thread-archive 0.0.2__py3-none-any.whl

thread_archive/_web/server.py

@@ -0,0 +1,299 @@
+"""The ``archive web`` server: a socket-free router + a thin stdlib HTTP adapter.
+
+The whole read surface the UI needs already exists as the plain Python library
+(:mod:`thread_archive._api`): ``search`` / ``read_thread`` / ``status``. This is a
+skin over it — no ranking/fusion logic is duplicated here. :func:`route` is a pure
+``(method, path, params) -> (status, content_type, body, headers)`` function so
+tests drive it without opening a socket. :func:`serve` runs it in the foreground
+(``archive web``); :func:`serve_in_thread` runs it in a background daemon thread so
+the always-on ``archive watch`` process can cohost the viewer (one process, one
+engine) — that's how the read surface gets a persistent URL with no extra daemon.
+"""
+
+from __future__ import annotations
+
+import json
+import threading
+from datetime import datetime
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from pathlib import Path
+from typing import Optional
+from urllib.parse import parse_qs, urlparse
+
+from .. import _api as api
+
+STATIC_DIR = (Path(__file__).parent / "static").resolve()
+
+# Content types for the built bundle (vite emits hashed assets/*.js|css + fonts).
+_CTYPES = {
+    ".html": "text/html; charset=utf-8",
+    ".js": "text/javascript; charset=utf-8",
+    ".mjs": "text/javascript; charset=utf-8",
+    ".css": "text/css; charset=utf-8",
+    ".json": "application/json",
+    ".map": "application/json",
+    ".svg": "image/svg+xml",
+    ".ico": "image/x-icon",
+    ".png": "image/png",
+    ".woff": "font/woff",
+    ".woff2": "font/woff2",
+}
+
+
+# ---------------------------------------------------------------------------
+# small helpers
+# ---------------------------------------------------------------------------
+def _json_default(obj):
+    """Serialize the non-JSON-native values the api hits carry (datetimes)."""
+    if isinstance(obj, datetime):
+        return obj.isoformat()
+    return str(obj)
+
+
+# Responses are uniform 4-tuples: (status, content_type, body, extra_headers).
+Response = tuple[int, str, bytes, dict]
+
+
+def _ok(obj) -> Response:
+    return 200, "application/json", json.dumps(obj, default=_json_default).encode(), {}
+
+
+def _text(status: int, msg: str) -> Response:
+    return status, "text/plain; charset=utf-8", msg.encode(), {}
+
+
+def _redirect(url: str) -> Response:
+    return 302, "text/plain; charset=utf-8", b"", {"Location": url}
+
+
+def _serve_file(p: Path) -> Response:
+    if not p.is_file():
+        return _text(404, "not found")
+    ctype = _CTYPES.get(p.suffix.lower(), "application/octet-stream")
+    return 200, ctype, p.read_bytes(), {}
+
+
+def _first(params: dict, key: str) -> Optional[str]:
+    v = params.get(key, [None])[0]
+    return v if v else None
+
+
+def _repeated(params: dict, key: str, limit: int = 10) -> list[str]:
+    """Every value given for ``key``, in order, deduped and capped."""
+    seen: list[str] = []
+    for v in params.get(key, []):
+        v = (v or "").strip()
+        if v and v not in seen:
+            seen.append(v)
+    return seen[:limit]
+
+
+def _int(params: dict, key: str, default: int) -> int:
+    try:
+        return int(params.get(key, [default])[0])
+    except (TypeError, ValueError):
+        return default
+
+
+def _bool(params: dict, key: str, default: bool) -> bool:
+    v = _first(params, key)
+    if v is None:
+        return default
+    return v.lower() in ("1", "true", "yes", "on")
+
+
+def _csv(params: dict, key: str) -> Optional[list[str]]:
+    v = _first(params, key)
+    if not v:
+        return None
+    return [s for s in (x.strip() for x in v.split(",")) if s] or None
+
+
+def _list_threads(*, limit: int, q: Optional[str]) -> list[dict]:
+    """Recent conversation threads for the sidebar (newest first). Topics and
+    archived threads are excluded; ``q`` filters on title/name substring."""
+    from sqlalchemy import select
+
+    from .._store import Thread, get_session
+
+    api.open_archive()  # ensure the engine is up for this process' home
+    stmt = (
+        select(Thread.id, Thread.title, Thread.name, Thread.source, Thread.updated_at)
+        .where(Thread.thread_type != "topic")
+        .where(Thread.archived.is_(False))
+    )
+    if q:
+        like = f"%{q}%"
+        stmt = stmt.where(Thread.title.ilike(like) | Thread.name.ilike(like))
+    stmt = stmt.order_by(Thread.updated_at.desc()).limit(limit)
+    with get_session() as s:
+        rows = s.execute(stmt).all()
+    return [
+        {
+            "id": r.id,
+            "title": r.title or r.name,
+            "source": r.source,
+            "updated_at": r.updated_at.isoformat() if r.updated_at else None,
+        }
+        for r in rows
+    ]
+
+
+def resolve_archive_link(link_id: str, source: Optional[str] = None) -> Optional[int]:
+    """Resolve a provider session id to its archive thread id, via ``ImportState``.
+
+    This is the archive-link lookup an editor needs ("I have a session uuid, open the
+    conversation"), and the same lookup that lets a bare uuid be pasted straight into
+    ``/archive/<uuid>``. The id passed is the tail of a longer ``source_id`` the watcher
+    stored, joined by a provider-specific separator: claude-code stores ``{project}:{uuid}``
+    (``:``), codex stores the rollout filename stem ``rollout-{ts}-{uuid}`` (``-``), and
+    cloth stores the bare session uuid. So we match exact **or** either separator-suffix —
+    a bare uuid finds ``{project}:{uuid}``, ``rollout-…-{uuid}``, *and* the bare cloth uuid.
+    ``source`` narrows the search to one provider (an editor knows its own); omit it (None)
+    to resolve across every provider — what pasting a bare uuid as a thread id wants, since
+    the paster rarely knows which harness it came from. Newest import wins. Owned here: the
+    watcher cohosts the persistent server, so the archive serves its own editor
+    links."""
+    from sqlalchemy import select
+
+    from .._store import ImportState, get_session
+
+    api.open_archive()
+    stmt = (
+        select(ImportState.thread_id)
+        .where(ImportState.thread_id.isnot(None))
+        .where(
+            (ImportState.source_id == link_id)
+            | (ImportState.source_id.like(f"%:{link_id}"))
+            | (ImportState.source_id.like(f"%-{link_id}"))
+        )
+        .order_by(ImportState.last_import_at.desc())
+    )
+    if source:
+        stmt = stmt.where(ImportState.source == source.replace("_", "-"))
+    with get_session() as s:
+        return s.execute(stmt).scalars().first()
+
+
+# ---------------------------------------------------------------------------
+# the router (socket-free, the testable core)
+# ---------------------------------------------------------------------------
+def route(method: str, path: str, params: dict) -> Response:
+    """Resolve one request to ``(status, content_type, body, extra_headers)``."""
+    if method != "GET":
+        return _text(405, "method not allowed")
+
+    # ---- JSON API: thin wrappers over thread_archive._api ----
+    if path == "/api/health":
+        # Cheap liveness for probes (the family manifest's health URL).
+        # /api/status is the real survey but counts the whole index — seconds,
+        # not the milliseconds a poller budgets.
+        paths = api.open_archive()
+        return _ok({"ok": paths.index_path.exists(), "home": str(paths.home)})
+
+    if path == "/api/status":
+        return _ok(api.status())
+
+    if path == "/api/archive-link":
+        # ``id`` may repeat: a caller that cannot tell which of the uuids it can
+        # see is the session id sends every candidate, best guess first, and the
+        # archive — the only party that knows what was actually imported — picks
+        # the first that resolves. An editor webview holds ids for turns, drafts
+        # and client-side threads that look exactly like a session uuid and can
+        # never resolve; making them harmless beats guessing right.
+        link_ids = _repeated(params, "id")
+        if not link_ids:
+            return _text(400, "missing id")
+        source = _first(params, "source")
+        for link_id in link_ids:
+            tid = resolve_archive_link(link_id, source)
+            if tid is not None:
+                url = f"/archive/{tid}"
+                if _bool(params, "redirect", False):
+                    return _redirect(url)
+                return _ok({"thread_id": tid, "url": url, "id": link_id})
+        tried = ", ".join(link_ids)
+        return 404, "application/json", json.dumps({"error": f"no thread for id={tried}"}).encode(), {}
+
+    if path == "/api/search":
+        q = (_first(params, "q") or "").strip()
+        if not q:
+            return _ok({"query": "", "hits": []})
+        hits = api.search(
+            q,
+            limit=_int(params, "limit", 30),
+            source=_csv(params, "source"),
+            content_types=_csv(params, "content_types"),
+            since=_first(params, "since"),
+            until=_first(params, "until"),
+        )
+        return _ok({"query": q, "hits": hits})
+
+    if path.startswith("/api/read/") or path.startswith("/api/thread/"):
+        try:
+            tid = int(path.rsplit("/", 1)[-1])
+        except ValueError:
+            return _text(404, "bad thread id")
+        thinking = _bool(params, "thinking", path.startswith("/api/thread/"))
+        tools = _bool(params, "tools", True)
+        if path.startswith("/api/thread/"):
+            # structured render blocks (the viewer's reader)
+            return _ok(api.read_thread_structured(tid, include_thinking=thinking, include_tools=tools))
+        # flat transcript string (CLI-shaped; kept for back-compat). 'full' shows
+        # tool calls (with thinking), 'chat' is the readable assistant text only.
+        transcript = api.read_thread(tid, mode="full" if tools else "chat")
+        return _ok({"thread_id": tid, "transcript": transcript})
+
+    if path == "/api/threads":
+        return _ok(
+            {"threads": _list_threads(limit=_int(params, "limit", 100), q=_first(params, "q"))}
+        )
+
+    # unmatched API path — don't fall through to the SPA shell
+    if path.startswith("/api/"):
+        return _text(404, "not found")
+
+    # ---- a real static asset from the built bundle (assets/*.js|css, favicon…) ----
+    rel = path.lstrip("/")
+    if rel:
+        candidate = (STATIC_DIR / rel).resolve()
+        if (candidate == STATIC_DIR or STATIC_DIR in candidate.parents) and candidate.is_file():
+            return _serve_file(candidate)
+
+    # ---- SPA fallback: every other path renders the app shell (client routing) ----
+    return _serve_file(STATIC_DIR / "index.html")
+
+
+# ---------------------------------------------------------------------------
+# the HTTP adapter (the only networked part)
+# ---------------------------------------------------------------------------
+class _Handler(BaseHTTPRequestHandler):
+    def do_GET(self):  # noqa: N802 — stdlib dispatch name
+        parsed = urlparse(self.path)
+        try:
+            status, ctype, body, headers = route("GET", parsed.path, parse_qs(parsed.query))
+        except Exception as exc:  # noqa: BLE001 — isolate per request; never kill the loop
+            status, ctype, headers = 500, "application/json", {}
+            body = json.dumps({"error": str(exc)}).encode()
+        self.send_response(status)
+        self.send_header("Content-Type", ctype)
+        self.send_header("Content-Length", str(len(body)))
+        for key, value in headers.items():
+            self.send_header(key, value)
+        self.end_headers()
+        self.wfile.write(body)
+
+    def log_message(self, *args):  # keep the foreground console quiet
+        pass
+
+
+def serve_in_thread(*, host: str = "127.0.0.1", port: int = 8787) -> ThreadingHTTPServer:
+    """Start the viewer on a background daemon thread and return the server.
+
+    For ``archive watch --web``: the always-on watcher process cohosts the read
+    surface so there's a persistent URL without a second daemon. Assumes the caller
+    already opened the archive (the watcher does). The thread is a daemon, so it dies
+    with the process; the caller may ``server_close()`` on shutdown for a clean stop."""
+    httpd = ThreadingHTTPServer((host, port), _Handler)
+    threading.Thread(target=httpd.serve_forever, name="archive-web", daemon=True).start()
+    return httpd

thread_archive/_web/static/assets/index-DECSH1Mz.css

@@ -0,0 +1,10 @@
+pre code.hljs{display:block;overflow-x:auto;padding:1em}code.hljs{padding:3px 5px}/*!
+  Theme: GitHub Dark
+  Description: Dark theme as seen on github.com
+  Author: github.com
+  Maintainer: @Hirse
+  Updated: 2021-05-15
+
+  Outdated base version: https://github.com/primer/github-syntax-dark
+  Current colors taken from GitHub's CSS
+*/.hljs{color:#c9d1d9;background:#0d1117}.hljs-doctag,.hljs-keyword,.hljs-meta .hljs-keyword,.hljs-template-tag,.hljs-template-variable,.hljs-type,.hljs-variable.language_{color:#ff7b72}.hljs-title,.hljs-title.class_,.hljs-title.class_.inherited__,.hljs-title.function_{color:#d2a8ff}.hljs-attr,.hljs-attribute,.hljs-literal,.hljs-meta,.hljs-number,.hljs-operator,.hljs-variable,.hljs-selector-attr,.hljs-selector-class,.hljs-selector-id{color:#79c0ff}.hljs-regexp,.hljs-string,.hljs-meta .hljs-string{color:#a5d6ff}.hljs-built_in,.hljs-symbol{color:#ffa657}.hljs-comment,.hljs-code,.hljs-formula{color:#8b949e}.hljs-name,.hljs-quote,.hljs-selector-tag,.hljs-selector-pseudo{color:#7ee787}.hljs-subst{color:#c9d1d9}.hljs-section{color:#1f6feb;font-weight:700}.hljs-bullet{color:#f2cc60}.hljs-emphasis{color:#c9d1d9;font-style:italic}.hljs-strong{color:#c9d1d9;font-weight:700}.hljs-addition{color:#aff5b4;background-color:#033a16}.hljs-deletion{color:#ffdcd7;background-color:#67060c}:root{--bg: #fbfbfa;--panel: #ffffff;--ink: #1b1b19;--muted: #6b6b66;--line: #e6e4df;--accent: #3b6ea5;--accent-soft: #e8eef6;--user: #f3f1ec;--assistant: #ffffff;--badge: #efece6;--code-bg: #0d1117}@media(prefers-color-scheme:dark){:root{--bg: #16161a;--panel: #1d1d22;--ink: #e9e8e4;--muted: #9a988f;--line: #2c2c33;--accent: #7aa7d6;--accent-soft: #243240;--user: #232329;--assistant: #1d1d22;--badge: #2a2a31;--code-bg: #0d1117}}*{box-sizing:border-box}html,body,#root{margin:0;height:100%}body{background:var(--bg);color:var(--ink);font:14px/1.55 -apple-system,BlinkMacSystemFont,Segoe UI,Roboto,sans-serif}button{font:inherit;color:inherit;background:none;border:none;cursor:pointer;text-align:left}.app{display:flex;height:100vh}aside{width:300px;min-width:300px;border-right:1px solid var(--line);background:var(--panel);display:flex;flex-direction:column;height:100%}.brand{padding:14px 16px 10px;font-weight:600;letter-spacing:.2px}.searchbar{padding:0 12px 12px;border-bottom:1px solid var(--line)}.input{width:100%;padding:9px 11px;border:1px solid var(--line);border-radius:8px;background:var(--bg);color:var(--ink);font:inherit;outline:none}.input:focus{border-color:var(--accent)}.rail-head{padding:12px 16px 6px;font-size:11px;text-transform:uppercase;letter-spacing:.08em;color:var(--muted);display:flex;gap:8px;align-items:center}.rail-filter{margin-left:auto;width:110px;padding:3px 7px;font-size:11px;text-transform:none;letter-spacing:0;border:1px solid var(--line);border-radius:6px;background:var(--bg);color:var(--ink);outline:none}.rail{overflow-y:auto;flex:1;padding-bottom:12px}.rail-item{display:block;width:100%;padding:8px 16px;border-left:2px solid transparent}.rail-item:hover{background:var(--accent-soft)}.rail-item.active{background:var(--accent-soft);border-left-color:var(--accent)}.rail-item .t{display:block;white-space:nowrap;overflow:hidden;text-overflow:ellipsis}.rail-item .m{color:var(--muted);font-size:12px}.rail-empty{padding:16px;color:var(--muted)}main{flex:1;display:flex;flex-direction:column;min-width:0;height:100%}.statusbar{color:var(--muted);font-size:12px;padding:10px 28px;border-bottom:1px solid var(--line)}.statusbar.err{color:#c4543b}.content{flex:1;overflow-y:auto}.wrap{max-width:840px;margin:0 auto;padding:24px 28px 96px}h1.title{font-size:19px;margin:0 0 4px}.submeta{color:var(--muted);font-size:12px;margin-bottom:18px}.toggles{font-size:12px;color:var(--muted);margin-bottom:18px}.toggles label{margin-right:14px;cursor:pointer}.empty{color:var(--muted);padding:44px 0;text-align:center}.group{margin-bottom:22px}.gh{font-weight:600;padding:4px 0;display:flex;gap:8px;align-items:baseline;width:100%}.gh:hover{color:var(--accent)}.src{color:var(--muted);font-size:11px;font-weight:400}.hit{display:block;width:100%;border:1px solid var(--line);border-radius:8px;padding:9px 12px;margin:8px 0;background:var(--panel)}.hit:hover{border-color:var(--accent)}.snip{white-space:pre-wrap;word-break:break-word}.meta{margin-top:5px;display:flex;gap:7px;align-items:center;flex-wrap:wrap}.badge{background:var(--badge);color:var(--muted);border-radius:5px;padding:1px 6px;font-size:11px}.badge.sem{background:var(--accent-soft);color:var(--accent)}.msg{position:relative;border-radius:10px;padding:12px 16px;margin:12px 0;border:1px solid var(--line)}.msg.user{background:var(--user)}.msg.assistant{background:var(--assistant)}.msg .role{font-size:11px;text-transform:uppercase;letter-spacing:.08em;color:var(--muted);margin-bottom:8px}.msg.cont{margin-top:0;border-top:none;border-top-left-radius:0;border-top-right-radius:0;padding-top:4px}.msg:has(+.msg.cont){border-bottom-left-radius:0;border-bottom-right-radius:0}.md>:first-child{margin-top:0}.md>:last-child{margin-bottom:0}.md p{margin:0 0 10px;white-space:pre-wrap}.md pre{background:var(--code-bg);border-radius:8px;padding:12px 14px;overflow-x:auto}.md code{font-family:ui-monospace,SFMono-Regular,Menlo,monospace;font-size:12.5px}.md :not(pre)>code{background:var(--badge);padding:1px 5px;border-radius:4px}.md pre code{background:none;padding:0}.md a{color:var(--accent)}.md table{border-collapse:collapse;margin:8px 0}.md th,.md td{border:1px solid var(--line);padding:4px 9px}.md blockquote{margin:8px 0;padding-left:12px;border-left:3px solid var(--line);color:var(--muted)}.tool{margin:8px 0;border:1px solid var(--line);border-radius:8px;background:var(--bg)}.tool>summary{cursor:pointer;padding:7px 11px;display:flex;gap:8px;align-items:center;font-size:12px;color:var(--muted);list-style:none}.tool>summary::-webkit-details-marker{display:none}.tool>summary:before{content:"▸";font-size:10px}.tool[open]>summary:before{content:"▾"}.tool-name{color:var(--ink);font-weight:600;font-family:ui-monospace,Menlo,monospace}.tool-tag{text-transform:uppercase;letter-spacing:.06em;font-size:10px}.tool.error{border-color:#c4543b66}.tool.error .tool-tag{color:#c4543b}.tool-body{margin:0;padding:10px 13px;border-top:1px solid var(--line);white-space:pre-wrap;word-break:break-word;overflow-x:auto;font-family:ui-monospace,SFMono-Regular,Menlo,monospace;font-size:12px}.tool.thinking summary{font-style:italic}.msg-divider{margin:14px 0}.msg-divider .model-switch{margin:0}.model-switch{display:flex;align-items:center;gap:10px;margin:12px 0 8px}.model-switch:before,.model-switch:after{content:"";flex:1;border-top:1px dashed var(--line)}.model-switch-label{font-size:11px;letter-spacing:.04em;color:var(--muted);text-transform:uppercase;white-space:nowrap}.model-switch-label b{color:var(--ink);font-weight:600;font-family:ui-monospace,Menlo,monospace;text-transform:none}.safeguard{margin:8px 0;padding:9px 12px;border-radius:8px;border:1px solid #c9962e55;background:#c9962e14;font-size:13px}.safeguard-tag{display:block;margin-bottom:4px;font-size:10px;letter-spacing:.06em;text-transform:uppercase;color:#b07d1e}.safeguard .md p{margin:0}.md-raw{margin:0;white-space:pre-wrap;word-break:break-word;font-family:ui-monospace,SFMono-Regular,Menlo,monospace;font-size:12.5px;color:var(--ink)}.msg-foot{display:flex;justify-content:flex-end;margin-top:6px}.msg-info{width:22px;height:22px;border-radius:50%;display:flex;align-items:center;justify-content:center;color:var(--muted);font-size:14px;line-height:1;opacity:.35;transition:opacity .12s ease,color .12s ease}.msg:hover .msg-info{opacity:.7}.msg-info:hover,.msg-info[aria-expanded=true]{opacity:1;color:var(--accent)}.msg-meta{margin-top:8px;padding-top:10px;border-top:1px solid var(--line);display:flex;flex-direction:column;gap:8px;align-items:flex-start}.msg-meta-row{display:flex;flex-wrap:wrap;gap:10px;align-items:center;font-size:11px;color:var(--muted);font-family:ui-monospace,SFMono-Regular,Menlo,monospace}.msg-meta-row span{white-space:nowrap}.chip{background:var(--accent-soft);color:var(--accent);border-radius:5px;padding:1px 7px;font-size:11px;font-family:ui-monospace,SFMono-Regular,Menlo,monospace}.raw-toggle{padding:4px 10px;border:1px solid var(--line);border-radius:7px;background:var(--bg);color:var(--ink);font-size:12px}.raw-toggle:hover{border-color:var(--accent)}.raw-toggle.on{background:var(--accent-soft);border-color:var(--accent);color:var(--accent)}.model-tags{display:inline-flex;flex-wrap:wrap;gap:6px;margin-left:8px;vertical-align:middle}.model-tag{font-family:ui-monospace,SFMono-Regular,Menlo,monospace;font-size:11px;padding:1px 7px;border-radius:5px;color:hsl(var(--mc-h) 60% 40%);background:hsl(var(--mc-h) 62% 50% / .12);border:1px solid hsl(var(--mc-h) 55% 50% / .35)}.msg.has-model{box-shadow:inset 3px 0 hsl(var(--mc-h) 55% 50%)}.msg.has-model .role{color:hsl(var(--mc-h) 52% 42%)}.chip.model-chip{color:hsl(var(--mc-h) 60% 40%);background:hsl(var(--mc-h) 62% 50% / .13)}@media(prefers-color-scheme:dark){.model-tag{color:hsl(var(--mc-h) 70% 70%);background:hsl(var(--mc-h) 60% 60% / .16);border-color:hsl(var(--mc-h) 55% 60% / .4)}.msg.has-model{box-shadow:inset 3px 0 hsl(var(--mc-h) 58% 60%)}.msg.has-model .role{color:hsl(var(--mc-h) 62% 66%)}.chip.model-chip{color:hsl(var(--mc-h) 70% 70%);background:hsl(var(--mc-h) 55% 55% / .18)}}