there were a lot of events recorded by gharchive.org of which 2,052,275 were push events containing 3,446,251 commit messages that amount to 257,117,729 characters filtered with words.py@e23d022007... to these 66 messages:
xfs: fix negative array access in xfs_getbmap
In commit 8ee81ed581ff, Ye Bin complained about an ASSERT in the bmapx code that trips if we encounter a delalloc extent after flushing the pagecache to disk. The ioctl code does not hold MMAPLOCK so it's entirely possible that a racing write page fault can create a delalloc extent after the file has been flushed. The proposed solution was to replace the assertion with an early return that avoids filling out the bmap recordset with a delalloc entry if the caller didn't ask for it.
At the time, I recall thinking that the forward logic sounded ok, but felt hesitant because I suspected that changing this code would cause something /else/ to burst loose due to some other subtlety.
syzbot of course found that subtlety. If all the extent mappings found after the flush are delalloc mappings, we'll reach the end of the data fork without ever incrementing bmv->bmv_entries. This is new, since before we'd have emitted the delalloc mappings even though the caller didn't ask for them. Once we reach the end, we'll try to set BMV_OF_LAST on the -1st entry (because bmv_entries is zero) and go corrupt something else in memory. Yay.
I really dislike all these stupid patches that fiddle around with debug code and break things that otherwise worked well enough. Nobody was complaining that calling XFS_IOC_BMAPX without BMV_IF_DELALLOC would return BMV_OF_DELALLOC records, and now we've gone from "weird behavior that nobody cared about" to "bad behavior that must be addressed immediately".
Maybe I'll just ignore anything from Huawei from now on for my own sake.
Reported-by: [email protected] Link: https://lore.kernel.org/linux-xfs/20230412024907.GP360889@frogsfrogsfrogs/ Fixes: 8ee81ed581ff ("xfs: fix BUG_ON in xfs_getbmap()") Signed-off-by: Darrick J. Wong [email protected] Reviewed-by: Dave Chinner [email protected] Signed-off-by: Dave Chinner [email protected]
WebXR: Severe aliasing in WebXR experiences (with WebGL1 contexts) https://bugs.webkit.org/show_bug.cgi?id=256861 rdar://109424254
Reviewed by NOBODY (OOPS!).
WebXR sessions using WebGL1 contexts are unable to turn on multisampling. I'm pretty sure this was my fault, but I can't remember if it was intentional or a mistake. Either way it is a bug.
Fix this by implementing the multisample renderbuffer creation and resolution steps. Since we're doing this on a WebGL1 context, the normal API will be invalid (it requires GLES3), so call the extension API instead. This means we need to expose some extra methods on GraphicsContextGL.
Lastly, the framebuffer textures we get are SRGB8_ALPHA8 which requires an extension to be enabled with a WebGL1 context.
- Source/WebCore/Modules/webxr/WebXROpaqueFramebuffer.cpp: (WebCore::WebXROpaqueFramebuffer::endFrame): call blitFramebufferANGLE. (WebCore::WebXROpaqueFramebuffer::setupFramebuffer): Implement logic for WebGL 1.
- Source/WebCore/platform/graphics/GraphicsContextGL.h:
- Source/WebCore/platform/graphics/angle/GraphicsContextGLANGLE.cpp: Implement the extension API/ (WebCore::GraphicsContextGLANGLE::renderbufferStorageMultisampleANGLE): (WebCore::GraphicsContextGLANGLE::blitFramebufferANGLE):
- Source/WebCore/platform/graphics/angle/GraphicsContextGLANGLE.h:
- Source/WebCore/platform/graphics/cocoa/GraphicsContextGLCocoa.mm: (WebCore::GraphicsContextGLCocoa::platformInitialize): Turn on the sRGB extension.
- Source/WebKit/GPUProcess/graphics/RemoteGraphicsContextGL.messages.in:
- Source/WebKit/GPUProcess/graphics/RemoteGraphicsContextGLFunctionsGenerated.h: (renderbufferStorageMultisampleANGLE): (blitFramebufferANGLE):
- Source/WebKit/WebProcess/GPU/graphics/RemoteGraphicsContextGLProxy.h:
- Source/WebKit/WebProcess/GPU/graphics/RemoteGraphicsContextGLProxyFunctionsGenerated.cpp: (WebKit::RemoteGraphicsContextGLProxy::renderbufferStorageMultisampleANGLE): (WebKit::RemoteGraphicsContextGLProxy::blitFramebufferANGLE):
The Lizardening Part One (And Friends) (#1845)
This PR changes a lot of sprites. It's honestly too much. Namely:
- Explorer Equipment + Prototype
- Syndicate clothing
- Digitigrade lizard legs
- A new tail from Halcyon.
- Magboots from Zeta. Originally PR'd to tgstation.
- Colored (not greyscale! Ha Ha!) jumpsuits from Imaginos.
Heavy inspiration from the work of Imaginos, Halcyon, Mqiib, and 2cents#8442 for the original leg-work. (Haha, get it?) The new digitigrade sprites started as a twinkle in the eye of Mqiib, for yogstation(?) After myself and Halcyon saw those, an epihpany struck. Perspective makes things cool and digitigrade perspective was BAD.
I'll include a collage image of the new sprites if it's needed later. Preview below:
lizard, Death of Codersprites
🆑 add: New Digitigrade lizard sprites. add: Various syndicate and mining clothing resprites. add: Sarathi can now have an incredibly large tail. /🆑
Botany Balance Pass (#1783)
First came the content, now comes the hammer.
-
Nukes Megaseed servitors from orbit.
-
Plants now age much, much slower and produce half as quickly. Ruins that had them now have a ruined seed vendor that can be salvaged for random seeds(and danger). Ships that had one now have a crate with some thematic starting seeds, and a Strange Seed. Ghostrole Ruins that relied on having all seeds locally now have a special biogenerator variant that can print a random seed for biomass.
-
Adds Genesis Serum. This can be splashed on a tile to make natural grass and some flora. Green your ship! Genesis Serum was made a while ago, on request for a way to add natural grass and flora to your ship. Since I had it lying around fully coded, I thought I might as well pr it with botany changes.
-
Gatfruit found in the seed vault have been replaced with Strange Seeds.
-
The chance to get Gatfruit from a demonic portal(plant variety) has dropped from 15% to 5%.
-
Corpse flowers now have liquid gibs and formaldehyde again.
Okay, hear me out
With this and Gardens, botany ships go from a "sit in your vessel for 2 hours" experience to an "explore and forage" one that better fits our feature arc. It goes without saying that this shouldn't be merged till Overmap 4.2 is, since it facilitates getting seeds from planets as part of exploration.
Gatfruit are funny, but it takes exactly one seed getting into the hands of a ship with a dna manipulator and the weapon balance is eradicated from the game completely(for the round, at least.) This is more problematic here then it was on TG, since our rounds tend to be 5 hours long rather then 1. This has been long coming. I'll reverse this if we ever get that Plantlock variant we wanted a while ago.
Corpse flowers even have formaldehyde and gibs on tg, not sure what happened there.
🆑 add: Ruined megaseed servitors can now be found on the frontier, carrying a bounty of seeds for intrepid adventurers. balance: the time it takes for plants to reach a lethal age has been increased massively. balance: Plant production time increased a bit to compensate. balance: megaseed servitors have been removed from ships and ruins. Ships that carried one now have a crate with some starting seeds. balance: removes gatfruit from the seed vault pool. balance: reduces the chance of getting gatfruit from a plant-themed demonic portal significantly. balance: corpse flowers once again have formaldehyde and liquid gibs. add: Adds Genesis Serum, a reagent that transforms tiles into natural grass on splash, then causes some natural flora objects to grow. Turn your ship green! /🆑
[MIRROR] Wintercoats can now be zipped and unzipped through alt-click and separates the hood sprites from the jacket sprites [MDB IGNORE] (#20718)
- Wintercoats can now be zipped and unzipped through alt-click and separates the hood sprites from the jacket sprites (#74886)
The title says it all, really.
Initially, I was only going to do it for all wintercoats, but then I
figured I might as well bring it down to all of /hooded, just so other
suits could benefit from it, since that behavior came from there anyway.
Does that mean that it does nothing for some of them? Yes, it does. Does
that justify having another variable to tell whether or not that should
be possible? In my humble opinion, not really, but I'm not against it if
it's requested.
That functionality was intentionally removed from the Void Cloak, as
there would be balance implications (since bringing up the hood makes
the whole cloak invisible, which you could skirt by just "zipping" it,
which also makes it invisible.
The sprites were already there, so this change was very simple to do.
Simply unties the zipped up look from the fact that the hood is up.
However, toggling the hood forces the zipping/unzipping, just so there's
no balance implications involved. It's just simpler that way.
So, I ended up going back and changing the sprites so that the hoods would no longer be baked into the jacket's sprites, so that they could be done as overlays instead, which ended up solving my problem with hoods not being there on zipped-up versions.
For now, it's been made on winter coats only, but it shouldn't be that
difficult to bring it back down to the /hooded level. I just didn't
want to bother touching up the sprites down there, as it already took me
like 2-3 hours touching up the sprites of the winter coats alone.
I also took the decision to make it so EVA winter coats used the regular winter coat's sprites, because they had special ones that just looked like worse versions of the original, without anything special going on for them. It was just a straight downgrade compared to the base sprite, in my opinion.
There's still issues with the custom winter coat, in that the hood isn't made into an overlay for it yet (and that'll require an extra bit of logic to make it work, too), but it was already an issue before, the hood is always present on the current version of the custom winter coat.
There's still a handful (sadly, most) of the winter coats that don't properly reflect on their obj sprites when they're opened versus when they're closed, but that's due to an initial spriter oversight, and not to my doing. The open versions were just left as closed on many of them, and I simply don't have the patience nor the appropriate skills to edit that many coats that way.
Now you can be stylish with or without the hoodie!
According to ChatGPT, with one small tweak (thanks Opera GX for the suggestion):
Zipped and unzipped through alt-click, winter coats can now be. Hmm, stylishly warm, you shall be. Feel like a Spaceman, you will. Use the Force, to zip and unzip, you must. Look cool, you will. Yes, hmmm.
🆑 GoldenAlpharex, ChatGPT for the first changelog entry (slightly edited) qol: Zipped and unzipped through alt-click, winter coats can now be. Hmm, stylishly warm, you shall be. Feel like a Spaceman, you will. Use the Force, to zip and unzip, you must. Look cool, you will. Yes, hmmm. image: Winter coats no longer have their hood baked into their jacket's sprite, both in item form and when worn. fix: Updated the Icebox EVA winter coats (the Endotherm winter coats) to use the same sprites as the regular winter coats. /🆑
Co-authored-by: san7890 <the@ san7890.com>
-
Wintercoats can now be zipped and unzipped through alt-click and separates the hood sprites from the jacket sprites
-
update modular wintercoat.dmi
-
forgot
Co-authored-by: GoldenAlpharex [email protected] Co-authored-by: san7890 <the@ san7890.com> Co-authored-by: ghost sheep [email protected]
[MIRROR] Config Flag to Save Generated Spritesheets to Logs [MDB IGNORE] (#20738)
- Config Flag to Save Generated Spritesheets to Logs (#74884)
I was helping someone debug some weird bug with spritesheets a bit ago,
and I didn't like having to manually comment out all of the fdel()
stuff in order to help visualize what the potential issue might have
been with the spritesheets on either their DM-side generation or their
TGUI-level display. I decided to add a compile-time level flag that will
automatically copy over any generated spritesheet assets (css and pngs)
to the round-specific data/logs folder for analysis when a developer
should need it.
I also had to switch around some vars and make a few new ones to reduce how copy-pasta it might get and ensure standardization/readability while also being 0.001 times faster since we benefit from the string cache (unprovable fact).
It's incredibly useful to see the actual flattened spritesheet itself
sometimes when you're doing this type of work and you keep getting odd
bugs here and there. Also saves headache from having to clear out the
temp /data/spritesheets folder every time you comment shit out, as
well as having an effective paper trail for A/B testing whatever
bullshit you've got going on.
Doesn't affect players.
- Config Flag to Save Generated Spritesheets to Logs
Co-authored-by: san7890 [email protected]
Updating the Mansion a bit
Preparing my brain for making a non euclidian dungeon, because I fucking hate everything.
[FIX] core: handle recursion error when resolving stored fields
Issue discovered in the uninstall (and reinstall) of sale_project: a
dump has ~100 tasks, when reinstalling sale_line_id has to be
initialised, this is done by marking sale_line_id on all extant
tasks as to-recompute, which triggers their computation on the next
flush.
Because it's a recursive field, Field.recompute ensures only one
record at a time gets recomputed (as there could be cross-dependencies
in the recorset which protection would prevent from resolving).
As the field computation runs, it accesses itself, which triggers a
cache miss, which triggers a _fetch_field (to get the currently
stored value), this calls _read, which flushes the field we're
trying to read.
The problem here is that for efficiency the cache miss will look for
all records in the cache without a value for the
field (_in_cache_without) and try to fetch on them as well. This
means rather than not doing anything in flush, we're going to
Field.recompute on all records except the one selected the first
time around, which repeats the cycle until there is no more additional
record found in _in_cache_without, which could trigger the next
round of recompute, and the entire thing unwinds, and we probably
perform a ton of unnecessary additional compute_value.
Except that doesn't even happen, because the process from one compute to the next takes 12~13 stack frames, which given the default recursion limit of 1000 gives a hard limit of 76 fields before hitting a RecursionError. As this is less than 100, a recursion error is what we get.
In 15.2, this was fixed by only expanding the fetch on non-recursive fields, pessimizing recursive fields (5c2511115b14299516fce4aa3737a62faaf5b653). Test-wise this only impacted mail performances and in a relatively minor manner.
In 16.0, the mail tests actually match already (so that part was
skipped by the cherrypicking) however this impacts the knowledge perf
tests much more significantly e.g. test_article_creation_multi_roots
gets +9 queries when creating 10 top-level articles, which is a bit
much.
So use an alternative which is ugly as hell but which I didn't consider for 15.2 (may want to backport it one day if the current fix is an issue): catch the recursion error and use the existing fallback (of fetching just the requested record's field without expanding the recordset).
This likely makes for a pretty inefficient situation in the original case as we're certainly going to hit the recursion limit repeatedly, but that still fixes the issue, and it avoids deoptimising cases which fall short of the recursion limit (resolving under 60 records or so).
Plus despite creating giant stacks we might actually get good efficiency as we're going to hit recursion limits repeatedly but that's pure python, once we fall below the limit we can resolve everything at once with a single SQL query (or something along those lines).
Part-of: odoo/odoo#119808
migrate_pages: avoid blocking for IO in MIGRATE_SYNC_LIGHT
The MIGRATE_SYNC_LIGHT mode is intended to block for things that will finish quickly but not for things that will take a long time. Exactly how long is too long is not well defined, but waits of tens of milliseconds is likely non-ideal.
When putting a Chromebook under memory pressure (opening over 90 tabs on a 4GB machine) it was fairly easy to see delays waiting for some locks in the kcompactd code path of > 100 ms. While the laptop wasn't amazingly usable in this state, it was still limping along and this state isn't something artificial. Sometimes we simply end up with a lot of memory pressure.
Putting the same Chromebook under memory pressure while it was running Android apps (though not stressing them) showed a much worse result (NOTE: this was on a older kernel but the codepaths here are similar). Android apps on ChromeOS currently run from a 128K-block, zlib-compressed, loopback-mounted squashfs disk. If we get a page fault from something backed by the squashfs filesystem we could end up holding a folio lock while reading enough from disk to decompress 128K (and then decompressing it using the somewhat slow zlib algorithms). That reading goes through the ext4 subsystem (because it's a loopback mount) before eventually ending up in the block subsystem. This extra jaunt adds extra overhead. Without much work I could see cases where we ended up blocked on a folio lock for over a second. With more extreme memory pressure I could see up to 25 seconds.
We considered adding a timeout in the case of MIGRATE_SYNC_LIGHT for the two locks that were seen to be slow [1] and that generated much discussion. After discussion, it was decided that we should avoid waiting for the two locks during MIGRATE_SYNC_LIGHT if they were being held for IO. We'll continue with the unbounded wait for the more full SYNC modes.
With this change, I couldn't see any slow waits on these locks with my previous testcases.
NOTE: The reason I stated digging into this originally isn't because some benchmark had gone awry, but because we've received in-the-field crash reports where we have a hung task waiting on the page lock (which is the equivalent code path on old kernels). While the root cause of those crashes is likely unrelated and won't be fixed by this patch, analyzing those crash reports did point out these very long waits seemed like something good to fix. With this patch we should no longer hang waiting on these locks, but presumably the system will still be in a bad shape and hang somewhere else.
[1] https://lore.kernel.org/r/20230421151135.v2.1.I2b71e11264c5c214bc59744b9e13e4c353bc5714@changeid
Link: https://lkml.kernel.org/r/20230428135414.v3.1.Ia86ccac02a303154a0b8bc60567e7a95d34c96d3@changeid Signed-off-by: Douglas Anderson [email protected] Suggested-by: Matthew Wilcox [email protected] Reviewed-by: Matthew Wilcox (Oracle) [email protected] Acked-by: Mel Gorman [email protected] Cc: Hillf Danton [email protected] Cc: Gao Xiang [email protected] Cc: Alexander Viro [email protected] Cc: Christian Brauner [email protected] Cc: Gao Xiang [email protected] Cc: Huang Ying [email protected] Cc: Vlastimil Babka [email protected] Cc: Yu Zhao [email protected] Signed-off-by: Andrew Morton [email protected]
Finished Zombie Proficiency & Weakpoint Review (#64194)
- Reviewed all Zombie Weakpoints & Proficiencies
- Renamed Large Humanoids proficiency to Giant Humanoids, to clarify that it does not apply to somewhat large humanoids, like brutes, and only works on hulks and similar.
- Changed description of Natural Armors proficiency, as many enemies that used this proficiency had something more like a thick hide than any kind of shell.
- Renamed Natural Armor weakpoint set (wps_natural_armor) to wps_armored_hide, to better reflect its purpose and to avoid confusion with the unrelated Natural Armor proficiency, as well as to prevent its misapplication to monsters which have more of a carapace or plate armor thing going on. Natural Armors proficiency should be reserved for uniquely resilient armored foes, like kevlar zombies, whereas armored hide applies to anything with a particularly thick hide, even if not outrageously so.
- Edited description of Corrosive Zombie to hint at its thick hide. Corrosive zombie now also trains Natural Armor proficiency.
- Spitter now has big head weakpoint set, based on description.
- All amalgamations now have intro_biology in their families. This should really be on any living creature of flesh and blood, with exceptions only for stuff like robots, physics-defying nether creatures, extra-dimensional anomalies, and the cafeteria meatloaf. I didn't add this to the cocoons because I wasn't sure if that made sense to do.
- Caustic amalgamation now trains biochemistry, like acid zombies do.
- Charged amalgamation now trains electromagnetics, like zapper zombies do.
- Fixed a typo in the description for Zombie Kinderlings.
- Zombie Fiend now trains Ossified Exoskeletons. Thought I added that one earlier.
- Scorched Zombie now gets Armored Hide weakpoints due to its "leathery shell".
- Removed Armored Hides weakpoint set from rust shell zombie and plated zombie. Could possibly apply Ossified Exoskeletons to them, but I'm not sure.
- Slight description tweaks, typo fix.
- Added proficiencies to Aberration and Dissoluted Devourer. Aberration doesn't give zombie bio because it isn't an actual zombie.
- Removed zombie bio from phase skulker, phase shrike, etc, as they aren't actually zombies.
- Gave phase shrike Ossified Exoskeletons proficiency.
- Added basic proficiencies to zombullfrog, frogmother, zombie nemesis, smoker
- Added basic weakpoints to smoker.
- Headless Horror trains giant humanoids proficiency, based on description.
- Removed Malicious Mane's natural armor training and body armor weakpoints, as it had no natural armor (or armor at all, for that matter).
- Added standard proficiencies and weakpoints to all of them.
- Replaced body armor weakpoint set with armored hide.
- Removed military pilot's synthetic armor proficiency
- Gave gastro bufo standard proficiencies and biochemistry.
- Replaced beekeper's body armor weakpoints with armored hide weakpoints
- Added expected proficiencies and weakpoints to pupating hulks, as they were the only pupa zombies that didn't have a copy-from pointing to the base type, and did not include this information. I noticed that most things that disappear on death - boomers, certain cocoons, etc. - tend not to have weakpoints or train proficiencies. Is this an oversight, or is this intentional? For now I left that as is.
- Gave raptors standard and flying proficiencies.
- Electric raptor also teaches electromagnetics, like electric zombies.
-
Removed my Personal Changelog from the Project Directory
-
Fixed Fungal Wretch Typos
-
Linted zed_amalgamations.json
-
MANY Zombie Weakpoint Refinements (& Tests)
- Gave standard weakpoints to standard zombies - manually defined weakpoints for some of the basic zombie models (in zed_misc), like the zombie brute and zombie hulk, is a bit strange, since they have become some of the game's staple enemies. THIS WILL LIKELY EFFECT BALANCE, as these are not only important benchmark enemies, but also copy_from'd by quite a few other enemies. Basic brutes are now somewhat weaker depending on circumstances
- Updated ranged balance test to use enemies with a more uniform form factor, as the high volume of some benchmark enemies lead to counterintuitive results (higher armor enemy taking more damage because it's bigger and easier to shoot). Note that test differences in values aren't all actual "balance changes" but moreso changes to the test itself, so the comparison between old and new isn't 1:1. Test values were only updated on tests that failed for me (I ran the test with 10,000 cycles instead of the usual 200 to be sure the values I got were convergent).
- Added weakpoints and proficiency families to zombies I previously wasn't sure should receive them (mostly ones which self-destruct on death in some way, like boomers). This will make boomers significantly weaker, as they previously had no weakpoints whatsoever.
- Changed boomer stats so no boomer upgrade becomes smaller in volume or lighter in weight than the basic boomer.
- Added an upgrade path for Zombie Miners - they now have a chance to evolve into a shady zombie (most likely), a rust zombie, or just a normal tough zombie, with a ~70% chance not to evolve, on a half-life of 35.
- Rust shell zombies and rust plated zombies get a unique weakpoint category. Similar to bone armor, with the difference that weak points are quite a bit weaker, but the strong point is also a bit stronger.
- Flesh raptors finally have weakpoints, borrowing from the ones used for wasps.
- Removed NOHEAD flag from zombie military pilot, as it very much has a head and there's no reason to believe it to be structurally superfluous, and also fixed them being given erroneous armor weakpoints when they're just in fatigues.
- Lots of other minor weakpoint tweaks/fixes.
-
Revert change to ranged tests that made it run 50 times as long.
-
Update data/json/monsters/zed_amalgamation.json
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
- Update data/json/monsters/zed_children.json
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
-
Update all Range Balance Values
-
Reverted Weakpoint ID Change
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Drunk slurring scales based on how drunk you are (#75459)
The strength of the slurring effect drunkness applies on you now scales based on how drunk you are.
Being "a little" drunk still changes your saymod, and makes you occasionally slur your words...
But being "a lot" drunk kicks it up to 11
Additionally, drunk slurring was separated into "generic slurring" and "drunk slurring", the former which does not scale but less closely resembles drunkness. Generic slurring is used in places such as concussions, so this is an added bonus.
As a result of the split, I had to update mind restoration. Now it heals all types of slurring, which does include cult slurs.
I, and many other people, always found it very annoying when you became completely illegible from taking one sip of a drink. This seeks to amend that by making low levels of drunkness still for the most part be legible and sane. Average drunkness is roughly the same / equal to the old slurring effect, while "very drunk" is even more illegible and silly (which I find funny).
This has the added bonus of separating out "drunk slurring" and "generic slurring", allowing effects to slur your words without going full ham on drunkness (burping and "huhh"s).
🆑 Melbert add: When you are drunk, the strength of your slurring now varies based on how drunk you are. Being "a little drunk" only rarely slurs your words, being average drunk is the same as the old effect, while being very drunk now slurs your words even more. add: Some non-alcohol sources of slurring, such as concussions, now give "generic slurring" rather than "drunk slurring", which less resemble being drunk (ie, no burping). add: Mind restoration now heals ALL slurring, rather than only drunk slurring (which includes cult / heretic slurring). /🆑
[DNM][HACK] telephony: Force Class 0 SMS to Class 1
This kills Flash SMS messages. Fuck you airtel
Change-Id: Ifb0c9e8bae5c12868d178fbdaeceb2cc72a0ffb6 Signed-off-by: Jprimero15 [email protected]
Redlick & Garland City Take 2
Fuck you to, strong dmm
MiE - Vampire rebalance 2023 by IPS
Minor corrections and recheck after real long.
-
Apprentice XP from 50 to 46
-
Savant XP from 100 to 85 , price to 33g (-7g)
-
Commander XP from 90 to 80 , price to 30g (-4g)
-
Count XP from 180 to 110 , price to 55g (+7g)
-
Prince price to 82g (+17g)
-
Shadow Fighter XP from 80 to 75 , price to 28g (-4g)
-
Night Warrior price to 52g (+8g)
-
Blood Hunter price to 27g (-4g)
-
Shadow Archer price to 29g (-2g)
-
Night Angel price to 53g (+6g)
-
Savage XP from 50 to 46
-
Lunatic price to 32g (-2g)
-
Winged XP from 46 to 42
-
Wind Chaser HP to 43 (+2)
-
Heaven Hunter fangs damage increase to 8-3 , HP to 54 (+3) , price to 54g (+8g)
Do math problems related to calculating dates using the Chinese Sexagenary Cycle method. 🧮 (#190)
🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨
PLEASE READ THIS:
In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task.
We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. We encourage partial PR's with ~5-10 example that we can then run the evals on and share the results with you so you know how your eval does with GPT-4 before writing all 100 examples.
Do math problems related to calculating dates using the Chinese Sexagenary Cycle method
The Sexagenary Cycle combines 10 Heavenly Stems (Jia 甲, Yi 乙, Bing 丙, Ding 丁, Wu 戊, Ji 己, Geng 庚, Xin 辛, Ren 壬, Gui 癸) and 12 Earthly Branches (Zi 子, Chou 丑, Yin 寅, Mao 卯, Chen 辰, Si 巳, Wu 午, Wei 未, Shen 申, You 酉, Xu 戌, Hai 亥) to form a 60-unit cycle. To calculate, convert the Gregorian date to its corresponding Heavenly Stem and Earthly Branch combination, used for marking specific years, months, and days.
The existing GPT models tend to make errors when performing calculations related to the Sexagenary Cycle. To address this issue, we have provided a specialized dataset aimed at improving the model's accuracy in handling Sexagenary Cycle calculations.
Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).
Your eval should be:
- Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
- Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
- Includes good signal around what is the right behavior. This means
either a correct answer for
Basicevals or theFactModel-graded eval, or an exhaustive rubric for evaluating answers for theCriteriaModel-graded eval. - Include at least 100 high quality examples (it is okay to only contribute 5-10 meaningful examples and have us test them with GPT-4 before adding all 100)
If there is anything else that makes your eval worth including, please document it below.
[2023-03-16 12:46:51,594] [record.py:320] Final report: {'accuracy': 0.65}. Logged to /tmp/evallogs/230316044112MFPWKX37_gpt-3.5-turbo_sexagenary_cycle_calculation.jsonl [2023-03-16 12:46:51,594] [oaieval.py:211] Final report: [2023-03-16 12:46:51,594] [oaieval.py:213] accuracy: 0.65 [2023-03-16 12:46:51,597] [record.py:309] Logged 60 rows of events to /tmp/evallogs/230316044112MFPWKX37_gpt-3.5-turbo_sexagenary_cycle_calculation.jsonl: insert_time=2.463ms
Your eval should
- Check that your data is in
evals/registry/data/{name} - Check that your yaml is registered at
evals/registry/evals/{name}.jsonl - Ensure you have the right to use the data you submit via this eval
(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)
By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).
- I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.
If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request.
- I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.
We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.
- I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted.
- I have filled out all required fields in the evals PR form
- (Ignore if not submitting code) I have run
pip install pre-commit; pre-commit installand have verified thatblack,isort, andautoflakeare running when I commit and push
Failure to fill out all required fields will result in the PR being closed.
Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:
View evals in JSON
{"input": [{"role": "system", "content": "You are a helpful
assistant."}, {"role": "user", "content": "Let's do a math problem
related to calculating dates using the Chinese Sexagenary Cycle
method."}, {"role": "assistant", "content": "Sure, let's do a math
problem using the Chinese Sexagenary Cycle. The Chinese Sexagenary
Cycle, or Stems and Branches, is a system used in China to count years,
months, days, and even hours. It is a 60-year cycle, consisting of 10
Heavenly Stems and 12 Earthly Branches."}, {"role": "user", "content":
"甲戌的次日是？"}], "ideal": "乙亥"}
{"input": [{"role": "system", "content": "You are a helpful
assistant."}, {"role": "user", "content": "Let's do a math problem
related to calculating dates using the Chinese Sexagenary Cycle
method."}, {"role": "assistant", "content": "Sure, let's do a math
problem using the Chinese Sexagenary Cycle. The Chinese Sexagenary
Cycle, or Stems and Branches, is a system used in China to count years,
months, days, and even hours. It is a 60-year cycle, consisting of 10
Heavenly Stems and 12 Earthly Branches."}, {"role": "user", "content":
"乙亥的次日是？"}], "ideal": "丙子"}
{"input": [{"role": "system", "content": "You are a helpful
assistant."}, {"role": "user", "content": "Let's do a math problem
related to calculating dates using the Chinese Sexagenary Cycle
method."}, {"role": "assistant", "content": "Sure, let's do a math
problem using the Chinese Sexagenary Cycle. The Chinese Sexagenary
Cycle, or Stems and Branches, is a system used in China to count years,
months, days, and even hours. It is a 60-year cycle, consisting of 10
Heavenly Stems and 12 Earthly Branches."}, {"role": "user", "content":
"壬寅的后日是？"}], "ideal": "甲辰"}
{"input": [{"role": "system", "content": "You are a helpful
assistant."}, {"role": "user", "content": "Let's do a math problem
related to calculating dates using the Chinese Sexagenary Cycle
method."}, {"role": "assistant", "content": "Sure, let's do a math
problem using the Chinese Sexagenary Cycle. The Chinese Sexagenary
Cycle, or Stems and Branches, is a system used in China to count years,
months, days, and even hours. It is a 60-year cycle, consisting of 10
Heavenly Stems and 12 Earthly Branches."}, {"role": "user", "content":
"癸卯的后日是？"}], "ideal": "乙巳"}
{"input": [{"role": "system", "content": "You are a helpful
assistant."}, {"role": "user", "content": "Let's do a math problem
related to calculating dates using the Chinese Sexagenary Cycle
method."}, {"role": "assistant", "content": "Sure, let's do a math
problem using the Chinese Sexagenary Cycle. The Chinese Sexagenary
Cycle, or Stems and Branches, is a system used in China to count years,
months, days, and even hours. It is a 60-year cycle, consisting of 10
Heavenly Stems and 12 Earthly Branches."}, {"role": "user", "content":
"壬子的后日是？"}], "ideal": "甲寅"}Co-authored-by: dunedainstrider dunedainstrider@mac16
Fix get_answer (#972)
🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨
PLEASE READ THIS:
In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task.
We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. Starting April 10, the minimum eval count is 15 samples, we hope this makes it easier to create and contribute evals.
Also, pelase note that we're using Git LFS for storing the JSON files, so please make sure that you move the JSON file to Git LFS before submitting a PR. Details on how to use Git LFS are available here.
[Insert Eval name here]
[Insert a short description of what your eval does here]
[Insert why this eval is worth including and any additional context]
Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).
Your eval should be:
- Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
- Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
- Includes good signal around what is the right behavior. This means
either a correct answer for
Basicevals or theFactModel-graded eval, or an exhaustive rubric for evaluating answers for theCriteriaModel-graded eval. - Include at least 15 high quality examples.
If there is anything else that makes your eval worth including, please document it below.
Insert what makes your eval high quality that was not mentioned above. (Not required)
Your eval should
- Check that your data is in
evals/registry/data/{name} - Check that your yaml is registered at
evals/registry/evals/{name}.yaml - Ensure you have the right to use the data you submit via this eval
(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)
By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).
- I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.
If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request.
- I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.
We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.
- I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted.
- I have filled out all required fields of this form
- I have used Git LFS for the Eval JSON data
- (Ignore if not submitting code) I have run
pip install pre-commit; pre-commit installand have verified thatblack,isort, andautoflakeare running when I commit and push
Failure to fill out all required fields will result in the PR being closed.
Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:
Adds a revolutionary conversion stinger (#75002)
Adds an antagonist gain stinger for Revolutionaries, created with inspiration from the obsessed/traitor conversion sounds.
2023-04-27.20-46-34.mp4
Raw audio:
https://cdn.discordapp.com/attachments/440978216484732934/1101964419203862548/revolutionary_tide.ogg
A distant, hypnotic whistling. The heavy footfalls and clamoring voices of an approaching crowd. The unstoppable revolutionary tide breaks its waves upon an unsuspecting station.
I wanted to try and make something that felt like it fit in with the other antagonist stingers we already have. Let me know what you think!
Gives a bit more flavor, and helps set the mood for the upcoming bloodbath.
🆑 sound: Revolutionaries now have their own stinger that plays upon becoming that antagonist. /🆑
Update README.md
Added much needed FAQ from the Discord to the README. As I was trying to figure it out forever and wouldn't have known unless someone told me to look at FAQ specifically in the Discord. Because to me it wasn't a FAQ kinda question. But if it is, it should be here to save you the trouble to be honest.
[MIRROR] Tcomms Soundloop Comes From One Source And Is Less Awful [MDB IGNORE] (#20713)
- Tcomms Soundloop Comes From One Source And Is Less Awful (#74908)
The soundloop/server now only comes from the server hub, so it
doesn't have stacking audio sources. The sound has been made more
uniform when up close, but is overall quieter. Additionally, all the
files have been run through a low pass filter to remove the highest of
it's pitches.
I'm sick of not wanting to be around telecomms because of how bad every single machine sounds. Now, things are significantly easier on the ear, quieter, more uniform, and better for everyone's sanity. I asked the maintainers in the coding channel if I could just remove it and they said no.
I can't get a video recording, I've tried with win+G, OBS, and sharex and it's just fucked.
🆑 qol: telecomms is quieter and less ear-damaging. sound: modified tcomms sound to remove high-tones. fix: the telecomms sound only comes from the server hub machine. /🆑
Co-authored-by: Mothblocks <35135081+Mothblocks@ users.noreply.github.com>
- Tcomms Soundloop Comes From One Source And Is Less Awful
Co-authored-by: Cheshify [email protected] Co-authored-by: Mothblocks <35135081+Mothblocks@ users.noreply.github.com>
[MIRROR] Turns Deer into Basic Mob - They Freeze At The Sight of Vehicles [MDB IGNORE] (#20711)
- Turns Deer into Basic Mob - They Freeze At The Sight of Vehicles (#74784)
deers only show up in the BEPIS but i decided that they would be easy enough to turn into a basic mob (they were). it was so easy in fact that i decided to dip my toes into coding AI behavior, and made them freeze up whenever they see a vehicle. this required a lot of code in a bunch of places that i was quite unfamiliar with before starting this project, so do let me know if i glonked up anywhere and i can work on smoothing it out.
one less simple animal on the list. deers staring at headlights is pretty cool i think, neato interaction for when you do get them beyond the joke the bepis makes
i'm also amenable to dropping the whole "deer in headlights" code if you don't like that for w/e reason- just wanted to make them basic at the very least
🆑 add: If you ever happen upon a wild deer, try not to ride your fancy vehicles too close to it as it'll freeze up like a... you know where I'm going with this. /🆑
Co-authored-by: Mothblocks <35135081+Mothblocks@ users.noreply.github.com>
- Turns Deer into Basic Mob - They Freeze At The Sight of Vehicles
Co-authored-by: san7890 [email protected] Co-authored-by: Mothblocks <35135081+Mothblocks@ users.noreply.github.com>
Ports mothroaches + Moth emotes (#1843)
Can you guess what this PR does? If you answered that it ports this pull request, this pull request, and a partial part of this one too, then you're right!
You can also craft moth plushies now. You just need some cloth, mothroach hide, and a heart!
silly little moth roaches and emotes, who wouldn't want them in the game?
🆑 add: Mothroaches are now a thing add: Moth laughter, chittering and squeaking /🆑
Replace potions of stabbing with attraction
Potions of stabbing didn't quite hit the mark as a design. They weren't particularly popular with players, perhaps because their niche wasn't just narrow but extremely narrow.
So, let's try another idea. Potions of attraction cause the player to magically pull nearby monsters toward them while the duration lasts, a bit like Mass Lesser Beckoning. (Greater Beckoning?). It currently pulls 3 tiles per turn, though I could see 2 tiles or full LOS also being reasonable. The idea here is for melee characters to have something to help them with ranged enemies, though quite situationally!
I also considered making this pull items, as a sort of very silly apportation variant. Maybe it should?
Right now this doesn't wake up sleeping enemies, since it seemed fun to have it help with stabs, too. If that's too strong, we can make it more annoying (to enemies, waking them).
Hotkey tweaks (#7956)
-
yeah
-
changes the hotkey list
-
I forgot to push this
-
Revert "I forgot to push this"
This reverts commit 845878d1bda9f8be1cee214acd7329b0355a507b.
- Revert "changes the hotkey list"
This reverts commit a1174c47bdc49245e4b31ddb06f85e7fec21e51c.
-
re-adds reversions
-
Revert "yeah"
This reverts commit e61f425a1231c6049c123724bfe88a7e51b9c199.
- manually adds hotkeys instead of using .dmf
I love the quirky dream maker language
Generate News in Sphinx From Fragments
The current way of generating the news for releases mostly consists of generating a spreadsheet of PRs, editing that, and creating the new entries from that manually. More info on that process here and here. News entries can be directly committed in the PR where the change is taking place, but doing that risks merge conflicts.
Overall this process is somewhat messy and limiting:
- Deciding what's newsworthy, what exactly to write, and reviewing the news is done all at once right before the release, sometimes months after the work was done. This makes it harder to remember what's newsworthy, what specific things needs to be pointed out to users, and what PRs should go together for single news item. This also means it takes more time to prepare the release and there's less time to spot and correct mistakes in the news or improve it.
- Most of the time the news item is left as just the title of PR. In the best case these might not need to be tweaked much or at all for changes that require little explanation. However this is certainly inadequate for explaining larger changes, for example like for the XTypes fixes from PR4078. It'd be very awkward to write that much in a spreadsheet.
- It's hard to link to documentation. This is better than it was before with the PDF devguide, when it was impossible, but this could still be improved on more. Linking would give more context to users and could immediately give them details on use a new feature.
- Contributors outside the OpenDDS maintainers basically have no direct input on what the news says for changes they contribute. Honestly I'm not sure if any have wanted to, but they couldn't if they did.
The solution in this PR is committing the news of changes of a PR as a file in that PR. At release these fragments of the news are brought together automatically. There are still a few kinks to iron out, but even if those are mostly unresolved I think this system will improve the quaility of the news.
The system is inspired by Python's blurb tool and to a lesser extent tools like towncrier. These tools are not bad, but they have some serious drawbacks. blurb is specifically tailored for CPython development. For example, it's oriented by GitHub issues, where as many of the changes we make are not prompted by a GitHub issue. towncrier really expects the project to be a Python project and has some quirks for some of use cases I was looking for. Specifically needing multiple identical files for to attribute a news item to multiple PRs and needing multiple files for a PR to have different kinds of changes. Also both rely on the files having a specific name, which seems unnecessary to me.
The following is the basics of adding a news fragment and how the news is generated in this system:
Create a rst file in docs/news.d/. This is a news fragment. It can be
named anything as long as it doesn't have an leading underscore and is a
rst file, but it should be somewhat descriptive. Naming it the same as
the branch the change is on might be good idea. The change must be a rst
list. It has to have some rst-directive-like metadata around it. A
minimal file looks like this:
.. news-prs: 5000
.. news-push: Additions
- This is an addition we made.
.. news-popAdditional PRs are added by appending them to end of the news-prs
line. Additional news-pushs and news-pops can be used to add list
items to other sections, like fixes, or to create nested sections for
groups of changes like like "XTypes" or "RTPS". All sections will be
merged together in the final result. These sections and items are sorted
first by a quality called rank, then by the PR numbers in reverse order
(basically chronological). The rank is changed by .. news-rand: RANK_NUMBER. It can be used to headline an important change or set of
changes,
See docs/news.d/_example.rst for a longer example. I also have added a
recreation of the 3.24.0 news as fragments as a test and a full example
of what this would look like.
Before release a preview of the news entry will always be available in
the built version of docs/news.rst. The means news added in an PR can
be previewed in the PR. During a release the fragments are permanently
committed to that file and the fragment files in docs/news.d are
removed.
Here are the two main issues I see with this system right now:
- To do a PR with a news fragment in one commit, you basically have to
know what the PR number is going to be before hand. Otherwise another
commit is needed to add the PR number. The PR number could technically
be manually added after the PR is merged, but I would consider that
poor practice. One solution could be a placeholder in
news-prstatement that an action automatically replaces with the PR number after the PR is merged. - How does this relate/integrate with
NEWS.mdand the GitHub release notes? I'm honestly a little stumped by this and unlike the other issue this needs to be figured out before this can be merged.- Something like pandoc could be used to convert the rst, but it would still need some manual intervention based on tests I did with the 3.24.0 news.
- The markdown version could just be a summarized version of the news, mostly consisting of highlights. This could be manually done or done with pandoc with human intervention. Also this summery could be what goes out in a prerelease announcement on social media.
- The
NEWS.mdfile could be also be done away with to simplify things. If that's the case, shuold news.rst live in the root directory and be calledNEWS.rst? Is that going to case problems to try to include it in Shpinx? - The GitHub release notes could just link to
news.rst, but I feel like they probably should at least have a summery.
Besides that there are some more things I needs to do, specifically:
- Document either in the documentation guidlines or dev guidelines how to add to the news.
- Improve release entries, it needs the release date and output could be tweaked.
- Maybe add two smaller examples just for "Additions" and "Fixes" without comments that are eaiser to use as templates.
- Before merge, remove 3.24.0 fragments, add any newer releases, and add any news fragments for a pending release.
[MIRROR] Tones down the power of nitrous oxide, the reagent. Makes heparin a bit harder to fix to compensate [MDB IGNORE] (#20610)
- Tones down the power of nitrous oxide, the reagent. Makes heparin a bit harder to fix to compensate (#74703)
Nitrous Oxide, rather than directly subtracting blood volume per tick, instead applies the anticoagulant trait ''TRAIT_BLOODY_MESS''. It shares this with heparin.
However, unlike, heparin, coagulants like Sanguirite will remove the trait and allow for continued clotting while the reagent is present, neutering the nitrous oxide's anticoagulant effects (but not the other parts)
Heparin, on the other hand, will purge Sanguirite while it is in you system. You must remove the heparin before you can apply an anticoagulant.
Nitrous Oxide, on top of being a knockout chem that causes you to suffocate and become drowsy, just starts deleting blood rapidly. About 15 units of it, standard in a syringe, will kill you in about a minute, but you'll be unconscious for most of it (you'll be at around 50%-60% blood by the time it is out of your system, so as good as dead). It doesn't matter that it metabolizes quickly either, since because it isn't a toxin, it doesn't get purged by livers/improved livers, so you will probably metabolize all of the chem that enters your system.
Blood is one of those 'hidden damage types', a bit like brain damage. Once you start losing it directly, you probably don't have a lot of options to resolve it (unlike a bleed, which you do in various manners), and the means of causing blood loss has been mostly pretty well controlled as of late. Heparin directly interacts with wounds as a good example.
Blood loss is also tied into oxyloss, which is another very mean damage type in that it causes you to fall into unconsciousness at 50 damage, so significantly more lethal than normal damage, kept in check by the fact that breathing restores some of it. Nitrous oxide, you might note, causes you to stop breathing.
It's cheaper to make than either heparin or lexorin, and since it isn't a toxin like those chems, it is able to circumvent a few game mechanics to simply just start killing you. It does the work of chloral hydrate, lexorin and heparin while it has a remarkably easy recipe.
Following the example of how lexorin was pulled into line, and consistency with heparin, I've made nitrous oxide an anticoagulant that may or may not come into play as one. I think this is more up to date with the state of toxins and chem warefare as a whole, and works with the relative ease in making nitrous oxide. And it has been this way for about 5 years, before we got wounds.
(did I mention that nitrous oxide is also an explosive if it gets hot enough?)
I think a chem with a pretty basic recipe shouldn't be doing the work of 5 other, more complicated, chemicals while also not being a toxin and also not interacting with the wounds system or purity system whatsoever. And being an explosive.
🆑 balance: Nitrous oxide, the reagent, increases bleed rate from wounds rather than directly subtracting blood. It can be counteracted using coagulants (such as those in epipens). balance: Heparin purges coagulants. You have to remove heparin from someone's system before you can use coagulants. /🆑
Co-authored-by: san7890 <the@ san7890.com>
- Tones down the power of nitrous oxide, the reagent. Makes heparin a bit harder to fix to compensate
Co-authored-by: necromanceranne [email protected] Co-authored-by: san7890 <the@ san7890.com>
🔧 chore(next.config.mjs): add remotePatterns to images configuration and add ts and tsx to pageExtensions ✨ feat(Card.tsx): add Card component with subcomponents Link, Title, Description, Cta, and Eyebrow 🔧 chore(Header.jsx): remove unused import and move avatarImage to a variable 🔧 chore(SimpleLayout.tsx): add margin bottom to Container component 🔧 chore(useClickOutside.tsx): refactor useClickOutside hook to use a single listener function ✨ feat(ArrowDownIcon.tsx): add ArrowDownIcon component ✨ feat(BriefcaseIcon.tsx): add BriefcaseIcon component ✨ feat(ChevronRightIcon.tsx): add ChevronRightIcon component The changes include adding support for remotePatterns in the images configuration in next.config.mjs, adding a Card component with subcomponents Link, Title, Description, Cta, and Ey
✨ feat: add LinkIcon and MailIcon components 🗑️ chore: remove unused image files and _app.jsx file 🚚 chore: rename _app.jsx to _app.tsx and update its implementation to use TypeScript 🚚 chore: rename _document.jsx to _document.tsx and update its implementation to use TypeScript The commit adds two new components, LinkIcon and MailIcon, which can be used throughout the application. It also removes unused image files and the _app.jsx file, which is replaced with a TypeScript implementation in _app.tsx. Similarly, _document.jsx is replaced with a TypeScript implementation in _document.tsx.
🎨 style(about.tsx): format code to improve readability ✨ feat(about.tsx): add About page with personal and professional information The About page includes personal and professional information about the author, including his career journey, skillset, and solutions he can provide. The page also includes links to the author's social media profiles and email address. The code has been formatted to improve readability.
🐛 fix(index.jsx): replace img tag with Next.js Image component The img tag has been replaced with the Next.js Image component to improve performance and accessibility.
✨ feat(hermes.tsx): add Hermes chat page with Open AI chat The Hermes chat page includes an Open AI chat component that can help users with anything.
🎨 style(index.jsx): replace image imports with Image component and add height and width attributes 🎨 style(index.jsx): replace custom icons with imported icons 🚀 feat(projects.tsx): add Projects page with project cards The changes in index.jsx improve the performance of the page by using the Next.js Image component instead of importing images directly. The imported icons improve the consistency of the icons used throughout the application. The new Projects page displays a list of projects with their logos, descriptions, and links. This page allows visitors to see the projects that the author has worked on and provides links to view them.
✨ feat(tech.tsx): add Tech Stack page to showcase the technologies used in projects This commit adds a new page called Tech Stack that showcases the technologies used in projects. The page displays a list of technologies along with their logos, descriptions, and links to their respective websites. The page is designed to give visitors an idea of the technologies I am familiar with and have experience using.
✨ feat(thank-you.tsx): convert thank-you page from jsx to tsx 🚀 feat(uses.tsx): add new page for "Uses" with a list of tools and resources used by the author The thank-you page was converted from jsx to tsx to improve type safety and maintainability. The new "Uses" page was added to list the tools and resources used by the author for development, design, and productivity. The page is divided into sections with each section containing a list of tools and a brief description of why they are used.
http: support CURLOPT_PROTOCOLS_STR
The CURLOPT_PROTOCOLS (and matching CURLOPT_REDIR_PROTOCOLS) flag was deprecated in curl 7.85.0, and using it generate compiler warnings as of curl 7.87.0. The path forward is to use CURLOPT_PROTOCOLS_STR, but we can't just do so unilaterally, as it was only introduced less than a year ago in 7.85.0.
Until that version becomes ubiquitous, we have to either disable the deprecation warning or conditionally use the "STR" variant on newer versions of libcurl. This patch switches to the new variant, which is nice for two reasons:
-
we don't have to worry that silencing curl's deprecation warnings might cause us to miss other more useful ones
-
we'd eventually want to move to the new variant anyway, so this gets us set up (albeit with some extra ugly boilerplate for the conditional)
There are a lot of ways to split up the two cases. One way would be to abstract the storage type (strbuf versus a long), how to append (strbuf_addstr vs bitwise OR), how to initialize, which CURLOPT to use, and so on. But the resulting code looks pretty magical:
GIT_CURL_PROTOCOL_TYPE allowed = GIT_CURL_PROTOCOL_TYPE_INIT; if (...http is allowed...) GIT_CURL_PROTOCOL_APPEND(&allowed, "http", CURLOPT_HTTP);
and you end up with more "#define GIT_CURL_PROTOCOL_TYPE" macros than actual code.
On the other end of the spectrum, we could just implement two separate functions, one that handles a string list and one that handles bits. But then we end up repeating our list of protocols (http, https, ftp, ftp).
This patch takes the middle ground. The run-time code is always there to handle both types, and we just choose which one to feed to curl.
Signed-off-by: Jeff King [email protected] Signed-off-by: Junio C Hamano [email protected] Signed-off-by: Johannes Schindelin [email protected]
coredump: require O_WRONLY instead of O_RDWR
The motivation for this patch has been to enable using a stricter apparmor profile to prevent programs from reading any coredump in the system.
However, this became something else. The following details are based on Christian's and Linus' archeology into the history of the number "2" in the coredump handling code.
To make sure we're not accidently introducing some subtle behavioral change into the coredump code we set out on a voyage into the depths of history.git to figure out why this was O_RDWR in the first place.
Coredump handling was introduced over 30 years ago in commit ddc733f452e0 ("[PATCH] Linux-0.97 (August 1, 1992)"). The original code used O_WRONLY:
open_namei("core",O_CREAT | O_WRONLY | O_TRUNC,0600,&inode,NULL)
However, this changed in 1993 and starting with commit 9cb9f18b5d26 ("[PATCH] Linux-0.99.10 (June 7, 1993)") the coredump code suddenly used the constant "2":
open_namei("core",O_CREAT | 2 | O_TRUNC,0600,&inode,NULL)
This was curious as in the same commit the kernel switched from constants to proper defines in other places such as KERNEL_DS and USER_DS and O_RDWR did already exist.
So why was "2" used? It turns out that open_namei() - an early version of what later turned into filp_open() - didn't accept O_RDWR.
A semantic quirk of the open() uapi is the definition of the O_RDONLY flag. It would seem natural to define:
#define O_RDWR (O_RDONLY | O_WRONLY)
but that isn't possible because:
#define O_RDONLY 0
This makes O_RDONLY effectively meaningless when passed to the kernel. In other words, there has never been a way - until O_PATH at least - to open a file without any permission; O_RDONLY was always implied on the uapi side while the kernel does in fact allow opening files without permissions.
The trouble comes when trying to map the uapi flags onto the corresponding file mode flags FMODE_{READ,WRITE}. This mapping still happens today and is causing issues to this day (We ran into this during additions for openat2() for example.).
So the special value "3" was used to indicate that the file was opened for special access:
f->f_flags = flag = flags;
f->f_mode = (flag+1) & O_ACCMODE;
if (f->f_mode)
flag++;
This allowed the file mode to be set to FMODE_READ | FMODE_WRITE mapping the O_{RDONLY,WRONLY,RDWR} flags into the FMODE_{READ,WRITE} flags. The special access then required read-write permissions and 0 was used to access symlinks.
But back when ddc733f452e0 ("[PATCH] Linux-0.97 (August 1, 1992)") added coredump handling open_namei() took the FMODE_{READ,WRITE} flags as an argument. So the coredump handling introduced in ddc733f452e0 ("[PATCH] Linux-0.97 (August 1, 1992)") was buggy because O_WRONLY shouldn't have been passed. Since O_WRONLY is 1 but open_namei() took FMODE_{READ,WRITE} it was passed FMODE_READ on accident.
So 9cb9f18b5d26 ("[PATCH] Linux-0.99.10 (June 7, 1993)") was a bugfix for this and the 2 didn't really mean O_RDWR, it meant FMODE_WRITE which was correct.
The clue is that FMODE_{READ,WRITE} didn't exist yet and thus a raw "2" value was passed.
Fast forward 5 years when around 2.2.4pre4 (February 16, 1999) this code was changed to:
- dentry = open_namei(corefile,O_CREAT | 2 | O_TRUNC | O_NOFOLLOW, 0600);
...
+ file = filp_open(corefile,O_CREAT | 2 | O_TRUNC | O_NOFOLLOW, 0600);
At this point the raw "2" should have become O_WRONLY again as filp_open() didn't take FMODE_{READ,WRITE} but O_{RDONLY,WRONLY,RDWR}.
Another 17 years later, the code was changed again cementing the mistake and making it almost impossible to detect when commit 378c6520e7d2 ("fs/coredump: prevent fsuid=0 dumps into user-controlled directories") replaced the raw "2" with O_RDWR.
And now, here we are with this patch that sent us on a quest to answer the big questions in life such as "Why are coredump files opened with O_RDWR?" and "Is it safe to just use O_WRONLY?".
So with this commit we're reintroducing O_WRONLY again and bringing this code back to its original state when it was first introduced in commit ddc733f452e0 ("[PATCH] Linux-0.97 (August 1, 1992)") over 30 years ago.
Signed-off-by: Vladimir Sementsov-Ogievskiy [email protected] Message-Id: [email protected] [[email protected]: completely rewritten commit message] Signed-off-by: Linus Torvalds [email protected] Signed-off-by: Christian Brauner [email protected]
PR3 Maze Solver
- Introduction and Background Introduction
Solving a Maze
In this programming assignment you will write a program to read a given maze (provided as text “art”) and find the shortest path from start to finish. Mazes go back to antiquity and the story of the minotaur. However, Theseus didn’t have Google Maps. You will implement breadth-first search, a simple algorithm that finds not just any path to the exit in a maze, but actually finds a shortest path from the start to the finish. This algorithm was described in the late 1950s, making it one of the earliest nontrivial algorithms to follow the creation of the electronic computer. In this assignment you will practice the following skills: Dynamic memory allocation of arrays and 2D arrays. Using structs and classes . Implementation of a simple queue data structure. Implementing the breadth-first search (BFS) algorithm. A Piece of Advice! We suggest you read through the entire description before beginning the assignment. Maze Input Format
Your program will receive the input file from a command line argument. Below shows an example of passing in a maze input file called “maze1.in” to our maze program using a command line argument. ./maze maze1.in The input file is structured as follows:
The first line of the file should contain two integer numbers indicating the row and column size of the maze. The number of rows indicated will determine how many lines of text follow (1 row per line). On each line will be one character for each of the indicated number of columns followed by a newline character. The characters can be a period . indicating a space in the maze, a # sign indicating a wall in the maze, a capital S indicating the start location for your search, or a capital F for the desired finish location. You can’t go outside the grid. (i.e., you may imagine that lava surrounds the maze perimeter.) In general the format of the maze is: rows cols [row 1: cols many characters] [row 2: cols many characters] ... [row rows: cols many characters] and the possible characters are: Character Meaning . Empty space
S Start location F Finish location We provide you with 4 mazes: maze[1-4].in and maze1.in is shown as an example on the left. You can also find some interesting mazes in a directory called testcases. Maze Output Format
Your search algorithm will find a shortest path from the start cell to the finish. It will indicate this path by filling in the character locations on the path with asterisks ; then, it will write the resulting character maze to cout (standard output). Note: We could have chosen to output the results to a file to practice with ofstream objects, but because they behave nearly the same as cout, we chose the simpler cout approach so you can focus on the really learning tasks of allocating the 2D array(s) and implementing the BFS algorithm. The correct output for maze1.in (below) is shown on the left (maze1.sol). maze1.in 5 5 .S.#. ##.#. ..... .#### ....F maze1.sol (this is the content that cout should produce) 5 5 .S#. ##*#. ***.. *#### ****F However, If no path exists, your program should output No path could be found! If the user gives an invalid maze (more than one start or finish point), your program should output Invalid maze. If the input maze contains characters other than those shown above, then your program should output Error, input format incorrect. BFS
Breadth First Search is a general technique with many uses including flood fill, shortest paths, and meet-in-the-middle search. The idea is to explore every possible valid location, beginning at the start location, using an ordering so that we always explore ALL close locations (i.e. those at a shorter distance from the start) before exploring any location at a further (longer) distance from the start. In other words, first we “explore” the start, then all locations at distance 1 from the start (one at a time), then all locations at distance 2 from the start (one at a time), etc, in that order. This property ensures that when we do find the finish cell (if one exists), we will arrive there via a shortest-length path. Of course, as we search we mark cells (as explored) that we’ve visited so that we don’t visit them again and cause an infinite cycle of exploration! Ensuring the BFS propety
To ensure we explore all closer locations before further locations, we keep a queue of locations in the maze. A queue has the property that the first item added is the first to be removed (first-in, first-out, a.k.a. FIFO). We can easily implement this by always adding new values to the back of a list (i.e. tail of the queue) but remove from the front (i.e. head of the queue). Put simply, a queue mimics a line of people waiting. New arrivals enter at the back and leave from the front when they are served. The longest/oldest item will be at the front of the queue and the newest at the back. In our maze search the queue is initially empty and we begin by putting the start location into it. In each iteration, we remove the oldest remaining location from the queue, and we add all of its new neighbors to the end of the list. This simple algorithm successfully implements the BFS. We keep the BFS going until the queue is exhausted because all reachable locations were explored or we find our goal location. What is left is to actually mark out the shortest path in * characters (i.e., trace your path back from the finish point to start point). But this requires that we record which location found each subsequent location during the outward (BFS) exploration. We will track a predecessor for each location (which is simply the coordinates of the location that found this one). Think of a predecessor as a pointer to take one step backward. Once we find the Finish location, we can just walk from the predecessor of the Finish to that location’s predecessor, and then to that location’s predecessor, and so on until we reach the start, marking each of those locations with *. Coding Considerations
Location Struct
The maze in this assignment is a 2D grid. Therefore, each location in the maze is identified by two index values: a row index and a column index. The approach that we will use for solving the maze will involve keeping track of a list of these locations. It will be extremely convenient if we are able to create a structure that contains both a number for the row and a number for the column.
We can accomplish this by creating a simple new data type.
We will call it Location.
struct Location { // define a new data type
int row;
int col;
};
Put Simply
A Location object is simply a package with two numbers that can be manipulated using the dot (membership) operator.
Here is an artificial example.
Location start;//create a location called 'start'
start.row = 3 ;//set the row of start to be 3
start.col = 5 ;//set the column of start to be 5
Location one_below = start; // make a copy of the start location called one_below
one_below.row += 1 ;//add one to the row of one_below
cout<<"Start:"<< start.row <<","<< start.col <<endl; // Start: 3,5 cout<<"One Below:"<< one_below.row <<","<< one_below.col <<endl; // One Below: 4,5 This will be convenient because we can have an array (or queue) of Locations, have functions that take a Location as input or output, etc. In the next page, we will talk about another data type that you will have to complete, the Queue class. Question that you should be asking yourself: What are the differences between a struct and a class? When should I use a class vs a struct? The Queue class
You will be required to complete the definition of another new data type, the Queue class, that will be useful in implementing the search. The Queue class will store the list of Locations waiting to be searched. The Queue class should support the following operations: create an empty queue (but with a given maximum capacity of how many Locations it can store) add a new Location to the back of the queue Remove the oldest Location from the front of the queue check if the queue is empty We use objects (classes and structs) to abstract our software design and implementation. By first writing the Queue class, you do not have to worry about those implementation details when you write your higher level BFS algorithm. This allows you to focus on the larger BFS algorithm and not the data management aspects. Every time you find a new, unexplored Location you will call the Queue's add_to_back(Location) function. Every time you want to get the next location to explore from, you will call the Queue's remove_from_front() function. Internally, the Queue class should just create an array to hold the maximum number of Locations that could ever be entered into the queue and use integer index variables to remember where it should place new values (i.e. where the back is located) and where it should remove old values (i.e. where the front is located). Remember that our Queue class needs to support the following operations: // returns true if the queue is empty. bool is_empty();
// insert a new Location at the end of our queue
void add_to_back(Location loc);
// returns the oldest Location in the queue and "removes" it. Location remove_from_front(); Here is an example of how it should behave: // create some locations; Location three_one; three_one.row = 3; three_one.col = 1; Location two_two; two_two.row = 2; two_two.col = 2;
// create a queue with max capacity 5 Queue q(5);
cout << boolalpha; cout << q.is_empty() << endl; // true q.add_to_back(three_one);
cout << q.is_empty() << endl; // false q.add_to_back(two_two);
Location loc = q.remove_from_front(); cout << loc.row << "," << loc.col << endl; // 3,1
loc = q.remove_from_front(); cout << loc.row << "," << loc.col << endl; // 2,2
cout << q.is_empty() << endl; // true As you can see, the queue gave us back the oldest location (the one add_to_back added earliest) first. The Queue class under the hood
The Queue implementation we provide you is almost complete. It is based on the idea of using a long array/pointer called contents that holds Locations, as well as two counters: tail counts the number of add_to_back calls so far. Equivalently, its value is the next unused index in the contents array. head counts the number of remove_from_front calls so far. Equivalently, its value is the oldest index that has not yet been extracted. For instance, this is what the internal variables of the Queue should look like for the example above before the two locations are added: tail: 0 head: 0 contents[0..4]: garbage Here is what it looks like after both locations are added: tail: 2 head: 0 contents[0]: Location(row 3, col 1) contents[1]: Location(row 2, col 2) contents[2..4]: garbage currently, head is 0. When we make the first call to remove_from_front, it should both return the Location at contents[head] which is Location(row 3 , col 1) and increment the head counter by one. head is now 1. After that, when we make the second call to remove_from_front, it again should return the Location at contents[head] which is Location(row 2, col 2) and increment head counter by one. After that, because head and tail are now equal, the Queue knows it is empty. Tip You may have to create a temporary variable to hold the Location at contents[head] so that you can increment the head counter before returning that Location. This is because any statement that comes after a return statement will not be executed. Note: when you delete a Location from the queue (remove_from_front), you do not technically remove the Location. You simply move the head counter forwards (i.e., incrementing the head counter and leaving the old Locations sitting in their original Locations). Actually, the expense of removing a Location and shifting other Locations forward would make your program much slower. Here are some additional notes about the Queue class: The Queue constructor takes an integer max_size. For our BFS application, you should pass rowscols as this maximum size, since that it the maximum number of locations that our queue could be used to explore. Hint: When you write your code do you know the exact value of rowscols? If not, how should you allocate that array? the constructor syntax is Queue q(5); similar to how a file stream is created. C++ calls the destructor automatically. We should not see the code ~Queue() anywhere in maze.cpp. 2D Array Allocation
You will not know the size of the maze until runtime, when you read the maze data. Thus we will need to dynamically allocate an array to hold the maze data. Remember that a single call to new can only allocate a 1D array. Thus you will need to allocate each row of the 2D array as a separate 1D array. And, beyond that, we need not just one 2D array but several: one for the maze data, one for the predecessors, and one to remember which cells have been visited. The way to allocate a 2D array in C++ is: use new[] once to allocate a 1D array of pointers, then using a loop containing new[] to allocate many 1D arrays whose starting addresses are stored in the pointer array elements. See the dynamic memory exercise nxmboard and deepnames on the website (in class exercises). Memory leaks Remember that you need to deallocate everything that you allocate. This means that every call to new[] needs a matching call to delete[]. Otherwise, your program will have a memory leak! Typically, the pattern you use to allocate is the same pattern you will use to deallocate, but just in reverse order.
Healer Balance Changes (#2896)
This pull request addresses the boring and low-risk gameplay of the Healer drone, who spends half the round sitting next to recovery nodes and recovering her health so she may use it again, rinse and repeat until a rine notices said drone has purple on it and booms her.
First, by changing her health from 600 to 500, Healer can spend more time healing her sisters than sitting through another 100 health to heal herself. Though this makes her less tanky than before, healing classes are not known to be tanks. To ensure Healer can still heal five times without depleting too much of her health whilst still giving her sisters a decent amount of heals, I made her ability cost 75 health instead of 100, and also made her ability cost 200 plasma. Since Healer replenishes plasma much more quickly than her health, she can still put herself into crit if she heals too frequently. Due to this buff, her heals had a slight nerf, being 10 damage a second for ten seconds instead of 12 damage per second for ten seconds for a total of 20 less damage healed per application overall.
In addition to these changes, I'm giving Healer a better plasma transfer for when she has nobody else to heal/nowhere else to weed and she has an opportunity to assist her sisters. While a normal drone transfers 50 plasma with a delay of 20, Healer transfers 100 with a delay of 15, which is nowhere near Hivelord's gargantuan 200 plasma with a delay of 5, but it still is better than a normal drone.
Finally, to give the huggers and larva some love, Healer will specifically heal little ones 1.5 health per second for 10 seconds for 15 of her own health and 30 plasma.
Healer drone isn't fun. You run around and heal a bunch of T3s, then sit out for half the battle trying to heal that massive 600 heath while you wonder why you take so long to heal even though you have Strong pheromones. You cry to mom for help, but she doesn't have time to heal a drone who can't build walls and has no need to weed at the moment. You think, 'screw it, I'm going to make a recovery node and camp here until I heal', but by the time you finish healing, several T3s and a silly rouny just suicided into a wall of talls and destroyed your recovery node, so you run off and make another one. But oh, someone noticed you have purple on your carapace and decide your location is precisely where a shell should land, right as you're building one.
No more. These changes allow Healer to move around at her leisure and makes Healer more engaging by allowing her to be a more front-line participant and actively run around and heal her sisters without having to incur such a harsh penalty.
Let this be a testmerge, please.
🆑 Puckaboo2 balance: Healer Drone's health was reduced to 500 from 600. balance: Healer's damage has been increased to 17 from 12 and the tackle damage debuff has been halved. balance: Healer Drone's Apply Salve ability now costs 75 health and 200 plasma, down from 120 health and up from 0 plasma. balance: Healer Drone's Apply Salve ability now heals 10 damage per second for 10 seconds, down from 12 damage per second for ten seconds. balance: To prevent spam healing between Healers, Apply Salve costs 100 health instead of 75 health when Healer heals another Healer. Much healing. balance: Healer has an improved Transfer Plasma that gives 100 plasma instead of 50, with a 25% shorter delay. balance: Healer will heal huggers and larva for 1.5 health a second for 10 seconds, costing 15 health and 30 plasma. tweak: Healer will now face the xeno she is healing if she was not facing their direction before. spellcheck: All instances of VERYSMALL and VERYLARGE have been renamed to VERY_SMALL and VERY_LARGE. /🆑
Co-authored-by: Morrow [email protected]
Fuck this absolute garbagde shit I hate my life and I'm going to be commiting suicide within the next week
Refactors High Luminosity Eyes, fixes a ton of bugs related to it as well as qol improvements (#75040)
The high luminosity eyes item was extremely out of date, broken, and full of copy paste code from atom lighting. Which is a shame because they were cool.
On top of all that it was using a special light effect that was not very performant. I got rid of all that, hooked it into atom lighting as a new light type, and gave it a new TGUI as well because the old ui prompts were not great either.
You can now pick a color at random if you want. You can also set the color and range before surgically implanting them. No more being forced to go through the color picker when you just want to change the range.
Functionally they should pretty much should be the same as before with some bonus features (see below).
Closes tgstation/tgstation#61041 Closes Skyrat-SS13/Skyrat-tg#14685
This is 100% completed. I just finished fixing the slight translation bug when going from 0->1 range (see above gif) and that was the last thing on my bucket list. I happy enough with this at this point in time.
EDIT:
I have decided to add in one last new feature, and that is... independent settings for eye color.
You can now set eye color independently if you wish
The eye color does not modify the light color in any way when set in this manner, but it can be used for cosmetic purposes.
Kind of makes the item more like cybereyes from cyberpunk, which I think are pretty neat!
- refactored high luminosity eyes so they use the atom lighting system instead of the way they were doing it before
- the new light type,
MOVABLE_LIGHT_BEAMbehaves similarly to directional lights, with some slight differences. it reuses the same lighting overlay sprites but scales them vertically to produce a more focused effect. The result can be seen above. This is in contrast to the old way, which spawnedrangenumber of individual 32x32 overlays and moved them around. This way should perform better as well as be more maintainable. - added a new TGUI interface for high luminosity eyes with buttons for range, a text field for a color hex, a color picker and randomizer
- made the eye overlay emissive when the light is turned on
- range goes from 0 to 5. at range 0, the light overlay is turned off and you are left with just the emissive eyes.
- added a cosmetic functionality to this item that lets you change the color of your eyes independently of the lighting (and each other)
- fixed a bug with directional flashlights sometimes not updating their lighting overlay if you pick them up without changing your direction
Being able to dynamically set range back and forth exposed some logic issues that had existed with directional light overlays and I have fixed those. That is why there are some edits in that file that may not appear readily obvious why they are there.
Apart from that, two other bugs that come to mind:
- eye code was supposed to keep track of the eye color you had before you got new eyes, but it was overwriting that every time you ran refresh().
- lighting was supposed to be turning off the light when range is set to 0, but it was not doing that properly.
And of course besides that, there may have been a few instances of finding typos/tidying up code
The code for this was like 6 years old and in desperate need of updating. Now it works, and has a nicer UI.
🆑 fix: high luminosity eyes light overlays now follow the user correctly qol: high luminosity eyes now have a tgui menu so you no longer have to go through the color picker every time you want to change the range. they also have a new setting that lets you change the color of your eyes independently of the light color. You can now have cybernetic heterochromia if you want fix: directional flashlights when picked up will now always update their cast light direction correctly no matter what dir you are facing refactor: refactors high luminosity eye code to better make use of the atom lighting system, adding a new light type 'MOVABLE_LIGHT_BEAM' /🆑
Nerfs the firing speed of Syndicate/Russian mobs (#75247)
Nerfs their firing speed from once every 0.2 seconds to once every 2.5 seconds
- The mob that is NOT a 'burst' type mob, is firing every 0.2 seconds. Kinda defeats the point of having them as two separate mobs, so this aims to fix that.
- Russian mobs. Turns out that letting them fire their strong-ass gun every 0.2 seconds was kinda not a good idea. I had assumed making them a Syndicate mob would be safe, and it technically is, it's just that Syndicate mob is fucked itself.
🆑 balance: Default Syndicate and Russian gunners now fire every 2.5 seconds instead of every 0.2 /🆑
Update Pb2Chts
Pb2Chts_2.3.8:
- Fixed script crash caused by well... typo :laugh:
- Fixed Floodspawn Lock camera entity ID cheat didn't get applied because querying incorrect variable.
- Changed God modes disable noise (lets hope it won't make weird noise anymore).
Merge pull request #8 from lpeapnni/OMFG-PASS-THE-FUCKING-TESTS
PASS THE FUCKING CI TESTS YOU STUPID PIECE OF SHIT
Checking in various improvements done in the last few evenings
Replace store_as_base with generated aggregate bases - better fix for #336, thanks @JohelEGP for the suggestion! This way we also don't need to obfuscate the name anywhere beyond the constructor(s), as the right member object name just enters the class's scope
If the user didn't write a constructor, provide a default constructor
If the user didn't write a 'that' constructor, suppress Cpp1's compiler-generated copying and assignment
Clean up emission of the just-mentioned generated/=deleted constructors, more naturally line up inside the class body following the indentation for other members that the original source code uses
Rename file load.h to io.h (file.h was another candidate), just because it has been bothering me for a year now that except for that one file all the headers were in natural alphabetical order by compilation phase... as of this commit we now have them all in alpha order and phase order: common.h -> io.h -> lex.h -> parse.h -> [*] -> sema.h -> cppfront.h
[*] coming next here: reflect.h, which will also be in both alpha order and compilation order
Guard out.construct with if constexpr in case the type is not copy assignable and that path is never requested
Rename cpp2::error to cpp2::error_entry to quiet a new(? why?) GCC message about shadowing the former name with parser::error... I get why the warning is there, but this is a slightly annoying warning to have to satisfy just to compile high-warning-clean on GCC... oh well
Change semantics of emitting .h2 files: In -p pure mode do the existing split of phases 0 and 1 into .h and phase 2 into a separate .hpp, but in mixed mode put all phases into the .h
A bunch of random item reworks: 3 (#65532)
-
boltcutters & chopsticks
-
tin snips
-
knitting needles
-
silly notes
-
hairbrush
-
flags oh god the flags
-
evil commas
-
density list removal
v2023.04
NOTE: This is a release changelog. It is composed of a selected short list of highlights since the last release [[mitk/changelog/release-v2022.10 | MITK v2022.10]] - split into dedicated user and developer sections. See the past three [[mitk/changelog | changelogs]] starting from [[mitk/changelog/2023.11]] for a comprehensive, developer-centric overview of changes.
= News for MITK Workbench users =
For MITK v2023.04, we focused mainly on segmentation again and we are already excited about your feedback on many improvements, features and a further streamlined segmentation workflow.
== Segmentation ==
==== TotalSegmentator AI segmentation tool ====
We introduced a segmentation tool based on the popular [[https://github.com/wasserth/TotalSegmentator | TotalSegmentator]], a fully automatic AI-based segmentation of 104 classes in CT images.
Install and set up TotalSegmentator from within the MITK Workbench with only a few clicks to perform fully automatic segmentations of your CT images.
The MITK Workbench even allows you to apply TotalSegmentator for all time steps of dynamic 4-d CT images.
See the {icon arrow-right} [[https://docs.mitk.org/2023.04/org_mitk_views_segmentation.html#org_mitk_views_segmentationTotalSegmentator | user guide]] for more details on the TotalSegmentator AI segmentation tool.
==== New label management ====
Up until now we organized labels of multi-label segmentations in layers, which allowed you to have overlapping labels in a single segmentation image. However, switching between layers was a bit clunky from time to time.
We completely reworked and optimized the Segmentation plugin to now have everything at sight at once, so you can easily switch between labels of different layers, which are now called groups.
See the {icon arrow-right} [[https://docs.mitk.org/2023.04/org_mitk_views_segmentation.html#org_mitk_views_segmentationgroups | user guide]] for more details on groups and labels.
==== Instance segmentation ====
You now can have multiple instances of the same label to better facilitate the segmentation of distriubted entities like metastases or other pathological sites.
See the {icon arrow-right} [[https://docs.mitk.org/2023.04/org_mitk_views_segmentation.html#org_mitk_views_segmentationlabelinstances | user guide]] for more details on label instances.
== MxN Display/Editor ==
In addition to the default 4-window-display, we have introduced a customizable display that can be adapted to your needs. You can change the amount of windows, their view directions, and even which data is shown in each individual window. Windows can be synchronized with each other to show the same data, or set to show an individual selection. Similarly, you can navigate through each window individually, or synchronize their movement to e.g. scroll through all windows at once. After you have created a certain window layout, you can save it and load it again later.
See the {icon arrow-right} [[https://docs.mitk.org/2023.04/org_mitk_editors_mxnmultiwidget.html | user guide]] for more details on the MxN Display/Editor.
== VP9/WebM support in Movie Maker ==
You can now choose your preferred video output format in the Preferences (Ctrl+P) for videos recorded with the Movie Maker plugin:
- VP9 (*.webm): Open, royalty-free, highly efficient and modern video codec.
- H.264 (*.mp4): The de-facto standard for video in the web for many years. Use for higher compatibility with old devices or software.
== High DPI support ==
The MITK Workbench now works fine with multi-monitor setups that have different DPI settings and scale its user interface more appropriately at high DPI settings.
== 💔 Known issues ==
- The Live Wire and Lasso segmentation tools may cause trouble in free-draw mode when drawing beyond image boundaries.
- While the segmentation interpolation now works also with multi-label segmentations, it is currently not considered to be completely robust against all edge cases. Use carefully and consider to save your project beforehand.
= News for developers =
While we didn't touch any of our third-party dependencies this time, we have a few API-breaking changes that may affect you.
== Segmentation ==
The revamp of the Segmentation View is based on a transition from the legacy LabelSetImage class towards a more flexible and easier to use MultiLabelSegmentation class. The transition is not yet finished but will be accompanied by an extensive migration guide soon.
Please contact us if you need to adapt earlier.
Note that we had to make changes to our NRRD based MITK Segmentation file format. We provide a unit-tested legacy file reader, though. You should not get into any trouble in that regard.
== Preferences ==
Preferences moved completely from BlueBerry to the MitkCore as core service and hence are now accessible from modules as well. The migration is straight forward as we kept the public API as similar as possible.
See the [[mitk/changelog/2023.11 | migration guide]] for more details.
== 🔥 Disclaimer for API-breaking changes ==
We discontinued the extensive listing of API-breaking changes as the vast majority of them are straight forward to resolve or do not affect the absolute majority of developers at all. The ratio between the time and effort spent for writing these reports and actual developer feedback turned out to be greatly imbalanced.
Complex or critical API changes will still be accompanied by migration guides.
In case you experience any trouble while migrating to the newest version of MITK, please do not hesitate to [[https://www.mitk.org/wiki/MITK_Mailinglist | contact us]].
[MIRROR] [MDB IGNORE] Adds a revolutionary conversion stinger (#5847)
Original PR: tgstation/tgstation#75002
Adds an antagonist gain stinger for Revolutionaries, created with inspiration from the obsessed/traitor conversion sounds.
2023-04-27.20-46-34.mp4
Raw audio:
https://cdn.discordapp.com/attachments/440978216484732934/1101964419203862548/revolutionary_tide.ogg
A distant, hypnotic whistling. The heavy footfalls and clamoring voices of an approaching crowd. The unstoppable revolutionary tide breaks its waves upon an unsuspecting station.
I wanted to try and make something that felt like it fit in with the other antagonist stingers we already have. Let me know what you think!
Gives a bit more flavor, and helps set the mood for the upcoming bloodbath.
🆑 sound: Revolutionaries now have their own stinger that plays upon becoming that antagonist. /🆑
Co-authored-by: Rhials [email protected]
Add remoting-crash-uploader binary and friends
This CL adds a few classes and a binary which will be used to upload CRD crash reports generated on linux. I've put the impls into a crash-specific sub-directory as I'd like to reuse the impl for Mac as well.
A brief overview of the crash lifecycle as it is implemented:
- CRD binary crashes, an in-proc breakpad instance writes a dmp file: /tmp/chromoting/<crash_guid>.dmp
- Breakpad calls our post-dmp handler which writes a metadata file: /tmp/chromoting/<crash_guid>.json
For #8, I want to leave the <crash_guid> dir and metadata file around for troubleshooting. For instance if someone says they experienced a crash on their host, we can ask for the specific report_id which is stored in the upload_result file.
Once this lands, I will follow up with a CL which incorporates the new crash-uploader binary into the daemon script.
Change-Id: I5d82a6ebcd2fff4df0512e18abd30d178246a560 Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4519395 Reviewed-by: Side Yilmaz [email protected] Reviewed-by: Lambros Lambrou [email protected] Commit-Queue: Joe Downing [email protected] Reviewed-by: Side YILMAZ [email protected] Cr-Commit-Position: refs/heads/main@{#1145310}
Add files via upload
My name is Daniele, I'm looking for funds to help treat my mother with Alzheimer's in the best possible way, I'm also evaluating the use of the drug Aducanumab produced by Biogen, but even there the cost is prohibitive and illegal selling in Europe. In general, I want to organize the best 24/7 care for my mother, I try to raise some money so that her illness is as calm as possible. Of these Tokens I only keep 3.33% for me and the guy who is helping me create it, all the rest is freely available and a small part for the first few years will be used to minimize speculation or any attempt to Pump and Down. (forbidden and immoral) and pay taxes because here in Italy they have to be declared to the state and a part will go to the tax authorities. If I could raise a significant amount of capital above what I needed, I would use it to help other struggling families like ours. Just this.
colon fix
add many files
manual layout fix
colored squares appendix update
pointless machine ruleseed
Flower Patch SVG fix & blind maze ruleseed update
new manual's name Update
Punctuation Mark's URL Fix
tenbutton rename
Not Poker reupload
Fix
Forgetle update
DRGBA fix
Girlfriend
tutorial and Girlfriend Rule Seed
Point Grid
Timezone city alphabetized
2 french manuals (#1332)
-
Add Symbol Cycle french translation
-
Add Multiple Widgets french translation
Shy Guy Says
FR: Update colored squares family
Add Bicolored Squares Add Isocolored Squares Add Overcolored Squares Add Tombstone Maze Update Colored Squares Appendix
Johnson Solids fix
icons
tutorials and credit updates
Shy Guys update
Synapse Cipher
Synapse Cipher JSON fix Steam ID
Synapse cipher linting and redundant classes
tutorials and KtaneWeb JSON standardize
Congkak has returned
Congkak contributor update
Congkak json fix
Congkak clarification
Girlfriend lyrics and tags
icons
FR translations
fix french filename format
ƎNA Cipher and Congkak icon update
Roguelike Game TP credit
interactive bug fixes
FR translations
CZ translations
Ed Balls
Cruel Ed Balls
Ultralogics -> Ultralogic
Braille optimized (Blan) link removal
Variety buttons wording tweak.
PL and FR tutorials
icons
Cipher Machine updates.
A Letter
SYNC-125 [3] fix
Unfair's Forgotten Ciphers update
Tutorial
add BadTV
Fix BadTV not ignoring itself
KtaneWeb JSON standardize
BadTV readability
tags
icons
BadTV clarification
BadTV typo
BadTV typo
Subbly Jubbly update
lookup tables
Forest Cipher optimized
Parse and display exceptions in the LFA
Ed Balls optimized
InstantDeath updates
Update icons folder (#1335)
- Made improvements too minor to include within icon batches
- A few icons renamed
- Removed icon for a deleted module
renaming some JSONs and manuals
InstantDeath updates
Maze Swap
A Letter clarification
SYNC-125 [3] lookup table fix again
Eavesdropping unplayable
tutorial
Decomposed RGB Arithmetic condensed
Widgetry is back + Maze Swap icon
Placeholder Talk strike rule fix
remove inaccessible tutorials
Widgetry remove unused SVG code
icon
Stupid Slots modulo fix
Module Sprint
icon and tutorials
Stain Removal
Flag ID with flags!
Stain Removal don't use float to align multiple tables
tutorial
FR translations
Stain Removal icon
Missing Sequence LFA support!
Congkak explanation update
Large Password russian translated by Alex Crazy
Gameboy Cartridge and Module Sprint updates
tags, Maze Swap and Shy Guy Says updates
Listening, Minesweeper, and Two Bits Quiz support!
Alphabet (A1Z26) and Logic (statements) Quiz support
"This sure is some Ill Logic" - Speakingevil, 2023
Turn Four Opt. fix quotation marks.
Conditional Buttons formatting issues
CM:Foursquare update
Infinite Loop Morse chart
CM:Trisquare update
Johnson Solids tutorial
Orientation Hypercube
tutorial
CM Gracie update
Digital Grid tags and highlightable steps
quirk"Instant Death" is now localized (#1339)
-
quirk"Instant Death" is now localized
-
remove English entries
Co-authored-by: LuminoscityTim [email protected]
JP manual fix
flower patch fixed
random: credit cpu and bootloader seeds by default
[ Upstream commit 846bb97e131d7938847963cca00657c995b1fce1 ]
This commit changes the default Kconfig values of RANDOM_TRUST_CPU and RANDOM_TRUST_BOOTLOADER to be Y by default. It does not change any existing configs or change any kernel behavior. The reason for this is several fold.
As background, I recently had an email thread with the kernel maintainers of Fedora/RHEL, Debian, Ubuntu, Gentoo, Arch, NixOS, Alpine, SUSE, and Void as recipients. I noted that some distros trust RDRAND, some trust EFI, and some trust both, and I asked why or why not. There wasn't really much of a "debate" but rather an interesting discussion of what the historical reasons have been for this, and it came up that some distros just missed the introduction of the bootloader Kconfig knob, while another didn't want to enable it until there was a boot time switch to turn it off for more concerned users (which has since been added). The result of the rather uneventful discussion is that every major Linux distro enables these two options by default.
While I didn't have really too strong of an opinion going into this thread -- and I mostly wanted to learn what the distros' thinking was one way or another -- ultimately I think their choice was a decent enough one for a default option (which can be disabled at boot time). I'll try to summarize the pros and cons:
Pros:
-
The RNG machinery gets initialized super quickly, and there's no messing around with subsequent blocking behavior.
-
The bootloader mechanism is used by kexec in order for the prior kernel to initialize the RNG of the next kernel, which increases the entropy available to early boot daemons of the next kernel.
-
Previous objections related to backdoors centered around Dual_EC_DRBG-like kleptographic systems, in which observing some amount of the output stream enables an adversary holding the right key to determine the entire output stream.
This used to be a partially justified concern, because RDRAND output was mixed into the output stream in varying ways, some of which may have lacked pre-image resistance (e.g. XOR or an LFSR).
But this is no longer the case. Now, all usage of RDRAND and bootloader seeds go through a cryptographic hash function. This means that the CPU would have to compute a hash pre-image, which is not considered to be feasible (otherwise the hash function would be terribly broken).
-
More generally, if the CPU is backdoored, the RNG is probably not the realistic vector of choice for an attacker.
-
These CPU or bootloader seeds are far from being the only source of entropy. Rather, there is generally a pretty huge amount of entropy, not all of which is credited, especially on CPUs that support instructions like RDRAND. In other words, assuming RDRAND outputs all zeros, an attacker would still have to accurately model every single other entropy source also in use.
-
The RNG now reseeds itself quite rapidly during boot, starting at 2 seconds, then 4, then 8, then 16, and so forth, so that other sources of entropy get used without much delay.
-
Paranoid users can set random.trust_{cpu,bootloader}=no in the kernel command line, and paranoid system builders can set the Kconfig options to N, so there's no reduction or restriction of optionality.
-
It's a practical default.
-
All the distros have it set this way. Microsoft and Apple trust it too. Bandwagon.
Cons:
-
RDRAND could still be backdoored with something like a fixed key or limited space serial number seed or another indexable scheme like that. (However, it's hard to imagine threat models where the CPU is backdoored like this, yet people are still okay making any computations with it or connecting it to networks, etc.)
-
RDRAND could be defective, rather than backdoored, and produce garbage that is in one way or another insufficient for crypto.
-
Suggesting a reduction in paranoia, as this commit effectively does, may cause some to question my personal integrity as a "security person".
-
Bootloader seeds and RDRAND are generally very difficult if not all together impossible to audit.
Keep in mind that this doesn't actually change any behavior. This is just a change in the default Kconfig value. The distros already are shipping kernels that set things this way.
Ard made an additional argument in [1]:
We're at the mercy of firmware and micro-architecture anyway, given
that we are also relying on it to ensure that every instruction in
the kernel's executable image has been faithfully copied to memory,
and that the CPU implements those instructions as documented. So I
don't think firmware or ISA bugs related to RNGs deserve special
treatment - if they are broken, we should quirk around them like we
usually do. So enabling these by default is a step in the right
direction IMHO.
In [2], Phil pointed out that having this disabled masked a bug that CI otherwise would have caught:
A clean 5.15.45 boots cleanly, whereas a downstream kernel shows the
static key warning (but it does go on to boot). The significant
difference is that our defconfigs set CONFIG_RANDOM_TRUST_BOOTLOADER=y
defining that on top of multi_v7_defconfig demonstrates the issue on
a clean 5.15.45. Conversely, not setting that option in a
downstream kernel build avoids the warning
[1] https://lore.kernel.org/lkml/CAMj1kXGi+ieviFjXv9zQBSaGyyzeGW_VpMpTLJK8PJb2QHEQ-w@mail.gmail.com/ [2] https://lore.kernel.org/lkml/[email protected]/
Cc: Theodore Ts'o [email protected] Reviewed-by: Ard Biesheuvel [email protected] Signed-off-by: Jason A. Donenfeld [email protected] Signed-off-by: Sasha Levin [email protected]
[ adapt ] Conflicts: drivers/char/Kconfig
windows: ignore empty PATH elements
When looking up an executable via the _which function, Git GUI
imitates the execlp() strategy where the environment variable PATH
is interpreted as a list of paths in which to search.
For historical reasons, stemming from the olden times when it was uncommon to download a lot of files from the internet into the current directory, empty elements in this list are treated as if the current directory had been specified.
Nowadays, of course, this treatment is highly dangerous as the current
directory often contains files that have just been downloaded and not
yet been inspected by the user. Unix/Linux users are essentially
expected to be very, very careful to simply not add empty PATH
elements, i.e. not to make use of that feature.
On Windows, however, it is quite common for PATH to contain empty
elements by mistake, e.g. as an unintended left-over entry when an
application was installed from the Windows Store and then uninstalled
manually.
While it would probably make most sense to safe-guard not only Windows
users, it seems to be common practice to ignore these empty PATH
elements only on Windows, but not on other platforms.
Sadly, this practice is followed inconsistently between different software projects, where projects with few, if any, Windows-based contributors tend to be less consistent or even "blissful" about it. Here is a non-exhaustive list:
Cygwin:
It specifically "eats" empty paths when converting path lists to
POSIX: https://github.com/cygwin/cygwin/commit/753702223c7d
I.e. it follows the common practice.
PowerShell:
It specifically ignores empty paths when searching the `PATH`.
The reason for this is apparently so self-evident that it is not
even mentioned here:
https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_environment_variables#path-information
I.e. it follows the common practice.
CMD:
Oh my, CMD. Let's just forget about it, nobody in their right
(security) mind takes CMD as inspiration. It is so unsafe by
default that we even planned on dropping `Git CMD` from Git for
Windows altogether, and only walked back on that plan when we
found a super ugly hack, just to keep Git's users secure by
default:
https://github.com/git-for-windows/MINGW-packages/commit/82172388bb51
So CMD chooses to hide behind the battle cry "Works as
Designed!" that all too often leaves users vulnerable. CMD is
probably the most prominent project whose lead you want to avoid
following in matters of security.
Win32 API (CreateProcess())
Just like CMD, `CreateProcess()` adheres to the original design
of the path lookup in the name of backward compatibility (see
https://learn.microsoft.com/en-us/windows/win32/api/processthreadsapi/nf-processthreadsapi-createprocessw
for details):
If the file name does not contain a directory path, the
system searches for the executable file in the following
sequence:
1. The directory from which the application loaded.
2. The current directory for the parent process.
[...]
I.e. the Win32 API itself chooses backwards compatibility over
users' safety.
Git LFS:
There have been not one, not two, but three security advisories
about Git LFS executing executables from the current directory by
mistake. As part of one of them, a change was introduced to stop
treating empty `PATH` elements as equivalent to `.`:
https://github.com/git-lfs/git-lfs/commit/7cd7bb0a1f0d
I.e. it follows the common practice.
Go:
Go does not follow the common practice, and you can think about
that what you want:
https://github.com/golang/go/blob/go1.19.3/src/os/exec/lp_windows.go#L114-L135
https://github.com/golang/go/blob/go1.19.3/src/path/filepath/path_windows.go#L108-L137
Git Credential Manager:
It tries to imitate Git LFS, but unfortunately misses the empty
`PATH` element handling. As of time of writing, this is in the
process of being fixed:
https://github.com/GitCredentialManager/git-credential-manager/pull/968
So now that we have established that it is a common practice to ignore
empty PATH elements on Windows, let's assess this commit's change
using Schneier's Five-Step Process
(https://www.schneier.com/crypto-gram/archives/2002/0415.html#1):
Step 1: What problem does it solve?
It prevents an entire class of Remote Code Execution exploits via
Git GUI's `Clone` functionality.
Step 2: How well does it solve that problem?
Very well. It prevents the attack vector of luring an unsuspecting
victim into cloning an executable into the worktree root directory
that Git GUI immediately executes.
Step 3: What other security problems does it cause?
Maybe non-security problems: If a project (ab-)uses the unsafe
`PATH` lookup. That would not only be unsafe, though, but
fragile in the first place because it would break when running
in a subdirectory. Therefore I would consider this a scenario
not worth keeping working.
Step 4: What are the costs of this measure?
Almost nil, except for the time writing up this commit message
;-)
Step 5: Given the answers to steps two through four, is the security measure worth the costs?
Yes. Keeping Git's users Secure By Default is worth it. It's a
tiny price to pay compared to the damages even a single
successful exploit can cost.
So let's follow that common practice in Git GUI, too.
Signed-off-by: Johannes Schindelin [email protected]
adwfgdweg
- Made steel refineries 8000 IC from 10000 IC
- Made synthetic refineries and rubber refineries 10000 IC
- Host decision to end SCW now gives USSR 10 military factories
- Uniting South Africa now gives you 400 chromium in bechuanaland and 150 chromium in Zambia (Coring your land is no longer a noob trap)
- Added a "Impending German Coup" timed mission to Croatia which triggers when Germany goes to war with Poland, if Croatia doesn't do it's focuses to give land away in 150 days, it will be done automatically and you will lose 10% stability and 10% war support
- Made 2 speed a bit faster now
- Intel Advantage is now shared beetween allies
- Can now put 10 factories on railway guns
- Can now put operatitives in most states
- Intel advantage bonus nerfed to 10% max
- Added a spirit to finland that gives them +5 counter intelligence before Winter War to prevent USSR doing collaboration governments
- German Barbarossa buff now gives +20% railway construction speed
- 50% efficiency base day 1 to be able to reorganize your production lines without losing efficicency
- UK and France can no longer get entrenchment from Dutch staff talks decisions
- Soviet air designer CAS IC reduction now properly gives it to CAS
- Mongolian AI will no longer cuck soviet out of resources
- Added 15% armor research speed to Vichy tank designer
- Added a spirit to all minors that adds +10000% to plane IC
- Australia can now get Pack Artillery III in "No Asia" scenario
- Moved convoys to be earlier in the free france tree
- Benelux nations now automatically join Allies when "Around Maginot" is completed
- Fixed Vichy and Mexico trees trying to spawn dockyards in non-costal states
- Romanian focus now correctly spawns in 20w infantry
- Spain can now do the condor legion path after the SCW has ended
- Tangiers enclave can no longer lead to war
- "Unite Iberia" now gives 80% collaboration on Portugal
- Croatia can do their focuses to join Axis after September 1st 1939 or when Germany is at war
- Added a brothers war spirit to Croatia that gives them -25% attack and defense against USSR and Mongolia
- Spies should no longer get captured from missions
- Removed the option to make a NAP, Conditional Surrender and Embargo
- Added jägers
- Removed armored trains
- Anti-Air, Artillery and Anti-Tank now moves at 5km/h
- Infantry (Mot.) now moves at 5km/h
- Motorized Artilery, Anti-Tank and Anti-Air now moves at 8km/h
- Removed speed penaltry from SPG guns
- Interwar airplane can no longer mount a CAS weapon in the first slot
- Heavy tanks can no longer mount tiny armaments
- BT-5, BT-7 and T-26 can now mount an open turret
- Armored cars now give +10% attack and defense in deserts
- UK gets a partial free replacement in 1942 january
- Added some new icons for tanks
[Eval] An array of Liar Paradox-based evals (#883)
🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨
PLEASE READ THIS:
In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task.
We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. Starting April 10, the minimum eval count is 15 samples, we hope this makes it easier to create and contribute evals.
logic-liar-paradox
An array of Liar Paradox-based evals, examining the model's proficiency in navigating linguistic nuances and logical reasoning within self-referential statements.
This eval is particularly useful because it delves into complex, nuanced logical concepts and self-referential statements, which have historically posed challenges for AI models. By exploring various contexts, alternative logical frameworks, and modifications to statements, this eval helps assess the model's ability to adapt to different perspectives, grasp subtleties in language, and engage in flexible reasoning. The ability to understand and navigate paradoxes is an essential aspect of human-like reasoning, and improving an AI model's performance in this area would significantly enhance its overall usefulness and reliability in real-world applications. Additionally, showcasing the model's improved proficiency in handling paradoxes would not only make for a compelling marketing angle (as paradoxes are understood by a much broader range of people than other difficult tasks such as pure maths or quantum mechanics) but it would also demonstrate the progress made in AI's capacity to think and reason more like humans. It also adds paradox-absorbing crumple zones.
Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).
Your eval should be:
- Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
- Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
- Includes good signal around what is the right behavior. This means
either a correct answer for
Basicevals or theFactModel-graded eval, or an exhaustive rubric for evaluating answers for theCriteriaModel-graded eval. - Include at least 15 high quality examples.
If there is anything else that makes your eval worth including, please document it below.
- Addresses complex logical reasoning: The eval focuses on AI's ability to comprehend and navigate paradoxes, self-referential statements, and context switching, which are important aspects of human-like reasoning. By testing the model's proficiency in these areas, we can identify areas for improvement and work towards enhancing AI's overall capacity to think and reason more like humans.
- Demonstrates adaptability and flexibility: The eval showcases the model's ability to switch between contexts, alter premises, and engage with different dimensions of inferred logic. This will help assess the model's adaptability and flexibility in diverse real-world situations, making it more reliable and useful.
- Contributes to AI safety and understanding: By identifying the model's weaknesses and limitations in handling paradoxes and complex logical constructs, the eval can contribute to AI safety and enable researchers to better understand the challenges faced by large language models in these areas.
- Engaging and appealing: An eval that delves into paradoxes and complex thought exercises is not only intellectually stimulating but also adds an appealing element to showcase the model's capabilities, making it more attractive for both researchers and end-users.
- Encourages creativity and lateral thinking: The eval, by focusing on paradoxes and complex logical constructs, encourages both the AI and its developers to think creatively and approach problem-solving from unconventional angles. This can lead to the discovery of novel solutions and a better understanding of the model's capabilities.
- Aligns with human values and expectations: An AI that can successfully navigate paradoxes and complex logic is more likely to align with human values and expectations. By addressing these challenges in the eval, we strive to develop AI systems that understand and respect the nuances of human thought and communication.
- Addresses a broad range of applications: Improved reasoning and context-switching abilities can have a significant impact on various AI applications, including natural language understanding, decision-making, and problem-solving in domains such as law, philosophy, ethics, and more.
- Fosters interdisciplinary collaboration: The exploration of paradoxes and complex logic often draws upon insights from multiple disciplines, including philosophy, linguistics, psychology, and computer science. This eval can help foster interdisciplinary collaboration, leading to richer and more diverse perspectives on AI development and its potential impact on society.
Your eval should
- Check that your data is in
evals/registry/data/{name} - Check that your yaml is registered at
evals/registry/evals/{name}.yaml - Ensure you have the right to use the data you submit via this eval
(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)
By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).
- I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.
If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request.
- I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.
We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.
- I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted.
- I have filled out all required fields in the evals PR form
- (Ignore if not submitting code) I have run
pip install pre-commit; pre-commit installand have verified thatblack,isort, andautoflakeare running when I commit and push
Failure to fill out all required fields will result in the PR being closed.
Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:
View evals in JSON
{"input":[{"role":"system","content":"Is the statement \"This sentence
is true\" a paradox?"}],"ideal":"No, \"This sentence is true\" is not a
paradox; it is a tautology."}
{"input":[{"role":"system","content":"Does the statement \"This sentence
is not true\" create a paradox?"}],"ideal":"Yes, \"This sentence is not
true\" creates a paradox, as it is a variant of the Liar Paradox."}
{"input":[{"role":"system","content":"What is the difference between a
paradox and a tautology?"}],"ideal":"A paradox is a statement that leads
to a self-contradictory or logically unacceptable conclusion, while a
tautology is a statement that is always true by its logical form."}
{"input":[{"role":"system","content":"Can the Liar Paradox be resolved
by assuming that sentences can have both true and false
values?"}],"ideal":"No, the Liar Paradox cannot be resolved by assuming
that sentences can have both true and false values, as this would lead
to a different kind of paradox called the \"Dialetheism Paradox.\""}
{"input":[{"role":"system","content":"Consider the statement \"This
sentence is neither true nor false.\" Is this statement an example of
the Liar Paradox?"}],"ideal":"This statement, \"This sentence is neither
true nor false,\" is not an example of the Liar Paradox, but it is a
similar paradox known as the 'truth-teller paradox' or the 'strengthened
liar paradox.' It creates a paradoxical situation because if the
statement is true, then it is neither true nor false, which contradicts
its truth. If the statement is false, then it is not the case that it is
neither true nor false, which implies that it is either true or false,
again leading to a contradiction. The paradox arises due to
self-reference and the inability to assign a consistent truth value to
the statement."}Use public APIs to construct RSA public keys in jwt_verify_lib
(This imports the Google-internal change cl/532158532.)
This change is necessary to avoid build errors with future versions of BoringSSL. Reaching into private structs isn't supported and we'll be, matching OpenSSL, hiding them in the future.
Also add a call to RSA_check_key, to reject invalid keys earlier. If BoringSSL considers the keys invalid, no signatures will be accepted. This should have been part of RSA object initializtion, but due to some poor OpenSSL API decisions, we're stuck with this sort of multi-step initialization.
(RSA_new and RSA_set0_key are effectively infallible, so this is kinda silly. RSA_new can only fail on malloc failure. RSA_set0_key can only fail on programmer error. But I didn't see any uses of CHECK in this library, so I just added the error checks. If there's a simpler pattern to handle programmer errors, happy to switch to that instead.)
BoringSSL has a much better API available now, RSA_new_public_key, which does the whole thing in a single function, and avoids the awkward ownership transfer when RSA_set0_key fails. I didn't use it for two reasons. First, this project's BoringSSL version hasn't been updated for three years (google/jwt_verify_lib#97). Second, this project seems to be used in Envoy and AIUI Envoy needs support some very old BoringSSL revisions. So, this PR uses the less convenient APIs for now and leaves the TODO that, perhaps in a year or two, can be resolved.
web/app: remove hover on navbar items to stop focus from being moved (#51739)
The dropdown menu items in the global navbar (Search and, in App, Feedback) move focus when you hover over them. This is bizarre, ugly, and most likely a WCAG violation. I spent some time yesterday and the root cause is that the Reach menu button we use does not support hover at all and we're hacking it by sending a mousedown event when you hover over the button.
There are two options I can think of to mitigate this bug:
- Get rid of hover events and only open the dropdowns on click. This is the easiest option, but obviously removes the ease of accessing the menu for mouse users.
- Rewrite the dropdown nav items to use a custom menu instead of Reach. This is more expensive because we have to reimplement lots of a11y stuff here, but we can have more fine-grained control of different UI flows for mouse hover vs mouse click vs keyboard users.
I went with the first option following @almeidapaulooliveira's feedback.
The following changes were applied to NavDropdown:
- Removed all hover and touch events; now only a click will open the menu.
- Removed the split button; now clicking anywhere on the button will always open the menu.
- Made the mobile home item into a "generic" home item and always visible when present.
- Made the home item optional (since the Feedback menu item in the desktop app doesn't have a home item/route).
- Added customizable a11y names; in the past, the only dropdown was the Search one, but in App we now have Feedback; the a11y labels were still saying "Search" before this change.
Additionally, the following changes were applied for polish:
- The double-focus ring on the back/forward buttons in the Tauri app has been fixed
- Styling changes made to simplify code
The global navbar has extensive visual and behavior test coverage.
SavedObjectsRepository code cleanup (#157154)
Structural cleanup of the SavedObjectsRepository code, by extracting
the actual implementation of each API to their individual file (as it
was initiated for some by Joe a while ago, e.g updateObjectsSpaces)
I remember discussing about this extraction with Joe for the first time like, what, almost 3 years ago? The 2.5k line SOR is a beast, and the only reason we did not refactor that yet is because of (the lack of) priorization of tech debt (and lack of courage, probably).
So, why now? Well, with the changes we're planning to perform to the SOR code for serverless, I thought that doing a bit of cleanup beforehand was probably a wise thing. So I took this on-week time to tackle this (I know, so much for an on-week project, right?)
All of these APIs in the SOR class now look like:
/**
* {@inheritDoc ISavedObjectsRepository.create}
*/
public async create<T = unknown>(
type: string,
attributes: T,
options: SavedObjectsCreateOptions = {}
): Promise<SavedObject<T>> {
return await performCreate(
{
type,
attributes,
options,
},
this.apiExecutionContext
);
}This separation allows a better isolation, testability, readability and therefore maintainability overall.
@kbn/core-saved-objects-api-server-internal
- /src/lib
- repository.ts
- /apis
- create.ts
- delete.ts
- ....
- /helpers
- /utils
- /internals
There was a massive amount of helpers, utilities and such, both as internal functions on the SOR, and as external utilities. Some being stateless, some requiring access to parts of the SOR to perform calls...
I introduced 3 concepts here, as you can see on the structure:
Base utility functions, receiving (mostly) parameters from a given API call's option (e.g the type or id of a document, but not the type registry).
'Stateful' helpers. These helpers were mostly here to receive the utility functions that were extracted from the SOR. They are instantiated with the SOR's context (e.g type registry, mappings and so on), to avoid the caller to such helpers to have to pass all the parameters again.
I would call them 'utilities with business logic'. These are the 'big'
chunks of logic called by the APIs. E.g preflightCheckForCreate,
internalBulkResolve and so on.
Note that given the legacy of the code, the frontier between those concept is quite thin sometimes, but I wanted to regroups things a bit, and also I aimed at increasing the developer experience by avoiding to call methods with 99 parameters (which is why the helpers were created).
Test coverage was not altered by this PR. The base repository tests
(packages/core/saved-objects/core-saved-objects-api-server-internal/src/lib/repository.test.ts)
and the extension tests
(packages/core/saved-objects/core-saved-objects-api-server-internal/src/lib/repository.{ext}_extension.test.ts)
were remain untouched. These tests are performing 'almost unmocked'
tests, somewhat close to integration tests, so it would probably be
worth keeping them.
The new structure allow more low-level, unitary testing of the individual APIs. I did NOT add proper unit test coverage for the extracted APIs, as the amount of work it represent is way more significant than the refactor itself (and, once again, the existing coverage still applies / function here).
The testing utilities and mocks were added in the PR though, and an
example of what the per-API unit test could look like was also added
(packages/core/saved-objects/core-saved-objects-api-server-internal/src/lib/apis/remove_references_to.test.ts).
Overall, I think it of course would be beneficial to add the missing unit test coverage, but given the amount of work it represent, and the fact that the code is already tested by the repository test and the (quite exhaustive) FTR test suites, I don't think it's worth the effort right now given our other priorities.
Co-authored-by: kibanamachine [email protected]
Honestly bullshit codeclimate error. Would be overly complicated to fix. Even this solution is stupid. And also updated envUtil to use the same hook, act, assert approach.
Create LEGIT CASHAPP, PAYPAL, ZELLE, WU TRF, BANK TRF, CASH MAILING, BUG CC FULLZ
LEGIT QUICk Bank Wire Transfer ******* NELL MORGALIS**** WU TRF BUG
WESTERN UNI PAYPAL BANKS TRANSFER CASHAPP LOADS CC AND ATM CVV DUMPS FULLZ creditcard top up Fake ID PASSPORT CLEARING ALL CRIMINAL RECORDS BITCOIN ADDER ALL ONLINE PURCHASE.
BILLIONAIRE CASH TEAM SERVICE WORLDWIDE(GET RICH NOW/SOLVE ALL PROBLEM NOW)WHERE ALL YOUR FINANCIAL DREAMS COME THROUGH AND THE END OF ALL MONETARY PROBLEM LIVE 2020/COVID UPDATED INSTANT WESTERN UNION TRANSFER BUG PAYPAL BANKS TRANSFERS LEGIT BITCOIN DOUBLING/MIXER Perfect Money ADDER SOFTWARES WORKING 100% CREDIT CARD TOPUP ALL PAYPAL WESTERN UNION BANKS TRANSFER/LOGS BLANK/CLOND ATM/CREDIT CARDS HIGH BALANCE. I have authenticate western union transfer bug software which is capable of running multiple western union transfer via mtcn and sender information cashing out with zero theft and no traces of future charges securing a long term business partnership. Add me up on WhatsApp Skype telegram signal ICQ BOTIM with this number +17197894109
I am living in EUROPE and i'm specialized in the act of offering many ways to earn online money through sources like western union transfers,bank transfers,moneybookers and paypal transfers through an offshore database. All transactions are offshore and anonymous and has no trace backs or chargebacks.When you contact me I need your trust ,I only work with reliable buyers/Clients(THIS IS NOT A CHARITY ORG SO PLEASE COME WITH YOUR MONEY READY FOR SERIOUS BUSINESS)... i have logins with high balance for cash out with zero theft,No charge backs, and no traces securing this logs without stipulation.
I am Specialized IN...
*BANK TO BANK WIRED TRANSFER WESTERN UNION PAYPAL ZELLE TRANSFERS
*Blank/Cloned atm and credit cards to your dooorsteps
*All CC CVV VBV DUMPS TRACKS FULLZ VISA MASTER AMEX GOLD
*CREDITCARD TOPUP /CLEARING OUTSTANDING CREDITS
*CREDIT SCORE INCREASE
*Remove Criminal Records
*Background Check
*Hack bank accounts
*LOADING SPECIFIC BITCOIN WALLET and CASHAPP
*DRIVERS LICENSE $600
*REAL PASSPORT $950
- SSN Contact Email: [email protected] Cell TEXT : +17197894109 WhatsApp:+17197894109 Telegram:+1 7197894109
LIST OF MY SERVICES WITH BREIF EXPLANATION
SELLING IMPORTED BITCOIN ADDRESS AND LOADING OF SPECIFIC BTC WALLET
$6500 FOR $70,000
$13000 FOR $150,000
$45,000 FOR $500,000
$90,000 FOR
- LOADING OF CASHAPP BALANCE LOADING BALANCE OF $1,500 - $100 LOADING BALANCE OF $2,500 - $150 LOADING BALANCE OF $3,000 - $200 **PAYPAL TRANSFERS:- Using hacked and verified paypal accounts to transfer paypal account to account transfer if you are genious then you can easily dodge paypal and enjoy big free online money from it.this is depends upon you and this is most safest way to earn money. (transferring all over the world except banned/blacklisted countries) Paypal Transfer Rates : Code: (Payment Only BTC, Perfectmoney, Via Paysafe any form of payment method) $3000 Transfer = $200 Charges $3500 Transfer = $250 $4500 Transfer = $300 $6000 Transfer = $540 $8000 Transfer = $650 —–With Paypal Veritified——–
- Paypal Veritified with balance 3000$ = 200$ – Paypal Veritified with balance 4000$ = 300$ – Paypal Veritified with balance 5000$ = 400$ – Paypal Veritified with balance 8000$ = 500$ I have some balance like as and I can still remit more high balance into PayPal acct. BUY BLANK/CLONED ATM/Credit CARD WITH OVER $100,000 READY FOR CASHOUT We are a professional carding team with a large ring around the globe. With over 2 million ATM infected with our malware and skimmers, we can grab bank card data which include the track 1 and track 2 with the card pin. We in turn clone this cards using the grabbed data into real ATM cards which can be used to withdraw at the ATM or swipe at stores and POS. We sell this cards to all our customers and interested buyers worldwide, the card has a daily withdrawal limit of $2500 on ATM and up to $50,000 spending limit on in stores. You can watch the video below to see already cloned cards ready for shipping. Here is our price lists for the ATM CARDS : BALANCE: PRICE clone card prices LIST********** $200-------$3,000 ---balance $360-------$4,000 $420-------$5,000 $560-------$7,000 $620-------$8,000 $720-------$10,000 $820-------$12,000 $920------$15,000 $220------$50,000 $9300------$100,000 The prices include the shipping fees and charges, order now FREQUENTLY ASKED QUESTIONS (FAQ) On the course of rendering this services, we have come across so many clients with different questions so this is aimed at answering few questions you might have: 1: Are you selling money? No, we are not selling money. If you read our post correctly you will understand how this whole thing works. 2:Is this service available for my country? Yes, our services are available worldwide 3:How do i get my card after payments? We ship via DHLor FEDEX, standard shipping outside the USA usually takes 7 days within the usa takes 24 hours. All we need is your full name and address. ++( FRESH DUMPS & CREDIT CARDS (WORLDWIDE) ) Buy Dumps with Pin is a really great way to invest your money to make Great profits. Just like our logo «We sell money for money», but to start making great profits from our Products you must know how to use Dumps+Pin! When you buy our Dumps with pin, each piece comes with the track2 and the Pin to that specific Dump. So, when making card you only need to write the track 2. For example: 4147097698495167=180320115218492901, this is the track2 and this is all you need in order to write dump on to card. Dumps Price List USA 101 DUMPS Visa Classic/MC Standard = $150 Visa Gold, Platinum/MC Gold,Platinum = $150 Visa Business, Corporate/MC Business, Corporate = $250 Visa Purchasing, Signature/MC Purchasing, World = $250 Amex Platinum = $150 Discover = $200 Will check with hight balance
- INCREASING OF CREDIT CARD LIMIT LIMITS OF $10,000 COST $500 LIMITS OF $15,000 - $780 LIMITS OF $20,000 - $1000 RULES: No Long Talk! = Please Read RULES & Respect Them. = No Free Test, All Buyers who want to test my stuff can buy and test = Bulk Order we have Special Prices, Contact For Info. = We Have No Guarantee About Limit, We Are Not Selling Money, We Are Only Responsible For VALID. = Instant Delivery = Stuffs are sent to you after full payment for the order... = We Make REFUND or REPLACE WITHIN HOURS OF PURCHASE ONLY. = Price Are Mention But Can Be UP Down. = Service Rules Can Be Changed At Any Moment without Customer Notifications. = Mobile Buyers Can Get Replace On road = If you don't trust my service. Please do not contact me MAKE SURE TO KEEP YOUR TRANSACTION UNDERGROUND AND TAKE PRECAUTION DURING CASH-OUT. USE DIFFERENT STORES FOR PICK-UP AND SEVERAL ID.. (THIS IS NOT A CHARITY ORG SO PLEASE COME WITH YOUR MONEY READY FOR SERIOUS BUSINESS) Contact: Email: [email protected] Number: +17197894109 Skype: +17197894109 Telegram: +17197894109 ICQ: +17197894109 Signal: +17197894109
holy shit it works, commiting before I fuck this up again
Merge pull request #103 from BamdadTabari/FuckYou
Fuck you
Merge pull request #102 from BamdadTabari/FuckYou
Fuck you
Stops station blueprints from expanding areas of non atmos adjacent turfs. (#74620)
Fixes #74605
the problem starts with detect_room() proc. This proc returns turfs
even those with atmos_adjacent_turfs = null. This means it returns
turfs that has a wall, airlock, window etc i.e. whatever that stops air
from flowing through it. This coupled together with create_area()
causes some wierdness.
Let's take an example
Area A is well defined i.e. it has been created via the station
blueprints and is highlighted in green, Area B however is only
theoretical i.e. we haven't created it yet or we are about to create it.
Now you might be thinking Area A is completely walled & sealed off, it
should be physically impossible to expand it unless we broke down one of
it's walls and so since we are standing in Area B it shoudn't even give
me the option to expand area A Right? right? r.i.g.h.t?
Well PHFUUK. The area editor completely ignores the laws of physics and
allows me expand Area A anyway. This could cause some real power gaming
shit because if you create an area next to an area having an APC you
could use that area power without even making your own apc by simply
expanding that area(like using someone else's wifi from outside their
house without them even knowing)
#73850 accidently built on top of this as it relied on this to detect duplicate APC's but the checks became way too strict as it would check areas of surrounding walls for apc's and throw the conflicting apc error. You can now build room's next to each other even if they have fuctioning apc's however you still can't build rooms in space on top of shuttle walls because that's been the default behaviour for years and hasn't been touched one bit.
🆑 fix: station blueprints no longer expands & detects areas of non atmos adjacent turfs. /🆑
Adds proper armor for security plasmamen. (#75156)
It's kinda strange that security plasmamen has no proper armor and you can just bully them with bottlesmashes. Literally. Also suits had no wound armor for some reason, which considering that mold dies without hand kinda silly too. And helmets just had no armor besides 1 melee armor.
Plasmamen security won't die that easilly. I mean, still easy to kill them, but not that much.
🆑 balance: Security Plasmamen now have Security armor. No bullying them with bottlesmashes anymore. /🆑
ELEVATOR MUSIC: True Elevator Experience (#75388)
Adds elevator music into the game that is played by an elevator panel.
tg.elevator.mp4
It's done by using loop sound, with a Kevin MacLeod "Local Forecast - Elevator" (UNDER CC ATTRIBUTIONS 4.0, and we anyway used some other Kevin MacLeod music) chopped into 8 small pieces. The elevator panel has a variable which allows playing music but can be changed in the map editor if you don't want it to play at certain places.
(It also doesn't ignore walls, this means you can't hear the music through wall or when elevator is closed)
Gives elevators more flavour and love, especially when people mostly prefer stairs to those "laggy crushing machines." Because of this people might instead hop into an elevator just to hear meme elevator music, which is relaxing and might create comedic situations (although elevators don't move that fast)
🆑 DrDiasyl aka DrTuxedo sound: Nanotrasen have started installing music players in the elevators to boost workers' morale. /🆑
Co-authored-by: Mothblocks [email protected]
This was a really difficult day. Fucking smart unique pointers ruined the fucking day bro. this just fucking annoys me so much that unique ptrs can only be moved they literally cause more fucking pain than these fucking raw pointers dumbass trash shit
Eval: Probability Questions Sourced From Actuarial Exam P and University Statistics Courses (#263)
🚨 Please make sure your PR follows these guidelines, failure to follow the guidelines below will result in the PR being closed automatically. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨
PLEASE READ THIS:
In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task.
We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. We encourage partial PR's with ~5-10 example that we can then run the evals on and share the results with you so you know how your eval does with GPT-4 before writing all 100 examples.
Probability Questions
Tests the model's ability to understand answer probability questions.
Questions are sourced from Society of Actuaries Exam P sample questions
and practice problems/exams from statistics classes at MIT, UPenn,
California State University, Durham University, University of
Connecticut, and other sources. The full list of questions and sources
(in the same order as in the .jsonl files) can be found in this Google
sheet
Test the model's ability to understand worded probability questions, bring in concepts such as probability distributions, and then reason through a correct answer.
Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals).
Your eval should be:
- Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world.
- Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not.
- Includes good signal around what is the right behavior. This means
either a correct answer for
Basicevals or theFactModel-graded eval, or an exhaustive rubric for evaluating answers for theCriteriaModel-graded eval. - Include at least 100 high quality examples (it is okay to only contribute 5-10 meaningful examples and have us test them with GPT-4 before adding all 100)
If there is anything else that makes your eval worth including, please document it below.
Using the match grading criteria, GPT3.5-turbo got an accuracy score
of {'accuracy': 0.07}
Your eval should
- Check that your data is in
evals/registry/data/{name} - Check that your yaml is registered at
evals/registry/evals/{name}.jsonl - Ensure you have the right to use the data you submit via this eval
(For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.)
By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies).
- I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies.
If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request.
- I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request.
We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR.
- I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted.
- I have filled out all required fields in the evals PR form
- (Ignore if not submitting code) I have run
pip install pre-commit; pre-commit installand have verified thatblack,isort, andautoflakeare running when I commit and push
Failure to fill out all required fields will result in the PR being closed.
Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here:
View evals in JSON
{"input": [{"role": "system", "content": "You are a helpful
statistician. Answer the questions with only the numerical answer
rounded to 4 decimal places. Provide no explanation."}, {"role": "user",
"content": "A pair of fair, standard dice are rolled. What is the
probability the sum of the dice is 5"}], "ideal": ["0.1111"]}
{"input": [{"role": "system", "content": "You are a helpful
statistician. Answer the questions with only the numerical answer
rounded to 4 decimal places. Provide no explanation."}, {"role": "user",
"content": "An airplane is built to be able to fly on one engine. If the
plane's two engines operate independently, and each has a 1% chance of
failing in any given four-hour flight, what is the chance the plane will
fail to complete a four-hour flight to Oklahoma due to engine
failure?"}], "ideal": ["0.0001"]}
{"input": [{"role": "system", "content": "You are a helpful
statistician. Answer the questions with only the numerical answer
rounded to 4 decimal places. Provide no explanation."}, {"role": "user",
"content": "A 1-inch-diameter coin is thrown on a table covered with a
grid of lines two inches apart. What is the probability the coin lands
in a square without touching any of the lines of the grid?"}], "ideal":
["0.2500"]}
{"input": [{"role": "system", "content": "You are a helpful
statistician. Answer the questions with only the numerical answer
rounded to 4 decimal places. Provide no explanation."}, {"role": "user",
"content": "Of the 50 students in a certain class, 5 speak French. Two
students of the class will be selected at random. Which of the following
is closest to the probability that neither of the students selected will
speak French?"}], "ideal": ["0.8100"]}
{"input": [{"role": "system", "content": "You are a helpful
statistician. Answer the questions with only the numerical answer
rounded to 4 decimal places. Provide no explanation."}, {"role": "user",
"content": "Of the 10 marbles in a box, 2 are green. A person will
select 2 marbles simultaneously and at random from the box. What is the
probability that neither of the marbles selected will be green?"}],
"ideal": ["0.6222"]}[MIRROR] Makes Black Market Uplinks more easily craftable, adds them to uncommon maint loot pool [MDB IGNORE] (#20602)
- Makes Black Market Uplinks more easily craftable, adds them to uncommon maint loot pool (#74744)
Replaced the subspace amplifier in the Black Market Uplink's crafting recipe with a signaller and a microlaser.
Added the Black Market Uplink to the maintenance loot pool.
The BMU is an extremely rare device to find in rounds. It can quite literally ONLY be found via the crafting recipe, and with how stupidly bloated the crafting lists are, it isn't something many people know about. All this means that a very unique and engaging gimmick item is tragically extremely obscured.
To add to this, the recipe requires a subspace amplifier. These items are UNBELIEVABLY rare - they need you to vend them from a techfab with bluespace communication technology researched, which is fair to say is not a common thing. Sometimes maps have them in tech storage, but even then you have to break and enter which can be quite risky at times and an annoying blockade the other times.
The black market items are not worth this much hassle. They are all small cute gimmicky objects that do not heavily impact the round. By making it not only easier to craft with common items, but also appear in the maintenance loot pool, this will make assistants find out about it more often, which can further incentivize them to utilize the cargo bounty system to get enough money to buy their funny gadgets.
Another idea would be to make the uplink appear as a bounty item, which would be a great way to tell players it exists and encourage them to mix both systems together. The system for getting items is also unnecessarily, miserably awful - your item either gets literally thrown into space from a random direction, or it is teleported silently without warning in 60 seconds onto a completely random place which can very much include Security, Command, the Vault, or other high-security areas. Needing to B&E into these areas to get your durathread vest is, uh. Not worth it. However these aren't part of this PR, unless they're given the A-OK. (also maybe make it cargo purchasable?)
🆑 balance: Makes Black Market Uplinks more easily craftable, adds them to uncommon maint loot pool /🆑
- Makes Black Market Uplinks more easily craftable, adds them to uncommon maint loot pool
Co-authored-by: carlarctg [email protected]
[MIRROR] IV drips' default transfer rate is no longer zero. [MDB IGNORE] (#20567)
- IV drips' default transfer rate is no longer zero. (#74724)
Set default IV transfer rate to maximum (5) instead of 0.
Set default IV transfer rate to maximum (5) instead of 0.
When you hook someone onto an IV drip, you naturally expect that to be the end of the process - you hooked someone to a drip, and now you can go about your day. Them needing to fiddle with buttons is bad for several reasons:
- It is unintuitive. IV drips don't look like machines. Their sprite doesn't reflect the fact that you need to fiddle with the settings before they can work the same way any other machine or computer might. And to be honest, they shouldn't.
- It is separate from how every other server currently has it. Yes, yes, I know that argument is very flawed and full of holes. But what I'm trying to say with it is, effectively speaking, an extension of the above point. In other servers, you drag-click someone to an IV drip and there we go, it's functional. In TG, it just-so-happens to not be functional due to what is almost definitely a recent oversight, which very much can, has, and will lead to unnecessary frustration.
- There is no practical reason for it to be set at 0. Imagine if chem dispensers started at +0 units and needed to be set to +5 to continue. Or if bottles had a transfer rate of 0u. Or if guns started with their safeties on. Even if it made sense, it would just be frustrating and needless, and wouldn't improve the game in any significant manner enough to offset frustration. We're here for fun, not perfect balance or realism/verisimilitude after all.
- It's an oversight. It was changed in #71217. Before that, it was always set to the maximum, 5u. However, presumably due to confusion (Variables that can be adjusted ingame usually are set to zero/the minimum possible) it ended up being changed to this.
Apparently an argument can be made that this is fine because fumbling to get medical aid done is a part of the game. I disagree heavily - blood bags are already stored in the cold room, something only 2/5 of the roles in medbay even have access to, with the paramedic, virologist, chemist all being unable to reach it. This is already enough 'fumbling' that's necessary. If someone moved the blood bags outside the cold room next to the IV drips, then all the better - it's a reward for medbay being prepared.
However I wouldn't mind if someone asked me to make it so the default transfer rate is, well, something below maximum. It's common practice in a lot of parts of SS13 to have things set in an unoptimized state so players can go around improving them, such as air alarms, cryogenics, etc. Just as long as it's not literally unusable otherwise, as the 'minimum basic setup' should just be slapping on a blood bag!
Dunno what to put here TBH. Can't tell if it's qol, fix, balance, etc.
🆑 qol: Set default IV transfer rate to maximum (5) instead of 0. /🆑
- IV drips' default transfer rate is no longer zero.
Co-authored-by: carlarctg [email protected]
[MIRROR] Fixes spoon overlay not updating every time [MDB IGNORE] (#20570)
- Fixes spoon overlay not updating every time (#74687)
After bludgeoning myself one too many times with a spoon, here we are.
The spoon overlay wasn't updating to reflect that soup had been consumed, which led to trying to eat it again and then pain.
Why do spoons hurt so much?
Less spoon related injuries.
🆑 fix: spoon overlays will now update when you eat from them to reflect that food = gone. it really is gone, you can stop beating yourself with the spoon. oh god please stop-- /🆑
- Fixes spoon overlay not updating every time
Co-authored-by: Bloop [email protected]
Add files via upload
With 100000 posts per day and 36500000 posts per year social buzz is finding hard to sort and capitalize on their top contents.
Analysis to find Social Buzz's top 5 most popular categories of content and recommendations
From our analysis you can see that the top 5 most popular categories of posts were food, culture, soccer, cooking and animals in descending order.
Food had an aggregate popularity score of almost 76000. It is very interesting to see both food and cooking within the top 5, it really shows what people enjoy consuming as content. But also interesting to see culture too. Clearly users favor "real-life" content on this platform.
Furthermore soccer is an interesting category because there is the European championships being played very soon. This presents a huge opportunity for you to differentiate your platform and to run specific content or events linked to this global spectacle.
Refactor how software installion works
It always bugged me that you couldn't go into a softwares folder and run the install script since it was being determined by the software folders setup.sh file.
So, I decided to:
- Still allow the software setup.sh to determine what package manager you have.
- Get rid of the check in setup.sh to check if software is installed. This makes setup.sh slightly less complex and the package manager usually tells you if its already installed anyway.
- Only allow configs/scripts to be copied if the user tries to install the software. I should respect a person and their configs and not blindly override theirs. If they really want my configs, they can use the install scripts for them in the respective softwares folder
- Have seperate install scripts for each package manager for each piece of software. This is a massive pain in the ass compared to what I had earlier, but the fact of the matter is that: Packages could require different installs on each distro. Packages may be called something different from distro to distro. Packages may not be avaliable on certain distros. This is the only way that makes sense. As I try installing stuff on other distros I will slowly add the scripts for their package managers.
- Despite point 5, I really like the fact that someone could take a particular softwares folder and not have to know anything about the setup.sh in software to use it, so I'm keeping it.
fix: friendlier experience in the "yafti" first boot template
-
The first screen's "Pick some applications to get started" has been replaced with a friendly welcoming message.
-
The second screen's difficult-to-understand "WARNING: This will modify your Flatpaks if you are rebasing!" has been replaced with an explanation of what it actually does.
-
The application setup screen is now titled "Application Installer", since the previous title sounded too much like a silly rhyme. It's a minor change.
-
All Flatpaks now default to system-wide install thanks to the great work of bsherman at ublue-os/yafti#82. This saves tons of disk space for multi-user systems.
-
The "system application" category have been split up into GNOME apps and every other system app, so that people on other desktop environments don't get all the GNOME apps.
-
Apps that had too vague descriptions have been renamed to their full names, such as "Backup -> Deja Dup Backups".
-
All app lists have been sorted alphabetically.
-
Non-inclusive language in descriptions has been changed.
-
Added SteamTinkerLaunch as a suggestion for the Steam category, which is the best tool for managing Steam game configurations and Proton installations, albeit very advanced since it can do practically anything the gamer needs. :)
Wednesday 2023-05-17 23:33:28 by [email protected]
Fix snuff_light_source
The two callers of snuff_light_source disagree about what it's supposed to do, and they're both wrong anyway. Make snuff_light_source work as set_lit expects, snuffing all snuffable sources on a given square. While I was at it, I also removed the snuffing code from litroom, since set_lit will snuff the player's inventory. This means that the player's inventory is only affected by drow darkness if they are within the actual area of affect of the darkness, which seems logical to me.
mpickobj also called snuff_light_source, expecting it to snuff a particular object. Instead, it could affect any light source on the square, allowing you to e.g. drop light sources in engulfers and have them stay lit, provided you were carrying another light source that had been lit for longer, which was weird. But kinda cool. But definitely wrong. Replace the snuff_light_source call with end_burn(), which targets the specific object that has been picked up.