+ jHjI{Rt^RIHt^RIt^RIt^RIt^RIt^RIHt^RI t ^RI t ^RI t RRlt RRlt RR ltR R ltR R ltRRltRRltRRltRRltRRltRRltRtRRltRRltRR ltR!R"ltR#R$ltR%R&lt!R'R(4tR)R*ltR+R,lt R-R.lt!R/R0lt"]#R18Xd]$!]!!44hR#)2zFresh-checkout source update completion and its stdlib-only parent transport. Imported before a swap; executed by path from the selected tree afterward. The parent never imports application helpers from the replacement checkout. ) annotationsN)Pathc$V^8dQhRRRRRR/#)pathrdatadictreturnNone)formats"A/home/ubuntu/.hermes/hermes-agent/hermes_cli/update_completion.py __annotate__rs!d$4cVPR4pVP\P!V4RR7VP V4R#).tmputf-8encodingN) with_suffix write_textjsondumpsreplace)rr temporarys&& r _write_jsonrs<  (I D)G< drc V^8dQhRRRR/#)rcodeintr r )r s"r rrs--s-s-rc&V^8dV#^V, #)r )rs&r _exit_statusr!s194,#*,rc ,V^8dQhRRRRRRRRR R/#) rrequestr result_pathrrrreceipt dict | Noner r )r s"r rrs1TCR]bercp\VR^RVR,R,RVRVRRRVPR4/4V#)z[The bootstrap child's answer; windows_resume stays the parent's token (and its atexit net).schema update_idr% exit_codewindows_resumeN pm_receipt)rget)r#r$rr%s&&&&r _write_bootstrap_resultr.sG ![')"4["A;PT7,dL'++lB[ Krc(V^8dQhRRRRRRRR/#) rproczsubprocess.Popenr#r request_pathrr r r )r s"r rr&s*!!+!d!$!SW!rc^RIHp^RIHpHpV!V4pVeRV R2p\ RV 24VP RRV4VPP4pVeqVPPR 4VR ,R ,8XdC\P!\P!VP44VR &\W!4V!V4R#) aEWindows: bind the suspended completion child to the update's job, then resume it. Post-commit, refusing the child would fail a committed update, so a child the job refuses still runs, fenced by its own checkout lease instead (R5b: it joins our lock holding a lease byte, which keeps the checkout busy after a killed owner until the child exits). Never silently: the refusal is printed and recorded as a failed ``update_custody`` step in this run's receipt and, before the child runs (it has not read its request yet), in the receipt it resumes, so the terminal receipt carries it.update_receipt)bind_child_to_update_treeresume_suspended_childNz6the update's job would not take the completion child (z=), so it runs outside the job, holding its own checkout leaseu ⚠ Update completion: update_custodyFr)r%) hermes_clir4hermes_cli.update_lockr5r6print record_step_currentr-rrloadsrr) r0r#r1r4r5r6refusaldetailcurrents &&& r _bind_and_resumerA&s*X'-GJ7)TDD )&23""#3UFC ))--/  7<<#3#3K#@GIDVWbDc#c!%DJJw||,D!EGI   .4 rc V^8dQhRRRR/#)rr#rr r )r s"r rr?s\\D\T\rc \VR,4p\PPR4'dH\P !\P RRRRR\V4.VR\PR R 7\\PVR ,R R 7pR@FpVPVR4K \P!RR7;_uu_4p\V4R, p\V4R, p/VCR\PP!4/Cp\\V4R, 4VR&\#WP4\P RRRRRRRVR, 2\VR, 4\V4\V4. p^RIHpHp \P*!V3RVRVR\P,R\P./\P0R8Xd R R!R"V !V4/MR#\P2V,/Bp \4P6!R$4!R%4p \P0R8wd \9WV4V PP;R&4p \PP=V P?W'*R'74\PPA4V 'dKxT PC4p T PPU4\WT 4p \XPZ!TP]R/R074pTR1,^8wgTR2,TR3,R2,8wd \_R44hTR5,T 8wd \_R64hTPaR34pTecTPaR24TR3,R2,8wg4TPaR74'dT ^8HTPaR84R98H8wd \_R:4hT ^8XdTf \_R;4hTuuRRR4# \DEdlp\P0R8XdC\PF!T PH\JPL4M~ \NdMqi;i\P !R(R)R*R+\T PH4.\P\P\P^ R!\P2R,7T PQ4T PC^ R-7h T PC^ R-7i;i T PQ4T PC^ R-7i T PC^ R-7i;i;i \DdpTPSR.4YhRp?ii;iRp?ii;i T PPU4i;i \b\^\d\f3d/pR5T ;'g^R3RRT R?2/uRp?uuRRR4#Rp?ii;i +'giR#;i)AzOWait for new code; zero exit without a correlated terminal result fails closed.sourcelinux-I-Sz-Bz-czpimport sys; sys.path.insert(0, sys.argv[1]); from pm.libatomic import install_before_lock; install_before_lock()NF)cwdstdoutstderrcheckhome1) HERMES_HOMEPYTHONUNBUFFEREDzhermes-completion-)prefixz request.jsonz result.json stdout_isattybytecodebytecode_cache-u-Xutf8pycache_prefix=hermes_cli/update_completion.py)CREATE_SUSPENDEDcheckout_lock_fdsrHenvrIrJposixstart_new_sessionTpass_fds creationflagsrri )finaltaskkillz/Tz/Fz/PID)stdinrIrJtimeoutrKr_)rcz@Completion cleanup failed; child processes may still be running. utf-8-sigrr(r)r%z%completion response identity mismatchr*z/completion response disagrees with process exit finished_atoutcomesuccessz/completion receipt does not attest this outcomez-completion did not publish a terminal receiptr+errorz (the completion process exited )) PYTHONPATH PYTHONHOME VIRTUAL_ENV)4rsysplatform startswith subprocessrun executablestrDEVNULLrosenvironpoptempfileTemporaryDirectoryrIisattyrr9rYrZPopenPIPESTDOUTnameCREATE_NO_WINDOWcodecsgetincrementaldecoderrAread1writedecodeflushwait BaseExceptionkillpgpidsignalSIGKILLProcessLookupErrorkilladd_notecloser!rr= read_text ValueErrorr-OSErrorKeyError TypeError)r#rootr[key directoryr1r$commandrYrZr0decoderchunkrexc cleanup_errorresultr%s& r run_completionr?s  ! "D  ||w''dD$^_bcg_hj Z5G5Gu V rzzwv MC: T;  $ $,@ A AYI7 9o 5 CWCoszz/@/@/BC$'Y*(D$E !L*>>4tT64?[bcs[tZuIvt??@|$c+&68 O  SS#&S/9SGQGXGXSSUSZSZ^eSe#T:7H7NO!{,KK{{=11AI7;;y#9Y#FG !RSSqyW_ !PQQ [ B A<  -.ww')!IIdhh?1! !# D$DHH 'V-7-?-? HZHZ.8.@.@"TX5?5P5PR .  " -   " -.  " - " -  - _`, -+ 4 KK   "Xy9 NIt=Mtu$DTF!LN NW B AR NS B A AsEW0-BP P %W0>CV$=V$?W0 VT//QT Q- *T,Q- -A0TT.UU=T UU T= *U =U U U U: "U5 5U: :U==VVV!!W0$W-?W(W-W0(W--W00 X c V^8dQhRRRR/#)rrrr r r )r s"r rrsAA$A4Arc^RIHp\PVP4pWnVR,VnVPPV4Vn R#)r r3r)N) r8r4object__new__ UpdateReceiptrcorrelation_idr<set current_token)rr4r%s& r _resume_receiptrsF)nn^99:GL!+.G*3377@Grc V^8dQhRRRR/#rr#rr r&r )r s"r rrs 7 7D 7[ 7rc^RIHpVR,R,pVPV4pVe(V^,PR4'd V^,#VP V4#)r r3r%r)re)r8r4read_run_recordr-finalized_receipt)r#r4r)founds& r _read_terminal_receiptrsX) ";/I  * *9 5E U1X\\-88Qx  + +I 66rc V^8dQhRRRR/#rr )r s"r rrsTkrc^RIHpV!VR,R,4pVeV^,PR4R8wdR#V^,PRR4V^,#)r )rr%r)Nrfrunning writer_pid)hermes_cli.update_receiptrr-rw)r#rrs& r _running_recordrsP9 GI.{; rc V^8dQhRRRR/#rr#rr r r )r s"r rrsMMdMtMrcFVPR4'gR#\VR,4R, pVPVPR,4pVP RRR7VP V4R# \ dp\R T 24Rp?R#Rp?ii;i) ahThe gateway /update watcher's status: the code is committed (same as _complete_selected). Guarded like ``update_cmd_fleet._write_gateway_update_exit_code`` (which this stdlib-only process cannot import): an unwritable home must not turn the committed run into an error before its result is written (review P1). The watcher then falls back to its own timeout. gateway_modeNrLz.update_exit_coder0rru5 ⚠ Could not publish the gateway's update status: )r-r with_namer~rrrr:)r#markerrrs& r _publish_gateway_successrs ;;~ & &Mgfo&)<<$$V[[6%9: S73&! M EcUKLLMsAA<< B BB zDesktop app build owed:c(V^8dQhRRRRRRRR/#) rr#rrecordr&reasonrsr r r )r s"r rrs) : : :k :3 :SW :rcjVPR4'gR#\;QJd?RT;'g/PR4;'gR 4F 'gK RM4 RM0!RT;'g/PR4;'gR 44'dR#\R\ RV 2RR 7R#) aName the owed Desktop build when this run never got the build to its own verdict (F6, review P4). The builder prints the line only when the build itself fails; a run that never reached it (dependencies owed, the tail refused by the lock, a completion process lost) would otherwise end as a plain "Update complete." in the hand-off. A ``build`` stage on the run's record means the build ran and printed this line itself when the Desktop app failed. desktopNc3z"TF1p\V\4;'dVPR4R8HxK3 R#5i)r~buildN) isinstancerr-).0marks& r -_report_desktop_build_owed..s2 qNpd:dD ! A Adhhv&6'&A ANps;;stagesTFz  rr )r-anyr:DESKTOP_BUILD_OWED)r#rrs&&&r _report_desktop_build_owedrs ;;y ! ! s qv||Y[N`N`aiNjNpNpnpNp qsss qv||Y[N`N`aiNjNpNpnpNp qqq B!"!F8 ,D9rc$V^8dQhRRRRRR/#)rr#rrrsr r )r s"r rrs&!_!_D!_#!_$!_rc^RIHp\RV 2RR7\V4pVf^RIHpV!\ VR,44\V\V4R 4VP4VPR VR R 7\V4'*pV'd\\V44TPV'd^M^R 4\V4pT;'g/PR4pVR8Xd^M VR8Xd^M^pV^8Xd \!V4RVRVRRRR/# \dp\RT 24Rp?LRp?ii;i)atThe parent's side of C3: the code is committed, but the completion never answered (review P2). Reached when spawning or reading the completion failed (an OSError from the temporary directory, the request file or Popen) or the child died without a correlated terminal result (OOM, SIGKILL). The tail is owed, never a failed update: it is armed for the next launch, the run's receipt is a success that names the ``completion`` follow-up, and the exit is 0 -- unless the user's local changes are still parked (#122557: ``partial``, exit 1). Runs in the parent (pre-swap) interpreter: everything it uses is already imported there. r3u/ ⚠ Source update completion did not finish: TrNarm_completionrD4 ⚠ Could not record the owed source-update tail: z$the update completion did not finish completionz.the next launch or `hermes update` finishes itretryz*completion owed after the code was updatedrf interruptedpartialr*r%r+r,)r8r4r:rhermes_cli.venv_syncrr Exceptionrrresume_run_recordrecord_followuprrfinalize_pending_update_receiptr-r) r#rr4r%rrparkedrfrs && r settle_lost_completionrs2* ;F8 DDQ$W-G P ; 4 12 3 #7OG,DFlm((*&&|VCs&t-g66  #G, -66FqKwx(1}}"!!),G]*3W 5IqD qy ) y'3CT)(K1QRD qy ) "7 GG% L DSEJKKLsD## E.EEc(V^8dQhRRRRRRRR/#rr#rr1rr$r rr )r s"r rr@s(66d6$6T6c6rc ^RIp^RIHp^RIHp^RIHpHp\VR,4pVR,R,p \PR8XdC^R I H p V !V4p V e1\R V P;'gR V P 2 R 24h^R IHp Hp V !V4^RIHpV!V4VP)V 4;_uu_4V!4VP+RVRR7V !V4VP-V 4VR&\/W4RRR4\1V!V44RRRRRRRVR, 2\1VR, 4\1V4\1V4R. p^RI Hp\5\6P8!WV!V4V!V4R74pVP;4'g\=WRRV R24#V# TP-T 4TR&\/Y4i;i +'giL;i) r N)r%)ensure_tools_for_sync)activation_environmentproject_pythonrDr%r)win32)_acquire_checkoutz+could not join the update's checkout lock (zheld by process ri)rcollect_superseded_generations)convert_treeless_checkout_firstT)explicit project_rootevict_incompatible_pluginsr,rFrGrTrUrVrWrSrX --prepared)rZ)rHr[r^rzthe completion process exited z without a result)pmr% pm.clientrpm.environmentsrrrrmrnr9r RuntimeErrorrrrrrhermes_cli.gitlockrworker_context sync_venvlast_for_updaterrsrZr!rpcallexistsr)r#r1r$rr%rrrrr)rrefusedrrrrrZrs&&& r _preparer@s /F ! "D ";/I ||w =#D)   ##*>>#U#U7G }5U"VVW YZ ZS4B#D)    * * / " # LL$TVZL [ *4 0$+$;$;I$FGL !  . +>$'(T4vtwO_G`Fa5b4;;<< #k"2LBG 9  ?UVZ?[1B41HJ KD     #G,&DTFJ[$\^ ^ K%%,$;$;I$FGL !  . + *s$ G  #F<.G $_complete_selected..s 6QRY6Zr) rrrpre_update_snapshot_idr!rannounce followups before_buildz"the source-update tail did not runrc32<"TF wrVS9xK R#5ir#r )rr_rs& r r%_complete_selected..sD)wtDN*)sT)clear_completionno_gateway_restartgateway_restartz+--no-gateway-restart: deferred, marker keptrestartskippeduQ→ Gateway restart deferred (--no-gateway-restart); restart gateways separately.u# → Gateway restart not skipped (z,): gateways are still off the checkout code.u ✓ Gateway restart skipped: . incompleterrgr+)_pre_update_plan_windows_gateway_resumeupdate_completer )(r8rrrhermes_cli.source_completionrhermes_cli.update_inventoryrrrrrrrrr PROJECT_ROOTr_LAST_SIBLING_SNAPSHOTSr-_sweep_bytecode_after_updater SystemExitrstype__name__rappendrrr._write_gateway_update_exit_coder:_fleet_restart_skip_reason_pending_fleet_restart_needed#_restart_gateway_fleet_after_updategetattr*_resume_windows_gateways_and_merge_outcome_verify_fleet_after_updater)r#rrrrrrrrrrrcomplete plan_datarowrr) runtime_saferr tail_owedr.skipr1rsf @r _complete_selectedrNys?>EEXX ! "D'0H078K0L-I$4*+c8b 8b)--PZ\^B_!`B_3-"6#"6B_!`8b+cD++GH,=>(*I L 1/ '),9N 0Q^I_&';<&{{+?@$[[)=>>TDY.Z \ D)DD)DDIy! 9~228<{{'//%'TUY * ab  0 0 6D 88:: 3D69efg %t,Y * /vQ78<CZ[Gu(M(MHS\]99'7K[C\^eft^uv))$HX@Y\*77x*9 Oq"a$ z "1S//T#Y//"7D2VW f-,/00 1s&J=9AK KL/AL**L/c V^8dQhRRRR/#rr )r s"r rrs\\\$\rc VPR4pV'd.VPR4'dVPR4R8wdR#^RIHpV!4PV4R# \dp\ RT R24Rp?R#Rp?ii;i) aRestart the gateways the POSIX pause stopped: the dependencies are synced and the launchers published, the product builds (minutes) have not started. A gateway needs nothing later to boot. A failure stays owed: the post-build resume retries it and records the outcome. POSIX only: the fleet restart skips what this restarted (``already_restarted``). A Windows set keeps resuming after the fleet restart, which would otherwise drain it a second time.r+ resume_neededrnr\N)_mu) ⚠ Paused gateway restart incomplete (z); retrying after the build)r-hermes_cli.update_cmdrR%_resume_windows_gateways_after_updaterr:)r#tokenrRrs& r r$r$sx KK( )E  /22eii 6Kw6V(\ 2259 \ 9#>YZ[[\sA(( B 3BB c2]tRtRtRtRRltRtRtRtR#) _ForwardedOutputizHThe parent's pipe preserves its terminal's prompt policy and log mirror.cV^8dQhRR/#)rrzrr )r s"r r_ForwardedOutput.__annotate__s44t4rc "WuVnVnR#r#streamterminal)selfr\rzs&&&r __init___ForwardedOutput.__init__s%+" T]rc VP#r#)r])r^s&r rz_ForwardedOutput.isattys }}rc .\VPV4#r#)rEr\)r^r~s&&r __getattr___ForwardedOutput.__getattr__st{{D))rr[N) r? __module__ __qualname____firstlineno____doc__r_rzrd__static_attributes__r rr rWrWsR4*rrWc V^8dQhRRRR/#rr )r s"r rrsccTcdcrc~^RIHpVPP4pVe\ WP R4R#R#)z[The tail raised out of ``_complete_selected``: before the build stage, the build never ran.r3Nz%the source-update tail did not finish)r8r4r<r-rr)r#r4r@s& r _report_unbuilt_desktoprms3)%%))+G"7LL:abrc$V^8dQhRRRRRR/#)rr#rr$rr rr )r s"r rrs!11T111rcv^RIHp^RIHp\ VR,4V!VP R4VR,R,4VP RR4^RrT\V4'g^R rT\"P$P R4pT'dTeTP'4MR pT'd \)T4^RIHp T !TR,4TP3YE4\5T4p T 'dT P R4R8Xd^pT'dTR,'d^RIHp T !R4T 'g T;'g^p\9TR^RTR,R,RTRT RTR,/4T# \d^R rTTPYTR 7EL+\dGpTPR9d+\T4TPR RTP 24R p?ELvR p?i\dAp\T4TPR \T4P RT 24R p?ELR p?ii;i \.d0pTP1TR,R,RRT 24R p?ELR p?ii;i \"P$P R4pT'dTeTP'4MR pT'd \)T4^RIHp T !TR,4M< \.d/pTP1TR,R,RRT 24R p?MR p?ii;iTP3YE4\5T4p T 'dT P R4R8Xd^pT'dTR,'d^RIHp T !R4T 'g T;'g^p\9TR^RTR,R,RTRT RTR,/4i;i)r r3)accept_worker_receiptr%r,r)rrgzsource update completionz6local changes left in the stash; re-apply them by handz9KeyboardInterrupt: interrupted after the code was updatedr*Nrzcompletion exited z: zhermes_cli.update_custody)rTr+z!Windows gateway recovery failed: rfr)rAFr(r*)r N)r8r4 pm.receiptrprr-rrNKeyboardInterrupt#finalize_interrupted_update_receiptr=rrmrrr>r?rmmodulesrefusal_noticer:rSrTr owe_followuprrrAr) r#r$r4rprrrcustodyr>rTterminal_receiptrAs && r _finishrzsT)0GI&''++l3WY5G 5TU 20&( !'**V&++//"=>.2w7J'((*PT  'N S S 1':J2K L 66tD1':  0 4 4Y ?9 LD GN++ M +E 2991DK agi&8&E{TX ')97CS;T"   KM SWf::6:R Z 889 $ #G ,  * *<;MchhZ9X Y U(&&|S 8J8J7K2cU5STTU S  ' ' (:;(GIY*KC5(Q S S S++//"=>.2w7J'((*PT  'N S S 1':J2K L S  ' ' (:;(GIY*KC5(Q S S S 66tD1':  0 4 4Y ?9 LD GN++ M +E 2991DK agi&8&E{TX ')97CS;T"  sE7H67 H3I3H3#H3$;G%I3% H32H335H.(I3.H33I36 I0$I++I03AN89 N8KN8 L&$L N8LAN8N8*N8N8 /N8cV^8dQhRR/#)rr rr )r s"r rr$s::c:rc\\\PR,4wr\P !VP RR74pVR,^8wd \R4h\\4P4P^,pV\VR,4P48wd \R4h\PP^\V44\\PVP!RR44\nR \PR ,9d^R IHpV!V4\'W!4#\)W V4#) :Nrdrr(z%unsupported source completion requestrDz*completion checkout does not match requestrQFr:r~NN)activate_dependencies)maprrmargvrr=rr__file__resolveparentsrinsertrsrWrIr-rrrz _bootstrap)r1r$r#rrs r rr$s #D#((3- 8Ljj///EFGxA@AA > ! ! # + +A .D tGH%&..00EFFHHOOAs4y!!#**gkk/5.QRCJsxx|# :d#w,, g[ 99rc(V^8dQhRRRRRRRR/#rr )r s"r rr8s. j j jD jt j jrc X\WV4# \dE\TR,4^RIHpTP R^R7\ Y^\T44u#\\3d>p\YR\T4;'g\T4P4uRp?#Rp?ii;i)zTDependency preparation in the parent's interpreter; its failure is a follow-up (A6).r%r3zr?)r#r1r$r4rs&&& r rr8s j{;; c *+-:: JVY ; [&wSBXY`Babb z "j$G.#c(JhJhVZ[^V_VhVhiijs( A B)B)+B),2B$B)$B)__main__)%ri __future__rrrrurpathlibrrprmrxrr!r.rArrrrrrrrrrrrrNr$rWrmrzrrr?r=r rr rs #   -!2\~A 7JM,/ :!_HHD6rM`\" * *c1h:( j  z TV r