I'd prefer to be specific [00:19]
JohnTalenI don't have my code viewer set up for .js, only .c* at this time.
i'll just follow the code from the well written interface.
[00:20]
db48xJohnTalen: which one is the well-written interface, specifically? [00:21]
JohnTalenthe html pages, that's about *it*. [00:22]
db48xthat's good! A lot of the choices I made were so that the API seen by the users was as nice as possible [00:24]
JohnTalenEmulator.Start is a good example of how not to indent.
particularly when it gets to loading.then(
[00:30]
db48xfair enough; I'm well aware that my indentation style is not widely liked
aside from that, is there anything wrong with it?
[00:36]
JohnTalenif I was to write loadFiles, I would have to rewrite it using the files to be reused as you've set them up. That is one small example of lack of reuse. [00:38]
baihaving some of the promise handling functions as member methods on those objects might help
the inline functions get a bit hard to follow, and they're nested a bit deep in some cases
[00:38]
db48xJohnTalen: which are the files to be reused, specifically? [00:39]
JohnTalendb48x: loadFiles( under Promise. That's another thing. I have to page up or down dozens of screens to find out which function I'm in. This is huge detriment to some code which isn't that bad. I'm not saying it's all bad. But being a new comer is a huge set back for me to have to write in. [00:41]
db48xwhat line number? [00:42]
JohnTalenideally. you'll want to just have a structure specific to the platform and associate it to code hand in hand. And have the innards handle the differences through a case mechanism. [00:42]
db48xI _think_ I understand that, possibly
but it's quite vague
oh shoot, I'm supposed to be meeting some folks right now
can we pick this up in a few hours?
[00:43]
JohnTalenline 95 in my version.
yeah that's okay.
[00:44]
db48xthanks [00:44]
JohnTalenyw
thanks for hearing me.
[00:44]
baian example would be something like, instead of loading.then(function(_game_data) { ... }.then(function(game_files) { ... }.then(function() { ... }.then(function() { ... }) would be something like:
loading.then(this.handleGameData.bind(this)).then(this.handleGameFiles.bind(this)).then(this.postinit.bind(this)).then(this.run.bind(this))
or if you pre-bind your functions it's even cleaner
then all that code gets moved off into its own member functions
[00:45]
JohnTalenWell, the whole program is a start. But the focus of what needs to be done is there. It's just a matter of broking up the structures into a hierachical consistency that lends itself to reuse.
Perhaps if I rewrite loadFiles if something a little further in I can possibly be of some added value benefit.
s/if/or
[00:51]
baisomething like this https://pastebin.com/3np0jNyS [00:53]
JohnTalenfor example. why should I need to getElementsByTagName in an almost directly forward facing function? It's nickels and dimes from top to bottom in that sense. But that's *OKAY*. It's the first program of it's kind. It's allowable. :) [00:53]
baithat way when you inherit from Emulator, you don't have to redefine the whole start() function, you can just override the specific parts [00:54]
JohnTalenbai: yes. that is indeed a great step in the right direction. 'Think' hierachy of case statements of resuable code. Each level is a layer. [00:55]
................................ (idle for 2h37mn)
db48xbai: that's a good, specific example
my commendations on both points
though I wouldn't bother making them member functions, since there's generally only one "instance" of the IALoader, and these functions are all about dealing with the data flowing in from the xhr requests
I've never considered that anyone might want to inherit from Emulator, though it could make sense to structure it that way
JohnTalen: what's your specific concern with getElementsByTagName? you'd never need to call that to support a new emulator
[03:32]
........ (idle for 38mn)
JohnTalendb48x: What specifically did you want me to do?
I mean to say how much and what did I need to touch in Emularity. I've only just brushed over it's surface.
.=?
I mean when I integrate it. I can make it as humanly readble as possible. Without touching your code. Or porting what functions make sense. Prolly low level ones since that all there seems to be in Emularity.
s/that/thats
[04:15]
exhausted, I'll be here for 5 more minutes then I'll be back tommorow. Hopefully the whole day. [04:24]
bbt [04:32]
***JohnTalen has quit IRC (Quit: leaving) [04:32]
.......................... (idle for 2h5mn)
balrog has quit IRC (Read error: Operation timed out) [06:37]
.... (idle for 19mn)
balrog has joined #jsmess [06:56]
...................................................................... (idle for 5h45mn)
db48x has quit IRC (ny.us.hub hub.efnet.us)
devesine has quit IRC (ny.us.hub hub.efnet.us)
balrog has quit IRC (ny.us.hub hub.efnet.us)
azakai has quit IRC (ny.us.hub hub.efnet.us)
zino has quit IRC (ny.us.hub hub.efnet.us)
gsathya has quit IRC (ny.us.hub hub.efnet.us)
SketchCow has quit IRC (ny.us.hub hub.efnet.us)
arkiver has quit IRC (ny.us.hub hub.efnet.us)
bai has quit IRC (ny.us.hub hub.efnet.us)
bwn has quit IRC (ny.us.hub hub.efnet.us)
datajerk has quit IRC (ny.us.hub hub.efnet.us)
Lord_Nigh has quit IRC (ny.us.hub hub.efnet.us)
balrog has joined #jsmess
azakai has joined #jsmess
bwn has joined #jsmess
db48x has joined #jsmess
zino has joined #jsmess
datajerk has joined #jsmess
gsathya has joined #jsmess
devesine has joined #jsmess
bai has joined #jsmess
Lord_Nigh has joined #jsmess
SketchCow has joined #jsmess
arkiver has joined #jsmess
hub.efnet.us sets mode: +o arkiver
[12:41]
......................................... (idle for 3h21mn)
azakai has quit IRC (Quit: Ex-Chat) [16:02]
............................. (idle for 2h20mn)
SketchCowSketchCow walks into channel
HA HA OH WOW
SketchCow walks out
SketchCow walks back
Hey, how's it going
[18:22]
db48xsomewhat tired
it's too early to be awake, really
[18:24]
SketchCowSo, we established Talent will not be sending Emularity a christmas card
You're invited to come hang out at IA, db48x (and bai too, if you want)
everyone else, stay away
[18:27]
db48xheh [18:29]
SketchCowalso, I think you're all in not the bay area [18:29]
db48xwhile I think that my indentation style is perfect, for several reasons, I was already aware that it's an acquired taste so that didn't upset me
and dealing with IA XML metadata, and a JSON file, _and_ a CFG file really is a lot of hoops to jump through to start up an emulator, but that wasn't a design choice
just what's necessary in order to deal with IA
[18:34]
SketchCowI had a great BBQ years back where I tore apart the editing of a film in front of the editor
To the editor, really
[18:46]
db48xheh [18:47]
............. (idle for 1h0mn)
***JohnTalen has joined #jsmess [19:47]
........... (idle for 52mn)
DFJustinyeah dunno when I'll be down in that neck of the woods again
wasn't there going to be an Archive Team Con at some point
[20:39]
...... (idle for 28mn)
SketchCowWe discussed it
db48x: Did you want to compile a pile of WASMs? That seems an easy win
[21:07]
........... (idle for 54mn)
db48xI have to do the work on Emularity first [22:01]
SketchCowI know they won't load, but it takes days to compile anyway [22:10]
db48xI didn't time it last time, but it was finished the next morning [22:11]
SketchCowWell, when you can; we'll be the webassembly vanguards
And then drop a big MAME trouble ticket
I've been doing that critical work of scanning in AOL CDs.
Now matching up the first batch
[22:20]
..... (idle for 21mn)
Batch being uploaded into IA. 65 CD-ROMs, mostly AOL [22:45]
Vito`SketchCow: since you're there, would you mind re-fixing https://archive.org/details/SPACECOMMERCE-cdrom so it's in the CD-ROM collections [22:46]
SketchCowDone [22:50]
Vito`thanks [22:50]
.............. (idle for 1h7mn)
JohnTalenPloding through Vice. Have to recompile, get ONE error message. Change the duplicates, run a script to copy files and recompile. This is for each single error message. emcc only reports one at a time. [23:57]

↑back Search ←Prev date Next date→ Show only urls(Click on time to select a line by its url)