Initial attempt to merge playerbots from zero into two#188
Conversation
|
Bots module compiles and loads from config as expected (now that I've setup my project with the correct directory structure: binary in /usr/local/bin and configs in /usr/local/etc). Pulled in some code from https://github.com/celguar/mangosbot-bots to fill out the deathknight logic. As per @Foereaper the modules need additional TLC across all cores. |
leprasmurf
changed the title
|
@leprasmurf checked and eliminated most of the complaints the checks had, there are two mediums remaining: https://app.codacy.com/gh/mangostwo/server/pullRequest?prid=9108498 |
billy1arm
left a comment
billy1arm
left a comment
There was a problem hiding this comment.
Reviewed 606 of 614 files at r1, 8 of 8 files at r2, all commit messages.
Reviewable status:complete! all files reviewed, all discussions resolved (waiting on @leprasmurf)
Most of the changes are pulled from mangoszero/server@ee33683 or mangos zero HEAD.
Whitespace and buffer overflow fix (PR checks)
Pull in some of the updates from Celguar's playerbot repo.
Post-increments converted to pre-increments for efficiency
* Convert <string>.find -> <string>.starts_with * Pass string by constant reference instead of value * Pre-increment in the for loop post-action
* Built-in and module auctionhouse bots conflict on the .ahbot command. Reconfigure module integration for .auctionbot command and finish integrating. * cppcheck fixes * Bot module documentation
|
Player, Random, and the Auction House Bots are all building and configurable. The ahbot module config disables the built-in ahbot's buyer and seller agents before enabling its own. There is a final assertion that I've yet to track down:
Otherwise the auction house bot responds to the console or chat commands |
|
There's a memory leak in here somewhere, my instance crashes within 24 hours of running. I'm going to recompile with the |
|
Is this partly merged? |
|
❌ Build Mangos TWO Server 22.02.147 failed (commit c4afe52bab by @billy1arm) |
Attempting to fix compile issues after recent changes
|
@leprasmurf - My apologies for my extended absence, trying to get things moving again |
|
✅ Build Mangos TWO Server 22.02.149 completed (commit 461809ce29 by @billy1arm) |
|
✅ Build Mangos TWO Server 22.02.151 completed (commit c2637af477 by @billy1arm) |
…ame code Co-authored-by: billy1arm <2219118+billy1arm@users.noreply.github.com>
…ellAura*.cpp (#219) * [Decomp][Spell] Cohesion split: extract SpellAuras.cpp handlers to SpellAura*.cpp Mirror of mangosthree #188-195 (bundled, same source file). Pure cohesion split of SpellAuras.cpp (12,928 -> 6,069 lines) with no behaviour change. The Aura::HandleXxx / SpellAuraHolder handlers move to dedicated TUs: - SpellAuraDummy.cpp - HandleAuraDummy. - SpellAuraControl.cpp - crowd-control + immunity handlers (carries the static frozenAuraTypes table). - SpellAuraMisc.cpp - shapeshift-boost/absorb/misc handlers. - SpellAuraShapeshift.cpp - shapeshift/movement handlers. - SpellAuraPeriodic.cpp - periodic + stat handlers. - SpellAuraTrailing.cpp - trailing handlers. - SpellAuraLate.cpp - late crit/mastery/vehicle handlers. - SpellAuraAddModifier.cpp - HandleAddModifier. SpellAuras.h is unchanged; aura holder lifecycle / ApplyModifier dispatch stay in SpellAuras.cpp. CMake file(GLOB) auto-compiles the new TUs. * [Decomp][Spell] Drop redundant caster null-check in HandleSchoolAbsorb Codacy/Cppcheck flagged the `caster &&` condition as always-true: the method early-returns on `if (!caster)` near the top, so caster cannot be null at this point. No behaviour change.
4 participants
Most of the changes are pulled from mangoszero/server@ee33683 or mangos zero HEAD.
WIP:
Almost there, compilation fails towards the end
This change is