🧭 Update ROADMAP and add configurable logging; fix autochat syntax

- Refresh ROADMAP.md to reflect current Forgejo milestones and open feature issues\n- Add .env logging controls and update src/logger.py to honor them (console/file/error toggles, per-handler levels)\n- Fix indentation bug in src/autochat.py so emoji reactions work\n- Truncate and standardize logs; document logging usage in .env

.env

@@ -1,11 +1,31 @@
 DISCORD_TOKEN=MTM2OTc3NDY4OTYzNDg4MTU4Ng.G9Nrgz.akHoOO9SrXCDwiOCI3BUXfdR4bpSNb9zrVx9UI
 # This is using the TailScale IP
-OLLAMA_API=http://100.107.221.83:11434/api/
+OLLAMA_API=http://192.168.0.100:11434/
 MODEL_NAME=gemma3:12b
-CHANNEL_ID=1370420592360161393
+CHANNEL_ID=1380999713272238151
 SHOW_THINKING_BLOCKS=false
 DEBUG_MODE=true
 AUTOREPLY_ENABLED=true
+# ---------------------------
+# Logging configuration
+# - LOG_LEVEL: global base level (INFO recommended)
+# - LOG_CONSOLE: enable console logs (true/false)
+# - LOG_CONSOLE_LEVEL: level for console output (INFO, DEBUG, etc.)
+# - LOG_CONSOLE_TO_STDOUT: if true, console logs go to STDOUT (useful for container logging)
+# - LOG_TO_FILE: enable writing to a rotating file
+# - LOG_FILE_LEVEL: level stored in the file (set to DEBUG to capture full LLM payloads)
+# - LOG_FILE: filename for rotated logs
+# - LOG_ERROR_FILE: enable writing errors to a separate file (filename is derived)
+# Example: to capture full prompts/responses in logs, set LOG_FILE_LEVEL=DEBUG
+LOG_LEVEL=INFO
+LOG_CONSOLE=true
+LOG_CONSOLE_LEVEL=INFO
+LOG_CONSOLE_TO_STDOUT=true
+LOG_TO_FILE=true
+LOG_FILE_LEVEL=DEBUG
+LOG_FILE=bot.log
+LOG_ERROR_FILE=true
+# ---------------------------
 # Cooldown in seconds
 AUTOREPLY_COOLDOWN=0
 # Used for Codex to reach my repo on Forgejo

ROADMAP.md

@@ -1,5 +1,76 @@
 # 📍 DeltaBot Development Roadmap
 
+This roadmap is an actionable summary of current milestones and feature issues. It mirrors the milestones on the Forgejo repository and links to issues (local issue numbers).
+
+---
+
+## 🚀 Alpha Build Ready — Core features (Milestone)
+Focus: deliver core capabilities so Delta is useful and reliably persona-driven.
+
+Open high-priority items (Alpha):
+- #37 — 🧠 LoRA Support — improve model fine-tuning/load-time behavior
+- #36 — Memory — persistence for context beyond immediate messages
+- #26 — Web usage — optional web-enabled features (codex/integration)
+- #25 — 🔁 Enable Modelfile Support — support alternate model packaging
+- #24 — 💸 Set up Monetization — billing/paid features plumbing
+- #22 — 📡 Remote Admin Panel — admin web UI / remote control
+- #17 — 🖼️ Image Generation — generate images via local models
+- #16 — 👀 Image Interpretation — describe and analyze posted images
+- #10 — Post "Reply" — post as reply-to instead of plain message
+
+Closed/implemented Alpha items: #8, #9, #15, #30, #31
+
+Suggested next steps for Alpha:
+- Break large items (Remote Admin Panel, Image Generation) into sub-tasks.
+- Prioritize Memory (#36) and Post "Reply" (#10) to stabilize context handling.
+- Add clear acceptance criteria to each open Alpha issue.
+
+---
+
+## 🧪 Beta Build — Polishing & optional features (Milestone)
+Focus: polish, scaling, and user-experience improvements.
+
+Open Beta items:
+- #35 — Respect token budget (~1000 tokens max)
+- #34 — 📌 Pin system messages
+- #33 — 🧠 Add memory persistence (overlaps with Alpha Memory)
+- #27 — Multi model support — support local switching/multiple endpoints
+- #18 — 🎭 Multi Personality — multiple personas selectable per server/channel
+
+Suggested next steps for Beta:
+- Decide which Alpha items must land before Beta starts.
+- Resolve overlaps (Memory appears in both Alpha and Beta) and consolidate the plan.
+
+---
+
+## 📦 Backlog / Unmilestoned Features
+Lower-priority, exploratory, or undecided items to consider batching into future milestones.
+
+- #23 — 📢 Broadcast / Announcement Mode
+- #21 — 📈 Analytics Dashboard
+- #20 — Shopper Assist
+- #19 — 🚨 Content-aware moderation assist
+- #14 — 📊 Engagement-based adjustment
+- #13 — Context-aware scheduling
+- #12 — Bot can be given some power to change the server a bit
+- #11 — 🗓️ Scheduled specific times/dates
+
+Suggested backlog housekeeping:
+- Group these into thematic milestones: e.g. "Admin Tools", "Analytics", "Media & Vision".
+- Add rough estimates (S/M/L) and owners for each item.
+
+---
+
+## How to use this roadmap
+- Update issue bodies with acceptance criteria and subtask checklists.
+- Assign owners and estimate effort for each Alpha item.
+- Use the Forgejo milestone due dates to prioritize development sprints.
+
+---
+
+_Generated: Sep 19, 2025 — synchronized with repository milestones and open feature issues._
+# 📍 DeltaBot Development Roadmap
+
 This roadmap tracks the major phases of DeltaBot — from MVP to full AI companion chaos. ✅ = complete
 
 ---

ROADMAP.md.bak

@@ -0,0 +1 @@
+Backup of previous ROADMAP.md

src/autochat.py

@@ -89,7 +89,7 @@ async def maybe_react_to_message(message, persona):
 
     roll = random.random()
     if roll > EMOJI_REACTION_CHANCE:
-        logger.info(f"🎲 Reaction skipped (chance {EMOJI_REACTION_CHANCE:.2f}, roll {roll:.2f})")
+        logger.debug("🎲 Reaction skipped (chance %.2f, roll %.2f)", EMOJI_REACTION_CHANCE, roll)
         return
 
     try:
@@ -108,10 +108,11 @@ async def maybe_react_to_message(message, persona):
         )
 
         emoji_reply = get_ai_response(prompt).strip()
-        logger.info(f"🎭 Emoji suggestion from LLM: {emoji_reply}")
+
+        # Log the raw emoji suggestion at DEBUG
+        logger.debug("🎭 Emoji suggestion from LLM: %s", emoji_reply)
 
         # Extract valid emojis
-        import re
         emojis = re.findall(r'[\U0001F300-\U0001F6FF\U0001F900-\U0001F9FF\U0001F1E0-\U0001F1FF]', emoji_reply)
         unique_emojis = list(dict.fromkeys(emojis))[:3]
 

src/logger.py

@@ -1,22 +1,163 @@
 import logging
 import os
+import sys
+from logging.handlers import RotatingFileHandler
 
-def setup_logger(name: str = "bot", level=logging.INFO, log_file: str = "bot.log"):
-    formatter = logging.Formatter(
+
+def setup_logger(name: str = "bot"):
+    """Create a logger with rotating file handler, separate error log, and
+    a concise console output. Behavior is controlled by environment vars:
+      - LOG_LEVEL (default INFO)
+      - LOG_FILE (default bot.log)
+      - LOG_MAX_BYTES (default 5_000_000)
+      - LOG_BACKUP_COUNT (default 5)
+
+    If `colorlog` is installed, console output will be colorized.
+    """
+
+    # Config from environment (all values can be set in .env)
+    level_name = os.getenv("LOG_LEVEL", "INFO").upper()
+    try:
+        base_level = getattr(logging, level_name)
+    except Exception:
+        base_level = logging.INFO
+
+    # Handlers toggle + per-handler levels
+    console_enabled = os.getenv("LOG_CONSOLE", "true").lower() in ("1", "true", "yes")
+    console_level_name = os.getenv("LOG_CONSOLE_LEVEL", level_name).upper()
+    console_to_stdout = os.getenv("LOG_CONSOLE_TO_STDOUT", "true").lower() in ("1", "true", "yes")
+
+    file_enabled = os.getenv("LOG_TO_FILE", "true").lower() in ("1", "true", "yes")
+    file_level_name = os.getenv("LOG_FILE_LEVEL", "DEBUG").upper()
+    log_file = os.getenv("LOG_FILE", "bot.log")
+    max_bytes = int(os.getenv("LOG_MAX_BYTES", 5_000_000))
+    backup_count = int(os.getenv("LOG_BACKUP_COUNT", 5))
+
+    error_file_enabled = os.getenv("LOG_ERROR_FILE", "true").lower() in ("1", "true", "yes")
+
+    # File formatter: include module and line number for easier debugging
+    file_formatter = logging.Formatter(
+        "[%(asctime)s] [%(levelname)s] [%(name)s:%(lineno)d] %(message)s",
+        "%Y-%m-%d %H:%M:%S",
+    )
+
+    # Console formatter: shorter. Try to use colorlog if available.
+    try:
+        import colorlog
+
+        console_formatter = colorlog.ColoredFormatter(
+            "%(log_color)s[%(asctime)s] [%(levelname)s]%(reset)s %(message)s",
+            datefmt="%Y-%m-%d %H:%M:%S",
+            log_colors={
+                "DEBUG": "cyan",
+                "INFO": "green",
+                "WARNING": "yellow",
+                "ERROR": "red",
+                "CRITICAL": "red",
+            },
+        )
+    except Exception:
+        console_formatter = logging.Formatter(
             "[%(asctime)s] [%(levelname)s] %(message)s", "%Y-%m-%d %H:%M:%S"
         )
 
-    stream_handler = logging.StreamHandler()
-    stream_handler.setFormatter(formatter)
+    # Handlers
+    # Console handler (stdout by default, configurable)
+    console_handler = logging.StreamHandler(sys.stdout if console_to_stdout else sys.stderr)
+    console_handler.setFormatter(console_formatter)
+    try:
+        console_level = getattr(logging, console_level_name)
+    except Exception:
+        console_level = base_level
+    console_handler.setLevel(console_level)
 
-    file_handler = logging.FileHandler(log_file, encoding='utf-8')
-    file_handler.setFormatter(formatter)
+    file_handler = RotatingFileHandler(
+        log_file, maxBytes=max_bytes, backupCount=backup_count, encoding="utf-8"
+    )
+    file_handler.setFormatter(file_formatter)
+    try:
+        file_level = getattr(logging, file_level_name)
+    except Exception:
+        file_level = logging.DEBUG
+    file_handler.setLevel(file_level)
+
+    # Separate error-only rotating file
+    error_log_file = os.path.splitext(log_file)[0] + ".error.log"
+    error_handler = RotatingFileHandler(
+        error_log_file, maxBytes=max_bytes, backupCount=backup_count, encoding="utf-8"
+    )
+    error_handler.setLevel(logging.ERROR)
+    error_handler.setFormatter(file_formatter)
 
     logger = logging.getLogger(name)
-    logger.setLevel(level)
 
+    # Set logger base level to the most permissive of configured levels so handlers can filter
+    levels = [base_level]
+    if console_enabled:
+        levels.append(console_level)
+    if file_enabled:
+        levels.append(file_level)
+    # minimum numeric level means more verbose (DEBUG=10)
+    logger.setLevel(min(levels))
+
+    # Avoid adding duplicate handlers if logger already configured
     if not logger.handlers:
-        logger.addHandler(stream_handler)
+        if console_enabled:
+            logger.addHandler(console_handler)
+        if file_enabled:
             logger.addHandler(file_handler)
+        if error_file_enabled:
+            logger.addHandler(error_handler)
 
     return logger
+
+
+def generate_req_id(prefix: str = "r") -> str:
+    """Generate a short request id for correlating logs."""
+    import uuid
+
+    return f"{prefix}{uuid.uuid4().hex[:8]}"
+
+
+def mask_secret(value: str | None, head: int = 6, tail: int = 4) -> str | None:
+    """Mask a secret for safe logging."""
+    if not value:
+        return value
+    if len(value) <= head + tail + 3:
+        return "***"
+    return f"{value[:head]}...{value[-tail:]}"
+
+
+def log_llm_request(logger: logging.Logger, req_id: str, model: str, user: str | None, context_len: int | None):
+    logger.info("%s LLM request start model=%s user=%s context_len=%s", req_id, model, user or "-", context_len or 0)
+
+
+def log_llm_payload(logger: logging.Logger, req_id: str, payload: object):
+    logger.debug("%s LLM full payload: %s", req_id, payload)
+
+
+def log_llm_response(logger: logging.Logger, req_id: str, model: str, duration_s: float, short_text: str | None, raw: object | None = None):
+    logger.info("%s LLM response model=%s duration=%.3fs summary=%s", req_id, model, duration_s, (short_text or "[no text]").replace("\n", " ")[:160])
+    if raw is not None:
+        logger.debug("%s LLM raw response: %s", req_id, raw)
+
+
+class SamplingFilter(logging.Filter):
+    """Filter that randomly allows a fraction of records through.
+
+    Useful for noisy logs like 'reaction skipped'.
+    """
+
+    def __init__(self, sample_rate: float = 0.1):
+        super().__init__()
+        try:
+            self.rate = float(sample_rate)
+        except Exception:
+            self.rate = 0.1
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        import random
+
+        if self.rate >= 1.0:
+            return True
+        return random.random() < self.rate