updated my ssh_config.nix

sunshine is a cool idea but has been causing annoying ui glitches and started preventing the display manning for starting
Its a cool idea in theory but not useful enough for me to want to debug

.github/workflows/build_systems.yml

@@ -23,6 +23,6 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - name: Build default package
-        run: "nixos-rebuild build --flake ./#${{ matrix.system }}"
+        run: "nixos-rebuild build --accept-flake-config --flake ./#${{ matrix.system }}"
       - name: copy to nix-cache
         run: nix copy --accept-flake-config --to unix:///host-nix/var/nix/daemon-socket/socket .#nixosConfigurations.${{ matrix.system }}.config.system.build.toplevel

.github/workflows/fix_eval_warnings.yml

@@ -1,30 +0,0 @@
-name: fix_eval_warnings
-on:
-  workflow_run:
-    workflows: ["build_systems"]
-    types: [completed]
-
-jobs:
-  check-warnings:
-    if: >-
-      github.event.workflow_run.conclusion != 'cancelled' &&
-      github.event.workflow_run.head_branch == 'main' &&
-      (github.event.workflow_run.event == 'push' || github.event.workflow_run.event == 'schedule')
-    runs-on: self-hosted
-    permissions:
-      contents: write
-      pull-requests: write
-
-    steps:
-      - uses: actions/checkout@v4
-
-      - name: Fix eval warnings
-        env:
-          GH_TOKEN: ${{ secrets.GH_TOKEN_FOR_UPDATES }}
-        run: >-
-          nix develop .#devShells.x86_64-linux.default -c
-          python -m python.eval_warnings.main
-          --run-id "${{ github.event.workflow_run.id }}"
-          --repo "${{ github.repository }}"
-          --ollama-url "${{ secrets.OLLAMA_URL }}"
-          --run-url "${{ github.event.workflow_run.html_url }}"

.github/workflows/merge_flake_lock_update.yml

@@ -6,24 +6,18 @@ on:
 
 jobs:
   merge:
-    runs-on: ubuntu-latest
+    runs-on: self-hosted
 
     permissions:
       contents: write
       pull-requests: write
 
     steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
       - name: merge_flake_lock_update
-        run: |
-          pr_number=$(gh pr list --state open --author RichieCahill --label flake_lock_update --json number --jq '.[0].number')
-          echo "pr_number=$pr_number" >> $GITHUB_ENV
-          if [ -n "$pr_number" ]; then
-            gh pr merge "$pr_number" --rebase
-          else
-            echo "No open PR found with label flake_lock_update"
-          fi
+        run: >-
+          nix develop .#devShells.x86_64-linux.default -c
+          python -m python.gitea_flake_lock merge
+          --repo "${{ github.repository }}"
         env:
-          GITHUB_TOKEN: ${{ secrets.GH_TOKEN_FOR_UPDATES }}
+          GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }}
+          GITEA_URL: https://gitea.tmmworkshop.com

.github/workflows/pytest.yml

@@ -7,7 +7,6 @@ on:
   pull_request:
     branches:
       - main
-  merge_group:
 
 jobs:
   pytest:

.github/workflows/update-flake-lock.yml

@@ -6,18 +6,20 @@ on:
 
 jobs:
   lockfile:
-    runs-on: ubuntu-latest
+    runs-on: self-hosted
+    permissions:
+      contents: write
+      pull-requests: write
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
-      - name: Install Nix
-        uses: DeterminateSystems/nix-installer-action@main
       - name: Update flake.lock
-        uses: DeterminateSystems/update-flake-lock@main
-        with:
-          token: ${{ secrets.GH_TOKEN_FOR_UPDATES }}
-          pr-title: "Update flake.lock"
-          pr-labels: |
-            dependencies
-            automated
-            flake_lock_update
+        run: nix flake update
+      - name: Create or update flake.lock PR
+        env:
+          GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }}
+          GITEA_URL: https://gitea.tmmworkshop.com
+        run: >-
+          nix develop .#devShells.x86_64-linux.default -c
+          python -m python.gitea_flake_lock update
+          --repo "${{ github.repository }}"

.gitignore

@@ -169,3 +169,6 @@ test.*
 # Frontend build output
 frontend/dist/
 frontend/node_modules/
+
+# data from testing llms
+data/*

.vscode/settings.json

@@ -40,7 +40,6 @@
     "cgroupdriver",
     "charliermarsh",
     "Checkpointing",
-    "cloudflared",
     "codellama",
     "codezombiech",
     "compactmode",
@@ -204,6 +203,7 @@
     "peerconnection",
     "PESKYFOX",
     "PGID",
+    "pgvector",
     "pipewire",
     "pkgs",
     "plugdev",
@@ -308,6 +308,7 @@
     "usernamehw",
     "userprefs",
     "vaninventory",
+    "vdev",
     "vfat",
     "victron",
     "virt",

common/global/default.nix

@@ -37,10 +37,17 @@
 
   nixpkgs = {
     overlays = builtins.attrValues outputs.overlays;
-    config.allowUnfree = true;
+    config = {
+      allowUnfree = true;
+      permittedInsecurePackages = [
+        "openssl-1.1.1w" # This is for discord-canary
+      ];
+    };
   };
 
   services = {
+    dbus.implementation = "dbus";
+
     # firmware update
     fwupd.enable = true;
 

common/global/nix.nix

@@ -34,6 +34,7 @@ in
       warn-dirty = false;
       flake-registry = ""; # disable global flake registries
       connect-timeout = 10;
+      download-buffer-size = 536870912;
       fallback = true;
     };
 

common/optional/monitoring-agent.nix

@@ -0,0 +1,256 @@
+{
+  config,
+  lib,
+  pkgs,
+  ...
+}:
+let
+  monitoringInterface = "ztwfunumly";
+  nodeTextfileDir = "/var/lib/prometheus-node-exporter-textfile";
+
+  mkProcessNameTemplate =
+    perPid: template: if perPid then "${template}:{{.PID}}:{{.StartTime}}" else template;
+
+  mkProcessMatchers = perPid: [
+    {
+      name = mkProcessNameTemplate perPid "{{.Username}}:{{.Matches.Module}}";
+      cmdline = [ "^/nix/store[^ ]*/bin/python[^ ]* -m (?P<Module>[^ ]+)" ];
+    }
+    {
+      name = mkProcessNameTemplate perPid "{{.Username}}:{{.Matches.Wrapped}}";
+      cmdline = [
+        "^/nix/store[^ ]*/bin/python[^ ]* /nix/store[^ ]*/bin/\\.?(?P<Wrapped>[^ /]+?)(?:-wrapped)?(?:\\s|$)"
+      ];
+    }
+    {
+      name = mkProcessNameTemplate perPid "{{.Username}}:{{.Matches.Wrapped}}";
+      cmdline = [
+        "^/nix/store[^ ]*/bin/node /nix/store[^ ]*-(?P<Wrapped>[A-Za-z0-9._+-]+)-[0-9][^ /]*/"
+      ];
+    }
+    {
+      name = mkProcessNameTemplate perPid "{{.Username}}:{{.Matches.Wrapped}}";
+      cmdline = [ "^/nix/store[^ ]*/(?:bin/|lib/[^ ]*/)?\\.?(?P<Wrapped>[^ /]+?)(?:-wrapped)?(?:\\s|$)" ];
+    }
+    {
+      name = mkProcessNameTemplate perPid "{{.Username}}:{{.ExeBase}}";
+      cmdline = [ ".+" ];
+    }
+  ];
+
+  perPidConfig = pkgs.writeText "process-exporter-per-pid.yaml" (
+    builtins.toJSON {
+      process_names = mkProcessMatchers true;
+    }
+  );
+
+  zpoolLatencyScript = pkgs.writeShellScript "zpool-latency-exporter" ''
+        set -euo pipefail
+
+        out_dir=${lib.escapeShellArg nodeTextfileDir}
+        host=${lib.escapeShellArg config.networking.hostName}
+        tmp_file="$(mktemp "$out_dir/zpool.prom.XXXXXX")"
+        trap 'rm -f "$tmp_file"' EXIT
+
+        pools="$(zpool list -H -o name | paste -sd, -)"
+
+        cat >"$tmp_file" <<'EOF'
+    # HELP zpool_iostat_total_wait_read_ns Average total read wait time reported by zpool iostat.
+    # TYPE zpool_iostat_total_wait_read_ns gauge
+    # HELP zpool_iostat_total_wait_write_ns Average total write wait time reported by zpool iostat.
+    # TYPE zpool_iostat_total_wait_write_ns gauge
+    # HELP zpool_iostat_disk_wait_read_ns Average disk read wait time reported by zpool iostat.
+    # TYPE zpool_iostat_disk_wait_read_ns gauge
+    # HELP zpool_iostat_disk_wait_write_ns Average disk write wait time reported by zpool iostat.
+    # TYPE zpool_iostat_disk_wait_write_ns gauge
+    # HELP zpool_iostat_syncq_wait_read_ns Average synchronous queue read wait time reported by zpool iostat.
+    # TYPE zpool_iostat_syncq_wait_read_ns gauge
+    # HELP zpool_iostat_syncq_wait_write_ns Average synchronous queue write wait time reported by zpool iostat.
+    # TYPE zpool_iostat_syncq_wait_write_ns gauge
+    # HELP zpool_iostat_asyncq_wait_read_ns Average asynchronous queue read wait time reported by zpool iostat.
+    # TYPE zpool_iostat_asyncq_wait_read_ns gauge
+    # HELP zpool_iostat_asyncq_wait_write_ns Average asynchronous queue write wait time reported by zpool iostat.
+    # TYPE zpool_iostat_asyncq_wait_write_ns gauge
+    EOF
+
+        zpool iostat -Hplvy -y 1 1 | awk -F '\t' -v host="$host" -v pools="$pools" '
+          function esc(str, out) {
+            out = str
+            gsub(/\\/, "\\\\", out)
+            gsub(/"/, "\\\"", out)
+            return out
+          }
+
+          function emit(metric, pool, vdev, value) {
+            if (value == "" || value == "-") {
+              return
+            }
+
+            printf "%s{host=\"%s\",pool=\"%s\",vdev=\"%s\"} %s\n",
+              metric,
+              esc(host),
+              esc(pool),
+              esc(vdev),
+              value
+          }
+
+          BEGIN {
+            split(pools, pool_names, ",")
+            for (idx in pool_names) {
+              if (pool_names[idx] != "") {
+                known_pools[pool_names[idx]] = 1
+              }
+            }
+          }
+
+          NF == 0 {
+            next
+          }
+
+          {
+            row_name = $1
+
+            if (row_name in known_pools) {
+              current_pool = row_name
+              current_vdev = "_pool"
+            } else if (current_pool == "") {
+              next
+            } else {
+              current_vdev = row_name
+            }
+
+            emit("zpool_iostat_total_wait_read_ns", current_pool, current_vdev, $8)
+            emit("zpool_iostat_total_wait_write_ns", current_pool, current_vdev, $9)
+            emit("zpool_iostat_disk_wait_read_ns", current_pool, current_vdev, $10)
+            emit("zpool_iostat_disk_wait_write_ns", current_pool, current_vdev, $11)
+            emit("zpool_iostat_syncq_wait_read_ns", current_pool, current_vdev, $12)
+            emit("zpool_iostat_syncq_wait_write_ns", current_pool, current_vdev, $13)
+            emit("zpool_iostat_asyncq_wait_read_ns", current_pool, current_vdev, $14)
+            emit("zpool_iostat_asyncq_wait_write_ns", current_pool, current_vdev, $15)
+          }
+        ' >>"$tmp_file"
+
+        mv "$tmp_file" "$out_dir/zpool.prom"
+        trap - EXIT
+  '';
+in
+{
+  networking.firewall.interfaces.${monitoringInterface}.allowedTCPPorts = [
+    9100
+    9134
+    9256
+    9257
+    9633
+  ];
+
+  services.prometheus.exporters = {
+    node = {
+      enable = true;
+      enabledCollectors = [
+        "pressure"
+        "processes"
+        "systemd"
+      ];
+      extraFlags = [ "--collector.textfile.directory=${nodeTextfileDir}" ];
+    };
+
+    process = {
+      enable = true;
+      user = "root";
+      group = "root";
+      settings.process_names = mkProcessMatchers false;
+      extraFlags = [
+        "-gather-smaps=false"
+        "-remove-empty-groups=true"
+        "-threads=false"
+      ];
+    };
+
+    smartctl.enable = true;
+    zfs.enable = true;
+  };
+
+  programs.atop = {
+    enable = true;
+    atopService.enable = true;
+    atopRotateTimer.enable = true;
+    atopacctService.enable = true;
+    settings.interval = 30;
+  };
+
+  systemd = {
+    services = {
+      prometheus-process-pid-exporter = {
+        description = "Prometheus process exporter with per-PID naming";
+        wantedBy = [ "multi-user.target" ];
+        after = [ "network.target" ];
+        serviceConfig = {
+          ExecStart = ''
+            ${pkgs.prometheus-process-exporter}/bin/process-exporter \
+              --web.listen-address 0.0.0.0:9257 \
+              --config.path ${perPidConfig} \
+              -children=false \
+              -gather-smaps=false \
+              -remove-empty-groups=true \
+              -threads=false
+          '';
+          User = "root";
+          Group = "root";
+          Restart = "always";
+          WorkingDirectory = "/tmp";
+          CapabilityBoundingSet = [ "" ];
+          DeviceAllow = [ "" ];
+          LockPersonality = true;
+          MemoryDenyWriteExecute = true;
+          NoNewPrivileges = true;
+          PrivateDevices = true;
+          PrivateTmp = true;
+          ProtectClock = true;
+          ProtectControlGroups = true;
+          ProtectHome = true;
+          ProtectHostname = true;
+          ProtectKernelLogs = true;
+          ProtectKernelModules = true;
+          ProtectKernelTunables = true;
+          ProtectSystem = "strict";
+          RemoveIPC = true;
+          RestrictAddressFamilies = [
+            "AF_INET"
+            "AF_INET6"
+          ];
+          RestrictNamespaces = true;
+          RestrictRealtime = true;
+          RestrictSUIDSGID = true;
+          SystemCallArchitectures = "native";
+          UMask = "0077";
+        };
+      };
+
+      zpool-latency-exporter = {
+        description = "Exports ZFS latency metrics for node_exporter textfile collection";
+        after = [ "zfs-import.target" ];
+        requires = [ "zfs-import.target" ];
+        path = [
+          config.boot.zfs.package
+          pkgs.coreutils
+          pkgs.gawk
+        ];
+        serviceConfig = {
+          Type = "oneshot";
+          ExecStart = zpoolLatencyScript;
+        };
+      };
+    };
+
+    timers.zpool-latency-exporter = {
+      wantedBy = [ "timers.target" ];
+      timerConfig = {
+        OnBootSec = "2m";
+        OnUnitActiveSec = "60s";
+        Unit = "zpool-latency-exporter.service";
+      };
+    };
+
+    tmpfiles.rules = [ "d ${nodeTextfileDir} 0755 root root - -" ];
+  };
+}

common/optional/syncthing_base.nix

@@ -12,7 +12,7 @@
       brain.id = "SSCGIPI-IV3VYKB-TRNIJE3-COV4T2H-CDBER7F-I2CGHYA-NWOEUDU-3T5QAAN"; # cspell:disable-line
       ipad.id = "KI76T3X-SFUGV2L-VSNYTKR-TSIUV5L-SHWD3HE-GQRGRCN-GY4UFMD-CW6Z6AX"; # cspell:disable-line
       jeeves.id = "ICRHXZW-ECYJCUZ-I4CZ64R-3XRK7CG-LL2HAAK-FGOHD22-BQA4AI6-5OAL6AG"; # cspell:disable-line
-      phone.id = "TBRULKD-7DZPGGZ-F6LLB7J-MSO54AY-7KLPBIN-QOFK6PX-W2HBEWI-PHM2CQI"; # cspell:disable-line
+      phone.id = "JPVQKQW-CFXOJXT-Q5G5F3H-QIDHDRE-GKHPTQB-GXZUQSP-U7FR7F7-INP3AAH"; # cspell:disable-line
       rhapsody-in-green.id = "ASL3KC4-3XEN6PA-7BQBRKE-A7JXLI6-DJT43BY-Q4WPOER-7UALUAZ-VTPQ6Q4"; # cspell:disable-line
     };
   };

common/optional/update.nix

@@ -4,7 +4,7 @@
     flags = [ "--accept-flake-config" ];
     randomizedDelaySec = "1h";
     persistent = true;
-    flake = "github:RichieCahill/dotfiles";
+    flake = "git+https://gitea.tmmworkshop.com/richie/dotfiles?ref=main";
     allowReboot = true;
     dates = "Sat *-*-* 06:00:00";
   };

flake.lock

@@ -8,11 +8,11 @@
       },
       "locked": {
         "dir": "pkgs/firefox-addons",
-        "lastModified": 1773979456,
-        "narHash": "sha256-9kBMJ5IvxqNlkkj/swmE8uK1Sc7TL/LIRUI958m7uBM=",
+        "lastModified": 1780027372,
+        "narHash": "sha256-LQ3CUdVZoKQqWzS2eEpY0rp9bJuzqydNFJUiJ6De9r8=",
         "owner": "rycee",
         "repo": "nur-expressions",
-        "rev": "81e28f47ac18d9e89513929c77e711e657b64851",
+        "rev": "ef18b76eabdf4f9b2ce8e99e78ce698923693300",
         "type": "gitlab"
       },
       "original": {
@@ -29,11 +29,11 @@
         ]
       },
       "locked": {
-        "lastModified": 1774007980,
-        "narHash": "sha256-FOnZjElEI8pqqCvB6K/1JRHTE8o4rer8driivTpq2uo=",
+        "lastModified": 1780099287,
+        "narHash": "sha256-efIPwVGtIWIjWcznhaop6XN6HxnOL8800hF6CBNvlqQ=",
         "owner": "nix-community",
         "repo": "home-manager",
-        "rev": "9670de2921812bc4e0452f6e3efd8c859696c183",
+        "rev": "7d8127d308c3fb9664f7e643eec944be74ebb37d",
         "type": "github"
       },
       "original": {
@@ -43,12 +43,15 @@
       }
     },
     "nixos-hardware": {
+      "inputs": {
+        "nixpkgs": "nixpkgs"
+      },
       "locked": {
-        "lastModified": 1774018263,
-        "narHash": "sha256-HHYEwK1A22aSaxv2ibhMMkKvrDGKGlA/qObG4smrSqc=",
+        "lastModified": 1780065812,
+        "narHash": "sha256-SCSLUKBmwlSLGQ8Xbr8PjRFtiHNk0l9ktqkcmqdBkfE=",
         "owner": "nixos",
         "repo": "nixos-hardware",
-        "rev": "2d4b4717b2534fad5c715968c1cece04a172b365",
+        "rev": "b76b5639c0593e0aeb0b5879ad62d4b30596c144",
         "type": "github"
       },
       "original": {
@@ -60,27 +63,24 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1773821835,
-        "narHash": "sha256-TJ3lSQtW0E2JrznGVm8hOQGVpXjJyXY2guAxku2O9A4=",
-        "owner": "nixos",
-        "repo": "nixpkgs",
-        "rev": "b40629efe5d6ec48dd1efba650c797ddbd39ace0",
-        "type": "github"
+        "lastModified": 1767892417,
+        "narHash": "sha256-8bW3q88CEg2u4hSP66Vf4lpbLonHz7hqDNBMcCY7E9U=",
+        "rev": "3497aa5c9457a9d88d71fa93a4a8368816fbeeba",
+        "type": "tarball",
+        "url": "https://releases.nixos.org/nixos/unstable/nixos-26.05pre924538.3497aa5c9457/nixexprs.tar.xz"
       },
       "original": {
-        "owner": "nixos",
-        "ref": "nixos-unstable",
-        "repo": "nixpkgs",
-        "type": "github"
+        "type": "tarball",
+        "url": "https://channels.nixos.org/nixos-unstable/nixexprs.tar.xz"
       }
     },
     "nixpkgs-master": {
       "locked": {
-        "lastModified": 1774051532,
-        "narHash": "sha256-d3CGMweyYIcPuTj5BKq+1Lx4zwlgL31nVtN647tOZKo=",
+        "lastModified": 1780101106,
+        "narHash": "sha256-VcvUdRb9rzKBbF6oMaMiAt+6HZQ1gom9b2dUybhVTVY=",
         "owner": "nixos",
         "repo": "nixpkgs",
-        "rev": "8620c0b5cc8fbe76502442181be1d0514bc3a1b7",
+        "rev": "26b82d423c4f6fda4a8015182516c938f8104337",
         "type": "github"
       },
       "original": {
@@ -106,12 +106,28 @@
         "type": "github"
       }
     },
+    "nixpkgs_2": {
+      "locked": {
+        "lastModified": 1779560665,
+        "narHash": "sha256-tpyBcxPpcQb8ukyNF7DoCwfSY3VPsxHoYwj00Cayv5o=",
+        "owner": "nixos",
+        "repo": "nixpkgs",
+        "rev": "64c08a7ca051951c8eae34e3e3cb1e202fe36786",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nixos",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
     "root": {
       "inputs": {
         "firefox-addons": "firefox-addons",
         "home-manager": "home-manager",
         "nixos-hardware": "nixos-hardware",
-        "nixpkgs": "nixpkgs",
+        "nixpkgs": "nixpkgs_2",
         "nixpkgs-master": "nixpkgs-master",
         "nixpkgs-stable": "nixpkgs-stable",
         "sops-nix": "sops-nix",
@@ -125,11 +141,11 @@
         ]
       },
       "locked": {
-        "lastModified": 1773889674,
-        "narHash": "sha256-+ycaiVAk3MEshJTg35cBTUa0MizGiS+bgpYw/f8ohkg=",
+        "lastModified": 1777944972,
+        "narHash": "sha256-VfGRo1qTBKOe3s2gOv8LSoA6Fk19PvBlwQ1ECN0Evn8=",
         "owner": "Mic92",
         "repo": "sops-nix",
-        "rev": "29b6519f3e0780452bca0ac0be4584f04ac16cc5",
+        "rev": "c591bf665727040c6cc5cb409079acb22dcce33c",
         "type": "github"
       },
       "original": {

overlays/default.nix

@@ -21,7 +21,6 @@
         alembic
         apprise
         apscheduler
-        confluent-kafka
         fastapi
         fastapi-cli
         httpx
@@ -42,6 +41,7 @@
         sqlalchemy
         tenacity
         textual
+        tiktoken
         tinytuya
         typer
         websockets

pyproject.toml

@@ -26,6 +26,7 @@ dependencies = [
 [project.scripts]
 database = "python.database_cli:app"
 van-inventory = "python.van_inventory.main:serve"
+whisper-transcribe = "python.tools.whisper.transcribe:main"
 
 [dependency-groups]
 dev = [
@@ -50,6 +51,7 @@ lint.ignore = [
     "COM812", # (TEMP) conflicts when used with the formatter
     "ISC001", # (TEMP) conflicts when used with the formatter
     "S603",   # (PERM) This is known to cause a false positive
+    "S607",   # (PERM) This is becoming a consistent annoyance
 ]
 
 [tool.ruff.lint.per-file-ignores]
@@ -78,9 +80,7 @@ lint.ignore = [
 "python/congress_tracker/**" = [
     "TC003", # (perm) this creates issues because sqlalchemy uses these at runtime
 ]
-"python/eval_warnings/**" = [
-    "S607", # (perm) gh and git are expected on PATH in the runner environment
-]
+
 "python/alembic/**" = [
     "INP001", # (perm) this creates LSP issues for alembic
 ]

python/alembic/data_science_dev/versions/2026_03_25-attach_partitions_to_posts.py

@@ -46,12 +46,7 @@ ALREADY_ATTACHED_QUERY = text("""
 def upgrade() -> None:
     """Attach all weekly partition tables to the posts parent table."""
     connection = op.get_bind()
-    already_attached = {
-        row[0]
-        for row in connection.execute(
-            ALREADY_ATTACHED_QUERY, {"parent": f"{schema}.posts"}
-        )
-    }
+    already_attached = {row[0] for row in connection.execute(ALREADY_ATTACHED_QUERY, {"parent": f"{schema}.posts"})}
 
     for year in range(PARTITION_START_YEAR, PARTITION_END_YEAR + 1):
         for week in range(1, iso_weeks_in_year(year) + 1):
@@ -74,7 +69,4 @@ def downgrade() -> None:
     for year in range(PARTITION_START_YEAR, PARTITION_END_YEAR + 1):
         for week in range(1, iso_weeks_in_year(year) + 1):
             table_name = f"posts_{year}_{week:02d}"
-            op.execute(
-                f"ALTER TABLE {schema}.posts "
-                f"DETACH PARTITION {schema}.{table_name}"
-            )
+            op.execute(f"ALTER TABLE {schema}.posts DETACH PARTITION {schema}.{table_name}")

python/alembic/data_science_dev/versions/2026_03_27-adding_congress_data_83bfc8af92d8.py

@@ -0,0 +1,153 @@
+"""adding congress data.
+
+Revision ID: 83bfc8af92d8
+Revises: a1b2c3d4e5f6
+Create Date: 2026-03-27 10:43:02.324510
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import sqlalchemy as sa
+from alembic import op
+
+from python.orm import DataScienceDevBase
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+# revision identifiers, used by Alembic.
+revision: str = "83bfc8af92d8"
+down_revision: str | None = "a1b2c3d4e5f6"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+schema = DataScienceDevBase.schema_name
+
+
+def upgrade() -> None:
+    """Upgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table(
+        "bill",
+        sa.Column("congress", sa.Integer(), nullable=False),
+        sa.Column("bill_type", sa.String(), nullable=False),
+        sa.Column("number", sa.Integer(), nullable=False),
+        sa.Column("title", sa.String(), nullable=True),
+        sa.Column("title_short", sa.String(), nullable=True),
+        sa.Column("official_title", sa.String(), nullable=True),
+        sa.Column("status", sa.String(), nullable=True),
+        sa.Column("status_at", sa.Date(), nullable=True),
+        sa.Column("sponsor_bioguide_id", sa.String(), nullable=True),
+        sa.Column("subjects_top_term", sa.String(), nullable=True),
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("created", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_bill")),
+        sa.UniqueConstraint("congress", "bill_type", "number", name="uq_bill_congress_type_number"),
+        schema=schema,
+    )
+    op.create_index("ix_bill_congress", "bill", ["congress"], unique=False, schema=schema)
+    op.create_table(
+        "legislator",
+        sa.Column("bioguide_id", sa.Text(), nullable=False),
+        sa.Column("thomas_id", sa.String(), nullable=True),
+        sa.Column("lis_id", sa.String(), nullable=True),
+        sa.Column("govtrack_id", sa.Integer(), nullable=True),
+        sa.Column("opensecrets_id", sa.String(), nullable=True),
+        sa.Column("fec_ids", sa.String(), nullable=True),
+        sa.Column("first_name", sa.String(), nullable=False),
+        sa.Column("last_name", sa.String(), nullable=False),
+        sa.Column("official_full_name", sa.String(), nullable=True),
+        sa.Column("nickname", sa.String(), nullable=True),
+        sa.Column("birthday", sa.Date(), nullable=True),
+        sa.Column("gender", sa.String(), nullable=True),
+        sa.Column("current_party", sa.String(), nullable=True),
+        sa.Column("current_state", sa.String(), nullable=True),
+        sa.Column("current_district", sa.Integer(), nullable=True),
+        sa.Column("current_chamber", sa.String(), nullable=True),
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("created", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_legislator")),
+        schema=schema,
+    )
+    op.create_index(op.f("ix_legislator_bioguide_id"), "legislator", ["bioguide_id"], unique=True, schema=schema)
+    op.create_table(
+        "bill_text",
+        sa.Column("bill_id", sa.Integer(), nullable=False),
+        sa.Column("version_code", sa.String(), nullable=False),
+        sa.Column("version_name", sa.String(), nullable=True),
+        sa.Column("text_content", sa.String(), nullable=True),
+        sa.Column("date", sa.Date(), nullable=True),
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("created", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["bill_id"], [f"{schema}.bill.id"], name=op.f("fk_bill_text_bill_id_bill"), ondelete="CASCADE"
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_bill_text")),
+        sa.UniqueConstraint("bill_id", "version_code", name="uq_bill_text_bill_id_version_code"),
+        schema=schema,
+    )
+    op.create_table(
+        "vote",
+        sa.Column("congress", sa.Integer(), nullable=False),
+        sa.Column("chamber", sa.String(), nullable=False),
+        sa.Column("session", sa.Integer(), nullable=False),
+        sa.Column("number", sa.Integer(), nullable=False),
+        sa.Column("vote_type", sa.String(), nullable=True),
+        sa.Column("question", sa.String(), nullable=True),
+        sa.Column("result", sa.String(), nullable=True),
+        sa.Column("result_text", sa.String(), nullable=True),
+        sa.Column("vote_date", sa.Date(), nullable=False),
+        sa.Column("yea_count", sa.Integer(), nullable=True),
+        sa.Column("nay_count", sa.Integer(), nullable=True),
+        sa.Column("not_voting_count", sa.Integer(), nullable=True),
+        sa.Column("present_count", sa.Integer(), nullable=True),
+        sa.Column("bill_id", sa.Integer(), nullable=True),
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("created", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(["bill_id"], [f"{schema}.bill.id"], name=op.f("fk_vote_bill_id_bill")),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_vote")),
+        sa.UniqueConstraint("congress", "chamber", "session", "number", name="uq_vote_congress_chamber_session_number"),
+        schema=schema,
+    )
+    op.create_index("ix_vote_congress_chamber", "vote", ["congress", "chamber"], unique=False, schema=schema)
+    op.create_index("ix_vote_date", "vote", ["vote_date"], unique=False, schema=schema)
+    op.create_table(
+        "vote_record",
+        sa.Column("vote_id", sa.Integer(), nullable=False),
+        sa.Column("legislator_id", sa.Integer(), nullable=False),
+        sa.Column("position", sa.String(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["legislator_id"],
+            [f"{schema}.legislator.id"],
+            name=op.f("fk_vote_record_legislator_id_legislator"),
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["vote_id"], [f"{schema}.vote.id"], name=op.f("fk_vote_record_vote_id_vote"), ondelete="CASCADE"
+        ),
+        sa.PrimaryKeyConstraint("vote_id", "legislator_id", name=op.f("pk_vote_record")),
+        schema=schema,
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table("vote_record", schema=schema)
+    op.drop_index("ix_vote_date", table_name="vote", schema=schema)
+    op.drop_index("ix_vote_congress_chamber", table_name="vote", schema=schema)
+    op.drop_table("vote", schema=schema)
+    op.drop_table("bill_text", schema=schema)
+    op.drop_index(op.f("ix_legislator_bioguide_id"), table_name="legislator", schema=schema)
+    op.drop_table("legislator", schema=schema)
+    op.drop_index("ix_bill_congress", table_name="bill", schema=schema)
+    op.drop_table("bill", schema=schema)
+    # ### end Alembic commands ###

python/alembic/data_science_dev/versions/2026_03_29-adding_legislatorsocialmedia_5cd7eee3549d.py

@@ -0,0 +1,58 @@
+"""adding LegislatorSocialMedia.
+
+Revision ID: 5cd7eee3549d
+Revises: 83bfc8af92d8
+Create Date: 2026-03-29 11:53:44.224799
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import sqlalchemy as sa
+from alembic import op
+
+from python.orm import DataScienceDevBase
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+# revision identifiers, used by Alembic.
+revision: str = "5cd7eee3549d"
+down_revision: str | None = "83bfc8af92d8"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+schema = DataScienceDevBase.schema_name
+
+
+def upgrade() -> None:
+    """Upgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table(
+        "legislator_social_media",
+        sa.Column("legislator_id", sa.Integer(), nullable=False),
+        sa.Column("platform", sa.String(), nullable=False),
+        sa.Column("account_name", sa.String(), nullable=False),
+        sa.Column("url", sa.String(), nullable=True),
+        sa.Column("source", sa.String(), nullable=False),
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("created", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.Column("updated", sa.DateTime(timezone=True), server_default=sa.text("now()"), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["legislator_id"],
+            [f"{schema}.legislator.id"],
+            name=op.f("fk_legislator_social_media_legislator_id_legislator"),
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_legislator_social_media")),
+        schema=schema,
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table("legislator_social_media", schema=schema)
+    # ### end Alembic commands ###

python/alembic/env.py

@@ -3,7 +3,6 @@
 from __future__ import annotations
 
 import logging
-import re
 import sys
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, Literal

python/alembic/richie/versions/2026_03_29-removed_ds_table_from_richie_db_c8a794340928.py

@@ -0,0 +1,187 @@
+"""removed ds table from richie DB.
+
+Revision ID: c8a794340928
+Revises: 6b275323f435
+Create Date: 2026-03-29 15:29:23.643146
+
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import sqlalchemy as sa
+from alembic import op
+from sqlalchemy.dialects import postgresql
+
+from python.orm import RichieBase
+
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+# revision identifiers, used by Alembic.
+revision: str = "c8a794340928"
+down_revision: str | None = "6b275323f435"
+branch_labels: str | Sequence[str] | None = None
+depends_on: str | Sequence[str] | None = None
+
+schema = RichieBase.schema_name
+
+
+def upgrade() -> None:
+    """Upgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table("vote_record", schema=schema)
+    op.drop_index(op.f("ix_vote_congress_chamber"), table_name="vote", schema=schema)
+    op.drop_index(op.f("ix_vote_date"), table_name="vote", schema=schema)
+    op.drop_index(op.f("ix_legislator_bioguide_id"), table_name="legislator", schema=schema)
+    op.drop_table("legislator", schema=schema)
+    op.drop_table("vote", schema=schema)
+    op.drop_index(op.f("ix_bill_congress"), table_name="bill", schema=schema)
+    op.drop_table("bill", schema=schema)
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    """Downgrade."""
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table(
+        "vote",
+        sa.Column("congress", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("chamber", sa.VARCHAR(), autoincrement=False, nullable=False),
+        sa.Column("session", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("number", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("vote_type", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("question", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("result", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("result_text", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("vote_date", sa.DATE(), autoincrement=False, nullable=False),
+        sa.Column("yea_count", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("nay_count", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("not_voting_count", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("present_count", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("bill_id", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("id", sa.INTEGER(), autoincrement=True, nullable=False),
+        sa.Column(
+            "created",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.Column(
+            "updated",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.ForeignKeyConstraint(["bill_id"], [f"{schema}.bill.id"], name=op.f("fk_vote_bill_id_bill")),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_vote")),
+        sa.UniqueConstraint(
+            "congress",
+            "chamber",
+            "session",
+            "number",
+            name=op.f("uq_vote_congress_chamber_session_number"),
+            postgresql_include=[],
+            postgresql_nulls_not_distinct=False,
+        ),
+        schema=schema,
+    )
+    op.create_index(op.f("ix_vote_date"), "vote", ["vote_date"], unique=False, schema=schema)
+    op.create_index(op.f("ix_vote_congress_chamber"), "vote", ["congress", "chamber"], unique=False, schema=schema)
+    op.create_table(
+        "vote_record",
+        sa.Column("vote_id", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("legislator_id", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("position", sa.VARCHAR(), autoincrement=False, nullable=False),
+        sa.ForeignKeyConstraint(
+            ["legislator_id"],
+            [f"{schema}.legislator.id"],
+            name=op.f("fk_vote_record_legislator_id_legislator"),
+            ondelete="CASCADE",
+        ),
+        sa.ForeignKeyConstraint(
+            ["vote_id"], [f"{schema}.vote.id"], name=op.f("fk_vote_record_vote_id_vote"), ondelete="CASCADE"
+        ),
+        sa.PrimaryKeyConstraint("vote_id", "legislator_id", name=op.f("pk_vote_record")),
+        schema=schema,
+    )
+    op.create_table(
+        "legislator",
+        sa.Column("bioguide_id", sa.TEXT(), autoincrement=False, nullable=False),
+        sa.Column("thomas_id", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("lis_id", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("govtrack_id", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("opensecrets_id", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("fec_ids", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("first_name", sa.VARCHAR(), autoincrement=False, nullable=False),
+        sa.Column("last_name", sa.VARCHAR(), autoincrement=False, nullable=False),
+        sa.Column("official_full_name", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("nickname", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("birthday", sa.DATE(), autoincrement=False, nullable=True),
+        sa.Column("gender", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("current_party", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("current_state", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("current_district", sa.INTEGER(), autoincrement=False, nullable=True),
+        sa.Column("current_chamber", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("id", sa.INTEGER(), autoincrement=True, nullable=False),
+        sa.Column(
+            "created",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.Column(
+            "updated",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_legislator")),
+        schema=schema,
+    )
+    op.create_index(op.f("ix_legislator_bioguide_id"), "legislator", ["bioguide_id"], unique=True, schema=schema)
+    op.create_table(
+        "bill",
+        sa.Column("congress", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("bill_type", sa.VARCHAR(), autoincrement=False, nullable=False),
+        sa.Column("number", sa.INTEGER(), autoincrement=False, nullable=False),
+        sa.Column("title", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("title_short", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("official_title", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("status", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("status_at", sa.DATE(), autoincrement=False, nullable=True),
+        sa.Column("sponsor_bioguide_id", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("subjects_top_term", sa.VARCHAR(), autoincrement=False, nullable=True),
+        sa.Column("id", sa.INTEGER(), autoincrement=True, nullable=False),
+        sa.Column(
+            "created",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.Column(
+            "updated",
+            postgresql.TIMESTAMP(timezone=True),
+            server_default=sa.text("now()"),
+            autoincrement=False,
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_bill")),
+        sa.UniqueConstraint(
+            "congress",
+            "bill_type",
+            "number",
+            name=op.f("uq_bill_congress_type_number"),
+            postgresql_include=[],
+            postgresql_nulls_not_distinct=False,
+        ),
+        schema=schema,
+    )
+    op.create_index(op.f("ix_bill_congress"), "bill", ["congress"], unique=False, schema=schema)
+    # ### end Alembic commands ###

python/data_science/bluesky_ids.py

@@ -1,104 +0,0 @@
-"""Utilities for converting Bluesky identifiers to numeric database IDs.
-
-Handles DID-to-user_id hashing, TID-to-post_id decoding, and AT-URI parsing.
-"""
-
-from __future__ import annotations
-
-import hashlib
-
-TID_CHARSET = "234567abcdefghijklmnopqrstuvwxyz"
-_TID_LENGTH = 13
-_BIGINT_MASK = 0x7FFFFFFFFFFFFFFF
-_AT_URI_SEGMENT_COUNT = 3
-
-
-def did_to_user_id(did: str) -> int:
-    """Convert a DID string to a deterministic 63-bit integer for user_id.
-
-    Uses SHA-256, truncated to 63 bits (positive signed BigInteger range).
-    Collision probability is negligible at Bluesky's scale (~tens of millions of users).
-
-    Args:
-        did: A Bluesky DID string, e.g. "did:plc:abc123".
-
-    Returns:
-        A positive 63-bit integer suitable for BigInteger storage.
-    """
-    digest = hashlib.sha256(did.encode()).digest()
-    return int.from_bytes(digest[:8], "big") & _BIGINT_MASK
-
-
-def tid_to_integer(tid: str) -> int:
-    """Decode a Bluesky TID (base32-sortbase) into a 64-bit integer for post_id.
-
-    TIDs are 13-character, base32-sortbase encoded identifiers that encode a
-    microsecond timestamp plus a clock ID. They are globally unique by construction.
-
-    Args:
-        tid: A 13-character TID string, e.g. "3abc2defghijk".
-
-    Returns:
-        A positive integer suitable for BigInteger storage.
-
-    Raises:
-        ValueError: If the TID is malformed (wrong length or invalid characters).
-    """
-    if len(tid) != _TID_LENGTH:
-        message = f"TID must be {_TID_LENGTH} characters, got {len(tid)}: {tid!r}"
-        raise ValueError(message)
-
-    result = 0
-    for char in tid:
-        index = TID_CHARSET.find(char)
-        if index == -1:
-            message = f"Invalid character {char!r} in TID {tid!r}"
-            raise ValueError(message)
-        result = result * 32 + index
-    return result
-
-
-def parse_at_uri(uri: str) -> tuple[str, str, str]:
-    """Parse an AT-URI into its components.
-
-    Args:
-        uri: An AT-URI string, e.g. "at://did:plc:abc123/app.bsky.feed.post/3abc2defghijk".
-
-    Returns:
-        A tuple of (did, collection, rkey).
-
-    Raises:
-        ValueError: If the URI doesn't have the expected format.
-    """
-    stripped = uri.removeprefix("at://")
-    parts = stripped.split("/", maxsplit=2)
-    if len(parts) != _AT_URI_SEGMENT_COUNT:
-        message = f"Expected {_AT_URI_SEGMENT_COUNT} path segments in AT-URI, got {len(parts)}: {uri!r}"
-        raise ValueError(message)
-    return parts[0], parts[1], parts[2]
-
-
-def post_id_from_uri(uri: str) -> int:
-    """Extract and decode the post_id (TID) from an AT-URI.
-
-    Args:
-        uri: An AT-URI pointing to a post.
-
-    Returns:
-        The post_id as an integer.
-    """
-    _did, _collection, rkey = parse_at_uri(uri)
-    return tid_to_integer(rkey)
-
-
-def user_id_from_uri(uri: str) -> int:
-    """Extract and hash the user_id (DID) from an AT-URI.
-
-    Args:
-        uri: An AT-URI pointing to a post.
-
-    Returns:
-        The user_id as an integer.
-    """
-    did, _collection, _rkey = parse_at_uri(uri)
-    return did_to_user_id(did)

python/data_science/bluesky_transform.py

@@ -1,143 +0,0 @@
-"""Transform Bluesky Jetstream messages into rows matching the Posts table schema."""
-
-from __future__ import annotations
-
-import json
-import logging
-from datetime import datetime
-
-from python.data_science.bluesky_ids import (
-    did_to_user_id,
-    post_id_from_uri,
-    tid_to_integer,
-    user_id_from_uri,
-)
-
-logger = logging.getLogger(__name__)
-
-INSTANCE = "bsky"
-POST_COLLECTION = "app.bsky.feed.post"
-EMBED_RECORD_TYPE = "app.bsky.embed.record"
-EMBED_RECORD_WITH_MEDIA_TYPE = "app.bsky.embed.recordWithMedia"
-
-
-def transform_jetstream_post(message: dict) -> dict:
-    """Transform a Jetstream commit message into a dict matching Posts table columns.
-
-    Expects a Jetstream message with kind=commit, operation=create,
-    collection=app.bsky.feed.post.
-
-    Args:
-        message: The full Jetstream JSON message.
-
-    Returns:
-        A dict with keys matching the Posts table columns.
-    """
-    did = message["did"]
-    commit = message["commit"]
-    record = commit["record"]
-
-    row: dict = {
-        "post_id": tid_to_integer(commit["rkey"]),
-        "user_id": did_to_user_id(did),
-        "instance": INSTANCE,
-        "date": datetime.fromisoformat(record["createdAt"]),
-        "text": record.get("text", ""),
-        "langs": _extract_langs(record),
-        "like_count": 0,
-        "reply_count": 0,
-        "repost_count": 0,
-        "reply_to": None,
-        "replied_author": None,
-        "thread_root": None,
-        "thread_root_author": None,
-        "repost_from": None,
-        "reposted_author": None,
-        "quotes": None,
-        "quoted_author": None,
-        "labels": _extract_labels(record),
-        "sent_label": None,
-        "sent_score": None,
-    }
-
-    _extract_reply_refs(record, row)
-    _extract_quote_refs(record, row)
-
-    return row
-
-
-def is_post_create(message: dict) -> bool:
-    """Check if a Jetstream message is a post creation event.
-
-    Args:
-        message: The full Jetstream JSON message.
-
-    Returns:
-        True if this is a create commit for app.bsky.feed.post.
-    """
-    if message.get("kind") != "commit":
-        return False
-    commit = message.get("commit", {})
-    return commit.get("operation") == "create" and commit.get("collection") == POST_COLLECTION
-
-
-def _extract_langs(record: dict) -> str | None:
-    """Extract langs array as a JSON string, or None if absent."""
-    langs = record.get("langs")
-    if langs is None:
-        return None
-    return json.dumps(langs)
-
-
-def _extract_labels(record: dict) -> str | None:
-    """Extract self-labels as a JSON string, or None if absent."""
-    labels_obj = record.get("labels")
-    if labels_obj is None:
-        return None
-    values = labels_obj.get("values", [])
-    if not values:
-        return None
-    label_strings = [label.get("val", "") for label in values]
-    return json.dumps(label_strings)
-
-
-def _extract_reply_refs(record: dict, row: dict) -> None:
-    """Populate reply_to, replied_author, thread_root, thread_root_author from record.reply."""
-    reply = record.get("reply")
-    if reply is None:
-        return
-
-    parent = reply.get("parent", {})
-    parent_uri = parent.get("uri")
-    if parent_uri:
-        row["reply_to"] = post_id_from_uri(parent_uri)
-        row["replied_author"] = user_id_from_uri(parent_uri)
-
-    root = reply.get("root", {})
-    root_uri = root.get("uri")
-    if root_uri:
-        row["thread_root"] = post_id_from_uri(root_uri)
-        row["thread_root_author"] = user_id_from_uri(root_uri)
-
-
-def _extract_quote_refs(record: dict, row: dict) -> None:
-    """Populate quotes and quoted_author from embed record references."""
-    embed = record.get("embed")
-    if embed is None:
-        return
-
-    embed_type = embed.get("$type", "")
-
-    if embed_type == EMBED_RECORD_TYPE:
-        _set_quote_from_record(embed.get("record", {}), row)
-    elif embed_type == EMBED_RECORD_WITH_MEDIA_TYPE:
-        inner_record = embed.get("record", {}).get("record", {})
-        _set_quote_from_record(inner_record, row)
-
-
-def _set_quote_from_record(record_ref: dict, row: dict) -> None:
-    """Set quotes and quoted_author from a record reference object."""
-    uri = record_ref.get("uri")
-    if uri and POST_COLLECTION in uri:
-        row["quotes"] = post_id_from_uri(uri)
-        row["quoted_author"] = user_id_from_uri(uri)

python/data_science/firehose_consumer.py

@@ -1,203 +0,0 @@
-"""Kafka consumer that ingests Bluesky posts into the partitioned Posts table.
-
-Consumes Jetstream messages from Kafka, transforms them into Posts rows,
-and batch-inserts them into PostgreSQL with manual offset commits.
-
-Usage:
-    firehose-consumer
-    firehose-consumer --kafka-servers kafka:9092 --batch-size 500
-"""
-
-from __future__ import annotations
-
-import json
-import logging
-import signal
-from os import getenv
-from threading import Event
-from typing import Annotated
-
-import typer
-from confluent_kafka import Consumer, KafkaError, KafkaException
-from sqlalchemy.orm import Session
-
-from python.data_science.bluesky_transform import is_post_create, transform_jetstream_post
-from python.data_science.ingest_posts import ingest_batch
-from python.orm.common import get_postgres_engine
-from python.orm.data_science_dev.posts.failed_ingestion import FailedIngestion
-
-logger = logging.getLogger(__name__)
-
-DEFAULT_TOPIC = "bluesky.firehose.posts"
-DEFAULT_KAFKA_SERVERS = "localhost:9092"
-DEFAULT_GROUP_ID = "bluesky-posts-ingestor"
-DEFAULT_BATCH_SIZE = 500
-POLL_TIMEOUT_SECONDS = 5.0
-
-shutdown_event = Event()
-
-app = typer.Typer(help="Consume Bluesky posts from Kafka and ingest into PostgreSQL.")
-
-
-@app.command()
-def main(
-    kafka_servers: Annotated[str, typer.Option(help="Kafka bootstrap servers")] = "",
-    topic: Annotated[str, typer.Option(help="Kafka topic to consume from")] = "",
-    group_id: Annotated[str, typer.Option(help="Kafka consumer group ID")] = "",
-    batch_size: Annotated[int, typer.Option(help="Messages per DB insert batch")] = DEFAULT_BATCH_SIZE,
-) -> None:
-    """Consume Bluesky posts from Kafka and ingest into the partitioned posts table."""
-    logging.basicConfig(
-        level=logging.INFO,
-        format="%(asctime)s %(levelname)s %(message)s",
-        datefmt="%H:%M:%S",
-    )
-
-    servers = kafka_servers or getenv("KAFKA_BOOTSTRAP_SERVERS", DEFAULT_KAFKA_SERVERS)
-    topic_name = topic or getenv("BLUESKY_FIREHOSE_TOPIC", DEFAULT_TOPIC)
-    group = group_id or getenv("KAFKA_GROUP_ID", DEFAULT_GROUP_ID)
-
-    signal.signal(signal.SIGTERM, _handle_shutdown)
-    signal.signal(signal.SIGINT, _handle_shutdown)
-
-    consumer = _create_consumer(servers, group)
-    consumer.subscribe([topic_name])
-
-    engine = get_postgres_engine(name="DATA_SCIENCE_DEV")
-    total_inserted = 0
-
-    logger.info("Starting firehose consumer: topic=%s group=%s batch_size=%d", topic_name, group, batch_size)
-
-    try:
-        with Session(engine) as session:
-            while not shutdown_event.is_set():
-                inserted = _consume_batch(consumer, session, batch_size)
-                total_inserted += inserted
-                if inserted > 0:
-                    logger.info("Batch inserted %d rows (total: %d)", inserted, total_inserted)
-    except KafkaException:
-        logger.exception("Fatal Kafka error")
-    finally:
-        logger.info("Closing consumer (total inserted: %d)", total_inserted)
-        consumer.close()
-
-
-def _consume_batch(consumer: Consumer, session: Session, batch_size: int) -> int:
-    """Poll a batch of messages, transform, and insert into the database.
-
-    Args:
-        consumer: The Kafka consumer instance.
-        session: SQLAlchemy database session.
-        batch_size: Maximum number of messages to consume per batch.
-
-    Returns:
-        Number of rows successfully inserted.
-    """
-    messages = consumer.consume(num_messages=batch_size, timeout=POLL_TIMEOUT_SECONDS)
-    if not messages:
-        return 0
-
-    rows: list[dict] = []
-    for message in messages:
-        error = message.error()
-        if error is not None:
-            if error.code() == KafkaError._PARTITION_EOF:  # noqa: SLF001 — confluent-kafka exposes this as a pseudo-private constant; no public alternative exists
-                continue
-            logger.error("Consumer error: %s", error)
-            continue
-
-        row = _safe_transform(message.value(), session)
-        if row is not None:
-            rows.append(row)
-
-    if not rows:
-        consumer.commit(asynchronous=False)
-        return 0
-
-    inserted = ingest_batch(session, rows)
-    consumer.commit(asynchronous=False)
-    return inserted
-
-
-def _safe_transform(raw_value: bytes | None, session: Session) -> dict | None:
-    """Transform a Kafka message value into a Posts row, logging failures.
-
-    Args:
-        raw_value: Raw message bytes from Kafka.
-        session: SQLAlchemy session for logging failures.
-
-    Returns:
-        A transformed row dict, or None if transformation failed.
-    """
-    if raw_value is None:
-        return None
-
-    try:
-        message = json.loads(raw_value)
-    except (json.JSONDecodeError, UnicodeDecodeError):
-        logger.exception("Failed to decode Kafka message")
-        _log_failed_ingestion(session, raw_value, "JSON decode error")
-        return None
-
-    if not is_post_create(message):
-        return None
-
-    try:
-        return transform_jetstream_post(message)
-    except (KeyError, ValueError, TypeError):
-        logger.exception("Failed to transform Jetstream message")
-        _log_failed_ingestion(session, raw_value, "Transform error")
-        return None
-
-
-def _log_failed_ingestion(session: Session, raw_value: bytes, error: str) -> None:
-    """Log a failed ingestion to the FailedIngestion table.
-
-    Args:
-        session: SQLAlchemy session.
-        raw_value: The raw message bytes.
-        error: Description of the error.
-    """
-    try:
-        session.add(
-            FailedIngestion(
-                raw_line=raw_value.decode(errors="replace")[:10000],
-                error=error,
-            )
-        )
-        session.commit()
-    except Exception:
-        session.rollback()
-        logger.exception("Failed to log ingestion failure")
-
-
-def _create_consumer(servers: str, group: str) -> Consumer:
-    """Create a configured Kafka consumer.
-
-    Args:
-        servers: Kafka bootstrap servers string.
-        group: Consumer group ID.
-
-    Returns:
-        A configured confluent_kafka.Consumer.
-    """
-    config = {
-        "bootstrap.servers": servers,
-        "group.id": group,
-        "auto.offset.reset": "earliest",
-        "enable.auto.commit": False,
-        "max.poll.interval.ms": 300000,
-        "fetch.min.bytes": 1024,
-        "session.timeout.ms": 30000,
-    }
-    return Consumer(config)
-
-
-def _handle_shutdown(_signum: int, _frame: object) -> None:
-    """Signal handler to trigger graceful shutdown."""
-    logger.info("Shutdown signal received")
-    shutdown_event.set()
-
-
-if __name__ == "__main__":
-    app()

python/data_science/firehose_producer.py

@@ -1,230 +0,0 @@
-"""Bluesky Jetstream firehose to Kafka producer.
-
-Connects to the Bluesky Jetstream WebSocket API with zstd compression,
-filters for post creation events, and produces them to a Kafka topic.
-
-Usage:
-    firehose-producer
-    firehose-producer --kafka-servers kafka:9092 --topic bluesky.firehose.posts
-"""
-
-from __future__ import annotations
-
-import json
-import logging
-import signal
-from os import getenv
-from threading import Event
-from typing import Annotated
-
-import typer
-from compression import zstd
-from confluent_kafka import KafkaError, KafkaException, Producer
-from websockets.exceptions import ConnectionClosed
-from websockets.sync.client import connect
-
-logger = logging.getLogger(__name__)
-
-JETSTREAM_URL = "wss://jetstream2.us-east.bsky.network/subscribe"
-DEFAULT_TOPIC = "bluesky.firehose.posts"
-DEFAULT_KAFKA_SERVERS = "localhost:9092"
-POLL_INTERVAL = 100
-POST_COLLECTION = "app.bsky.feed.post"
-
-shutdown_event = Event()
-
-app = typer.Typer(help="Stream Bluesky firehose posts into Kafka.")
-
-
-@app.command()
-def main(
-    kafka_servers: Annotated[str, typer.Option(help="Kafka bootstrap servers")] = "",
-    topic: Annotated[str, typer.Option(help="Kafka topic to produce to")] = "",
-    collections: Annotated[str, typer.Option(help="Comma-separated collections to subscribe to")] = POST_COLLECTION,
-) -> None:
-    """Connect to Bluesky Jetstream and produce post events to Kafka."""
-    logging.basicConfig(
-        level=logging.INFO,
-        format="%(asctime)s %(levelname)s %(message)s",
-        datefmt="%H:%M:%S",
-    )
-
-    servers = kafka_servers or getenv("KAFKA_BOOTSTRAP_SERVERS", DEFAULT_KAFKA_SERVERS)
-    topic_name = topic or getenv("BLUESKY_FIREHOSE_TOPIC", DEFAULT_TOPIC)
-
-    signal.signal(signal.SIGTERM, _handle_shutdown)
-    signal.signal(signal.SIGINT, _handle_shutdown)
-
-    producer = _create_producer(servers)
-    cursor: int | None = None
-
-    logger.info("Starting firehose producer → %s on %s", topic_name, servers)
-
-    while not shutdown_event.is_set():
-        try:
-            cursor = _stream_loop(producer, topic_name, collections, cursor)
-        except (ConnectionClosed, OSError):
-            logger.exception("WebSocket disconnected, reconnecting")
-        except KafkaException:
-            logger.exception("Kafka error, reconnecting")
-
-        if not shutdown_event.is_set():
-            logger.info("Reconnecting in 5 seconds (cursor=%s)", cursor)
-            shutdown_event.wait(timeout=5)
-
-    logger.info("Shutting down, flushing producer")
-    producer.flush(timeout=30)
-    logger.info("Producer shutdown complete")
-
-
-def _stream_loop(
-    producer: Producer,
-    topic: str,
-    collections: str,
-    cursor: int | None,
-) -> int | None:
-    """Connect to Jetstream and stream messages to Kafka until disconnected.
-
-    Args:
-        producer: The Kafka producer instance.
-        topic: Kafka topic name.
-        collections: Comma-separated AT Protocol collections to subscribe to.
-        cursor: Optional microsecond timestamp to resume from.
-
-    Returns:
-        The last processed time_us cursor value.
-    """
-    url = _build_jetstream_url(collections, cursor)
-    logger.info("Connecting to %s", url)
-
-    message_count = 0
-    last_cursor = cursor
-
-    with connect(url, additional_headers={"Accept-Encoding": "zstd"}) as websocket:
-        logger.info("Connected to Jetstream")
-
-        while not shutdown_event.is_set():
-            try:
-                raw_frame = websocket.recv(timeout=10)
-            except TimeoutError:
-                producer.poll(0)
-                continue
-
-            text = _decode_frame(raw_frame)
-            message = json.loads(text)
-
-            time_us = message.get("time_us")
-            if time_us is not None:
-                last_cursor = time_us
-
-            if not _is_post_create(message):
-                continue
-
-            did = message.get("did", "")
-
-            try:
-                producer.produce(
-                    topic,
-                    key=did.encode(),
-                    value=text.encode() if isinstance(text, str) else text,
-                    callback=_delivery_callback,
-                )
-            except BufferError:
-                logger.warning("Producer buffer full, flushing")
-                producer.flush(timeout=10)
-                producer.produce(
-                    topic,
-                    key=did.encode(),
-                    value=text.encode() if isinstance(text, str) else text,
-                    callback=_delivery_callback,
-                )
-
-            message_count += 1
-            if message_count % POLL_INTERVAL == 0:
-                producer.poll(0)
-
-            if message_count % 10000 == 0:
-                logger.info("Produced %d messages (cursor=%s)", message_count, last_cursor)
-
-    return last_cursor
-
-
-def _build_jetstream_url(collections: str, cursor: int | None) -> str:
-    """Build the Jetstream WebSocket URL with query parameters.
-
-    Args:
-        collections: Comma-separated collection names.
-        cursor: Optional microsecond timestamp for resumption.
-
-    Returns:
-        The full WebSocket URL.
-    """
-    params = ["compress=true"]
-    for raw_collection in collections.split(","):
-        cleaned = raw_collection.strip()
-        if cleaned:
-            params.append(f"wantedCollections={cleaned}")
-    if cursor is not None:
-        params.append(f"cursor={cursor}")
-    return f"{JETSTREAM_URL}?{'&'.join(params)}"
-
-
-def _decode_frame(frame: str | bytes) -> str:
-    """Decode a WebSocket frame, decompressing zstd if binary.
-
-    Jetstream with compress=true sends zstd-compressed binary frames.
-
-    Args:
-        frame: Raw WebSocket frame data.
-
-    Returns:
-        The decoded JSON string.
-    """
-    if isinstance(frame, bytes):
-        return zstd.decompress(frame).decode()
-    return frame
-
-
-def _is_post_create(message: dict) -> bool:
-    """Check if a Jetstream message is a post creation commit."""
-    if message.get("kind") != "commit":
-        return False
-    commit = message.get("commit", {})
-    return commit.get("operation") == "create" and commit.get("collection") == POST_COLLECTION
-
-
-def _create_producer(servers: str) -> Producer:
-    """Create a configured Kafka producer.
-
-    Args:
-        servers: Kafka bootstrap servers string.
-
-    Returns:
-        A configured confluent_kafka.Producer.
-    """
-    config = {
-        "bootstrap.servers": servers,
-        "linger.ms": 50,
-        "batch.size": 65536,
-        "compression.type": "zstd",
-        "acks": "all",
-        "retries": 5,
-        "retry.backoff.ms": 500,
-    }
-    return Producer(config)
-
-
-def _delivery_callback(error: KafkaError | None, _message: object) -> None:
-    """Log delivery failures from the Kafka producer."""
-    if error is not None:
-        logger.error("Kafka delivery failed: %s", error)
-
-
-def _handle_shutdown(_signum: int, _frame: object) -> None:
-    """Signal handler to trigger graceful shutdown."""
-    logger.info("Shutdown signal received")
-    shutdown_event.set()
-
-
-if __name__ == "__main__":
-    app()

python/data_science/ingest_congress.py

@@ -0,0 +1,613 @@
+"""Ingestion pipeline for loading congress data from unitedstates/congress JSON files.
+
+Loads legislators, bills, votes, vote records, and bill text into the data_science_dev database.
+Expects the parent directory to contain congress-tracker/ and congress-legislators/ as siblings.
+
+Usage:
+    ingest-congress /path/to/parent/
+    ingest-congress /path/to/parent/ --congress 118
+    ingest-congress /path/to/parent/ --congress 118 --only bills
+"""
+
+from __future__ import annotations
+
+import logging
+from pathlib import Path  # noqa: TC003 needed at runtime for typer CLI argument
+from typing import TYPE_CHECKING, Annotated
+
+import orjson
+import typer
+import yaml
+from sqlalchemy import select
+from sqlalchemy.orm import Session
+
+from python.common import configure_logger
+from python.orm.common import get_postgres_engine
+from python.orm.data_science_dev.congress import Bill, BillText, Legislator, LegislatorSocialMedia, Vote, VoteRecord
+
+if TYPE_CHECKING:
+    from collections.abc import Iterator
+
+    from sqlalchemy.engine import Engine
+
+logger = logging.getLogger(__name__)
+
+BATCH_SIZE = 10_000
+
+app = typer.Typer(help="Ingest unitedstates/congress data into data_science_dev.")
+
+
+@app.command()
+def main(
+    parent_dir: Annotated[
+        Path,
+        typer.Argument(help="Parent directory containing congress-tracker/ and congress-legislators/"),
+    ],
+    congress: Annotated[int | None, typer.Option(help="Only ingest a specific congress number")] = None,
+    only: Annotated[
+        str | None,
+        typer.Option(help="Only run a specific step: legislators, social-media, bills, votes, bill-text"),
+    ] = None,
+) -> None:
+    """Ingest congress data from unitedstates/congress JSON files."""
+    configure_logger(level="INFO")
+
+    data_dir = parent_dir / "congress-tracker/congress/data/"
+    legislators_dir = parent_dir / "congress-legislators"
+
+    if not data_dir.is_dir():
+        typer.echo(f"Expected congress-tracker/ directory: {data_dir}", err=True)
+        raise typer.Exit(code=1)
+
+    if not legislators_dir.is_dir():
+        typer.echo(f"Expected congress-legislators/ directory: {legislators_dir}", err=True)
+        raise typer.Exit(code=1)
+
+    engine = get_postgres_engine(name="DATA_SCIENCE_DEV")
+
+    congress_dirs = _resolve_congress_dirs(data_dir, congress)
+    if not congress_dirs:
+        typer.echo("No congress directories found.", err=True)
+        raise typer.Exit(code=1)
+
+    logger.info("Found %d congress directories to process", len(congress_dirs))
+
+    steps: dict[str, tuple] = {
+        "legislators": (ingest_legislators, (engine, legislators_dir)),
+        "legislators-social-media": (ingest_social_media, (engine, legislators_dir)),
+        "bills": (ingest_bills, (engine, congress_dirs)),
+        "votes": (ingest_votes, (engine, congress_dirs)),
+        "bill-text": (ingest_bill_text, (engine, congress_dirs)),
+    }
+
+    if only:
+        if only not in steps:
+            typer.echo(f"Unknown step: {only}. Choose from: {', '.join(steps)}", err=True)
+            raise typer.Exit(code=1)
+        steps = {only: steps[only]}
+
+    for step_name, (step_func, step_args) in steps.items():
+        logger.info("=== Starting step: %s ===", step_name)
+        step_func(*step_args)
+        logger.info("=== Finished step: %s ===", step_name)
+
+    logger.info("ingest-congress done")
+
+
+def _resolve_congress_dirs(data_dir: Path, congress: int | None) -> list[Path]:
+    """Find congress number directories under data_dir."""
+    if congress is not None:
+        target = data_dir / str(congress)
+        return [target] if target.is_dir() else []
+    return sorted(path for path in data_dir.iterdir() if path.is_dir() and path.name.isdigit())
+
+
+def _flush_batch(session: Session, batch: list[object], label: str) -> int:
+    """Add a batch of ORM objects to the session and commit. Returns count added."""
+    if not batch:
+        return 0
+    session.add_all(batch)
+    session.commit()
+    count = len(batch)
+    logger.info("Committed %d %s", count, label)
+    batch.clear()
+    return count
+
+
+# ---------------------------------------------------------------------------
+# Legislators — loaded from congress-legislators YAML files
+# ---------------------------------------------------------------------------
+
+
+def ingest_legislators(engine: Engine, legislators_dir: Path) -> None:
+    """Load legislators from congress-legislators YAML files."""
+    legislators_data = _load_legislators_yaml(legislators_dir)
+    logger.info("Loaded %d legislators from YAML files", len(legislators_data))
+
+    with Session(engine) as session:
+        existing_legislators = {
+            legislator.bioguide_id: legislator for legislator in session.scalars(select(Legislator)).all()
+        }
+        logger.info("Found %d existing legislators in DB", len(existing_legislators))
+
+        total_inserted = 0
+        total_updated = 0
+        for entry in legislators_data:
+            bioguide_id = entry.get("id", {}).get("bioguide")
+            if not bioguide_id:
+                continue
+
+            fields = _parse_legislator(entry)
+            if existing := existing_legislators.get(bioguide_id):
+                changed = False
+                for field, value in fields.items():
+                    if value is not None and getattr(existing, field) != value:
+                        setattr(existing, field, value)
+                        changed = True
+                if changed:
+                    total_updated += 1
+            else:
+                session.add(Legislator(bioguide_id=bioguide_id, **fields))
+                total_inserted += 1
+
+        session.commit()
+    logger.info("Inserted %d new legislators, updated %d existing", total_inserted, total_updated)
+
+
+def _load_legislators_yaml(legislators_dir: Path) -> list[dict]:
+    """Load and combine legislators-current.yaml and legislators-historical.yaml."""
+    legislators: list[dict] = []
+    for filename in ("legislators-current.yaml", "legislators-historical.yaml"):
+        path = legislators_dir / filename
+        if not path.exists():
+            logger.warning("Legislators file not found: %s", path)
+            continue
+        with path.open() as file:
+            data = yaml.safe_load(file)
+            if isinstance(data, list):
+                legislators.extend(data)
+    return legislators
+
+
+def _parse_legislator(entry: dict) -> dict:
+    """Extract Legislator fields from a congress-legislators YAML entry."""
+    ids = entry.get("id", {})
+    name = entry.get("name", {})
+    bio = entry.get("bio", {})
+    terms = entry.get("terms", [])
+    latest_term = terms[-1] if terms else {}
+
+    fec_ids = ids.get("fec")
+    fec_ids_joined = ",".join(fec_ids) if isinstance(fec_ids, list) else fec_ids
+
+    chamber = latest_term.get("type")
+    chamber_normalized = {"rep": "House", "sen": "Senate"}.get(chamber, chamber)
+
+    return {
+        "thomas_id": ids.get("thomas"),
+        "lis_id": ids.get("lis"),
+        "govtrack_id": ids.get("govtrack"),
+        "opensecrets_id": ids.get("opensecrets"),
+        "fec_ids": fec_ids_joined,
+        "first_name": name.get("first"),
+        "last_name": name.get("last"),
+        "official_full_name": name.get("official_full"),
+        "nickname": name.get("nickname"),
+        "birthday": bio.get("birthday"),
+        "gender": bio.get("gender"),
+        "current_party": latest_term.get("party"),
+        "current_state": latest_term.get("state"),
+        "current_district": latest_term.get("district"),
+        "current_chamber": chamber_normalized,
+    }
+
+
+# ---------------------------------------------------------------------------
+# Social Media — loaded from legislators-social-media.yaml
+# ---------------------------------------------------------------------------
+
+SOCIAL_MEDIA_PLATFORMS = {
+    "twitter": "https://twitter.com/{account}",
+    "facebook": "https://facebook.com/{account}",
+    "youtube": "https://youtube.com/{account}",
+    "instagram": "https://instagram.com/{account}",
+    "mastodon": None,
+}
+
+
+def ingest_social_media(engine: Engine, legislators_dir: Path) -> None:
+    """Load social media accounts from legislators-social-media.yaml."""
+    social_media_path = legislators_dir / "legislators-social-media.yaml"
+    if not social_media_path.exists():
+        logger.warning("Social media file not found: %s", social_media_path)
+        return
+
+    with social_media_path.open() as file:
+        social_media_data = yaml.safe_load(file)
+
+    if not isinstance(social_media_data, list):
+        logger.warning("Unexpected format in %s", social_media_path)
+        return
+
+    logger.info("Loaded %d entries from legislators-social-media.yaml", len(social_media_data))
+
+    with Session(engine) as session:
+        legislator_map = _build_legislator_map(session)
+        existing_accounts = {
+            (account.legislator_id, account.platform)
+            for account in session.scalars(select(LegislatorSocialMedia)).all()
+        }
+        logger.info("Found %d existing social media accounts in DB", len(existing_accounts))
+
+        total_inserted = 0
+        total_updated = 0
+        for entry in social_media_data:
+            bioguide_id = entry.get("id", {}).get("bioguide")
+            if not bioguide_id:
+                continue
+
+            legislator_id = legislator_map.get(bioguide_id)
+            if legislator_id is None:
+                continue
+
+            social = entry.get("social", {})
+            for platform, url_template in SOCIAL_MEDIA_PLATFORMS.items():
+                account_name = social.get(platform)
+                if not account_name:
+                    continue
+
+                url = url_template.format(account=account_name) if url_template else None
+
+                if (legislator_id, platform) in existing_accounts:
+                    total_updated += 1
+                else:
+                    session.add(
+                        LegislatorSocialMedia(
+                            legislator_id=legislator_id,
+                            platform=platform,
+                            account_name=str(account_name),
+                            url=url,
+                            source="https://github.com/unitedstates/congress-legislators",
+                        )
+                    )
+                    existing_accounts.add((legislator_id, platform))
+                    total_inserted += 1
+
+        session.commit()
+    logger.info("Inserted %d new social media accounts, updated %d existing", total_inserted, total_updated)
+
+
+def _iter_voters(position_group: object) -> Iterator[dict]:
+    """Yield voter dicts from a vote position group (handles list, single dict, or string)."""
+    if isinstance(position_group, dict):
+        yield position_group
+    elif isinstance(position_group, list):
+        for voter in position_group:
+            if isinstance(voter, dict):
+                yield voter
+
+
+# ---------------------------------------------------------------------------
+# Bills
+# ---------------------------------------------------------------------------
+
+
+def ingest_bills(engine: Engine, congress_dirs: list[Path]) -> None:
+    """Load bill data.json files."""
+    with Session(engine) as session:
+        existing_bills = {(bill.congress, bill.bill_type, bill.number) for bill in session.scalars(select(Bill)).all()}
+        logger.info("Found %d existing bills in DB", len(existing_bills))
+
+        total_inserted = 0
+        batch: list[Bill] = []
+        for congress_dir in congress_dirs:
+            bills_dir = congress_dir / "bills"
+            if not bills_dir.is_dir():
+                continue
+            logger.info("Scanning bills from %s", congress_dir.name)
+            for bill_file in bills_dir.rglob("data.json"):
+                data = _read_json(bill_file)
+                if data is None:
+                    continue
+                bill = _parse_bill(data, existing_bills)
+                if bill is not None:
+                    batch.append(bill)
+                    if len(batch) >= BATCH_SIZE:
+                        total_inserted += _flush_batch(session, batch, "bills")
+
+        total_inserted += _flush_batch(session, batch, "bills")
+    logger.info("Inserted %d new bills total", total_inserted)
+
+
+def _parse_bill(data: dict, existing_bills: set[tuple[int, str, int]]) -> Bill | None:
+    """Parse a bill data.json dict into a Bill ORM object, skipping existing."""
+    raw_congress = data.get("congress")
+    bill_type = data.get("bill_type")
+    raw_number = data.get("number")
+    if raw_congress is None or bill_type is None or raw_number is None:
+        return None
+    congress = int(raw_congress)
+    number = int(raw_number)
+    if (congress, bill_type, number) in existing_bills:
+        return None
+
+    sponsor_bioguide = None
+    sponsor = data.get("sponsor")
+    if sponsor:
+        sponsor_bioguide = sponsor.get("bioguide_id")
+
+    return Bill(
+        congress=congress,
+        bill_type=bill_type,
+        number=number,
+        title=data.get("short_title") or data.get("official_title"),
+        title_short=data.get("short_title"),
+        official_title=data.get("official_title"),
+        status=data.get("status"),
+        status_at=data.get("status_at"),
+        sponsor_bioguide_id=sponsor_bioguide,
+        subjects_top_term=data.get("subjects_top_term"),
+    )
+
+
+# ---------------------------------------------------------------------------
+# Votes (and vote records)
+# ---------------------------------------------------------------------------
+
+
+def ingest_votes(engine: Engine, congress_dirs: list[Path]) -> None:
+    """Load vote data.json files with their vote records."""
+    with Session(engine) as session:
+        legislator_map = _build_legislator_map(session)
+        logger.info("Loaded %d legislators into lookup map", len(legislator_map))
+        bill_map = _build_bill_map(session)
+        logger.info("Loaded %d bills into lookup map", len(bill_map))
+        existing_votes = {
+            (vote.congress, vote.chamber, vote.session, vote.number) for vote in session.scalars(select(Vote)).all()
+        }
+        logger.info("Found %d existing votes in DB", len(existing_votes))
+
+        total_inserted = 0
+        batch: list[Vote] = []
+        for congress_dir in congress_dirs:
+            votes_dir = congress_dir / "votes"
+            if not votes_dir.is_dir():
+                continue
+            logger.info("Scanning votes from %s", congress_dir.name)
+            for vote_file in votes_dir.rglob("data.json"):
+                data = _read_json(vote_file)
+                if data is None:
+                    continue
+                vote = _parse_vote(data, legislator_map, bill_map, existing_votes)
+                if vote is not None:
+                    batch.append(vote)
+                    if len(batch) >= BATCH_SIZE:
+                        total_inserted += _flush_batch(session, batch, "votes")
+
+        total_inserted += _flush_batch(session, batch, "votes")
+    logger.info("Inserted %d new votes total", total_inserted)
+
+
+def _build_legislator_map(session: Session) -> dict[str, int]:
+    """Build a mapping of bioguide_id -> legislator.id."""
+    return {legislator.bioguide_id: legislator.id for legislator in session.scalars(select(Legislator)).all()}
+
+
+def _build_bill_map(session: Session) -> dict[tuple[int, str, int], int]:
+    """Build a mapping of (congress, bill_type, number) -> bill.id."""
+    return {(bill.congress, bill.bill_type, bill.number): bill.id for bill in session.scalars(select(Bill)).all()}
+
+
+def _parse_vote(
+    data: dict,
+    legislator_map: dict[str, int],
+    bill_map: dict[tuple[int, str, int], int],
+    existing_votes: set[tuple[int, str, int, int]],
+) -> Vote | None:
+    """Parse a vote data.json dict into a Vote ORM object with records."""
+    raw_congress = data.get("congress")
+    chamber = data.get("chamber")
+    raw_number = data.get("number")
+    vote_date = data.get("date")
+    if raw_congress is None or chamber is None or raw_number is None or vote_date is None:
+        return None
+
+    raw_session = data.get("session")
+    if raw_session is None:
+        return None
+
+    congress = int(raw_congress)
+    number = int(raw_number)
+    session_number = int(raw_session)
+
+    # Normalize chamber from "h"/"s" to "House"/"Senate"
+    chamber_normalized = {"h": "House", "s": "Senate"}.get(chamber, chamber)
+
+    if (congress, chamber_normalized, session_number, number) in existing_votes:
+        return None
+
+    # Resolve linked bill
+    bill_id = None
+    bill_ref = data.get("bill")
+    if bill_ref:
+        bill_key = (
+            int(bill_ref.get("congress", congress)),
+            bill_ref.get("type"),
+            int(bill_ref.get("number", 0)),
+        )
+        bill_id = bill_map.get(bill_key)
+
+    raw_votes = data.get("votes", {})
+    vote_counts = _count_votes(raw_votes)
+    vote_records = _build_vote_records(raw_votes, legislator_map)
+
+    return Vote(
+        congress=congress,
+        chamber=chamber_normalized,
+        session=session_number,
+        number=number,
+        vote_type=data.get("type"),
+        question=data.get("question"),
+        result=data.get("result"),
+        result_text=data.get("result_text"),
+        vote_date=vote_date[:10] if isinstance(vote_date, str) else vote_date,
+        bill_id=bill_id,
+        vote_records=vote_records,
+        **vote_counts,
+    )
+
+
+def _count_votes(raw_votes: dict) -> dict[str, int]:
+    """Count voters per position category, correctly handling dict and list formats."""
+    yea_count = 0
+    nay_count = 0
+    not_voting_count = 0
+    present_count = 0
+
+    for position, position_group in raw_votes.items():
+        voter_count = sum(1 for _ in _iter_voters(position_group))
+        if position in ("Yea", "Aye"):
+            yea_count += voter_count
+        elif position in ("Nay", "No"):
+            nay_count += voter_count
+        elif position == "Not Voting":
+            not_voting_count += voter_count
+        elif position == "Present":
+            present_count += voter_count
+
+    return {
+        "yea_count": yea_count,
+        "nay_count": nay_count,
+        "not_voting_count": not_voting_count,
+        "present_count": present_count,
+    }
+
+
+def _build_vote_records(raw_votes: dict, legislator_map: dict[str, int]) -> list[VoteRecord]:
+    """Build VoteRecord objects from raw vote data."""
+    records: list[VoteRecord] = []
+    for position, position_group in raw_votes.items():
+        for voter in _iter_voters(position_group):
+            bioguide_id = voter.get("id")
+            if not bioguide_id:
+                continue
+            legislator_id = legislator_map.get(bioguide_id)
+            if legislator_id is None:
+                continue
+            records.append(
+                VoteRecord(
+                    legislator_id=legislator_id,
+                    position=position,
+                )
+            )
+    return records
+
+
+# ---------------------------------------------------------------------------
+# Bill Text
+# ---------------------------------------------------------------------------
+
+
+def ingest_bill_text(engine: Engine, congress_dirs: list[Path]) -> None:
+    """Load bill text from text-versions directories."""
+    with Session(engine) as session:
+        bill_map = _build_bill_map(session)
+        logger.info("Loaded %d bills into lookup map", len(bill_map))
+        existing_bill_texts = {
+            (bill_text.bill_id, bill_text.version_code) for bill_text in session.scalars(select(BillText)).all()
+        }
+        logger.info("Found %d existing bill text versions in DB", len(existing_bill_texts))
+
+        total_inserted = 0
+        batch: list[BillText] = []
+        for congress_dir in congress_dirs:
+            logger.info("Scanning bill texts from %s", congress_dir.name)
+            for bill_text in _iter_bill_texts(congress_dir, bill_map, existing_bill_texts):
+                batch.append(bill_text)
+                if len(batch) >= BATCH_SIZE:
+                    total_inserted += _flush_batch(session, batch, "bill texts")
+
+        total_inserted += _flush_batch(session, batch, "bill texts")
+    logger.info("Inserted %d new bill text versions total", total_inserted)
+
+
+def _iter_bill_texts(
+    congress_dir: Path,
+    bill_map: dict[tuple[int, str, int], int],
+    existing_bill_texts: set[tuple[int, str]],
+) -> Iterator[BillText]:
+    """Yield BillText objects for a single congress directory, skipping existing."""
+    bills_dir = congress_dir / "bills"
+    if not bills_dir.is_dir():
+        return
+
+    for bill_dir in bills_dir.rglob("text-versions"):
+        if not bill_dir.is_dir():
+            continue
+        bill_key = _bill_key_from_dir(bill_dir.parent, congress_dir)
+        if bill_key is None:
+            continue
+        bill_id = bill_map.get(bill_key)
+        if bill_id is None:
+            continue
+
+        for version_dir in sorted(bill_dir.iterdir()):
+            if not version_dir.is_dir():
+                continue
+            if (bill_id, version_dir.name) in existing_bill_texts:
+                continue
+            text_content = _read_bill_text(version_dir)
+            version_data = _read_json(version_dir / "data.json")
+            yield BillText(
+                bill_id=bill_id,
+                version_code=version_dir.name,
+                version_name=version_data.get("version_name") if version_data else None,
+                date=version_data.get("issued_on") if version_data else None,
+                text_content=text_content,
+            )
+
+
+def _bill_key_from_dir(bill_dir: Path, congress_dir: Path) -> tuple[int, str, int] | None:
+    """Extract (congress, bill_type, number) from directory structure."""
+    congress = int(congress_dir.name)
+    bill_type = bill_dir.parent.name
+    name = bill_dir.name
+    # Directory name is like "hr3590" — strip the type prefix to get the number
+    number_str = name[len(bill_type) :]
+    if not number_str.isdigit():
+        return None
+    return (congress, bill_type, int(number_str))
+
+
+def _read_bill_text(version_dir: Path) -> str | None:
+    """Read bill text from a version directory, preferring .txt over .xml."""
+    for extension in ("txt", "htm", "html", "xml"):
+        candidates = list(version_dir.glob(f"document.{extension}"))
+        if not candidates:
+            candidates = list(version_dir.glob(f"*.{extension}"))
+        if candidates:
+            try:
+                return candidates[0].read_text(encoding="utf-8")
+            except Exception:
+                logger.exception("Failed to read %s", candidates[0])
+    return None
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+
+def _read_json(path: Path) -> dict | None:
+    """Read and parse a JSON file, returning None on failure."""
+    try:
+        return orjson.loads(path.read_bytes())
+    except FileNotFoundError:
+        return None
+    except Exception:
+        logger.exception("Failed to parse %s", path)
+        return None
+
+
+if __name__ == "__main__":
+    app()

python/gitea.py

@@ -0,0 +1,335 @@
+"""Small Gitea API client for repository automation."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Self
+
+import httpx
+
+DEFAULT_PAGE_SIZE = 100
+EXPECTED_CREATED = 201
+EXPECTED_OK = 200
+
+
+@dataclass(frozen=True)
+class CreatedIssue:
+    """Issue data returned by Gitea."""
+
+    number: int | None
+    html_url: str | None
+    title: str
+
+
+@dataclass(frozen=True)
+class PullRequest:
+    """Pull request data returned by Gitea."""
+
+    number: int
+    title: str
+    html_url: str | None
+    labels: tuple[str, ...]
+    head_branch: str | None
+    base_branch: str | None
+
+
+@dataclass(frozen=True)
+class WorkflowJob:
+    """Workflow job data returned by Gitea Actions."""
+
+    id: int
+    name: str
+    run_id: int | None
+    status: str | None
+    conclusion: str | None
+
+
+class GiteaError(RuntimeError):
+    """Raised when Gitea rejects an API request."""
+
+
+def split_repo_name(repo: str) -> tuple[str, str]:
+    """Split an owner/repo string into its parts."""
+    owner, separator, repo_name = repo.partition("/")
+    if not separator or not owner or not repo_name:
+        msg = f"Invalid repository name: {repo}"
+        raise ValueError(msg)
+    return owner, repo_name
+
+
+class GiteaClient:
+    """HTTP client for the subset of Gitea APIs used in this repository."""
+
+    def __init__(
+        self,
+        *,
+        base_url: str,
+        token: str,
+        timeout: int = 30,
+        transport: httpx.BaseTransport | None = None,
+    ) -> None:
+        """Initialize the Gitea client."""
+        self._client = httpx.Client(
+            base_url=base_url.rstrip("/"),
+            timeout=timeout,
+            headers={"Authorization": f"token {token}"},
+            transport=transport,
+        )
+
+    def create_issue(
+        self,
+        *,
+        owner: str,
+        repo: str,
+        title: str,
+        body: str,
+        labels: list[int] | None = None,
+    ) -> CreatedIssue:
+        """Create a Gitea issue."""
+        payload: dict[str, object] = {"title": title, "body": body, "labels": labels or []}
+        response = self._request(
+            "POST",
+            f"/api/v1/repos/{owner}/{repo}/issues",
+            expected_statuses={EXPECTED_CREATED},
+            json=payload,
+        )
+        data = response.json()
+        return CreatedIssue(
+            number=_optional_int(data.get("number")),
+            html_url=_optional_str(data.get("html_url")),
+            title=str(data.get("title", title)),
+        )
+
+    def resolve_label_ids(self, *, owner: str, repo: str, labels: list[str]) -> list[int]:
+        """Resolve label names to Gitea label IDs."""
+        if not labels:
+            return []
+
+        available_labels: dict[str, int] = {}
+        page = 1
+        while True:
+            response = self._request(
+                "GET",
+                f"/api/v1/repos/{owner}/{repo}/labels",
+                params={"page": page, "limit": DEFAULT_PAGE_SIZE},
+            )
+            batch = response.json()
+            if not batch:
+                break
+            for label in batch:
+                label_name = str(label.get("name", ""))
+                label_id = _optional_int(label.get("id"))
+                if label_name and label_id is not None:
+                    available_labels[label_name] = label_id
+            if len(batch) < DEFAULT_PAGE_SIZE:
+                break
+            page += 1
+
+        missing = [label for label in labels if label not in available_labels]
+        if missing:
+            missing_names = ", ".join(sorted(missing))
+            msg = f"Missing Gitea labels: {missing_names}"
+            raise GiteaError(msg)
+
+        return [available_labels[label] for label in labels]
+
+    def list_open_pull_requests(
+        self,
+        *,
+        owner: str,
+        repo: str,
+        labels: list[str] | None = None,
+        head: str | None = None,
+    ) -> list[PullRequest]:
+        """List open pull requests for a repository."""
+        expected_labels = set(labels or [])
+        pull_requests: list[PullRequest] = []
+        page = 1
+        while True:
+            response = self._request(
+                "GET",
+                f"/api/v1/repos/{owner}/{repo}/pulls",
+                params={"state": "open", "page": page, "limit": DEFAULT_PAGE_SIZE},
+            )
+            batch = response.json()
+            if not batch:
+                break
+
+            for item in batch:
+                pull_request = _pull_request_from_api(item)
+                if head and pull_request.head_branch != head:
+                    continue
+                if expected_labels and not expected_labels.issubset(set(pull_request.labels)):
+                    continue
+                pull_requests.append(pull_request)
+
+            if len(batch) < DEFAULT_PAGE_SIZE:
+                break
+            page += 1
+
+        return pull_requests
+
+    def create_pull_request(
+        self,
+        *,
+        owner: str,
+        repo: str,
+        title: str,
+        body: str,
+        head: str,
+        base: str,
+        labels: list[str] | None = None,
+    ) -> PullRequest:
+        """Create a pull request."""
+        payload: dict[str, object] = {
+            "title": title,
+            "body": body,
+            "head": head,
+            "base": base,
+        }
+        if labels:
+            payload["labels"] = self.resolve_label_ids(owner=owner, repo=repo, labels=labels)
+
+        response = self._request(
+            "POST",
+            f"/api/v1/repos/{owner}/{repo}/pulls",
+            expected_statuses={EXPECTED_CREATED},
+            json=payload,
+        )
+        return _pull_request_from_api(response.json())
+
+    def merge_pull_request(
+        self,
+        *,
+        owner: str,
+        repo: str,
+        number: int,
+        merge_method: str = "rebase",
+        head_commit_id: str | None = None,
+        delete_branch_after_merge: bool = False,
+    ) -> None:
+        """Merge a pull request."""
+        payload: dict[str, object] = {
+            "Do": merge_method,
+            "delete_branch_after_merge": delete_branch_after_merge,
+        }
+        if head_commit_id:
+            payload["head_commit_id"] = head_commit_id
+
+        self._request(
+            "POST",
+            f"/api/v1/repos/{owner}/{repo}/pulls/{number}/merge",
+            json=payload,
+        )
+
+    def list_run_jobs(self, *, owner: str, repo: str, run_id: str | int) -> list[WorkflowJob]:
+        """List workflow jobs for a specific run."""
+        jobs: list[WorkflowJob] = []
+        page = 1
+        while True:
+            response = self._request(
+                "GET",
+                f"/api/v1/repos/{owner}/{repo}/actions/jobs",
+                params={"page": page, "limit": DEFAULT_PAGE_SIZE},
+            )
+            payload = response.json()
+            batch = payload.get("jobs", [])
+            if not batch:
+                break
+
+            for item in batch:
+                if str(item.get("run_id")) != str(run_id):
+                    continue
+                jobs.append(_workflow_job_from_api(item))
+
+            if len(batch) < DEFAULT_PAGE_SIZE:
+                break
+            page += 1
+
+        return jobs
+
+    def download_job_logs(self, *, owner: str, repo: str, job_id: int) -> str:
+        """Download logs for a workflow job."""
+        response = self._request(
+            "GET",
+            f"/api/v1/repos/{owner}/{repo}/actions/jobs/{job_id}/logs",
+        )
+        return response.text
+
+    def close(self) -> None:
+        """Close the underlying HTTP client."""
+        self._client.close()
+
+    def __enter__(self) -> Self:
+        """Enter the context manager."""
+        return self
+
+    def __exit__(self, *args: object) -> None:
+        """Close the HTTP client."""
+        self.close()
+
+    def _request(
+        self,
+        method: str,
+        path: str,
+        *,
+        expected_statuses: set[int] | None = None,
+        **kwargs: object,
+    ) -> httpx.Response:
+        """Send an HTTP request and validate the response status."""
+        response = self._client.request(method, path, **kwargs)
+        statuses = expected_statuses or {EXPECTED_OK}
+        if response.status_code not in statuses:
+            msg = f"Gitea request failed ({response.status_code}): {response.text}"
+            raise GiteaError(msg)
+        return response
+
+
+def _pull_request_from_api(data: dict[str, object]) -> PullRequest:
+    """Convert Gitea API pull-request data into a dataclass."""
+    number = _optional_int(data.get("number")) or _optional_int(data.get("index"))
+    if number is None:
+        msg = "Gitea pull request payload is missing a number"
+        raise GiteaError(msg)
+
+    labels = tuple(str(label.get("name", "")) for label in data.get("labels", []))
+    head = data.get("head", {})
+    base = data.get("base", {})
+    return PullRequest(
+        number=number,
+        title=str(data.get("title", "")),
+        html_url=_optional_str(data.get("html_url")),
+        labels=tuple(label for label in labels if label),
+        head_branch=_optional_str(head.get("ref")) or _optional_str(data.get("head_branch")),
+        base_branch=_optional_str(base.get("ref")) or _optional_str(data.get("base_branch")),
+    )
+
+
+def _workflow_job_from_api(data: dict[str, object]) -> WorkflowJob:
+    """Convert Gitea API workflow-job data into a dataclass."""
+    job_id = _optional_int(data.get("id"))
+    if job_id is None:
+        msg = "Gitea workflow job payload is missing an ID"
+        raise GiteaError(msg)
+
+    return WorkflowJob(
+        id=job_id,
+        name=str(data.get("name", "")),
+        run_id=_optional_int(data.get("run_id")),
+        status=_optional_str(data.get("status")),
+        conclusion=_optional_str(data.get("conclusion")),
+    )
+
+
+def _optional_int(value: object) -> int | None:
+    """Convert an API value to an integer when present."""
+    if value is None:
+        return None
+    return int(value)
+
+
+def _optional_str(value: object) -> str | None:
+    """Convert an API value to a string when present."""
+    if value is None:
+        return None
+    return str(value)

python/gitea_flake_lock.py

@@ -0,0 +1,138 @@
+"""Automation helpers for flake.lock pull requests on Gitea."""
+
+from __future__ import annotations
+
+import subprocess
+from os import getenv
+from typing import Annotated
+
+import typer
+
+from python.gitea import GiteaClient, PullRequest, split_repo_name
+
+DEFAULT_BASE_BRANCH = "main"
+DEFAULT_BRANCH = "automation/update-flake-lock"
+DEFAULT_GITEA_URL = "https://gitea.tmmworkshop.com"
+PR_LABELS = ["dependencies", "automated", "flake_lock_update"]
+PR_TITLE = "Update flake.lock"
+PR_BODY = "Automated flake.lock update."
+
+app = typer.Typer(add_completion=False)
+
+
+def run_cmd(cmd: list[str], *, check: bool = True) -> subprocess.CompletedProcess[str]:
+    """Run a subprocess command."""
+    return subprocess.run(cmd, capture_output=True, text=True, check=check)
+
+
+def ensure_flake_lock_pull_request(
+    client: GiteaClient,
+    *,
+    owner: str,
+    repo: str,
+    branch: str,
+    base: str,
+) -> PullRequest:
+    """Return an existing flake.lock PR for the branch or create one."""
+    pull_requests = client.list_open_pull_requests(owner=owner, repo=repo, head=branch)
+    if pull_requests:
+        return pull_requests[0]
+
+    return client.create_pull_request(
+        owner=owner,
+        repo=repo,
+        title=PR_TITLE,
+        body=PR_BODY,
+        head=branch,
+        base=base,
+        labels=PR_LABELS,
+    )
+
+
+def find_flake_lock_pull_request(client: GiteaClient, *, owner: str, repo: str) -> PullRequest | None:
+    """Find the first open flake.lock pull request."""
+    pull_requests = client.list_open_pull_requests(owner=owner, repo=repo, labels=["flake_lock_update"])
+    if not pull_requests:
+        return None
+    return pull_requests[0]
+
+
+def has_worktree_changes() -> bool:
+    """Return whether `flake.lock` has worktree changes."""
+    result = run_cmd(["git", "diff", "--quiet", "--", "flake.lock"], check=False)
+    return result.returncode != 0
+
+
+def commit_flake_lock_update(*, branch: str) -> None:
+    """Commit the updated lock file to the automation branch."""
+    run_cmd(["git", "config", "user.name", "gitea-actions[bot]"])
+    run_cmd(["git", "config", "user.email", "gitea-actions@tmmworkshop.com"])
+    run_cmd(["git", "checkout", "-B", branch])
+    run_cmd(["git", "add", "flake.lock"])
+    run_cmd(["git", "commit", "-m", "chore: update flake.lock"])
+
+
+def push_branch(*, branch: str) -> None:
+    """Push the automation branch to origin."""
+    run_cmd(["git", "push", "origin", f"HEAD:{branch}", "--force"])
+
+
+def _required_gitea_token() -> str:
+    """Read the required Gitea token from the environment."""
+    token = getenv("GITEA_TOKEN")
+    if token:
+        return token
+
+    msg = "GITEA_TOKEN environment variable is required"
+    raise RuntimeError(msg)
+
+
+@app.command()
+def update(
+    repo: Annotated[str, typer.Option("--repo", help="Gitea repository in owner/repo form")],
+    base: Annotated[str, typer.Option("--base", help="Base branch")] = DEFAULT_BASE_BRANCH,
+    branch: Annotated[str, typer.Option("--branch", help="Automation branch")] = DEFAULT_BRANCH,
+) -> None:
+    """Commit flake.lock changes and ensure a pull request exists."""
+    if not has_worktree_changes():
+        typer.echo("No flake.lock changes detected")
+        return
+
+    commit_flake_lock_update(branch=branch)
+    push_branch(branch=branch)
+
+    owner, repo_name = split_repo_name(repo)
+    with GiteaClient(
+        base_url=getenv("GITEA_URL", DEFAULT_GITEA_URL),
+        token=_required_gitea_token(),
+    ) as client:
+        pull_request = ensure_flake_lock_pull_request(
+            client,
+            owner=owner,
+            repo=repo_name,
+            branch=branch,
+            base=base,
+        )
+    typer.echo(pull_request.html_url or f"Pull request #{pull_request.number}")
+
+
+@app.command()
+def merge(
+    repo: Annotated[str, typer.Option("--repo", help="Gitea repository in owner/repo form")],
+) -> None:
+    """Merge the first open flake.lock pull request."""
+    owner, repo_name = split_repo_name(repo)
+    with GiteaClient(
+        base_url=getenv("GITEA_URL", DEFAULT_GITEA_URL),
+        token=_required_gitea_token(),
+    ) as client:
+        pull_request = find_flake_lock_pull_request(client, owner=owner, repo=repo_name)
+        if not pull_request:
+            typer.echo("No open PR found with label flake_lock_update")
+            return
+        client.merge_pull_request(owner=owner, repo=repo_name, number=pull_request.number, merge_method="rebase")
+    typer.echo(f"Merged PR #{pull_request.number}")
+
+
+if __name__ == "__main__":
+    app()

python/orm/data_science_dev/congress/__init__.py

@@ -0,0 +1,14 @@
+"""init."""
+
+from python.orm.data_science_dev.congress.bill import Bill, BillText
+from python.orm.data_science_dev.congress.legislator import Legislator, LegislatorSocialMedia
+from python.orm.data_science_dev.congress.vote import Vote, VoteRecord
+
+__all__ = [
+    "Bill",
+    "BillText",
+    "Legislator",
+    "LegislatorSocialMedia",
+    "Vote",
+    "VoteRecord",
+]

python/orm/data_science_dev/congress/bill.py

@@ -0,0 +1,66 @@
+"""Bill model - legislation introduced in Congress."""
+
+from __future__ import annotations
+
+from datetime import date
+from typing import TYPE_CHECKING
+
+from sqlalchemy import ForeignKey, Index, UniqueConstraint
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+
+from python.orm.data_science_dev.base import DataScienceDevTableBase
+
+if TYPE_CHECKING:
+    from python.orm.data_science_dev.congress.vote import Vote
+
+
+class Bill(DataScienceDevTableBase):
+    """Legislation with congress number, type, titles, status, and sponsor."""
+
+    __tablename__ = "bill"
+
+    congress: Mapped[int]
+    bill_type: Mapped[str]
+    number: Mapped[int]
+
+    title: Mapped[str | None]
+    title_short: Mapped[str | None]
+    official_title: Mapped[str | None]
+
+    status: Mapped[str | None]
+    status_at: Mapped[date | None]
+
+    sponsor_bioguide_id: Mapped[str | None]
+
+    subjects_top_term: Mapped[str | None]
+
+    votes: Mapped[list[Vote]] = relationship(
+        "Vote",
+        back_populates="bill",
+    )
+    bill_texts: Mapped[list[BillText]] = relationship(
+        "BillText",
+        back_populates="bill",
+        cascade="all, delete-orphan",
+    )
+
+    __table_args__ = (
+        UniqueConstraint("congress", "bill_type", "number", name="uq_bill_congress_type_number"),
+        Index("ix_bill_congress", "congress"),
+    )
+
+
+class BillText(DataScienceDevTableBase):
+    """Stores different text versions of a bill (introduced, enrolled, etc.)."""
+
+    __tablename__ = "bill_text"
+
+    bill_id: Mapped[int] = mapped_column(ForeignKey("main.bill.id", ondelete="CASCADE"))
+    version_code: Mapped[str]
+    version_name: Mapped[str | None]
+    text_content: Mapped[str | None]
+    date: Mapped[date | None]
+
+    bill: Mapped[Bill] = relationship("Bill", back_populates="bill_texts")
+
+    __table_args__ = (UniqueConstraint("bill_id", "version_code", name="uq_bill_text_bill_id_version_code"),)

python/orm/data_science_dev/congress/legislator.py

@@ -0,0 +1,66 @@
+"""Legislator model - members of Congress."""
+
+from __future__ import annotations
+
+from datetime import date
+from typing import TYPE_CHECKING
+
+from sqlalchemy import ForeignKey, Text
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+
+from python.orm.data_science_dev.base import DataScienceDevTableBase
+
+if TYPE_CHECKING:
+    from python.orm.data_science_dev.congress.vote import VoteRecord
+
+
+class Legislator(DataScienceDevTableBase):
+    """Members of Congress with identification and current term info."""
+
+    __tablename__ = "legislator"
+
+    bioguide_id: Mapped[str] = mapped_column(Text, unique=True, index=True)
+
+    thomas_id: Mapped[str | None]
+    lis_id: Mapped[str | None]
+    govtrack_id: Mapped[int | None]
+    opensecrets_id: Mapped[str | None]
+    fec_ids: Mapped[str | None]
+
+    first_name: Mapped[str]
+    last_name: Mapped[str]
+    official_full_name: Mapped[str | None]
+    nickname: Mapped[str | None]
+
+    birthday: Mapped[date | None]
+    gender: Mapped[str | None]
+
+    current_party: Mapped[str | None]
+    current_state: Mapped[str | None]
+    current_district: Mapped[int | None]
+    current_chamber: Mapped[str | None]
+
+    social_media_accounts: Mapped[list[LegislatorSocialMedia]] = relationship(
+        "LegislatorSocialMedia",
+        back_populates="legislator",
+        cascade="all, delete-orphan",
+    )
+    vote_records: Mapped[list[VoteRecord]] = relationship(
+        "VoteRecord",
+        back_populates="legislator",
+        cascade="all, delete-orphan",
+    )
+
+
+class LegislatorSocialMedia(DataScienceDevTableBase):
+    """Social media account linked to a legislator."""
+
+    __tablename__ = "legislator_social_media"
+
+    legislator_id: Mapped[int] = mapped_column(ForeignKey("main.legislator.id"))
+    platform: Mapped[str]
+    account_name: Mapped[str]
+    url: Mapped[str | None]
+    source: Mapped[str]
+
+    legislator: Mapped[Legislator] = relationship(back_populates="social_media_accounts")

python/orm/data_science_dev/congress/vote.py

@@ -0,0 +1,79 @@
+"""Vote model - roll call votes in Congress."""
+
+from __future__ import annotations
+
+from datetime import date
+from typing import TYPE_CHECKING
+
+from sqlalchemy import ForeignKey, Index, UniqueConstraint
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+
+from python.orm.data_science_dev.base import DataScienceDevBase, DataScienceDevTableBase
+
+if TYPE_CHECKING:
+    from python.orm.data_science_dev.congress.bill import Bill
+    from python.orm.data_science_dev.congress.legislator import Legislator
+    from python.orm.data_science_dev.congress.vote import Vote
+
+
+class VoteRecord(DataScienceDevBase):
+    """Links a vote to a legislator with their position (Yea, Nay, etc.)."""
+
+    __tablename__ = "vote_record"
+
+    vote_id: Mapped[int] = mapped_column(
+        ForeignKey("main.vote.id", ondelete="CASCADE"),
+        primary_key=True,
+    )
+    legislator_id: Mapped[int] = mapped_column(
+        ForeignKey("main.legislator.id", ondelete="CASCADE"),
+        primary_key=True,
+    )
+    position: Mapped[str]
+
+    vote: Mapped[Vote] = relationship("Vote", back_populates="vote_records")
+    legislator: Mapped[Legislator] = relationship("Legislator", back_populates="vote_records")
+
+
+class Vote(DataScienceDevTableBase):
+    """Roll call votes with counts and optional bill linkage."""
+
+    __tablename__ = "vote"
+
+    congress: Mapped[int]
+    chamber: Mapped[str]
+    session: Mapped[int]
+    number: Mapped[int]
+
+    vote_type: Mapped[str | None]
+    question: Mapped[str | None]
+    result: Mapped[str | None]
+    result_text: Mapped[str | None]
+
+    vote_date: Mapped[date]
+
+    yea_count: Mapped[int | None]
+    nay_count: Mapped[int | None]
+    not_voting_count: Mapped[int | None]
+    present_count: Mapped[int | None]
+
+    bill_id: Mapped[int | None] = mapped_column(ForeignKey("main.bill.id"))
+
+    bill: Mapped[Bill | None] = relationship("Bill", back_populates="votes")
+    vote_records: Mapped[list[VoteRecord]] = relationship(
+        "VoteRecord",
+        back_populates="vote",
+        cascade="all, delete-orphan",
+    )
+
+    __table_args__ = (
+        UniqueConstraint(
+            "congress",
+            "chamber",
+            "session",
+            "number",
+            name="uq_vote_congress_chamber_session_number",
+        ),
+        Index("ix_vote_date", "vote_date"),
+        Index("ix_vote_congress_chamber", "congress", "chamber"),
+    )

python/orm/data_science_dev/models.py

@@ -2,9 +2,15 @@
 
 from __future__ import annotations
 
+from python.orm.data_science_dev.congress import Bill, BillText, Legislator, Vote, VoteRecord
 from python.orm.data_science_dev.posts import partitions  # noqa: F401 — registers partition classes in metadata
 from python.orm.data_science_dev.posts.tables import Posts
 
 __all__ = [
+    "Bill",
+    "BillText",
+    "Legislator",
     "Posts",
+    "Vote",
+    "VoteRecord",
 ]

python/orm/richie/__init__.py

@@ -3,7 +3,6 @@
 from __future__ import annotations
 
 from python.orm.richie.base import RichieBase, TableBase, TableBaseBig, TableBaseSmall
-from python.orm.richie.congress import Bill, Legislator, Vote, VoteRecord
 from python.orm.richie.contact import (
     Contact,
     ContactNeed,
@@ -13,17 +12,13 @@ from python.orm.richie.contact import (
 )
 
 __all__ = [
-    "Bill",
     "Contact",
     "ContactNeed",
     "ContactRelationship",
-    "Legislator",
     "Need",
     "RelationshipType",
     "RichieBase",
     "TableBase",
     "TableBaseBig",
     "TableBaseSmall",
-    "Vote",
-    "VoteRecord",
 ]

python/orm/richie/congress.py

@@ -1,150 +0,0 @@
-"""Congress Tracker database models."""
-
-from __future__ import annotations
-
-from datetime import date
-
-from sqlalchemy import ForeignKey, Index, Text, UniqueConstraint
-from sqlalchemy.orm import Mapped, mapped_column, relationship
-
-from python.orm.richie.base import RichieBase, TableBase
-
-
-class Legislator(TableBase):
-    """Legislator model - members of Congress."""
-
-    __tablename__ = "legislator"
-
-    # Natural key - bioguide ID is the authoritative identifier
-    bioguide_id: Mapped[str] = mapped_column(Text, unique=True, index=True)
-
-    # Other IDs for cross-referencing
-    thomas_id: Mapped[str | None]
-    lis_id: Mapped[str | None]
-    govtrack_id: Mapped[int | None]
-    opensecrets_id: Mapped[str | None]
-    fec_ids: Mapped[str | None]  # JSON array stored as string
-
-    # Name info
-    first_name: Mapped[str]
-    last_name: Mapped[str]
-    official_full_name: Mapped[str | None]
-    nickname: Mapped[str | None]
-
-    # Bio
-    birthday: Mapped[date | None]
-    gender: Mapped[str | None]  # M/F
-
-    # Current term info (denormalized for query efficiency)
-    current_party: Mapped[str | None]
-    current_state: Mapped[str | None]
-    current_district: Mapped[int | None]  # House only
-    current_chamber: Mapped[str | None]  # rep/sen
-
-    # Relationships
-    vote_records: Mapped[list[VoteRecord]] = relationship(
-        "VoteRecord",
-        back_populates="legislator",
-        cascade="all, delete-orphan",
-    )
-
-
-class Bill(TableBase):
-    """Bill model - legislation introduced in Congress."""
-
-    __tablename__ = "bill"
-
-    # Composite natural key: congress + bill_type + number
-    congress: Mapped[int]
-    bill_type: Mapped[str]  # hr, s, hres, sres, hjres, sjres
-    number: Mapped[int]
-
-    # Bill info
-    title: Mapped[str | None]
-    title_short: Mapped[str | None]
-    official_title: Mapped[str | None]
-
-    # Status
-    status: Mapped[str | None]
-    status_at: Mapped[date | None]
-
-    # Sponsor
-    sponsor_bioguide_id: Mapped[str | None]
-
-    # Subjects
-    subjects_top_term: Mapped[str | None]
-
-    # Relationships
-    votes: Mapped[list[Vote]] = relationship(
-        "Vote",
-        back_populates="bill",
-    )
-
-    __table_args__ = (
-        UniqueConstraint("congress", "bill_type", "number", name="uq_bill_congress_type_number"),
-        Index("ix_bill_congress", "congress"),
-    )
-
-
-class Vote(TableBase):
-    """Vote model - roll call votes in Congress."""
-
-    __tablename__ = "vote"
-
-    # Composite natural key: congress + chamber + session + number
-    congress: Mapped[int]
-    chamber: Mapped[str]  # house/senate
-    session: Mapped[int]
-    number: Mapped[int]
-
-    # Vote details
-    vote_type: Mapped[str | None]
-    question: Mapped[str | None]
-    result: Mapped[str | None]
-    result_text: Mapped[str | None]
-
-    # Timing
-    vote_date: Mapped[date]
-
-    # Vote counts (denormalized for efficiency)
-    yea_count: Mapped[int | None]
-    nay_count: Mapped[int | None]
-    not_voting_count: Mapped[int | None]
-    present_count: Mapped[int | None]
-
-    # Related bill (optional - not all votes are on bills)
-    bill_id: Mapped[int | None] = mapped_column(ForeignKey("main.bill.id"))
-
-    # Relationships
-    bill: Mapped[Bill | None] = relationship("Bill", back_populates="votes")
-    vote_records: Mapped[list[VoteRecord]] = relationship(
-        "VoteRecord",
-        back_populates="vote",
-        cascade="all, delete-orphan",
-    )
-
-    __table_args__ = (
-        UniqueConstraint("congress", "chamber", "session", "number", name="uq_vote_congress_chamber_session_number"),
-        Index("ix_vote_date", "vote_date"),
-        Index("ix_vote_congress_chamber", "congress", "chamber"),
-    )
-
-
-class VoteRecord(RichieBase):
-    """Association table: Vote <-> Legislator with position."""
-
-    __tablename__ = "vote_record"
-
-    vote_id: Mapped[int] = mapped_column(
-        ForeignKey("main.vote.id", ondelete="CASCADE"),
-        primary_key=True,
-    )
-    legislator_id: Mapped[int] = mapped_column(
-        ForeignKey("main.legislator.id", ondelete="CASCADE"),
-        primary_key=True,
-    )
-    position: Mapped[str]  # Yea, Nay, Not Voting, Present
-
-    # Relationships
-    vote: Mapped[Vote] = relationship("Vote", back_populates="vote_records")
-    legislator: Mapped[Legislator] = relationship("Legislator", back_populates="vote_records")

python/signal_bot/device_registry.py

@@ -63,9 +63,9 @@ class DeviceRegistry:
             return
 
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if device:
                 if device.safety_number != safety_number and device.trust_level != TrustLevel.BLOCKED:
@@ -99,9 +99,9 @@ class DeviceRegistry:
         Returns True if the device was found and verified.
         """
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if not device:
                 logger.warning(f"Cannot verify unknown device: {phone_number}")
@@ -139,9 +139,9 @@ class DeviceRegistry:
     def grant_role(self, phone_number: str, role: Role) -> bool:
         """Add a role to a device. Called by admin over SSH."""
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if not device:
                 logger.warning(f"Cannot grant role for unknown device: {phone_number}")
@@ -150,7 +150,7 @@ class DeviceRegistry:
             if any(record.name == role for record in device.roles):
                 return True
 
-            role_record = session.execute(select(RoleRecord).where(RoleRecord.name == role)).scalar_one_or_none()
+            role_record = session.scalars(select(RoleRecord).where(RoleRecord.name == role)).one_or_none()
 
             if not role_record:
                 logger.warning(f"Unknown role: {role}")
@@ -165,9 +165,9 @@ class DeviceRegistry:
     def revoke_role(self, phone_number: str, role: Role) -> bool:
         """Remove a role from a device. Called by admin over SSH."""
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if not device:
                 logger.warning(f"Cannot revoke role for unknown device: {phone_number}")
@@ -182,16 +182,16 @@ class DeviceRegistry:
     def set_roles(self, phone_number: str, roles: list[Role]) -> bool:
         """Replace all roles for a device. Called by admin over SSH."""
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if not device:
                 logger.warning(f"Cannot set roles for unknown device: {phone_number}")
                 return False
 
             role_names = [str(role) for role in roles]
-            records = list(session.execute(select(RoleRecord).where(RoleRecord.name.in_(role_names))).scalars().all())
+            records = session.scalars(select(RoleRecord).where(RoleRecord.name.in_(role_names))).all()
             device.roles = records
             session.commit()
             self._update_cache(phone_number, device)
@@ -203,7 +203,7 @@ class DeviceRegistry:
     def list_devices(self) -> list[SignalDevice]:
         """Return all known devices."""
         with Session(self.engine) as session:
-            return list(session.execute(select(SignalDevice)).scalars().all())
+            return list(session.scalars(select(SignalDevice)).all())
 
     def sync_identities(self) -> None:
         """Pull identity list from signal-cli and record any new ones."""
@@ -226,9 +226,7 @@ class DeviceRegistry:
     def _load_device(self, phone_number: str) -> SignalDevice | None:
         """Fetch a device by phone number (with joined roles)."""
         with Session(self.engine) as session:
-            return session.execute(
-                select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            return session.scalars(select(SignalDevice).where(SignalDevice.phone_number == phone_number)).one_or_none()
 
     def _update_cache(self, phone_number: str, device: SignalDevice) -> None:
         """Refresh the cache entry for a device."""
@@ -244,9 +242,9 @@ class DeviceRegistry:
     def _set_trust(self, phone_number: str, level: str, log_msg: str | None = None) -> bool:
         """Update the trust level for a device."""
         with Session(self.engine) as session:
-            device = session.execute(
+            device = session.scalars(
                 select(SignalDevice).where(SignalDevice.phone_number == phone_number)
-            ).scalar_one_or_none()
+            ).one_or_none()
 
             if not device:
                 return False
@@ -269,7 +267,7 @@ def sync_roles(engine: Engine) -> None:
     expected = {role.value for role in Role}
 
     with Session(engine) as session:
-        existing = {record.name for record in session.execute(select(RoleRecord)).scalars().all()}
+        existing = set(session.scalars(select(RoleRecord.name)).all())
 
         to_add = expected - existing
         to_remove = existing - expected

python/tools/snapshot_manager.py

@@ -34,8 +34,9 @@ def main(config_file: Path) -> None:
                 logger.error(msg)
                 signal_alert(msg)
                 continue
-
-            get_snapshots_to_delete(dataset, get_count_lookup(config_file, dataset.name))
+            count_lookup = get_count_lookup(config_file, dataset.name)
+            logger.info(f"using {count_lookup} for {dataset.name}")
+            get_snapshots_to_delete(dataset, count_lookup)
     except Exception:
         logger.exception("snapshot_manager failed")
         signal_alert("snapshot_manager failed")
@@ -99,6 +100,7 @@ def get_snapshots_to_delete(
     """
     snapshots = dataset.get_snapshots()
 
+    logger.info(f"calculating snapshots for {dataset.name} to be deleted")
     if not snapshots:
         logger.info(f"{dataset.name} has no snapshots")
         return

python/tools/whisper/Dockerfile

@@ -0,0 +1,17 @@
+FROM nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04
+
+ENV DEBIAN_FRONTEND=noninteractive \
+    PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1
+
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends python3 python3-pip ffmpeg \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN pip3 install --no-cache-dir --upgrade pip \
+    && pip3 install --no-cache-dir faster-whisper requests
+
+WORKDIR /app
+COPY python/tools/whisper/inference.py /app/inference.py
+
+ENTRYPOINT ["python3", "/app/inference.py"]

python/tools/whisper/Dockerfile.dockerignore

@@ -0,0 +1,2 @@
+*
+!python/tools/whisper/inference.py

python/tools/whisper/__init__.py

@@ -0,0 +1 @@
+"""Whisper transcription tools (host orchestrator and container entrypoint)."""

python/tools/whisper/inference.py

@@ -0,0 +1,136 @@
+"""Container entrypoint that transcribes a directory of audio files with faster-whisper.
+
+Run inside the whisper-transcribe docker image; segment timestamps are grouped
+into one-minute buckets so the output reads as ``[HH:MM:00] text``.
+"""
+
+from __future__ import annotations
+
+import argparse
+import logging
+from pathlib import Path
+
+from faster_whisper import WhisperModel
+
+logger = logging.getLogger(__name__)
+
+AUDIO_EXTENSIONS = {".mp3", ".wav", ".m4a", ".flac", ".ogg", ".opus", ".mp4", ".mkv", ".webm", ".aac"}
+BUCKET_SECONDS = 60
+BEAM_SIZE = 5
+SECONDS_PER_HOUR = 3600
+SECONDS_PER_MINUTE = 60
+
+
+def format_timestamp(total_seconds: float) -> str:
+    """Render a whole-minute timestamp as ``HH:MM:00``.
+
+    Args:
+        total_seconds: Offset in seconds from the start of the audio.
+
+    Returns:
+        A zero-padded ``HH:MM:00`` string.
+    """
+    hours = int(total_seconds // SECONDS_PER_HOUR)
+    minutes = int((total_seconds % SECONDS_PER_HOUR) // SECONDS_PER_MINUTE)
+    return f"{hours:02d}:{minutes:02d}:00"
+
+
+def transcribe_file(model: WhisperModel, audio_path: Path, output_path: Path) -> None:
+    """Transcribe one audio file and write the bucketed transcript to disk.
+
+    Args:
+        model: Loaded faster-whisper model.
+        audio_path: Source audio file.
+        output_path: Destination ``.txt`` path.
+    """
+    logger.info("Transcribing %s", audio_path)
+    segments, info = model.transcribe(
+        str(audio_path),
+        language="en",
+        beam_size=BEAM_SIZE,
+        vad_filter=True,
+    )
+    logger.info("Duration %.1fs", info.duration)
+
+    buckets: dict[int, list[str]] = {}
+    for segment in segments:
+        bucket = int(segment.start // BUCKET_SECONDS)
+        buckets.setdefault(bucket, []).append(segment.text.strip())
+
+    lines = [f"[{format_timestamp(bucket * BUCKET_SECONDS)}] {' '.join(buckets[bucket])}" for bucket in sorted(buckets)]
+    output_path.write_text("\n\n".join(lines) + "\n", encoding="utf-8")
+    logger.info("Wrote %s", output_path)
+
+
+def find_audio_files(input_directory: Path) -> list[Path]:
+    """Collect every audio file under ``input_directory``.
+
+    Args:
+        input_directory: Directory to walk recursively.
+
+    Returns:
+        Sorted list of audio file paths.
+    """
+    return sorted(
+        path for path in input_directory.rglob("*") if path.is_file() and path.suffix.lower() in AUDIO_EXTENSIONS
+    )
+
+
+def configure_container_logger() -> None:
+    """Configure logging for the container (stdout, INFO)."""
+    logging.basicConfig(
+        level=logging.INFO,
+        format="%(asctime)s %(levelname)s %(message)s",
+    )
+
+
+def parse_arguments() -> argparse.Namespace:
+    """Parse CLI arguments for the container entrypoint.
+
+    Returns:
+        Parsed argparse namespace.
+    """
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--input", type=Path, default=Path("/audio"))
+    parser.add_argument("--output", type=Path, default=Path("/output"))
+    parser.add_argument("--model", default="large-v3")
+    parser.add_argument(
+        "--download-only",
+        action="store_true",
+        help="Download the model into the cache volume and exit without transcribing.",
+    )
+    return parser.parse_args()
+
+
+def main() -> None:
+    """Load the model, then either exit (download-only) or transcribe the directory."""
+    configure_container_logger()
+    arguments = parse_arguments()
+
+    logger.info("Loading model %s on CUDA", arguments.model)
+    model = WhisperModel(arguments.model, device="cuda", compute_type="float16")
+
+    if arguments.download_only:
+        logger.info("Model ready; exiting (download-only mode)")
+        return
+
+    arguments.output.mkdir(parents=True, exist_ok=True)
+
+    audio_files = find_audio_files(arguments.input)
+    if not audio_files:
+        logger.warning("No audio files found in %s", arguments.input)
+        return
+
+    logger.info("Found %d audio file(s)", len(audio_files))
+    for audio_path in audio_files:
+        relative = audio_path.relative_to(arguments.input)
+        output_path = arguments.output / relative.with_suffix(".txt")
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+        if output_path.exists():
+            logger.info("Skip %s (already transcribed)", relative)
+            continue
+        transcribe_file(model, audio_path, output_path)
+
+
+if __name__ == "__main__":
+    main()

python/tools/whisper/transcribe.py

@@ -0,0 +1,167 @@
+"""Build and run the whisper transcription docker container on demand.
+
+The container is started fresh for each invocation and removed on exit
+(``docker run --rm``). The model is cached in a named docker volume so
+only the first run pays the download cost.
+"""
+
+from __future__ import annotations
+
+import logging
+import subprocess
+from pathlib import Path
+from typing import Annotated
+
+import typer
+
+from python.common import configure_logger
+
+logger = logging.getLogger(__name__)
+
+
+class Config:
+    """Paths and names for the whisper-transcribe Docker workflow."""
+
+    image_tag = "whisper-transcribe:latest"
+    model_volume = "whisper-models"
+    repo_root = Path(__file__).resolve().parents[3]
+    dockerfile = Path(__file__).resolve().parent / "Dockerfile"
+    huggingface_cache = "/root/.cache/huggingface"
+
+
+def run_docker(arguments: list[str]) -> None:
+    """Run a docker subcommand, streaming output and raising on failure.
+
+    Args:
+        arguments: Arguments to pass to the ``docker`` binary.
+
+    Raises:
+        subprocess.CalledProcessError: If docker exits non-zero.
+    """
+    logger.info("docker %s", " ".join(arguments))
+    subprocess.run(["docker", *arguments], check=True)
+
+
+def build_image() -> None:
+    """Build the whisper-transcribe image using the repo root as build context."""
+    logger.info("Building image %s", Config.image_tag)
+    run_docker(
+        [
+            "build",
+            "--tag",
+            Config.image_tag,
+            "--file",
+            str(Config.dockerfile),
+            str(Config.repo_root),
+        ],
+    )
+
+
+def model_cache_present(model: str) -> bool:
+    """Check whether the given model is already downloaded in the cache volume.
+
+    Args:
+        model: faster-whisper model name (e.g. ``large-v3``).
+
+    Returns:
+        True if the HuggingFace cache directory for the model exists in the volume.
+    """
+    cache_directory = f"hub/models--Systran--faster-whisper-{model}"
+    completed = subprocess.run(
+        [
+            "docker",
+            "run",
+            "--rm",
+            "--volume",
+            f"{Config.model_volume}:/cache",
+            "alpine",
+            "test",
+            "-d",
+            f"/cache/{cache_directory}",
+        ],
+        check=False,
+    )
+    return completed.returncode == 0
+
+
+def download_model(model: str) -> None:
+    """Download the model into the cache volume and exit.
+
+    Args:
+        model: faster-whisper model name.
+    """
+    logger.info("Downloading model %s into volume %s", model, Config.model_volume)
+    run_docker(
+        [
+            "run",
+            "--rm",
+            "--device=nvidia.com/gpu=all",
+            "--ipc=host",
+            "--volume",
+            f"{Config.model_volume}:{Config.huggingface_cache}",
+            Config.image_tag,
+            "--model",
+            model,
+            "--download-only",
+        ],
+    )
+
+
+def transcribe(input_directory: Path, output_directory: Path, model: str) -> None:
+    """Run transcription on every audio file under ``input_directory``.
+
+    Args:
+        input_directory: Host path containing audio files (mounted read-only).
+        output_directory: Host path for ``.txt`` transcripts.
+        model: faster-whisper model name.
+    """
+    logger.info("Transcribing %s -> %s (model=%s)", input_directory, output_directory, model)
+    run_docker(
+        [
+            "run",
+            "--rm",
+            "--device=nvidia.com/gpu=all",
+            "--ipc=host",
+            "--volume",
+            f"{input_directory}:/audio:ro",
+            "--volume",
+            f"{output_directory}:/output",
+            "--volume",
+            f"{Config.model_volume}:{Config.huggingface_cache}",
+            Config.image_tag,
+            "--model",
+            model,
+        ],
+    )
+
+
+def main(
+    input_directory: Annotated[Path, typer.Argument(help="Directory of audio files to transcribe.")],
+    output_directory: Annotated[Path, typer.Argument(help="Directory to write .txt transcripts to.")],
+    model: Annotated[str, typer.Option(help="faster-whisper model name.")] = "large-v3",
+    *,
+    force_download: Annotated[
+        bool,
+        typer.Option("--force-download", help="Re-download the model even if already cached."),
+    ] = False,
+) -> None:
+    """Build the image, ensure the model is cached, then transcribe and stop."""
+    configure_logger()
+
+    resolved_input = input_directory.resolve(strict=True)
+    output_directory.mkdir(parents=True, exist_ok=True)
+    resolved_output = output_directory.resolve()
+
+    build_image()
+
+    if force_download or not model_cache_present(model):
+        download_model(model)
+    else:
+        logger.info("Model %s already cached in volume %s", model, Config.model_volume)
+
+    transcribe(resolved_input, resolved_output, model)
+    logger.info("Done. Container stopped.")
+
+
+if __name__ == "__main__":
+    typer.run(main)

systems/bob/default.nix

@@ -1,24 +1,30 @@
-{ inputs, ... }:
+{ inputs, pkgs, ... }:
 {
   imports = [
+    "${inputs.self}/users/math"
     "${inputs.self}/users/richie"
+    "${inputs.self}/users/steve"
     "${inputs.self}/common/global"
-    "${inputs.self}/common/optional/desktop.nix"
     "${inputs.self}/common/optional/docker.nix"
     "${inputs.self}/common/optional/scanner.nix"
+    "${inputs.self}/common/optional/monitoring-agent.nix"
     "${inputs.self}/common/optional/steam.nix"
     "${inputs.self}/common/optional/syncthing_base.nix"
     "${inputs.self}/common/optional/systemd-boot.nix"
     "${inputs.self}/common/optional/update.nix"
     "${inputs.self}/common/optional/yubikey.nix"
     "${inputs.self}/common/optional/zerotier.nix"
-    "${inputs.self}/common/optional/brain_substituter.nix"
     "${inputs.self}/common/optional/nvidia.nix"
     ./hardware.nix
     ./syncthing.nix
     ./llms.nix
   ];
 
+  boot = {
+    kernelPackages = pkgs.linuxPackages_6_18;
+    zfs.package = pkgs.zfs_2_4;
+  };
+
   networking = {
     hostName = "bob";
     hostId = "7c678a41";

systems/bob/hardware.nix

@@ -28,7 +28,6 @@
         allowDiscards = true;
         keyFileSize = 4096;
         keyFile = "/dev/disk/by-id/usb-Samsung_Flash_Drive_FIT_0374620080067131-0:0";
-        fallbackToPassword = true;
       };
     };
     kernelModules = [ "kvm-amd" ];

systems/bob/llms.nix

@@ -23,6 +23,7 @@
       "magistral:24b"
       "ministral-3:14b"
       "nemotron-3-nano:30b"
+      "nemotron-3-nano:4b"
       "nemotron-cascade-2:30b"
       "qwen3-coder:30b"
       "qwen3-embedding:0.6b"
@@ -41,11 +42,14 @@
       "qwen3:8b"
       "qwen3.5:27b"
       "qwen3.5:35b"
+      "qwen3.6:27b"
+      "qwen3.6:35b"
+      "rinex20/translategemma3:12b"
       "translategemma:12b"
       "translategemma:27b"
       "translategemma:4b"
     ];
-    models = "/zfs/models";
+    models = "/zfs/storage/models";
     openFirewall = true;
   };
 }

systems/bob/syncthing.nix

@@ -31,5 +31,15 @@
       ];
       fsWatcherEnabled = true;
     };
+    "recordings" = {
+      path = "/home/richie/recordings";
+      devices = [
+        "jeeves"
+        "phone"
+        "rhapsody-in-green"
+      ];
+      fsWatcherEnabled = true;
+    };
+
   };
 }

systems/brain/hardware.nix

@@ -26,7 +26,6 @@
         allowDiscards = true;
         keyFileSize = 4096;
         keyFile = "/dev/disk/by-id/usb-USB_SanDisk_3.2Gen1_03021630090925173333-0:0";
-        fallbackToPassword = true;
       };
     };
     kernelModules = [ "kvm-intel" ];

systems/jeeves/default.nix

@@ -4,17 +4,21 @@ let
 in
 {
   imports = [
-    "${inputs.self}/users/richie"
-    "${inputs.self}/users/math"
     "${inputs.self}/users/dov"
+    "${inputs.self}/users/math"
+    "${inputs.self}/users/richie"
+    "${inputs.self}/users/steve"
     "${inputs.self}/common/global"
     "${inputs.self}/common/optional/docker.nix"
+    "${inputs.self}/common/optional/monitoring-agent.nix"
     "${inputs.self}/common/optional/ssh_decrypt.nix"
     "${inputs.self}/common/optional/syncthing_base.nix"
     "${inputs.self}/common/optional/update.nix"
     "${inputs.self}/common/optional/zerotier.nix"
+    ./monitoring
     ./docker
     ./services
+    ./web_services
     ./hardware.nix
     ./networking.nix
     ./programs.nix
@@ -35,5 +39,10 @@ in
     zerotierone.joinNetworks = [ "a09acf02330d37b9" ];
   };
 
+  users.groups = {
+    nornsight = { };
+    nornsight-admin = { };
+  };
+
   system.stateVersion = "24.05";
 }

systems/jeeves/hardware.nix

@@ -9,7 +9,6 @@ let
     inherit device;
     keyFileSize = 4096;
     keyFile = "/dev/disk/by-id/usb-XIAO_USB_Drive_24587CE29074-0:0";
-    fallbackToPassword = true;
   };
   makeLuksSSD =
     device:

systems/jeeves/monitoring/dashboards/overview.json

@@ -0,0 +1,426 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 6,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "100 * (1 - avg by (instance) (rate(node_cpu_seconds_total{mode=\"idle\"}[5m])))",
+          "legendFormat": "{{instance}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "CPU Used",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 6,
+        "x": 6,
+        "y": 0
+      },
+      "id": 2,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "100 * (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes))",
+          "legendFormat": "{{instance}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "RAM Used",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 6,
+        "x": 12,
+        "y": 0
+      },
+      "id": 3,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "100 * (1 - (node_memory_SwapFree_bytes / node_memory_SwapTotal_bytes))",
+          "legendFormat": "{{instance}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Swap Used",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "short"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 6,
+        "x": 18,
+        "y": 0
+      },
+      "id": 4,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "node_load1",
+          "legendFormat": "{{instance}} load1",
+          "range": true,
+          "refId": "A"
+        },
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "node_load5",
+          "legendFormat": "{{instance}} load5",
+          "range": true,
+          "refId": "B"
+        },
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "node_load15",
+          "legendFormat": "{{instance}} load15",
+          "range": true,
+          "refId": "C"
+        }
+      ],
+      "title": "Load",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "Bps"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 0,
+        "y": 8
+      },
+      "id": 5,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "sum by (instance) (rate(node_disk_read_bytes_total[5m]))",
+          "legendFormat": "{{instance}} read",
+          "range": true,
+          "refId": "A"
+        },
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "sum by (instance) (rate(node_disk_written_bytes_total[5m]))",
+          "legendFormat": "{{instance}} write",
+          "range": true,
+          "refId": "B"
+        }
+      ],
+      "title": "Disk Throughput",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 12,
+        "y": 8
+      },
+      "id": 6,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "100 * (1 - (node_filesystem_avail_bytes{mountpoint=~\"(/|/home|/var|/zfs.*)\",fstype!=\"\"} / node_filesystem_size_bytes{mountpoint=~\"(/|/home|/var|/zfs.*)\",fstype!=\"\"}))",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{mountpoint}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Filesystem Usage",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percentunit"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 17
+      },
+      "id": 7,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, rate(namedprocess_namegroup_cpu_seconds_total[5m]))",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top Grouped CPU",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "bytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 17
+      },
+      "id": 8,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, namedprocess_namegroup_memory_bytes{memtype=\"resident\"})",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top Grouped Memory",
+      "type": "table"
+    }
+  ],
+  "refresh": "30s",
+  "schemaVersion": 39,
+  "style": "dark",
+  "tags": [
+    "monitoring"
+  ],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-24h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "Overview",
+  "uid": "monitor-overview",
+  "version": 1,
+  "weekStart": ""
+}

systems/jeeves/monitoring/dashboards/process-history-grouped.json

@@ -0,0 +1,216 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percentunit"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, rate(namedprocess_namegroup_cpu_seconds_total[5m]))",
+          "legendFormat": "{{instance}} {{groupname}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Grouped CPU",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "bytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 0
+      },
+      "id": 2,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, namedprocess_namegroup_memory_bytes{memtype=\"resident\"})",
+          "legendFormat": "{{instance}} {{groupname}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Grouped Resident Memory",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "Bps"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 10
+      },
+      "id": 3,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, rate(namedprocess_namegroup_read_bytes_total[5m]))",
+          "legendFormat": "{{instance}} {{groupname}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Grouped Read I/O",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "Bps"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 10
+      },
+      "id": 4,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(10, rate(namedprocess_namegroup_write_bytes_total[5m]))",
+          "legendFormat": "{{instance}} {{groupname}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Grouped Write I/O",
+      "type": "timeseries"
+    }
+  ],
+  "refresh": "30s",
+  "schemaVersion": 39,
+  "style": "dark",
+  "tags": [
+    "monitoring",
+    "process"
+  ],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-7d",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "Process History Grouped",
+  "uid": "monitor-process-history",
+  "version": 1,
+  "weekStart": ""
+}

systems/jeeves/monitoring/dashboards/process-live-pid.json

@@ -0,0 +1,224 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-pid-short"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percentunit"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-pid-short"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, rate(namedprocess_namegroup_cpu_seconds_total[2m]))",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top PID CPU",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-pid-short"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "bytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 0
+      },
+      "id": 2,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-pid-short"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, namedprocess_namegroup_memory_bytes{memtype=\"resident\"})",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top PID RSS",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-pid-short"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "Bps"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 0,
+        "y": 10
+      },
+      "id": 3,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-pid-short"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, rate(namedprocess_namegroup_read_bytes_total[2m]))",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top PID Read I/O",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-pid-short"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "Bps"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 10,
+        "w": 12,
+        "x": 12,
+        "y": 10
+      },
+      "id": 4,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-pid-short"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, rate(namedprocess_namegroup_write_bytes_total[2m]))",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{groupname}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top PID Write I/O",
+      "type": "table"
+    }
+  ],
+  "refresh": "15s",
+  "schemaVersion": 39,
+  "style": "dark",
+  "tags": [
+    "monitoring",
+    "process"
+  ],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-10m",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "Process Live PID",
+  "uid": "monitor-process-pid",
+  "version": 1,
+  "weekStart": ""
+}

systems/jeeves/monitoring/dashboards/storage-zfs.json

@@ -0,0 +1,351 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 8,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "100 * (zfs_pool_allocated_bytes / zfs_pool_size_bytes)",
+          "legendFormat": "{{instance}} {{pool}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Pool Usage",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "bytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 8,
+        "x": 8,
+        "y": 0
+      },
+      "id": 2,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "zfs_pool_free_bytes",
+          "legendFormat": "{{instance}} {{pool}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "Pool Free Bytes",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "bytes"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 8,
+        "x": 16,
+        "y": 0
+      },
+      "id": 3,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, zfs_dataset_used_bytes{type=\"filesystem\"})",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{name}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Top Filesystems by Used Bytes",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "ns"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 0,
+        "y": 8
+      },
+      "id": 4,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, zpool_iostat_total_wait_read_ns{vdev!=\"_pool\"})",
+          "legendFormat": "{{host}} {{pool}} {{vdev}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "ZFS Read Wait",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "ns"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 12,
+        "y": 8
+      },
+      "id": 5,
+      "options": {
+        "legend": {
+          "displayMode": "list",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "topk(20, zpool_iostat_total_wait_write_ns{vdev!=\"_pool\"})",
+          "legendFormat": "{{host}} {{pool}} {{vdev}}",
+          "range": true,
+          "refId": "A"
+        }
+      ],
+      "title": "ZFS Write Wait",
+      "type": "timeseries"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "celsius"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 0,
+        "y": 17
+      },
+      "id": 6,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": true,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "smartctl_device_temperature{temperature_type=\"current\"}",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{device}}",
+          "refId": "A"
+        }
+      ],
+      "title": "Disk Temperature",
+      "type": "table"
+    },
+    {
+      "datasource": {
+        "type": "prometheus",
+        "uid": "prom-main"
+      },
+      "fieldConfig": {
+        "defaults": {
+          "unit": "short"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 9,
+        "w": 12,
+        "x": 12,
+        "y": 17
+      },
+      "id": 7,
+      "options": {
+        "cellHeight": "sm",
+        "showHeader": true,
+        "sortBy": [
+          {
+            "desc": false,
+            "displayName": "Value"
+          }
+        ]
+      },
+      "targets": [
+        {
+          "datasource": {
+            "type": "prometheus",
+            "uid": "prom-main"
+          },
+          "editorMode": "code",
+          "expr": "smartctl_device_smart_status",
+          "format": "table",
+          "instant": true,
+          "legendFormat": "{{instance}} {{device}}",
+          "refId": "A"
+        }
+      ],
+      "title": "SMART Health",
+      "type": "table"
+    }
+  ],
+  "refresh": "30s",
+  "schemaVersion": 39,
+  "style": "dark",
+  "tags": [
+    "monitoring",
+    "zfs"
+  ],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-24h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "Storage and ZFS",
+  "uid": "monitor-storage",
+  "version": 1,
+  "weekStart": ""
+}

systems/jeeves/monitoring/default.nix

@@ -0,0 +1,186 @@
+{
+  lib,
+  pkgs,
+  ...
+}:
+let
+  vars = import ../vars.nix;
+
+  prometheusDataRoot = "${vars.database}/prometheus";
+  mainPrometheusDataDir = "${prometheusDataRoot}/main";
+  pidPrometheusDataDir = "${prometheusDataRoot}/pid-short";
+
+  prometheusYaml = pkgs.formats.yaml { };
+
+  mkPrometheusConfig =
+    name: cfg:
+    let
+      configFile = prometheusYaml.generate "${name}.yaml" cfg;
+    in
+    pkgs.runCommand "${name}-checked.yaml"
+      {
+        nativeBuildInputs = [ pkgs.prometheus.cli ];
+      }
+      ''
+        promtool check config ${configFile}
+        cp ${configFile} $out
+      '';
+
+  mkTarget = host: address: {
+    targets = [ address ];
+    labels.instance = host;
+  };
+
+  mainPrometheusConfig = mkPrometheusConfig "prometheus-main" {
+    global = {
+      scrape_interval = "30s";
+      scrape_timeout = "10s";
+      evaluation_interval = "30s";
+    };
+    scrape_configs = [
+      {
+        job_name = "node";
+        static_configs = [
+          (mkTarget "jeeves" "192.168.90.40:9100")
+          (mkTarget "bob" "192.168.90.25:9100")
+        ];
+      }
+      {
+        job_name = "process_grouped";
+        static_configs = [
+          (mkTarget "jeeves" "192.168.90.40:9256")
+          (mkTarget "bob" "192.168.90.25:9256")
+        ];
+      }
+      {
+        job_name = "smartctl";
+        static_configs = [
+          (mkTarget "jeeves" "192.168.90.40:9633")
+          (mkTarget "bob" "192.168.90.25:9633")
+        ];
+      }
+      {
+        job_name = "zfs";
+        static_configs = [
+          (mkTarget "jeeves" "192.168.90.40:9134")
+          (mkTarget "bob" "192.168.90.25:9134")
+        ];
+      }
+    ];
+  };
+
+  pidPrometheusConfig = mkPrometheusConfig "prometheus-pid-short" {
+    global = {
+      scrape_interval = "15s";
+      scrape_timeout = "10s";
+      evaluation_interval = "15s";
+    };
+    scrape_configs = [
+      {
+        job_name = "process_pid";
+        static_configs = [
+          (mkTarget "jeeves" "192.168.90.40:9257")
+          (mkTarget "bob" "192.168.90.25:9257")
+        ];
+      }
+    ];
+  };
+
+  mkPrometheusService =
+    {
+      dataDir,
+      configFile,
+      port,
+      retention,
+    }:
+    {
+      after = [
+        "zfs-media-database-prometheus.mount"
+        "network.target"
+      ];
+      requires = [ "zfs-media-database-prometheus.mount" ];
+      wantedBy = [ "multi-user.target" ];
+      unitConfig.RequiresMountsFor = [ dataDir ];
+      serviceConfig = {
+        ExecStart = "${lib.getExe pkgs.prometheus} ${
+          lib.escapeShellArgs [
+            "--config.file=${configFile}"
+            "--storage.tsdb.path=${dataDir}"
+            "--storage.tsdb.retention.time=${retention}"
+            "--web.listen-address=127.0.0.1:${toString port}"
+          ]
+        }";
+        User = "prometheus";
+        Group = "prometheus";
+        Restart = "always";
+        RestartSec = "5s";
+        WorkingDirectory = dataDir;
+        ReadWritePaths = [ dataDir ];
+        CapabilityBoundingSet = [ "" ];
+        DeviceAllow = [ "/dev/null rw" ];
+        DevicePolicy = "strict";
+        LockPersonality = true;
+        MemoryDenyWriteExecute = true;
+        NoNewPrivileges = true;
+        PrivateDevices = true;
+        PrivateTmp = true;
+        ProtectClock = true;
+        ProtectControlGroups = true;
+        ProtectHome = true;
+        ProtectHostname = true;
+        ProtectKernelLogs = true;
+        ProtectKernelModules = true;
+        ProtectKernelTunables = true;
+        ProtectProc = "invisible";
+        ProtectSystem = "strict";
+        RemoveIPC = true;
+        RestrictAddressFamilies = [
+          "AF_INET"
+          "AF_INET6"
+          "AF_UNIX"
+        ];
+        RestrictNamespaces = true;
+        RestrictRealtime = true;
+        RestrictSUIDSGID = true;
+        SystemCallArchitectures = "native";
+        SystemCallFilter = [
+          "@system-service"
+          "~@privileged"
+        ];
+      };
+    };
+in
+{
+  users = {
+    groups.prometheus = { };
+    users.prometheus = {
+      isSystemUser = true;
+      group = "prometheus";
+      description = "Prometheus daemon user";
+    };
+  };
+
+  systemd = {
+    services = {
+      prometheus-main = mkPrometheusService {
+        configFile = mainPrometheusConfig;
+        dataDir = mainPrometheusDataDir;
+        port = 9090;
+        retention = "90d";
+      };
+
+      prometheus-pid-short = mkPrometheusService {
+        configFile = pidPrometheusConfig;
+        dataDir = pidPrometheusDataDir;
+        port = 9092;
+        retention = "10m";
+      };
+    };
+
+    tmpfiles.rules = [
+      "d ${prometheusDataRoot} 0755 root root - -"
+      "d ${mainPrometheusDataDir} 0750 prometheus prometheus - -"
+      "d ${pidPrometheusDataDir} 0750 prometheus prometheus - -"
+    ];
+  };
+}

systems/jeeves/networking.nix

@@ -1,4 +1,13 @@
 {
+  # Docker loads br_netfilter on jeeves. Disable bridge netfilter so
+  # br-nix-builder behaves like a pure L2 bridge and bridged traffic
+  # does not hit the host firewall/rpfilter path.
+  boot.kernel.sysctl = {
+    "net.bridge.bridge-nf-call-arptables" = 0;
+    "net.bridge.bridge-nf-call-ip6tables" = 0;
+    "net.bridge.bridge-nf-call-iptables" = 0;
+  };
+
   networking = {
     hostName = "jeeves";
     hostId = "0e15ce35";
@@ -49,23 +58,10 @@
       "60-br-nix-builder" = {
         matchConfig.Name = "br-nix-builder";
         bridgeConfig = { };
-        address = [ "192.168.3.10/24" ];
-        routingPolicyRules = [
-          {
-            From = "192.168.3.0/24";
-            Table = 100;
-            Priority = 100;
-          }
-        ];
-        routes = [
-          {
-            Gateway = "192.168.3.1";
-            Table = 100;
-            GatewayOnLink = false;
-            Metric = 2048;
-            PreferredSource = "192.168.3.10";
-          }
-        ];
+        networkConfig = {
+          IPv6AcceptRA = false;
+          LinkLocalAddressing = "no";
+        };
         linkConfig.RequiredForOnline = "no";
       };
     };

systems/jeeves/runners/default.nix

@@ -1,20 +1,7 @@
-{ pkgs, ... }:
+{ ... }:
 {
   imports = [ ./nix_builder.nix ];
 
-  users = {
-    users.github-runners = {
-      shell = pkgs.bash;
-      isSystemUser = true;
-      group = "github-runners";
-      uid = 601;
-      openssh.authorizedKeys.keys = [
-        "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIA/S8i+BNX/12JNKg+5EKGX7Aqimt5KM+ve3wt/SyWuO github-runners" # cspell:disable-line
-      ];
-    };
-    groups.github-runners.gid = 601;
-  };
-
   services.nix_builder.containers = {
     nix-builder-00.enable = true;
     nix-builder-01.enable = true;

systems/jeeves/runners/nix_builder.nix

@@ -2,6 +2,7 @@
   config,
   lib,
   outputs,
+  utils,
   ...
 }:
 
@@ -9,6 +10,8 @@ with lib;
 let
   vars = import ../vars.nix;
   cfg = config.services.nix_builder;
+  runnerUsername = "gitea-runner";
+  runnerUserid = 601;
 in
 {
   options.services.nix_builder = {
@@ -23,37 +26,40 @@ in
         types.submodule (
           { name, ... }:
           {
-            options.enable = mkEnableOption "GitHub runner container";
+            options.enable = mkEnableOption "Gitea runner container";
           }
         )
       );
       default = { };
-      description = "GitHub runner container configurations";
+      description = "Gitea runner container configurations";
     };
   };
 
   config = {
+    users = {
+      users.${runnerUsername} = {
+        isSystemUser = true;
+        group = runnerUsername;
+        uid = runnerUserid;
+      };
+      groups.${runnerUsername}.gid = runnerUserid;
+    };
+
     containers = mapAttrs (
       name: containerCfg:
       mkIf containerCfg.enable {
         autoStart = true;
         privateNetwork = true;
         hostBridge = cfg.bridgeName;
-        ephemeral = true;
         bindMounts = {
-          storage = {
-            hostPath = "/zfs/media/github-runners/${name}";
-            mountPoint = "/zfs/media/github-runners/${name}";
-            isReadOnly = false;
-          };
           host-nix = {
             mountPoint = "/host-nix/var/nix/daemon-socket";
             hostPath = "/nix/var/nix/daemon-socket";
             isReadOnly = false;
           };
-          pat = {
-            hostPath = "${vars.secrets}/services/github-runners/runner_pat";
-            mountPoint = "${vars.secrets}/services/github-runners/runner_pat";
+          token = {
+            hostPath = "${vars.secrets}/services/gitea-runners";
+            mountPoint = "/run/secrets/gitea-runners";
             isReadOnly = true;
           };
         };
@@ -92,46 +98,69 @@ in
                 "nix-command"
               ];
               sandbox = true;
-              allowed-users = [ "github-runners" ];
+              allowed-users = [ "gitea-runner" ];
               trusted-users = [
                 "root"
-                "github-runners"
+                "gitea-runner"
               ];
             };
             nixpkgs = {
               overlays = builtins.attrValues outputs.overlays;
               config.allowUnfree = true;
             };
-            services.github-runners.${name} = {
+            users = {
+              users.${runnerUsername} = {
+                isSystemUser = true;
+                group = runnerUsername;
+                uid = runnerUserid;
+              };
+              groups.${runnerUsername}.gid = runnerUserid;
+            };
+            services.gitea-actions-runner.instances.${name} = {
               enable = true;
-              replace = true;
-              workDir = "/zfs/media/github-runners/${name}";
-              url = "https://github.com/RichieCahill/dotfiles";
-              extraLabels = [ "nixos" ];
-              tokenFile = "${vars.secrets}/services/github-runners/runner_pat";
-              user = "github-runners";
-              group = "github-runners";
-              extraPackages = with pkgs; [
+              name = "jeeves-${name}";
+              url = "http://192.168.99.14:6443/";
+              labels = [
+                "self-hosted:host"
+                "nixos:host"
+              ];
+              tokenFile = "/run/secrets/gitea-runners/registration-token";
+              hostPackages = with pkgs; [
+                bash
+                coreutils
+                curl
+                gawk
                 gitMinimal
-                gh
+                gnused
+                my_python
+                nix
                 nixfmt
                 nixos-rebuild
+                nodejs
                 treefmt
-                my_python
+                wget
               ];
             };
-            users = {
-              users.github-runners = {
-                shell = pkgs.bash;
-                isSystemUser = true;
-                group = "github-runners";
-                uid = 601;
+            systemd.services."gitea-runner-${utils.escapeSystemdPath name}" = {
+              serviceConfig = {
+                DynamicUser = mkForce false;
+                User = mkForce runnerUsername;
+                Group = mkForce runnerUsername;
               };
-              groups.github-runners.gid = 601;
             };
             system.stateVersion = "24.05";
           };
       }
     ) cfg.containers;
+
+    systemd.services = builtins.listToAttrs (
+      map (name: {
+        name = "container@${name}";
+        value = {
+          requires = [ "gitea.service" ];
+          after = [ "gitea.service" ];
+        };
+      }) (builtins.attrNames (filterAttrs (_: c: c.enable) cfg.containers))
+    );
   };
 }

systems/jeeves/scripts/zfs.sh

@@ -21,7 +21,9 @@ sudo zfs create media/secure/docker -o compression=zstd-9
 sudo zfs create media/secure/github-runners -o compression=zstd-9 -o sync=disabled
 sudo zfs create media/secure/home_assistant -o compression=zstd-19
 sudo zfs create media/secure/notes -o copies=2
-sudo zfs create media/secure/postgres -o recordsize=16k -o primarycache=metadata
+sudo zfs create media/secure/postgres -o mountpoint=/zfs/media/database/postgres -o recordsize=16k -o primarycache=metadata
+sudo zfs create media/secure/postgres-wal -o mountpoint=/zfs/media/database/postgres-wal -o recordsize=32k -o primarycache=metadata -o special_small_blocks=32K -o compression=lz4 -o secondarycache=none -o logbias=latency
+sudo zfs create media/secure/prometheus -o mountpoint=/zfs/media/database/prometheus -o compression=lz4
 sudo zfs create media/secure/services -o compression=zstd-9
 sudo zfs create media/secure/share -o mountpoint=/zfs/media/share -o exec=off
 
@@ -40,3 +42,4 @@ sudo zfs create storage/secure/plex -o recordsize=1M -o compression=zstd-19
 sudo zfs create storage/secure/secrets -o compression=zstd-19 -o copies=3
 sudo zfs create storage/secure/syncthing -o compression=zstd-19
 sudo zfs create storage/secure/transmission -o recordsize=1M -o compression=zstd-9 -o exec=off -o sync=disabled
+sudo zfs create storage/secure/important -o compression=zstd-19 -o copies=2 -o mountpoint=/zfs/storage/important

systems/jeeves/services/bluesky_firehose.nix

@@ -1,96 +0,0 @@
-{
-  pkgs,
-  inputs,
-  ...
-}:
-let
-  commonEnv = {
-    PYTHONPATH = "${inputs.self}";
-    KAFKA_BOOTSTRAP_SERVERS = "localhost:9092";
-    BLUESKY_FIREHOSE_TOPIC = "bluesky.firehose.posts";
-  };
-  commonServiceConfig = {
-    Type = "simple";
-    WorkingDirectory = "${inputs.self}";
-    User = "richie";
-    Group = "users";
-    Restart = "on-failure";
-    RestartSec = "10s";
-    StandardOutput = "journal";
-    StandardError = "journal";
-    NoNewPrivileges = true;
-    ProtectSystem = "strict";
-    ProtectHome = "read-only";
-    PrivateTmp = true;
-    ReadOnlyPaths = [ "${inputs.self}" ];
-  };
-in
-{
-  systemd.services.bluesky-firehose-topic-init = {
-    description = "Create Kafka topic for Bluesky firehose";
-    after = [ "apache-kafka.service" ];
-    requires = [ "apache-kafka.service" ];
-    wantedBy = [ "multi-user.target" ];
-
-    serviceConfig = {
-      Type = "oneshot";
-      RemainAfterExit = true;
-      ExecStart = pkgs.writeShellScript "create-bluesky-topic" ''
-        ${pkgs.apacheKafka}/bin/kafka-topics.sh \
-          --bootstrap-server localhost:9092 \
-          --create \
-          --if-not-exists \
-          --topic bluesky.firehose.posts \
-          --partitions 6 \
-          --replication-factor 1
-      '';
-    };
-  };
-
-  systemd.services.bluesky-firehose-producer = {
-    description = "Bluesky Jetstream to Kafka producer";
-    after = [
-      "network.target"
-      "apache-kafka.service"
-      "bluesky-firehose-topic-init.service"
-    ];
-    requires = [
-      "apache-kafka.service"
-      "bluesky-firehose-topic-init.service"
-    ];
-    wantedBy = [ "multi-user.target" ];
-
-    environment = commonEnv;
-
-    serviceConfig = commonServiceConfig // {
-      ExecStart = "${pkgs.my_python}/bin/python -m python.data_science.firehose_producer";
-    };
-  };
-
-  systemd.services.bluesky-firehose-consumer = {
-    description = "Bluesky Kafka to PostgreSQL consumer";
-    after = [
-      "network.target"
-      "apache-kafka.service"
-      "bluesky-firehose-topic-init.service"
-      "postgresql.service"
-    ];
-    requires = [
-      "apache-kafka.service"
-      "bluesky-firehose-topic-init.service"
-      "postgresql.service"
-    ];
-    wantedBy = [ "multi-user.target" ];
-
-    environment = commonEnv // {
-      DATA_SCIENCE_DEV_DB = "data_science_dev";
-      DATA_SCIENCE_DEV_USER = "richie";
-      DATA_SCIENCE_DEV_HOST = "/run/postgresql";
-      DATA_SCIENCE_DEV_PORT = "5432";
-    };
-
-    serviceConfig = commonServiceConfig // {
-      ExecStart = "${pkgs.my_python}/bin/python -m python.data_science.firehose_consumer";
-    };
-  };
-}

systems/jeeves/services/camofox-browser.nix

@@ -0,0 +1,80 @@
+{
+  ...
+}:
+let
+  vars = import ../vars.nix;
+in
+{
+  systemd.tmpfiles.rules = [
+    "d ${vars.docker_configs}/camofox-browser 0750 root root - -"
+  ];
+
+  containers.camofox-browser = {
+    autoStart = true;
+    privateNetwork = false;
+    bindMounts = {
+      camofox-browser = {
+        hostPath = "${vars.docker_configs}/camofox-browser";
+        mountPoint = "/var/lib/camofox-browser";
+        isReadOnly = false;
+      };
+    };
+    config =
+      {
+        pkgs,
+        lib,
+        ...
+      }:
+      {
+        networking.hostName = "camofox-browser";
+
+        environment.systemPackages = with pkgs; [
+          ffmpeg
+          git
+          nodejs
+          python3Packages.yt-dlp
+        ];
+
+        systemd.services.camofox-browser = {
+          description = "Camofox browser server";
+          wantedBy = [ "multi-user.target" ];
+          after = [ "network.target" ];
+          environment = {
+            CAMOFOX_HOST = "127.0.0.1";
+            CAMOFOX_PORT = "9377";
+            HOME = "/var/lib/camofox-browser";
+          };
+          path = with pkgs; [
+            bash
+            coreutils
+            git
+            nodejs
+          ];
+          serviceConfig = {
+            Restart = "always";
+            RestartSec = "5s";
+            WorkingDirectory = "/var/lib/camofox-browser";
+          };
+          script = ''
+            set -eu
+
+            app_dir=/var/lib/camofox-browser/app
+
+            if [ ! -d "$app_dir/.git" ]; then
+              git clone --depth 1 https://github.com/jo-inc/camofox-browser "$app_dir"
+            fi
+
+            cd "$app_dir"
+
+            if [ ! -d node_modules ]; then
+              npm install
+            fi
+
+            exec npm start
+          '';
+        };
+
+        system.stateVersion = lib.mkDefault "24.05";
+      };
+  };
+}

systems/jeeves/services/cloud_flare_tunnel.nix

@@ -1,17 +0,0 @@
-{ pkgs, ... }:
-let
-  vars = import ../vars.nix;
-in
-{
-  systemd.services.cloud_flare_tunnel = {
-    description = "cloud_flare_tunnel proxy's traffic through cloudflare";
-    after = [ "network.target" ];
-    wantedBy = [ "multi-user.target" ];
-    serviceConfig = {
-      Type = "simple";
-      EnvironmentFile = "${vars.secrets}/docker/cloud_flare_tunnel";
-      ExecStart = "${pkgs.cloudflared}/bin/cloudflared --no-autoupdate tunnel run";
-      Restart = "on-failure";
-    };
-  };
-}

systems/jeeves/services/gitea.nix

@@ -2,7 +2,10 @@ let
   vars = import ../vars.nix;
 in
 {
-  networking.firewall.allowedTCPPorts = [ 6443 ];
+  networking.firewall.allowedTCPPorts = [
+    6443
+    2223
+  ];
 
   services.gitea = {
     enable = true;
@@ -18,13 +21,17 @@ in
       createDatabase = false;
     };
     settings = {
+      actions = {
+        ENABLED = true;
+        DEFAULT_ACTIONS_URL = "github";
+      };
       service.DISABLE_REGISTRATION = true;
       server = {
         DOMAIN = "tmmworkshop.com";
         ROOT_URL = "https://gitea.tmmworkshop.com/";
         HTTP_PORT = 6443;
         SSH_PORT = 2223;
-        SSH_LISTEN_PORT = 2224;
+        SSH_LISTEN_PORT = 2223;
         START_SSH_SERVER = true;
         PUBLIC_URL_DETECTION = "auto";
       };

systems/jeeves/services/grafana.nix

@@ -0,0 +1,80 @@
+{
+  ...
+}:
+let
+  vars = import ../vars.nix;
+  grafanaDataDir = "${vars.services}/grafana";
+in
+{
+  networking.firewall.allowedTCPPorts = [ 3000 ];
+
+  services.grafana = {
+    enable = true;
+    dataDir = grafanaDataDir;
+    settings = {
+      database.type = "sqlite3";
+      security = {
+        admin_password = "$__file{${vars.secrets}/services/grafana/admin_password}";
+        admin_user = "admin";
+        secret_key = "$__file{${vars.secrets}/services/grafana/secret_key}";
+      };
+      server = {
+        http_addr = "192.168.90.40";
+        http_port = 3000;
+        root_url = "http://192.168.90.40:3000/";
+      };
+    };
+    provision = {
+      enable = true;
+      dashboards.settings = {
+        apiVersion = 1;
+        providers = [
+          {
+            name = "monitoring";
+            folder = "Monitoring";
+            type = "file";
+            disableDeletion = false;
+            editable = false;
+            allowUiUpdates = false;
+            updateIntervalSeconds = 30;
+            options.path = ../monitoring/dashboards;
+          }
+        ];
+      };
+      datasources.settings = {
+        apiVersion = 1;
+        prune = true;
+        datasources = [
+          {
+            access = "proxy";
+            editable = false;
+            isDefault = true;
+            name = "prom-main";
+            type = "prometheus";
+            uid = "prom-main";
+            url = "http://127.0.0.1:9090";
+          }
+          {
+            access = "proxy";
+            editable = false;
+            name = "prom-pid-short";
+            type = "prometheus";
+            uid = "prom-pid-short";
+            url = "http://127.0.0.1:9092";
+          }
+        ];
+      };
+    };
+  };
+
+  systemd = {
+    services.grafana.after = [
+      "prometheus-main.service"
+      "prometheus-pid-short.service"
+    ];
+
+    tmpfiles.rules = [
+      "d ${grafanaDataDir} 0750 grafana grafana - -"
+    ];
+  };
+}

systems/jeeves/services/kafka.nix

@@ -7,6 +7,13 @@ in
     settings = {
       listeners = [ "PLAINTEXT://localhost:9092" ];
       "log.dirs" = [ vars.kafka ];
+      "num.partitions" = 6;
+      "default.replication.factor" = 1;
+      "log.retention.hours" = 168;
+      "log.retention.bytes" = 10737418240;
+      "log.segment.bytes" = 1073741824;
+      "log.cleanup.policy" = "delete";
+      "auto.create.topics.enable" = false;
     };
   };
 }

systems/jeeves/services/postgress.nix

@@ -5,9 +5,14 @@ in
 {
   networking.firewall.allowedTCPPorts = [ 5432 ];
 
+  # Symlink pg_wal to a ZFS dataset on the special (metadata) vdev for fast WAL writes
+  # this is required for systemd sandboxing
+  systemd.services.postgresql.serviceConfig.ReadWritePaths = [ "/zfs/media/database/postgres-wal" ];
+
   services.postgresql = {
     enable = true;
     package = pkgs.postgresql_17_jit;
+    extensions = ps: with ps; [ pgvector ];
     enableTCPIP = true;
     enableJIT = true;
     dataDir = "${vars.database}/postgres";
@@ -33,6 +38,9 @@ in
       # signalbot
       local signalbot  signalbot   trust
 
+      # hedgedoc
+      local hedgedoc   hedgedoc    trust
+
       # math
       local postgres  math   trust
       host  postgres  math   127.0.0.1/32    trust
@@ -112,11 +120,19 @@ in
           login = true;
         };
       }
+      {
+        name = "hedgedoc";
+        ensureDBOwnership = true;
+        ensureClauses = {
+          login = true;
+        };
+      }
     ];
     ensureDatabases = [
       "data_science_dev"
       "hass"
       "gitea"
+      "hedgedoc"
       "math"
       "n8n"
       "richie"

systems/jeeves/services/signal_bot.nix

@@ -1,57 +0,0 @@
-{
-  pkgs,
-  inputs,
-  ...
-}:
-let
-  vars = import ../vars.nix;
-in
-{
-  users = {
-    users.signalbot = {
-      isSystemUser = true;
-      group = "signalbot";
-    };
-    groups.signalbot = { };
-  };
-
-  systemd.services.signal-bot = {
-    description = "Signal command and control bot";
-    after = [
-      "network.target"
-      "podman-signal_cli_rest_api.service"
-    ];
-    wants = [ "podman-signal_cli_rest_api.service" ];
-    wantedBy = [ "multi-user.target" ];
-
-    environment = {
-      PYTHONPATH = "${inputs.self}";
-      SIGNALBOT_DB = "signalbot";
-      SIGNALBOT_USER = "signalbot";
-      SIGNALBOT_HOST = "/run/postgresql";
-      SIGNALBOT_PORT = "5432";
-    };
-
-    serviceConfig = {
-      Type = "simple";
-      WorkingDirectory = "${inputs.self}";
-      User = "signalbot";
-      Group = "signalbot";
-      EnvironmentFile = "${vars.secrets}/services/signal-bot";
-      ExecStart = "${pkgs.my_python}/bin/python -m python.signal_bot.main";
-      StateDirectory = "signal-bot";
-      Restart = "on-failure";
-      RestartSec = "10s";
-      StandardOutput = "journal";
-      StandardError = "journal";
-      NoNewPrivileges = true;
-      ProtectSystem = "strict";
-      ProtectHome = "read-only";
-      PrivateTmp = true;
-      ReadWritePaths = [ "/var/lib/signal-bot" ];
-      ReadOnlyPaths = [
-        "${inputs.self}"
-      ];
-    };
-  };
-}

systems/jeeves/services/validate_system.toml

@@ -1,7 +1,6 @@
 zpool = ["root_pool", "storage", "media"]
 services = [
     "audiobookshelf",
-    "cloud_flare_tunnel",
     "haproxy",
     "docker",
     "home-assistant",

systems/jeeves/snapshot_config.toml

@@ -4,6 +4,7 @@ hourly = 24
 daily = 0
 monthly = 0
 
+# root_pool
 ["root_pool/home"]
 15_min = 8
 hourly = 24
@@ -27,57 +28,96 @@ monthly = 0
 hourly = 24
 daily = 30
 monthly = 6
+# storage
+["storage/ollama"]
+15_min = 2
+hourly = 0
+daily = 0
+monthly = 0
 
-["storage/plex"]
+["storage/secure"]
+15_min = 0
+hourly = 0
+daily = 0
+monthly = 0
+
+["storage/secure/plex"]
 15_min = 6
 hourly = 2
 daily = 1
 monthly = 0
 
-["media/plex"]
-15_min = 6
-hourly = 2
-daily = 1
+["storage/secure/transmission"]
+15_min = 4
+hourly = 0
+daily = 0
 monthly = 0
 
-["media/notes"]
+["storage/secure/secrets"]
 15_min = 8
 hourly = 24
 daily = 30
 monthly = 12
 
-["media/docker"]
-15_min = 3
-hourly = 12
-daily = 14
-monthly = 2
-
-["media/services"]
-15_min = 3
-hourly = 12
-daily = 14
-monthly = 2
-
-["media/home_assistant"]
+# media
+["media/temp"]
+15_min = 2
+hourly = 0
+daily = 0
+monthly = 0
+
+["media/secure"]
+15_min = 0
+hourly = 0
+daily = 0
+monthly = 0
+
+["media/secure/plex"]
+15_min = 6
+hourly = 2
+daily = 1
+monthly = 0
+
+["media/secure/postgres-wal"]
+15_min = 4
+hourly = 2
+daily = 0
+monthly = 0
+
+
+["media/secure/postgres"]
+15_min = 8
+hourly = 24
+daily = 7
+monthly = 0
+
+["media/secure/share"]
+15_min = 4
+hourly = 0
+daily = 0
+monthly = 0
+
+["media/secure/github-runners"]
+15_min = 6
+hourly = 2
+daily = 1
+monthly = 0
+
+["media/secure/notes"]
+15_min = 8
+hourly = 24
+daily = 30
+monthly = 12
+
+["media/secure/docker"]
 15_min = 3
 hourly = 12
 daily = 14
 monthly = 2
 
+# scratch
 ["scratch/transmission"]
-15_min = 0
-hourly = 0
-daily = 0
-monthly = 0
-
-["storage/transmission"]
-15_min = 0
-hourly = 0
-daily = 0
-monthly = 0
-
-["storage/ollama"]
-15_min = 0
+15_min = 2
 hourly = 0
 daily = 0
 monthly = 0

systems/jeeves/syncthing.nix

@@ -89,7 +89,16 @@ in
           ];
           fsWatcherEnabled = true;
         };
-        #
+        "recordings" = {
+          path = "/home/richie/recordings";
+          devices = [
+            "bob"
+            "phone"
+            "rhapsody-in-green"
+          ];
+          fsWatcherEnabled = true;
+        };
+        # davids-server
         "davids-backup1" = {
           id = "8229p-8z3tm"; # cspell:disable-line
           path = "${vars.syncthing}/davids_backups/1";

systems/jeeves/web_services/acme.nix

@@ -0,0 +1,74 @@
+let
+  domains = [
+    "audiobookshelf"
+    "cache"
+    "gitea"
+    "jellyfin"
+    "share"
+  ];
+  extraDomains = [ "www.norn-sight.com" ];
+
+  makeCert = name: {
+    name = "${name}.tmmworkshop.com";
+    value = {
+      webroot = "/var/lib/acme/.challenges";
+      group = "acme";
+      reloadServices = [ "haproxy.service" ];
+    };
+  };
+
+  makeExtraCert = name: {
+    inherit name;
+    value = {
+      webroot = "/var/lib/acme/.challenges";
+      group = "acme";
+      reloadServices = [ "haproxy.service" ];
+    };
+  };
+
+  acmeServices =
+    map (domain: "acme-${domain}.tmmworkshop.com.service") domains
+    ++ map (domain: "acme-${domain}.service") extraDomains;
+in
+{
+  users.users.haproxy.extraGroups = [ "acme" ];
+
+  security.acme = {
+    acceptTerms = true;
+    defaults.email = "Richie@tmmworkshop.com";
+    certs = builtins.listToAttrs ((map makeCert domains) ++ (map makeExtraCert extraDomains));
+  };
+
+  # Minimal nginx to serve ACME HTTP-01 challenge files for HAProxy
+  services.nginx = {
+    enable = true;
+    virtualHosts."acme-challenge" = {
+      listen = [
+        {
+          addr = "127.0.0.1";
+          port = 8402;
+        }
+      ];
+      locations."/.well-known/acme-challenge/" = {
+        root = "/var/lib/acme/.challenges";
+      };
+    };
+  };
+
+  # Ensure the challenge directory exists with correct permissions
+  systemd.tmpfiles.rules = [
+    "d /var/lib/acme/.challenges 0750 acme acme - -"
+    "d /var/lib/acme/.challenges/.well-known 0750 acme acme - -"
+    "d /var/lib/acme/.challenges/.well-known/acme-challenge 0750 acme acme - -"
+  ];
+
+  users.users.nginx.extraGroups = [ "acme" ];
+
+  # HAProxy needs certs to exist before it can bind :443.
+  # NixOS's acme module generates self-signed placeholders on first boot
+  # via acme-<domain>.service — just make HAProxy wait for them.
+  systemd.services.haproxy = {
+    after = acmeServices;
+    wants = acmeServices;
+  };
+}

systems/jeeves/web_services/default.nix

@@ -0,0 +1,9 @@
+{ lib, ... }:
+{
+  imports =
+    let
+      files = builtins.attrNames (builtins.readDir ./.);
+      nixFiles = builtins.filter (name: lib.hasSuffix ".nix" name && name != "default.nix") files;
+    in
+    map (file: ./. + "/${file}") nixFiles;
+}

systems/jeeves/web_services/haproxy.cfg

@@ -6,6 +6,7 @@ global
 defaults
   log global
   mode http
+  option httplog
   retries 3
   maxconn 2000
   timeout connect 5s
@@ -22,24 +23,38 @@ defaults
 #Application Setup
 frontend ContentSwitching
   bind *:80 v4v6
-  bind *:443 v4v6 ssl crt /zfs/storage/secrets/docker/cloudflare.pem
+  bind *:443 v4v6 ssl crt /var/lib/acme/audiobookshelf.tmmworkshop.com/full.pem crt /var/lib/acme/cache.tmmworkshop.com/full.pem crt /var/lib/acme/jellyfin.tmmworkshop.com/full.pem crt /var/lib/acme/share.tmmworkshop.com/full.pem crt /var/lib/acme/gitea.tmmworkshop.com/full.pem crt /var/lib/acme/www.norn-sight.com/full.pem
   mode  http
+
+  # ACME challenge routing (must be first)
+  acl is_acme path_beg /.well-known/acme-challenge/
+
   # tmmworkshop.com
   acl host_audiobookshelf  hdr(host) -i audiobookshelf.tmmworkshop.com
   acl host_cache  hdr(host) -i cache.tmmworkshop.com
   acl host_jellyfin  hdr(host) -i jellyfin.tmmworkshop.com
   acl host_share  hdr(host) -i share.tmmworkshop.com
-  acl host_gcw  hdr(host) -i gcw.tmmworkshop.com
-  acl host_n8n  hdr(host) -i n8n.tmmworkshop.com
   acl host_gitea  hdr(host) -i gitea.tmmworkshop.com
+  acl host_norn_sight  hdr(host) -i www.norn-sight.com
 
+  # Hosts allowed to serve plain HTTP (add entries to skip the HTTPS redirect)
+  acl allow_http hdr(host) -i __none__
+  # acl allow_http hdr(host) -i example.tmmworkshop.com
+
+  # Redirect all HTTP to HTTPS unless on the allow list or ACME challenge
+  http-request redirect scheme https code 301 if !{ ssl_fc } !allow_http !is_acme
+
+  use_backend acme_challenge if is_acme
   use_backend audiobookshelf_nodes if host_audiobookshelf
   use_backend cache_nodes  if host_cache
   use_backend jellyfin if host_jellyfin
   use_backend share_nodes  if host_share
-  use_backend gcw_nodes  if host_gcw
-  use_backend n8n  if host_n8n
   use_backend gitea  if host_gitea
+  use_backend norn_sight  if host_norn_sight
+
+backend acme_challenge
+  mode http
+  server acme 127.0.0.1:8402
 
 backend audiobookshelf_nodes
   mode http
@@ -60,14 +75,10 @@ backend share_nodes
   mode http
   server server 127.0.0.1:8091
 
-backend gcw_nodes
-  mode http
-  server server 127.0.0.1:8092
-
-backend n8n
-  mode http
-  server server 127.0.0.1:5678
-
 backend gitea
   mode http
-  server server 127.0.0.1:6443
+  server server 127.0.0.1:6443
+
+backend norn_sight
+  mode http
+  server server 192.168.90.49:8000

systems/rhapsody-in-green/agent_logger.nix

@@ -0,0 +1,35 @@
+{
+  pkgs,
+  inputs,
+  ...
+}:
+{
+  systemd.services.agent-logger = {
+    description = "Unified agent logger";
+    after = [ "local-fs.target" ];
+    wantedBy = [ "multi-user.target" ];
+
+    environment = {
+      AGENT_LOG_DB = "/var/lib/agent-logger/agent_log.sqlite";
+      HOME = "/home/richie";
+      PYTHONPATH = "${inputs.self}";
+    };
+
+    serviceConfig = {
+      Type = "simple";
+      User = "richie";
+      WorkingDirectory = "/home/richie";
+      ExecStart = "${pkgs.my_python}/bin/python -m python.agent_logger.main";
+      StateDirectory = "agent-logger";
+      Restart = "on-failure";
+      RestartSec = "5s";
+      StandardOutput = "journal";
+      StandardError = "journal";
+      NoNewPrivileges = true;
+      ProtectSystem = "strict";
+      ProtectHome = "read-only";
+      PrivateTmp = true;
+      ReadOnlyPaths = [ "${inputs.self}" ];
+    };
+  };
+}

systems/rhapsody-in-green/default.nix

@@ -11,7 +11,6 @@
     "${inputs.self}/common/optional/yubikey.nix"
     "${inputs.self}/common/optional/zerotier.nix"
     ./hardware.nix
-    ./llms.nix
     ./open_webui.nix
     ./qmk.nix
     ./syncthing.nix
@@ -23,11 +22,20 @@
     hostId = "6404140d";
     firewall = {
       enable = true;
-      allowedTCPPorts = [ ];
+      allowedTCPPorts = [
+        8000
+        8080
+        8081
+      ];
     };
     networkmanager.enable = true;
   };
 
+  programs.appimage = {
+    enable = true;
+    binfmt = true; # allows *.AppImage to be run directly
+  };
+
   services = {
     openssh.ports = [ 922 ];
     flatpak.enable = true;

systems/rhapsody-in-green/llms.nix

@@ -1,30 +0,0 @@
-{
-  services.ollama = {
-    user = "ollama";
-    enable = true;
-    host = "127.0.0.1";
-    syncModels = true;
-    loadModels = [
-      "deepscaler:1.5b"
-      "deepseek-r1:8b"
-      "gemma3:12b"
-      "gemma3:27b"
-      "gpt-oss:20b"
-      "lfm2:24b"
-      "qwen3:14b"
-      "qwen3.5:27b"
-    ];
-  };
-  systemd.services = {
-    ollama.serviceConfig = {
-      Nice = 19;
-      IOSchedulingPriority = 7;
-    };
-    ollama-model-loader.serviceConfig = {
-      Nice = 19;
-      CPUWeight = 50;
-      IOSchedulingClass = "idle";
-      IOSchedulingPriority = 7;
-    };
-  };
-}

systems/rhapsody-in-green/open_webui.nix

@@ -1,6 +1,7 @@
 {
   services.open-webui = {
     enable = true;
+    host = "0.0.0.0";
     environment = {
       ANONYMIZED_TELEMETRY = "False";
       DO_NOT_TRACK = "True";

systems/rhapsody-in-green/syncthing.nix

@@ -55,6 +55,15 @@
       ];
       fsWatcherEnabled = true;
     };
+    "recordings" = {
+      path = "/home/richie/recordings";
+      devices = [
+        "bob"
+        "jeeves"
+        "phone"
+      ];
+      fsWatcherEnabled = true;
+    };
     "vault" = {
       path = "/home/richie/vault";
       devices = [

tests/test_signal_bot.py

@@ -210,9 +210,9 @@ class TestContactCache:
             mock_session_cls.return_value.__exit__ = MagicMock(return_value=False)
             mock_device = MagicMock()
             mock_device.trust_level = TrustLevel.UNVERIFIED
-            mock_session.execute.return_value.scalar_one_or_none.return_value = mock_device
+            mock_session.scalars.return_value.one_or_none.return_value = mock_device
             registry.record_contact("+1234", "abc")
-            mock_session.execute.assert_called_once()
+            mock_session.scalars.assert_called_once()
 
 
 class TestLocationCommand:

users/dov/home/cli/git.nix

@@ -1,6 +1,7 @@
 {
   programs.git = {
     enable = true;
+    signing.format = null;
     settings = {
       user = {
         email = "dov.kruger@gmail.com";

users/elise/home/cli/git.nix

@@ -1,6 +1,7 @@
 {
   programs.git = {
     enable = true;
+    signing.format = null;
     settings = {
       user = {
         email = "DumbPuppy208@gmail.com";

users/math/default.nix

@@ -36,6 +36,8 @@ in
         "hass"
         "libvirtd"
         "networkmanager"
+        "nornsight"
+        "nornsight-admin"
         "plugdev"
         "scanner"
         "transmission"

users/math/home/cli/git.nix

@@ -1,6 +1,7 @@
 {
   programs.git = {
     enable = true;
+    signing.format = null;
     settings = {
       user = {
         email = "matthew.michal11@gmail.com";

users/math/systems/bob.nix

@@ -0,0 +1,5 @@
+{
+  imports = [
+    ../home/global.nix
+  ];
+}

users/richie/default.nix

@@ -36,6 +36,8 @@ in
         "hass"
         "libvirtd"
         "networkmanager"
+        "nornsight"
+        "nornsight-admin"
         "ollama"
         "plugdev"
         "scanner"

users/richie/home/cli/git.nix

@@ -1,6 +1,7 @@
 {
   programs.git = {
     enable = true;
+    signing.format = null;
     settings = {
       user = {
         email = "Richie@tmmworkshop.com";

users/richie/home/gui/default.nix

@@ -6,6 +6,7 @@
     "${inputs.self}/users/shared/sweet.nix"
     ./firefox
     ./kitty.nix
+    ./llm_tools.nix
     ./vscode
   ];
 
@@ -21,10 +22,8 @@
     # browser
     chromium
     # dev tools
-    claude-code
     gparted
     jetbrains.datagrip
     proxychains
-    opencode
   ];
 }

users/richie/home/gui/firefox/default.nix

@@ -1,8 +1,9 @@
-{ inputs, ... }:
+{ config, inputs, ... }:
 {
   imports = [ ./search_engines.nix ];
 
   programs.firefox = {
+    configPath = "${config.xdg.configHome}/mozilla/firefox";
     enable = true;
     profiles.richie = {
       extensions.packages = with inputs.firefox-addons.packages.x86_64-linux; [

users/richie/home/gui/kitty.nix

@@ -6,6 +6,7 @@
     settings = {
       allow_remote_control = "yes";
       shell = "${pkgs.zsh}/bin/zsh";
+      scrollback_lines = 50000;
       wayland_titlebar_color = "background";
       background_opacity = "0.75";
       tab_bar_edge = "top";

users/richie/home/gui/llm_tools.nix

@@ -0,0 +1,9 @@
+{ pkgs, ... }:
+{
+  home.packages = [
+    pkgs.claude-code
+    pkgs.master.codex
+    pkgs.master.opencode
+    pkgs.master.pi-coding-agent
+  ];
+}

users/richie/home/gui/vscode/keybindings.json

@@ -2,28 +2,32 @@
   {
     "key": "shift+alt+f",
     "command": "editor.action.formatDocument",
-    "when": "editorHasDocumentFormattingProvider && editorTextFocus && !editorReadonly && !inCompositeEditor"
+    "when": "editorHasDocumentFormattingProvider && editorTextFocus && !editorReadonly && !inCompositeEditor",
   },
   {
     "key": "alt+a d",
-    "command": "cSpell.addWordToWorkspaceSettings"
+    "command": "cSpell.addWordToWorkspaceSettings",
   },
   {
     "key": "ctrl+shift+`",
-    "command": "workbench.action.createTerminalEditor"
+    "command": "workbench.action.createTerminalEditor",
   },
   {
     "key": "ctrl+shift+`",
     "command": "-workbench.action.terminal.new",
-    "when": "terminalProcessSupported || terminalWebExtensionContributedProfile"
+    "when": "terminalProcessSupported || terminalWebExtensionContributedProfile",
   },
   {
     "key": "ctrl+shift+g r",
-    "command": "gitlens.git.rebase"
+    "command": "gitlens.git.rebase",
   },
   {
     "key": "ctrl+shift+g c",
     "command": "-gitlens.showQuickCommitFileDetails",
-    "when": "editorTextFocus && !gitlens:disabled && config.gitlens.keymap == 'chorded'"
-  }
+    "when": "editorTextFocus && !gitlens:disabled && config.gitlens.keymap == 'chorded'",
+  },
+  {
+    "key": "ctrl+shift+g p",
+    "command": "gitlens.pushRepositories",
+  },
 ]

users/richie/home/gui/vscode/settings.json

@@ -14,21 +14,21 @@
   "git.pruneOnFetch": true,
   "terminal.integrated.scrollback": 10000,
   "update.mode": "none",
-  "workbench.colorTheme": "Default Dark+",
+  "workbench.colorTheme": "Dark+",
   "workbench.secondarySideBar.showLabels": false,
 
   // turns off all sounds and announcements
   "accessibility.signals.terminalCommandFailed": {
     "sound": "off",
-    "announcement": "off"
+    "announcement": "off",
   },
   "accessibility.signals.terminalQuickFix": {
     "sound": "off",
-    "announcement": "off"
+    "announcement": "off",
   },
   "accessibility.signals.terminalBell": {
     "sound": "off",
-    "announcement": "off"
+    "announcement": "off",
   },
 
   // database settings
@@ -41,8 +41,8 @@
       "driver": "PostgreSQL",
       "name": "main",
       "database": "postgres",
-      "username": "richie"
-    }
+      "username": "richie",
+    },
   ],
 
   // formatters
@@ -55,7 +55,7 @@
   "[yaml]": { "editor.defaultFormatter": "redhat.vscode-yaml" },
   "[javascriptreact]": { "editor.defaultFormatter": "esbenp.prettier-vscode" },
   "[github-actions-workflow]": {
-    "editor.defaultFormatter": "redhat.vscode-yaml"
+    "editor.defaultFormatter": "redhat.vscode-yaml",
   },
   "[dockercompose]": {
     "editor.insertSpaces": true,
@@ -64,9 +64,9 @@
     "editor.quickSuggestions": {
       "other": true,
       "comments": false,
-      "strings": true
+      "strings": true,
     },
-    "editor.defaultFormatter": "redhat.vscode-yaml"
+    "editor.defaultFormatter": "redhat.vscode-yaml",
   },
 
   // spell check
@@ -78,7 +78,10 @@
     "Corvidae",
     "drivername",
     "fastapi",
-    "syncthing"
+    "Michal",
+    "Nornsight",
+    "sandboxing",
+    "syncthing",
   ],
 
   // nix
@@ -96,5 +99,6 @@
   // new
   "hediet.vscode-drawio.resizeImages": null,
   "hediet.vscode-drawio.appearance": "automatic",
-  "claudeCode.preferredLocation": "panel"
+  "claudeCode.preferredLocation": "panel",
+  "docker.extension.enableComposeLanguageServer": false,
 }

users/richie/home/ssh_config.nix

@@ -2,46 +2,46 @@
   programs.ssh = {
     enable = true;
     enableDefaultConfig = false;
-    matchBlocks = {
+    settings = {
       jeeves = {
-        hostname = "192.168.90.40";
-        user = "richie";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 629;
-        dynamicForwards = [ { port = 9050; } ];
-        compression = true;
+        HostName = "192.168.90.40";
+        User = "richie";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 629;
+        DynamicForward = [ { port = 9050; } ];
+        Compression = true;
       };
       unlock-jeeves = {
-        hostname = "192.168.99.14";
-        user = "root";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 2222;
+        HostName = "192.168.99.14";
+        User = "root";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 2222;
       };
       brain = {
-        hostname = "192.168.90.35";
-        user = "richie";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 129;
-        dynamicForwards = [ { port = 9050; } ];
+        HostName = "192.168.90.35";
+        User = "richie";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 129;
+        DynamicForward = [ { port = 9050; } ];
       };
       unlock-brain = {
-        hostname = "192.168.95.35";
-        user = "root";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 2222;
+        HostName = "192.168.95.35";
+        User = "root";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 2222;
       };
       bob = {
-        hostname = "192.168.90.25";
-        user = "richie";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 262;
-        dynamicForwards = [ { port = 9050; } ];
+        HostName = "192.168.90.25";
+        User = "richie";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 262;
+        DynamicForward = [ { port = 9050; } ];
       };
       rhapsody-in-green = {
-        hostname = "192.168.90.221";
-        user = "richie";
-        identityFile = "~/.ssh/id_ed25519";
-        port = 922;
+        HostName = "192.168.90.221";
+        User = "richie";
+        IdentityFile = "~/.ssh/id_ed25519";
+        Port = 922;
       };
     };
   };

users/richie/systems/bob.nix

@@ -1,6 +1,5 @@
 {
   imports = [
     ../home/global.nix
-    ../home/gui
   ];
 }

users/steve/default.nix

@@ -0,0 +1,44 @@
+{
+  pkgs,
+  config,
+  ...
+}:
+let
+  ifTheyExist = groups: builtins.filter (group: builtins.hasAttr group config.users.groups) groups;
+in
+{
+
+  users = {
+    users.steve = {
+      isNormalUser = true;
+
+      shell = pkgs.zsh;
+      group = "steve";
+      openssh.authorizedKeys.keys = [
+        "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIJH03VzDbUhzfhvwD+OsYh6GobODYaI9jdNdzWQoqFsp matth@Jove" # cspell:disable-line
+      ];
+      extraGroups = [
+        "audio"
+        "video"
+        "wheel"
+        "users"
+      ]
+      ++ ifTheyExist [
+        "dialout"
+        "docker"
+        "hass"
+        "libvirtd"
+        "networkmanager"
+        "plugdev"
+        "scanner"
+        "transmission"
+        "uaccess"
+        "wireshark"
+      ];
+      uid = 1005;
+    };
+
+    groups.steve.gid = 1005;
+  };
+  home-manager.users.steve = import ./systems/${config.networking.hostName}.nix;
+}

users/steve/home/cli/default.nix

@@ -0,0 +1,9 @@
+{
+  imports = [
+    ./direnv.nix
+    ./git.nix
+    ./zsh.nix
+  ];
+
+  programs.starship.enable = true;
+}

users/steve/home/cli/direnv.nix

@@ -0,0 +1,8 @@
+{
+
+  programs.direnv = {
+    enable = true;
+    enableZshIntegration = true;
+    nix-direnv.enable = true;
+  };
+}

users/steve/home/cli/git.nix

@@ -0,0 +1,15 @@
+{
+  programs.git = {
+    enable = true;
+    signing.format = null;
+    settings = {
+      user = {
+        email = "matthew.michal11@gmail.com";
+        name = "Matthew Michal";
+      };
+      pull.rebase = true;
+      color.ui = true;
+    };
+    lfs.enable = true;
+  };
+}

users/steve/home/cli/zsh.nix

@@ -0,0 +1,28 @@
+{
+  programs.zsh = {
+    enable = true;
+    syntaxHighlighting.enable = true;
+    history.size = 10000;
+    oh-my-zsh = {
+      enable = true;
+      plugins = [
+        "git"
+        "docker"
+        "docker-compose"
+        "colored-man-pages"
+        "rust"
+        "systemd"
+        "tmux"
+        "ufw"
+        "z"
+      ];
+    };
+    shellAliases = {
+      "lrt" = "eza --icons -lsnew";
+      "ls" = "eza";
+      "ll" = "eza --long --group";
+      "la" = "eza --all";
+
+    };
+  };
+}

users/steve/home/global.nix

@@ -0,0 +1,22 @@
+{ config, ... }:
+{
+  imports = [
+    ./cli
+    ./programs.nix
+    ./ssh_config.nix
+  ];
+
+  programs = {
+    home-manager.enable = true;
+    git.enable = true;
+  };
+
+  home = {
+    username = "steve";
+    homeDirectory = "/home/${config.home.username}";
+    stateVersion = "24.05";
+    sessionVariables = {
+      FLAKE = "$HOME/dotfiles";
+    };
+  };
+}

users/steve/home/programs.nix

@@ -0,0 +1,57 @@
+{ pkgs, ... }:
+{
+  home.packages = with pkgs; [
+    # cli
+    bat
+    btop
+    eza
+    fd
+    ffmpegthumbnailer
+    fzf
+    git
+    gnupg
+    imagemagick
+    jq
+    ncdu
+    ouch
+    p7zip
+    poppler
+    rar
+    ripgrep
+    starship
+    tmux
+    unzip
+    yazi
+    zoxide
+    # system info
+    hwloc
+    lynis
+    pciutils
+    smartmontools
+    usbutils
+    # networking
+    iperf3
+    nmap
+    wget
+    # python
+    poetry
+    ruff
+    uv
+    # nodejs
+    nodejs
+    # Rust packages
+    trunk
+    wasm-pack
+    cargo-watch
+    cargo-generate
+    cargo-audit
+    cargo-update
+    # nix
+    nix-init
+    nix-output-monitor
+    nix-prefetch
+    nix-tree
+    nixfmt
+    treefmt
+  ];
+}