Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Posted on 19-03-01, 22:56 in The Horstmann brace style
Full mod

Post: #141 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Posted by sureanem
How do you find out which block is which if the braces don't line up? With multiple levels of nesting, that gets ugly fast.


Indenting, obviously:
for (x = 1; x < 10; x++) {
if (x % 2 == 0) {
printf("Even!");
} else {
printf("Odd!");
}
}

If you tell me you can look up from "}" to find the aligned "{", but can't look up from "}" to find any other aligned text, I'm going to be very doubtful.

THAT SAID, much like "which font is the most readable", the biggest factor of all is what you're already used to. If you've spent twenty years finding blocks by looking for isolated "{" and "}" characters while ignoring all the text between them, finding blocks by any other pattern is going to be more effort and feel "wrong".

One of the happiest discoveries I've made while learning Rust is the rustfmt command for automatically formatting code. While automatic formatting can never be quite as nice as manual formatting, it's great to be able to type in any old thing, hit "save" and watch it magically become neat and tidy. Less dramatically but more significantly, I don't waste ten minutes trying to reformat something "just right" because it'll snap right back afterwards. Of course, it helps that the Standard Rust Style is pretty close to my preferred style anyway, but if more people used auto-formatting there'd be fewer threads like this one. ;)

The ending of the words is ALMSIVI.
Posted on 19-03-02, 01:00 in The Horstmann brace style (revision 1)
Full mod

Post: #142 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
> Vim supports it, and presumably emacs too. Which CLI-based editors don't, you mean?

nano doesn't highlight matching brackets, although it does have a "jump to matching bracket" command.

I know this because I was going to reply to tomman with something like "even nano highlights matching brackets!" and then decided to actually fact-check myself.

> How do you mean? It's true that in trivial examples like this, it's possible. But if you have more difficult examples, then matching them becomes considerably harder

This is the thing I was talking about, where we're looking at the same examples and seeing different things. I look at your examples and see something like:
##################
##################
######
###########
##################
####
#
###################
#
...and yeah, I can quite clearly see the edges of blocks just by looking at the shape of the left-hand margin. It gets more difficult to guess which line corresponds to a given closing-bracket with a bucket-shaped structure, where you have deeply-nested blocks and then many lines at the deepest level:
##################
##################
##################
##################
##################
##################
##################
##################
##################
##################
##################
##################
##################
##################
#
#
#
#
...especially if the inner-most block is taller than the screen height. Still, that's going to be taxing for *any* kind of visual recogition, since it's harder to extrapolate accurately the further you go.

> We'd still be discussing to what to set our autoformatters to, no?

Autoformatters aren't infinitely configurable, so the discussion only has to cover the available options, not every possible variation. Also, autoformatters have defaults, so one specific configuration is vastly easier to set up and maintain, and every alternative has additional costs. Most people get a lot less invested in their formatting choices when they have to spend an afternoon wrestling with config choices or trying to talk the maintainer into accepting their patches.

The ending of the words is ALMSIVI.
Posted on 19-03-05, 22:05 in Sales and giveaways
Full mod

Post: #143 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Humble Indie Bundle 20 is here. I think the most notable game so far is probably bunny murder simulator Overgrowth, the long-in-development sequel to 3D brawler Lugaru, which was in the first ever Humble Indie Bundle.

The ending of the words is ALMSIVI.
Posted on 19-03-07, 16:35 in Official higan changelogs? (revision 1)
Full mod

Post: #144 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
The higan git repository doesn't have "releases" because the git repository is an *effect* of releases, not a *cause*, so it would be misleading to imply that releases from there are in any way official.

However, it does have tags. You probably want v104b rather than v104, because byuu found a small problem in the original v104 and made a new release with the same number.

As you can see, the v104 official changelog doesn't go into much detail about how "integral scaling" does/should work. Unfortunately, I didn't have the unofficial docs for v104 finished by the time it was released, but I documented v104's video settings soon afterward.

EDIT: Forgot to mention, Kawa graciously hosts a mirror of the old forum under "Archive" in the header of this site (and the even older forum under "Older archive" beside it). That includes the v104 release thread.

The ending of the words is ALMSIVI.
Posted on 19-03-07, 17:14 in Official higan changelogs?
Full mod

Post: #145 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
The official higan page links to the git repo because byuu occasionally finds it useful, and perhaps it's useful to other people too. However, it's not a service byuu provides or is responsible for; if it went offline for some reason, higan development would continue without it. When the content on byuu.org disagrees with the content of the repository, byuu.org is always the official state of affairs, and the repo is a summary, or approximation, or compromise. As an example, the official v104 is the one tagged v104b in the git repo, because I had to compromise between accurately recording official releases, and accurately recording the development process. byuu's would have been well within his rights to ignore people trying to report problems with "v104b" since there was never an official release with that version number.

The "official repository", if you can call it that, would be byuu's personal archives of historic bsnes/higan releases, which are stored on his PC and not available to the public. I don't know if byuu kept a copy of the original v104 or not.

The ending of the words is ALMSIVI.
Full mod

Post: #146 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
The Cx4 emulation needs a program to run, and a lookup table of mathematical constants. The Cx4 program is stored in whatever SNES ROM you're playing, the lookup table is "cx4.data.rom" if you're using a modern version of higan, or "cx4.rom" if you're using an older version.

So the Cx4 does need two ROMs, but it doesn't need both "cx4.data.rom" and "cx4.rom", because those are the same.

The ending of the words is ALMSIVI.
Posted on 19-03-09, 04:31 in CGA or Hercules?
Full mod

Post: #147 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Colour is an important signifier, and obviously every later PC graphics card built on top of the CGA legacy, but I've got to give it to Hercules for higher-than-VGA-resolution graphics back in the days before VGA was even a thing.

The ending of the words is ALMSIVI.
Posted on 19-03-09, 05:04 in I have yet to have never seen it all.
Full mod

Post: #148 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Posted by sureanem
No, no, I wouldn't use it to refer to Chinese people, absolutely not. That would indeed be very offensive. But that seems to be the most common name for those kinds of products, at least that I've heard. Not sure what else you'd call them. Piracy-intended multimedia devices? China boxes? Chinese boxes? Two of these now start linking together a whole country/ethnicity with shoddy workmanship in a more 'official' sense, so I would argue it is much worse.

"linking together a whole country/ethnicity with shoddy workmanship" is exactly the reason why the term in question is a poor choice of words, yes. Well, one of the reasons.

I'm sure you don't have any aggressive or offensive intent, but consider that the same language is used by other people who do, and those people make you look bad by association.

The ending of the words is ALMSIVI.
Posted on 19-03-09, 05:42 in CGA or Hercules?
Full mod

Post: #149 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
I've only ever seen Hercules graphics on green phosphor, so green it is.

The ending of the words is ALMSIVI.
Posted on 19-03-09, 08:38 in Announcing the bsnes history kit
Full mod

Post: #150 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
I'm proud of the unofficial higan git repo; I like that there's a publically-accessible, easily searchable and reasonably reliable record of higan's development. However, I only started working on it after I'd been following the project (then known as bsnes) for some time, and bsnes had been under development for quite a while before that, so the oldest and juiciest parts of the story were lost.

Well, almost lost. Long before I showed up, a person named tukuyomi had been preserving what they could of the SNES emulation scene - releases, changelogs, test ROMs and everything else. They didn't get *everything* (who could?) but they got a *lot* of stuff, including bsnes releases from about v005 up to about v095, and usually with changelogs too. Unfortunately, this was all in separate tarballs with varying layouts, and a separate plain-text changelog file rather than anything machine-readable and searchable. But with a little scripting and elbow-grease, it should be possible to make a bsnes-history repository as easily-searchable as the main higan repo.

I have not made a bsnes-history repo. It would take a lot of time and effort, and probably manual-checking of individual commits, which I just don't have the patience for right now.

However, I have made a bsnes-history kit that does about 80% of the cleanup, and produces a bsnes-history repo that can be grafted into the higan repo.

There's still a few things I'd like to do before calling it "complete" (listed in TODO.md) but the most difficult bits are done: collating the changelogs and archives into a single chronological stream, and writing it out in the format "git fast-import" expects.

If anybody has any ideas for improvements, I'd love to hear about them. Also, if anybody has changelogs or archives that aren't present in this repo, I'd love to hear about that, too.

The ending of the words is ALMSIVI.
Posted on 19-03-09, 22:13 in Announcing the bsnes history kit
Full mod

Post: #151 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Right now, it just builds a repo that contains all available history and nothing else. If I uploaded that directly, anybody stumbling over it might assume it was the actual historical bsnes archive, rather than a reconstruction. Once I figure out how to put a README on the front that explains what's going on, without getting the README mixed up with the actual historical data, then I'll put up a pre-built repo for people to poke at. I've got some ideas on how to do that, I just don't have the time to test them all out yet.

The ending of the words is ALMSIVI.
Posted on 19-03-10, 10:34 in Announcing the bsnes history kit (revision 1)
Full mod

Post: #152 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
UPDATE: My hunches paid off, and I figured out a reasonably simple way to have the README separate from the history. I also figured out a reasonably simple way to take the full reconstructed history, and graft just the tail end onto the unofficial higan repo to make a full record.

The pre-generated bsnes-history repository is at https://gitlab.com/higan/bsnes-history/ but the master branch only contains the README. The actual history is in the history branch.

The ending of the words is ALMSIVI.
Posted on 19-03-10, 11:59 in Announcing the bsnes history kit
Full mod

Post: #153 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
I did ask byuu, and apparently he lost everything up to v018 in a tragic hard-drive mishap.

I confess I've never heard about byuu using LiveJournal before; I assumed the first bsnes releases were all on the ZSNES forums.

The ending of the words is ALMSIVI.
Posted on 19-03-14, 00:25 in Games You Played Today REVENGEANCE
Full mod

Post: #154 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Posted by Nicholas Steel
It's the runniest train in town.

I ded.

The ending of the words is ALMSIVI.
Posted on 19-03-15, 07:33 in Blackouts
Full mod

Post: #155 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Posted by wertigon
In WWII, the empire collapsed completely, which is why Australia and India now are their own countries (among many other things).

(a nitpick, but Australia federated its various colonies to become a new nation in 1901. A very Britain-aligned nation, to be sure, and with the British monarch as our head of state, but an independent nation. If you say we don't count as independent until we cut ties with the British monarchy, well, we *still* haven't cut ties with the British monarchy.

It *is* true that during WWII, Australia became less interested in European affairs and our alliance with Britain, and more concerned with Pacific affairs and our alliance with America)

The ending of the words is ALMSIVI.
Posted on 19-03-15, 12:57 in Announcing the bsnes history kit
Full mod

Post: #156 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
No project like this can ever be 100% complete, but I'd done a chunk of work, I'd made something presentable, and I was ready to set it aside and move on to something else.

Then somebody pointed out that the two oldest release archives in the collection were RAR files, not ZIP, and so they weren't included.

Then somebody else provided a copy of the changelog text for some of the archives that were missing it.

So I figured I should modify my scripts to take those things into account, and it turned out to be quite difficult, more or less a rewrite. So I got stuck in.

And while I was re-writing things, I decided to poke around the zboard, looking for any scraps of bsnes history, and discovered there was once a bsnes megathread, now deleted, but which is archived in the tukuyomi collection and I never noticed. And when I looked at it, I discovered a changelog for v045r09 that I'd been missing, and another set of changelogs for higan releases (which are already in the unofficial higan repo, but for completeness...). And then I was leafing through the old zboard threads and discovered apparently WIP releases used to be posted in the public forum (it was just the downloads that were private), and apparently some (most?) of those WIP announcements weren't recorded either.

So I've got a few more direct, near-machine-readable sources of changelog text that should be easy enough to integrate, and then some messy human-readable HTML forum dumps that I'll have to read through, or figure out how to index or something.

In the mean time, I've pushed a new version of bsnes-history-kit and of the example bsnes-history repo that includes the content of the RAR archives. It does mean a new dependency on the "unrar" tool (and also "unzip" and "tar", but those should be pretty widely available already), but I've removed the dependency on "awk", so that's a thing.

The ending of the words is ALMSIVI.
Posted on 19-03-16, 00:47 in Something about cheese!
Full mod

Post: #157 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Beto O’Rourke’s secret membership in America’s oldest hacking group

TL;DR: Presidential Candidate Beto O'Rourke was a member of the the Cult of the Dead Cow, a group probably most famous for releasing the Back Orifice and Back Orifice 2000 "remote administration" tools to highlight the security flaws in Windows 95-XP. It's also credited with coining the terms "31337" and "hacktivism".

The ending of the words is ALMSIVI.
Posted on 19-03-17, 00:09 in Mozilla, *sigh*
Full mod

Post: #160 of 443
Since: 10-30-18

Last post: 1101 days
Last view: 172 days
Old news, but Stay classy, Pale Moon. (why Pale Moon does not have an OpenBSD port)

The ending of the words is ALMSIVI.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
    Main » Screwtape » List of posts
    Yes, it's an ad.