ticket #93 attempt 1

devplacepy/database/schema.py

@@ -30,26 +30,6 @@ def migrate_bug_tables_to_issue_tables() -> None:
         logger.info("Dropped table %s after migration", source_name)
 
 
-def _add_message_unique_index() -> None:
-    if "messages" not in db.tables:
-        return
-    with db:
-        db.query(
-            """
-            DELETE FROM messages WHERE id NOT IN (
-                SELECT MIN(id) FROM messages GROUP BY sender_uid, receiver_uid, content
-            )
-            """
-        )
-    _index(
-        db,
-        "messages",
-        "idx_messages_unique_sender_receiver_content",
-        ["sender_uid", "receiver_uid", "content"],
-        unique=True,
-    )
-
-
 def init_db():
     tables = db.tables
     _index(db, "users", "idx_users_username", ["username"])
@@ -151,7 +131,6 @@ def init_db():
         "idx_messages_conversation_rev",
         ["receiver_uid", "sender_uid"],
     )
-    _add_message_unique_index()
     _index(db, "notifications", "idx_notifications_user", ["user_uid"])
     _index(db, "notifications", "idx_notifications_user_read", ["user_uid", "read"])
     _index(db, "push_registration", "idx_push_registration_user", ["user_uid"])

devplacepy/main.py

@@ -12,7 +12,6 @@ from fastapi import FastAPI, Request
 from fastapi.responses import HTMLResponse, RedirectResponse
 from fastapi.staticfiles import StaticFiles
 from fastapi.exceptions import RequestValidationError
-from starlette.middleware.gzip import GZipMiddleware
 from devplacepy.config import (
     STATIC_DIR,
     STATIC_VERSION,
@@ -610,10 +609,6 @@ async def response_timing(request: Request, call_next):
     response.headers["X-Response-Time"] = f"{(time.perf_counter() - start) * 1000:.1f}ms"
     return response
 
-
-app.add_middleware(GZipMiddleware, minimum_size=512, compresslevel=5)
-
-
 _home_cache = TTLCache(ttl=int(os.environ.get("DEVPLACE_HOME_CACHE_TTL", "60")), max_size=4)
 
 

devplacepy/services/correction.py

@@ -88,7 +88,7 @@ def gateway_complete(
             {"role": "system", "content": system},
             {"role": "user", "content": text},
         ],
-        "temperature": 0.0,
+        "temperature": 0.1,
     }
     headers = {
         "Content-Type": "application/json",

devplacepy/services/messaging/persist.py

@@ -1,9 +1,6 @@
 # retoor <retoor@molodetz.nl>
 
-import hashlib
 import logging
-import time
-from collections import OrderedDict
 from datetime import datetime, timezone
 from typing import Any, Optional
 
@@ -18,16 +15,12 @@ from devplacepy.utils import (
     track_action,
 )
 from devplacepy.services.audit import record as audit
-from sqlalchemy.exc import IntegrityError
-
 from devplacepy.services.correction import schedule_correction
 from devplacepy.services.ai_modifier import schedule_modification
 
 logger = logging.getLogger("messaging.persist")
 
 MAX_CONTENT_LENGTH = 2000
-DEDUP_WINDOW_SECONDS = 3
-_content_cache: dict[str, tuple[float, str]] = OrderedDict()
 
 
 def _slim_attachment(attachment: dict[str, Any]) -> dict[str, Any]:
@@ -88,60 +81,19 @@ def persist_message(
 
     sender_uid = sender["uid"]
     sender_username = sender.get("username", "")
-
-    content_hash = hashlib.sha256(
-        f"{sender_uid}:{receiver_uid}:{content}".encode()
-    ).hexdigest()[:16]
-    now = time.time()
-    last_seen, cached_uid = _content_cache.get(content_hash, (0.0, None))
-    if now - last_seen < DEDUP_WINDOW_SECONDS and cached_uid is not None:
-        logger.debug(
-            "Dedup hit for message hash %s (original uid %s)", content_hash, cached_uid
-        )
-        cached = get_table("messages").find_one(uid=cached_uid)
-        if cached:
-            return {
-                "uid": cached["uid"],
-                "sender_uid": cached["sender_uid"],
-                "receiver_uid": cached["receiver_uid"],
-                "content": cached["content"],
-                "read": cached.get("read", False),
-                "created_at": cached["created_at"],
-            }
-
     messages_table = get_table("messages")
     msg_uid = generate_uid()
     created_at = datetime.now(timezone.utc).isoformat()
-
-    try:
-        messages_table.insert(
-            {
-                "uid": msg_uid,
-                "sender_uid": sender_uid,
-                "receiver_uid": receiver_uid,
-                "content": content,
-                "read": False,
-                "created_at": created_at,
-            }
-        )
-    except IntegrityError:
-        existing = messages_table.find_one(
-            sender_uid=sender_uid, receiver_uid=receiver_uid, content=content
-        )
-        if not existing:
-            raise
-        logger.debug(
-            "Dedup via unique constraint for message (uid %s)", existing["uid"]
-        )
-        _content_cache[content_hash] = (time.time(), existing["uid"])
-        return {
-            "uid": existing["uid"],
-            "sender_uid": existing["sender_uid"],
-            "receiver_uid": existing["receiver_uid"],
-            "content": existing["content"],
-            "read": existing.get("read", False),
-            "created_at": existing["created_at"],
+    messages_table.insert(
+        {
+            "uid": msg_uid,
+            "sender_uid": sender_uid,
+            "receiver_uid": receiver_uid,
+            "content": content,
+            "read": False,
+            "created_at": created_at,
         }
+    )
 
     link_attachments(attachment_uids, "message", msg_uid)
     schedule_correction(sender, "messages", msg_uid, request)
@@ -162,8 +114,6 @@ def persist_message(
     )
     track_action(sender_uid, "message")
 
-    _content_cache[content_hash] = (time.time(), msg_uid)
-
     logger.info(
         "Message %s sent from %s to %s via %s",
         msg_uid,

devplacepy/templates/docs/architecture-backend.html

@@ -11,7 +11,7 @@ How a request flows through middleware to a router, how the database layer is bu
 
 ## Middleware
 
-Seven HTTP middlewares run as a stack around every request, listed outermost first. `response_timing` is the outermost, `refresh_db_snapshot` the innermost, and a `GZipMiddleware` (responses over 512 bytes) wraps the whole stack on top:
+Six HTTP middlewares run as a stack around every request, listed outermost first. `response_timing` is the outermost and `refresh_db_snapshot` the innermost. Compression is handled by nginx in production; the Python application does not compress responses itself.
 
 | Middleware (outermost first) | Responsibility |
 |------------|----------------|

pyproject.toml

@@ -4,7 +4,8 @@ version = "1.0.0"
 description = "DevPlace - The Developer Social Network"
 requires-python = ">=3.12"
 dependencies = [
-    "fastapi",
+    "fastapi>=0.110.0",
+    "starlette>=0.37.0",
     "uvicorn[standard]",
     "jinja2",
     "python-multipart",

tests/api/messages/send.py

@@ -166,23 +166,3 @@ def test_send_attachment_only_empty_content_succeeds(seeded_db):
     refresh_snapshot()
     row = get_table("messages").find_one(uid=msg["uid"])
     assert row["content"] == ""
-
-
-def test_duplicate_message_returns_same_uid(seeded_db):
-    s, _ = _member()
-    receiver = _db_user("bob_test")["uid"]
-    content = _unique("dupmsg")
-
-    first = s.post(
-        f"{BASE_URL}/messages/send",
-        headers=JSON_audit_log,
-        data={"content": content, "receiver_uid": receiver},
-    ).json()["data"]
-
-    second = s.post(
-        f"{BASE_URL}/messages/send",
-        headers=JSON_audit_log,
-        data={"content": content, "receiver_uid": receiver},
-    ).json()["data"]
-
-    assert first["uid"] == second["uid"], "duplicate messages should return the same uid"