Adding O8S timeline syncs #66
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The diff may not be super useful, so I'll summarize the changes (beyond simply adding syncs). Preserved the existing naming for the most part. - Added the Thunder/Fire hits of Celestriad - Fixed some slight +/- 1s timing issues - Added all non-starstrafe Soaks (The Path of Light, renamed for clarity), since they were included, albeit inconsistently, later on in the fight. Having specific timing for them is useful for dodging into it. - Added All Things Ending to Past/Future End during Starstrafe. The ATE cast on the normal ones is fast enough (3s) that it's not worth timing it, but there's a bigger gap in that one (8s). - Also added Meteor hit (the stack) during Starstrafe - Added some missing Ultimas (1328, 1358, 1543) - Added a missing Ultimate Embraces (1364) - Changed tether times to activation rather than appearance
Forgot to save these.
|
Tested it out tonight, worked well as far as we saw (end of third forsaken). Longest pull is here: https://www.twitch.tv/videos/234577658?t=2h20m42s |
|
Dang it, I forgot to try this myself tonight. Let me know if you're able to give this timeline a test all the way through and then I'll be happy to merge it. I feel like the existing one does pretty well, so mostly just want to give a new one a smoke test before changing it. |
|
Tested it all the way through: https://www.twitch.tv/videos/237540234 There was some weirdness right at the end: The corresponding events don't really line up very well: I decided to take out the Graven sync, since it's a separate unit and runs on a different timer, even though the earlier tether syncs were more or less right on the mark. The fact that one synced from .377 and the next synced from .733 indicates that the real boss is still operating on whole-second units. I bumped up the timeline since 09:55.196 to 00:10:01.560 is rightly a 6 second gap with rounding. I also discovered that the ending Ultima crescendo is a different ability ID, so I fixed that as well. The last line was likely just the drift from not having synced in a while. |
|
Congrats on the clear! Thanks again for the testing and all the timeline additions. I'll go ahead and merge this. Also, I wonder if the last ultimas are different because it gives a buff and isn't just damage on its own. |
CptDraco
added a commit
to CptDraco/cactbot
that referenced
this pull request
Apr 21, 2018
* Shortening O5 timeline names (quisquous#49) * Adding timelines for Sigmascape 6-8N (quisquous#53) * Cleanup o5678n reset lines * O8S naming cleanup (quisquous#55) * Triggers for o6s (quisquous#54) * Add Triggers in o6s * update severity for last kiss target * Fix tabbed indents and fix missing square bracket * More indenting fixes * o7n: replace missile with prey * raidboss: add preRun functions * raidboss: Allow overrides for condition and texts This allows users to specify shorter/longer outputs from triggers. Also, overriding conditions lets folks narrow (or broaden) when triggers show up, e.g. see provokes even when you're not a tank or turn off mechanics triggers for particular roles when they matter less. Style-wise the capitalization on all the options looks weird, but I'm just going to leave it for now for the sake of consistency. Closes quisquous#51. * Add tts for general provoke/shirk/etc triggers Also shorten the names. * ucu: Fix adds phase rp text mistake OOPS * raidboss: o6s triggers * o5s: Add more sync windows in case somebody misses the train coming back * fractal hard: don't call knockback on first death spin * o7s: make more program names specific * o5s: narrow sync times after add phase * Fix https update check Releases page is now https only, so needs some extra flags and should use the correct url. * Bump version to 0.6.3 * Create fr.js (quisquous#58) * Create fr.js New locale 'fr' * Update fr.js Added translation for Seal and Unseal. * Logic fixes in lang/fr * Move zone names out of locale-specific lang files These names come from ACT and not the game, so they're all in English. * Add sigmascape zones to well fed regex * Fix cactbot initialization race The log doesn't get set up until after the plugin has started. There's no event for this, so defer cactbot initialization until the first slow timer. This allows initialization error messages to not be dropped on the floor. * Support html entities in custom elements resource-bar, timer-bar, and timer-icon all currently only allow flat text. Allow html entities in these text strings as well. * raidboss: Allow timeline names to be overridden This closes issue quisquous#8. This also is intended to be used as the way for issue quisquous#57 to replace timeline names in English with French. * oopsy: fix test bootshine trigger misapplying * raidboss: Allow timeline syncs to be overridden This is intended to help solve issue quisquous#57. * raidboss: add support for localized triggers Triggers can now have localized regexes. For a given local (e.g. ko) if there's a regexKo entry on the trigger, that will take precedence over regex on its own. Additionally, any trigger field that can be treated as a function can now also return localized responses as an object, e.g. instead of: return 'Get Out'; ...triggers can instead do something like: return {en: 'Get Out', fr: 'something french here, etc'}; This change also avoids running Regexes.Parse on every single log line for every single trigger. @_@;; This will help issue quisquous#57. * oopsy: re-re-fix test bootshine typo This fixes a typo from 4dd1cc4, whoops. * oopsy: ability ids can only match entire ids This prevents Bootshine (35) from matching Hakaze (1D35). * raidboss: O8S timeline tweaks (quisquous#62) Adding targetable/untargetable lines where appropriate * Add DispatchToJS lock to prevent races These data races have always been there, but were only going to be exposed in PullCounter, which called an OverlayMessage that dispatched an event, which could collide with game events dispatched from the fast thread. These races result in bogus Javascript to execute, such as such as document.dispatchEvent(new CustomEvent('onSendSaveData', { detail: {document.dispatchEvent(new CustomEvent('onSendUserConfigLocation', { detail: "data":" etc ...because both of these events are trying to be sent simultaneously, and the string builder still has text from the previous event in progress. This leads to anonymous Javascript errors like "Unexpected token )", which are awfully confusing. It also sometimes but not always leads to JSON serialization C# errors. (This could explain why PullCounter never started when loaded remotely as well. onSendSaveData could be clobbering other events.) This should fix races that will become a lot more common in pull quisquous#64. * raidboss: Fix triggers returning integers This fixes a bug introduced in 6c2c560 where anything that returned an integer was incorrectly handled, so all delaySeconds became zero. One day I will learn enough Javascript not to make silly mistakes. * Custom User Config Local Directory (quisquous#64) Adds user-specified config directory override to the C# config panel. By default, it picks the user directory adjacent to the CactbotOverlay.dll (and not to the html). So if you are using cactbot out of a git repo outside of your ACT directory, you likely will need to specify a directory override. Closes quisquous#56. * Minor fixes after pull quisquous#64 * Allow blank as a custom user directory default This removes several levels of redundant default filling. * Simplify user location messages If remote user directories aren't going to check for existence, then there's no reason to do it in the local case either. This cleans up needing to send the overlay name along with the message and just allows the remote side to send the user directory alone. This also removes the checks for css and js, which allows for overriding css without overriding js in the case of local files. * Refactor SendFastRateEvents preamble/cleanup Since I'm about to add another early out, hoist the timer and semaphore code outside the SendFastRateEvents body so that nobody can mess it up. * Automatically add onDOMContentReady listener Every plugin will now get code injected to wait for DOMContentReady. Once this is done, it will message the overlay plugin back, which will automatically send the user data info, and the overlay can initialize itself. (Overlays currently pre-initialize and then reinitialize, but that can be cleaned up after this commit.) * Wait for user config before initializing overlay This is the followup to 476ea6d which changes overlays to wait on the onOverlayInitialize message before creating any of their objects, reading options, or hooking up to DOM elements. This should stop initialization races ONCE AND FOR ALL. This also reverts most of 65cf263 which half-heartedly tried to fix this. * util: make_timeline.py improvements (quisquous#67) Added FFlogs support, event filtering, and phases. Also supports importing from external scripts now. * oopsy: belated 4.2 flag value updates Many flag values changed in 4.2, most notably: * '0' is never an attack anymore (which will help issue quisquous#63) * instant death is 33 not 32 * mysterious 3C shift is now a 3D shift The 3D change explains why oopsy was flakily dropping damage messages. * oopsy: Add all nameless autos as ids Thanks to Nemekh on the fflogs discord. * oopsy: reset first puller on countdown start/cancel This should hopefully address issue quisquous#63. * readme: add more details to install instructions * Add fr.js to overlay html files This should fix some errors going on in issue quisquous#57. * raidboss: fix localization return values This fixes the return values from triggers to properly return localized results. Currently they only return the 'en' value. This should help with issue quisquous#57. * Read data files in unicode Previously, this was destroying any non-ASCII characters in read files, which meant that it was impossible to specify these properly, either as regexes or as outputs. This fixes a number of problems in issue quisquous#57. * raidboss: fr localization for test triggers From quisquous#57 (comment) * Add french engage replaceSync for test timeline This fixes the issue mentioned in: quisquous#57 (comment) * raidboss: add load/run o7s triggers * raidboss: O7S timeline fixes (quisquous#72) Minor oversights. They were close enough to not lose syncing but still enough to be misleading * raidboss: fix o7s visual continuity times Fixes quisquous#73. * raidboss: add suppressSeconds to triggers Too many triggers have had to do this manually, such as heavensfall or kefka triggers. Add this option that prevents retriggering for N seconds. * raidboss: Clean up triggers to use suppressSeconds * raidboss: better ucu fireball logic Don't use the debuff anywhere, just track if you took the fireball. Also, shorten the triggers back to not talk about thunder. * Adding O8S timeline syncs (quisquous#66) * raidboss: Adding O8S timeline syncs The diff may not be super useful, so I'll summarize the changes (beyond simply adding syncs). Preserved the existing naming for the most part. - Added the Thunder/Fire hits of Celestriad - Fixed some slight +/- 1s timing issues - Added all non-starstrafe Soaks (The Path of Light, renamed for clarity), since they were included, albeit inconsistently, later on in the fight. Having specific timing for them is useful for dodging into it. - Added All Things Ending to Past/Future End during Starstrafe. The ATE cast on the normal ones is fast enough (3s) that it's not worth timing it, but there's a bigger gap in that one (8s). - Also added Meteor hit (the stack) during Starstrafe - Added some missing Ultimas (1328, 1358, 1543) - Added a missing Ultimate Embraces (1364) - Changed tether times to activation rather than appearance * raidboss: O8S timeline minor fixes Forgot to save these. * raidboss: O8S end of fight syncs * Support mixing remote url with local user files For obvious security reasons, Chrome doesn't let you load files off the local disk from a remote url. To fix this, have the C# load all the user files from disk and pass them to JS if that directory exists locally. JS can then inject this text directly into the page. If the directory doesn't exist locally, then the overlay will link to the user page. This should let all combinations of local and remote for url and user directories work properly. Fixes quisquous#56. * Log dll versions and locations when starting cactbot This is a bit verbose, but I've had a ton of users who have old dlls lying around their system (quisquous#75), or the wrong OverlayPlugin (quisquous#70), or a misnamed FFXIV plugin (quisquous#50). This should help debug things. * raidboss: more o8s triggers * raidboss: fix o7s rot triggers * raidboss: improve mana release triggers for tts Closes quisquous#59. * Bump version to 0.6.4 * Fix null config directory preventing overlays from starting The Uri constructor apparently throws exceptions (oops) and this was causing the overlay to explode and not send the init message. This would happen if you hadn't set the user config directory. I had tested this earlier, but not after the most recent user file sending changes that broke it. SORRY T_T Fixes quisquous#76. * Bump version to 0.6.5 * de: Create de.js (quisquous#84) Add german locale * de: Update jobs.html (quisquous#79) Add de.js to html file * de: Update pull_counter.html (quisquous#80) Add de.js to html file * de: Update oopsyraidsy.html (quisquous#81) Add de.js to html file * de: Update xephero-cactbot.html (quisquous#82) Add de.js to html file * de: Update raidboss.html (quisquous#83) Add de.js (german lang) * Allow spaces in cactbot user config dir locations Uri unescaping, oops. /o\ Fixes quisquous#90. * Try loading user files html-relative before dll-relative This is the way it used to work and allows folks working in git repositories that aren't inside of the ACT folder to have their user folder work as they'd expect. If the html-relative doesn't exist, it'll fall back to DLL-relative. Fixes quisquous#88. * Localize o5s for de/fr/jp (quisquous#85) JEDF clients should now work with timeline sync and triggers. timeline translation for german client added * Clean up formatting from pull quisquous#85 Multi-line conditionals need {}. Two space indent everywhere. Line up }; with the { indent level. * Fix fr engage message in pull quisquous#85 See comments in quisquous#57 for more details. * Split out regexes by language after pull quisquous#85 * Register languages before initializing overlay * Add more ignore zones to xephero dps * oopsy: ignore pvp and eureka * raidboss: eureka triggers * Fix typo in o2s.js Fixes quisquous#94. * raidboss: add fafnir eureka trigger * o8s: Fix hyperdrive and ultimate embrace triggers Closes quisquous#98. * raidboss: pass locale through as data.lang This should help with issue quisquous#57. * o8s: add fr translation for mana release This uses the new data parameter from fcc101c to help issue quisquous#57. * jobs: Add option for showing hp/mp for other jobs * Add eureka map module * eureka: make resize helper match real thing * eureka: be more permissive with tracker regexes Tracker regexes can have hyphens in them too. * eureka: prevent console error when changing zones * rdmty: fix resize handle not appearing The path to resize_handle.js Fixes quisquous#102. * eureka: add ability to import times from web tracker (quisquous#103) * eureka: Split out tracker names explicitly * eureka: split out tracker vs local times If you've seen a mob die, then that's more accurate than a tracker time. * eureka: add more readme information * eureka: add star in NM readme text * raidboss: allow sound volume settings for infoText Previously you could only do this on named sounds. * eureka: add wraith counter
SiliconExarch
pushed a commit
to SiliconExarch/cactbot
that referenced
this pull request
Feb 19, 2024
Sorry for the fast follow, but quisquous#62 had a small error in one of the job steps that I didn't catch earlier. This should fix it (confirmed with re-testing).
github-actions bot
pushed a commit
to SiliconExarch/cactbot
that referenced
this pull request
Feb 19, 2024
Sorry for the fast follow, but quisquous#62 had a small error in one of the job steps that I didn't catch earlier. This should fix it (confirmed with re-testing). 72aa4ba
github-actions bot
pushed a commit
to SiliconExarch/cactbot
that referenced
this pull request
Feb 19, 2024
Sorry for the fast follow, but quisquous#62 had a small error in one of the job steps that I didn't catch earlier. This should fix it (confirmed with re-testing). 72aa4ba
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The diff may not be super useful, so I'll summarize the changes (beyond simply adding syncs). My group got to 1389 on this, so up to that is script-generated, past that is hand-made from fflogs. Preserved your existing naming for the most part.
This is untested so far. I go in again on Friday, I just wanted to put it up now in case there's anything glaring