opencode-db 0.6.1__tar.gz → 0.7.0__tar.gz

{opencode_db-0.6.1 → opencode_db-0.7.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: opencode-db
-Version: 0.6.1
+Version: 0.7.0
 Summary: OpenCode database CLI manager — browse, export, analyze, clean up sessions
 Project-URL: Homepage, https://github.com/VasilevNStas/opencode-db
 Project-URL: Source, https://github.com/VasilevNStas/opencode-db
@@ -8,6 +8,10 @@ Project-URL: Tracker, https://github.com/VasilevNStas/opencode-db/issues
 Requires-Python: >=3.12
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Provides-Extra: rich
+Requires-Dist: rich>=13.0; extra == "rich"
+Provides-Extra: all
+Requires-Dist: opencode-db[rich]; extra == "all"
 Dynamic: license-file
 
 # opencode-db
@@ -25,12 +29,29 @@ pip install opencode-db
 
 ```bash
 opencode-db list         # recent sessions
+opencode-db view         # interactive session viewer (colored, scrollable)
 opencode-db stats        # database summary
 opencode-db costs --total  # total token costs
 opencode-db export       # export dialog to .md (interactive)
 opencode-db help         # full reference
 ```
 
+## Features
+
+- **`view`** — просмотр сессии в терминале с ANSI-цветами, Markdown-рендерингом и прокруткой через `less`
+- **`list`** / **`info`** / **`search`** / **`tree`** — навигация и поиск по сессиям
+- **`export`** — экспорт диалога в Markdown (с поддержкой Obsidian)
+- **`delete`** / **`prune`** — удаление и массовая очистка с фильтрами
+- **`costs`** / **`stats`** — аналитика токенов и расходов
+- **`--db-path`** / **`OPENCODE_DB`** — кастомный путь к БД
+
+## Optional extras
+
+```bash
+pip install opencode-db[rich]   # enhanced Markdown rendering (recommended)
+pip install opencode-db         # zero external dependencies
+```
+
 ## Requirements
 
 - Python 3.12+

opencode_db-0.7.0/README_PYPI.md

@@ -0,0 +1,43 @@
+# opencode-db
+
+[![Python](https://img.shields.io/badge/python-3.12%2B-blue)](https://www.python.org/downloads/)
+[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)
+
+**CLI for OpenCode's local database.** OpenCode saves every agent session, conversation, token usage, and cost in a local SQLite file. This tool lets you inspect, export, analyze, and clean up that data.
+
+```bash
+pip install opencode-db
+```
+
+## Quick start
+
+```bash
+opencode-db list         # recent sessions
+opencode-db view         # interactive session viewer (colored, scrollable)
+opencode-db stats        # database summary
+opencode-db costs --total  # total token costs
+opencode-db export       # export dialog to .md (interactive)
+opencode-db help         # full reference
+```
+
+## Features
+
+- **`view`** — просмотр сессии в терминале с ANSI-цветами, Markdown-рендерингом и прокруткой через `less`
+- **`list`** / **`info`** / **`search`** / **`tree`** — навигация и поиск по сессиям
+- **`export`** — экспорт диалога в Markdown (с поддержкой Obsidian)
+- **`delete`** / **`prune`** — удаление и массовая очистка с фильтрами
+- **`costs`** / **`stats`** — аналитика токенов и расходов
+- **`--db-path`** / **`OPENCODE_DB`** — кастомный путь к БД
+
+## Optional extras
+
+```bash
+pip install opencode-db[rich]   # enhanced Markdown rendering (recommended)
+pip install opencode-db         # zero external dependencies
+```
+
+## Requirements
+
+- Python 3.12+
+- [OpenCode](https://opencode.ai) (must have been run at least once)
+- macOS / Linux

opencode_db-0.7.0/cmd_view.py

@@ -0,0 +1,363 @@
+"""Command view: просмотр сессии в терминале с прокруткой.
+
+Выводит отформатированный диалог с ANSI-цветами,
+автоматически использует $PAGER (less -R) для прокрутки.
+
+Режимы выбора сессии:
+  - view <session_id>          — по ID (полный или префикс)
+  - view --latest              — самая свежая сессия
+  - view (без аргументов)      — интерактивный выбор из списка
+"""
+
+import argparse
+import os
+import shutil
+import subprocess
+import sys
+
+from db import (
+    SessionError,
+    get_latest_session,
+    get_messages,
+    get_project_name,
+    get_recent_sessions,
+    get_session_info,
+    get_session_title,
+    parse_model,
+)
+from i18n import _
+from markdown_to_ansi import render as md_render
+from utils import build_help_epilog, format_cost, format_tokens, format_ts
+
+_VIEW_EXAMPLES = [
+    ("<session_id>", "help.view.e0"),
+    ("--latest", "help.view.e1"),
+    ("", "help.view.e2"),
+    ("--no-pager", "help.view.e3"),
+    ("--raw", "help.view.e4"),
+]
+
+
+def register(subparsers) -> None:
+    p = subparsers.add_parser(
+        "view",
+        help=_("help.cmd.view"),
+        formatter_class=argparse.RawDescriptionHelpFormatter,
+        epilog=build_help_epilog("view", _VIEW_EXAMPLES),
+    )
+    p.add_argument(
+        "session_id",
+        nargs="?",
+        help="Session ID (optional — interactive mode if omitted)",
+    )
+    p.add_argument("--latest", "-l", action="store_true", help=_("view.flag.latest"))
+    p.add_argument("--no-pager", "-P", action="store_true", help=_("view.flag.no_pager"))
+    p.add_argument("--raw", "-R", action="store_true", help=_("view.flag.raw"))
+
+
+def run(args, db) -> int:
+    if args.session_id:
+        return _view_by_id(db, args.session_id, args.no_pager, args.raw)
+    if args.latest:
+        return _view_latest(db, args.no_pager, args.raw)
+    return _view_interactive(db, args.no_pager, args.raw)
+
+
+# ======================================================================
+# Selection modes
+# ======================================================================
+
+
+def _view_by_id(db, session_id, no_pager, raw) -> int:
+    try:
+        info = get_session_info(db, session_id)
+    except SessionError as e:
+        print(e.message)
+        return 1
+    return _do_view(db, info, no_pager, raw)
+
+
+def _view_latest(db, no_pager, raw) -> int:
+    try:
+        session_id = get_latest_session(db)
+        info = get_session_info(db, session_id)
+    except SessionError as e:
+        print(e.message)
+        return 1
+    title = get_session_title(info)
+    print(_("session.latest", title=title))
+    return _do_view(db, info, no_pager, raw)
+
+
+def _view_interactive(db, no_pager, raw) -> int:
+    sessions = get_recent_sessions(db)
+    if not sessions:
+        print(_("view.no_sessions"))
+        return 1
+
+    print()
+    print(f"  {'─' * 60}")
+    print(f"  {_('view.interactive_header')}")
+    print(f"  {'─' * 60}")
+
+    for i, s in enumerate(sessions, 1):
+        title = get_session_title(s)
+        created = format_ts(s["time_created"], "%Y-%m-%d %H:%M")
+        cost = format_cost(s["cost"])
+        print(f"  {i:2d}. {s['id'][:24]}  {created}  {title[:40]:40s}  {cost}")
+
+    print(f"  {'─' * 60}")
+
+    while True:
+        try:
+            choice = input("  " + _("view.interactive_prompt")).strip()
+        except (EOFError, KeyboardInterrupt):
+            print()
+            return 1
+
+        if not choice:
+            print(_("view.interactive_abort"))
+            return 1
+
+        try:
+            idx = int(choice) - 1
+            if 0 <= idx < len(sessions):
+                break
+        except ValueError:
+            pass
+
+        print(f"  {_('view.interactive_error', n=len(sessions))}")
+
+    session_id = sessions[idx]["id"]
+    title = get_session_title(sessions[idx])
+    print(_("view.interactive_viewing", title=title))
+    try:
+        info = get_session_info(db, session_id)
+    except SessionError as e:
+        print(e.message)
+        return 1
+    return _do_view(db, info, no_pager, raw)
+
+
+# ======================================================================
+# Display engine
+# ======================================================================
+
+
+def _init_styles(raw=False):
+    """Return style dict. Empty dict = no ANSI codes."""
+    if raw or not sys.stdout.isatty():
+        return {}
+    return {
+        "R": "\033[0m",
+        "B": "\033[1m",
+        "D": "\033[2m",
+        "G": "\033[32m",
+        "C": "\033[36m",
+        "Y": "\033[33m",
+        "M": "\033[90m",
+    }
+
+
+def _do_view(db, info, no_pager, raw) -> int:
+    sty = _init_styles(raw)
+    messages = get_messages(db, info["id"])
+    text = _format_session(db, info, messages, sty)
+
+    use_pager = not no_pager and sys.stdout.isatty() and len(text) > 1000
+    if use_pager:
+        _page_text(text)
+    else:
+        sys.stdout.write(text)
+        if not text.endswith("\n"):
+            sys.stdout.write("\n")
+        sys.stdout.flush()
+
+    return 0
+
+
+def _format_session(db, info, messages, sty) -> str:
+    """Build the full formatted session view string."""
+    title = get_session_title(info)
+    model = parse_model(info)
+    created = format_ts(info["time_created"])
+    msg_count = len(messages)
+
+    duration = "—"
+    if info["time_updated"] and info["time_created"]:
+        duration_sec = (info["time_updated"] - info["time_created"]) / 1000
+        if duration_sec < 60:
+            duration = _("info.dur_sec", n=int(duration_sec))
+        elif duration_sec < 3600:
+            duration = _("info.dur_min", n=int(duration_sec / 60))
+        else:
+            hours = int(duration_sec / 3600)
+            mins = int((duration_sec % 3600) / 60)
+            duration = _("info.dur_hours", h=hours, m=mins)
+
+    ti = info["tokens_input"]
+    to = info["tokens_output"]
+    tr = info["tokens_reasoning"]
+    tokens_str = f"{format_tokens(ti)} in · {format_tokens(to)} out"
+    if tr:
+        tokens_str += f" · {format_tokens(tr)} reason"
+
+    cost = format_cost(info["cost"])
+
+    project_name = None
+    if info["project_id"]:
+        project_name = get_project_name(db, info["project_id"])
+
+    sep = f"{sty.get('M', '')}{'─' * 60}{sty.get('R', '')}"
+    lr = sty.get("R", "")
+
+    lines = []
+    lines.append("")
+    lines.append(f"{sty.get('B', '')}{sty.get('C', '')}🔍 {_('view.header')}{lr}")
+    lines.append(sep)
+    lines.append(f"  {_('info.id')}:       {info['id']}")
+    lines.append(f"  {_('list.header.title')}: {title}")
+    lines.append(f"  {_('info.model')}:   {model}")
+    lines.append(f"  {_('info.agent')}:    {info['agent'] or '—'}")
+    lines.append(f"  {_('info.created')}:  {created}")
+    lines.append(f"  {_('info.duration')}: {duration}")
+    lines.append(f"  {_('info.tokens_total')}: {tokens_str}")
+    lines.append(f"  {_('info.cost')}:       {cost}")
+    if project_name:
+        lines.append(f"  {_('info.project')}:   {project_name}")
+    lines.append(f"  {_('info.messages')}: {msg_count}")
+    lines.append(sep)
+    lines.append("")
+
+    if not messages:
+        lines.append(f"  {sty.get('M', '')}{_('view.no_messages')}{lr}")
+        lines.append("")
+        lines.append(sep)
+        lines.append("")
+    else:
+        for msg in messages:
+            role = msg.get("role", "unknown")
+            ts = format_ts(msg.get("time"), "%Y-%m-%d %H:%M:%S") if msg.get("time") else ""
+            agent = msg.get("agent", "")
+
+            if role == "user":
+                role_str = f"{sty.get('G', '')}👤 {_('view.role_user')} [{ts}]{lr}"
+            elif role == "assistant":
+                agent_tag = f" ({agent})" if agent else ""
+                role_str = f"{sty.get('C', '')}🤖 {_('view.role_assistant')}{agent_tag} [{ts}]{lr}"
+            else:
+                role_str = f"{sty.get('Y', '')}⚙️ {role} [{ts}]{lr}"
+
+            lines.append(sep)
+            lines.append(role_str)
+            lines.append(sep)
+
+            for i, part in enumerate(msg.get("parts", [])):
+                formatted = _format_part_ansi(part, sty)
+                if formatted:
+                    if i > 0:
+                        lines.append("")
+                    lines.append(formatted)
+
+    lines.append(f"{sty.get('M', '')}─── {_('view.footer')} ───{lr}")
+    lines.append("")
+
+    return "\n".join(lines)
+
+
+# ======================================================================
+# Part formatting (ANSI version of format_part_to_md)
+# ======================================================================
+
+
+def _format_part_ansi(part, sty) -> str:
+    """Форматирует часть сообщения с ANSI-цветами."""
+    t = part.get("type", "")
+
+    if t == "text":
+        text = part.get("text", "").strip()
+        if not text:
+            return ""
+        return md_render(text, sty, plain=not sty)
+
+    if t == "reasoning":
+        text = part.get("text", "").strip()
+        if not text:
+            return ""
+        title = f"{sty.get('Y', '')}{sty.get('D', '')}💭 {_('md.reasoning')}:{sty.get('R', '')}"
+        body = f"{sty.get('D', '')}{text}{sty.get('R', '')}"
+        return f"{title}\n{body}"
+
+    if t == "sound":
+        return f"{sty.get('Y', '')}*🔊 {_('md.sound')}*{sty.get('R', '')}"
+
+    if t == "tool":
+        tool_name = part.get("tool", "?")
+        t_input = part.get("input", {})
+        t_output = part.get("output", "")
+        t_status = part.get("status", "")
+
+        icon = {"completed": "✅", "error": "❌", "running": "⏳"}.get(t_status, "🔧")
+        y = sty.get("Y", "")
+        r = sty.get("R", "")
+
+        if tool_name == "read":
+            fp = t_input.get("filePath", "")
+            return f"{y}{icon} read:{r} `{fp}`"
+        if tool_name == "write":
+            fp = t_input.get("filePath", "")
+            return f"{y}{icon} write:{r} `{fp}`"
+        if tool_name == "edit":
+            fp = t_input.get("filePath", "")
+            return f"{y}{icon} edit:{r} `{fp}`"
+        if tool_name == "glob":
+            pat = t_input.get("pattern", "")
+            pth = t_input.get("path", "")
+            if pth:
+                return f"{y}{icon} glob:{r} `{pat}` in `{pth}`"
+            return f"{y}{icon} glob:{r} `{pat}`"
+        if tool_name == "grep":
+            pat = t_input.get("pattern", "")
+            return f"{y}{icon} grep:{r} `{pat}`"
+
+        lines = [f"{y}{icon} {tool_name}{r}"]
+        if t_input:
+            for k, v in t_input.items():
+                s = str(v)
+                if len(s) > 200:
+                    s = s[:200] + "..."
+                lines.append(f"  {k}: {s}")
+        if t_output:
+            out = str(t_output).strip()
+            if out:
+                if len(out) > 300:
+                    out = out[:300] + "..."
+                lines.append(f"  → {out}")
+        return "\n".join(lines)
+
+    return ""
+
+
+# ======================================================================
+# Pager
+# ======================================================================
+
+
+def _page_text(text: str) -> None:
+    """Pipe text through $PAGER or less -R."""
+    pager_cmd = os.environ.get("PAGER", "")
+    try:
+        if pager_cmd:
+            subprocess.run(pager_cmd, input=text, text=True, shell=True)
+        elif shutil.which("less"):
+            subprocess.run(["less", "-R"], input=text, text=True)
+        else:
+            sys.stdout.write(text)
+            if not text.endswith("\n"):
+                sys.stdout.write("\n")
+            sys.stdout.flush()
+    except OSError:
+        sys.stdout.write(text)
+        if not text.endswith("\n"):
+            sys.stdout.write("\n")
+        sys.stdout.flush()

{opencode_db-0.6.1 → opencode_db-0.7.0}/commands/__init__.py

@@ -23,6 +23,7 @@ import cmd_stats
 import cmd_todos
 import cmd_tree
 import cmd_vacuum
+import cmd_view
 
 # Словарь {имя_команды: модуль}
 # Модуль должен иметь register(subparsers) и run(args, db)
@@ -39,6 +40,7 @@ COMMANDS = {
     "stats": cmd_stats,
     "todos": cmd_todos,
     "vacuum": cmd_vacuum,
+    "view": cmd_view,
     "help": cmd_help,
 }
 

{opencode_db-0.6.1 → opencode_db-0.7.0}/formatters.py

@@ -1,43 +1,79 @@
 """Форматирование вывода: таблицы, JSON, Markdown."""
 
+from __future__ import annotations
+
 import json
 import os
+from typing import Any
 
 from config import IGNORE_DIRS
 from i18n import _
 
 
-def print_json(data) -> None:
+def print_json(data: Any) -> None:
     """Выводит данные в JSON-формате."""
     print(json.dumps(data, ensure_ascii=False, indent=2, default=str))
 
 
-def print_table_simple(headers, rows, sep="  ") -> None:
-    """Выводит таблицу с авто-шириной колонок."""
+def print_table_simple(
+    headers: list[str],
+    rows: list[list],
+    sep: str = "  ",
+) -> None:
+    """Выводит таблицу с рамкой и колонками.
+
+    Формат:
+      ┌──────┬───────────┐
+      │ ID   │ Title     │
+      ├──────┼───────────┤
+      │ abc  │ Session 1 │
+      └──────┴───────────┘
+    """
     if not rows:
         print(_("empty"))
         return
 
+    ncols = len(headers)
+    pad = 1  # padding inside each cell
+
+    # compute column widths
     widths = []
-    for ci in range(len(headers)):
+    for ci in range(ncols):
         max_w = len(headers[ci])
         for row in rows:
             max_w = max(max_w, len(str(row[ci])))
-        widths.append(min(max_w + 1, 80))
-
-    line = sep.join(h.ljust(widths[i])[: widths[i]] for i, h in enumerate(headers))
-    sep_line = "─" * len(line)
+        widths.append(min(max_w + pad * 2, 80))
+
+    tl = "┌"  # top-left
+    tm = "┬"  # top-middle
+    tr = "┐"  # top-right
+    ml = "├"  # middle-left
+    mm = "┼"  # middle-middle
+    mr = "┤"  # middle-right
+    bl = "└"  # bottom-left
+    bm = "┴"  # bottom-middle
+    br = "┘"  # bottom-right
+    ve = "│"  # vertical edge
+
+    top = tl + tm.join("─" * w for w in widths) + tr
+    sep = ml + mm.join("─" * w for w in widths) + mr
+    bot = bl + bm.join("─" * w for w in widths) + br
+
+    header_cells = [h.center(widths[ci])[: widths[ci]] for ci, h in enumerate(headers)]
+    header_line = ve + ve.join(header_cells) + ve
 
     print()
-    print(line)
-    print(sep_line)
+    print(top)
+    print(header_line)
+    print(sep)
     for row in rows:
-        line = sep.join(str(row[i]).ljust(widths[i])[: widths[i]] for i in range(len(headers)))
-        print(line)
+        cells = [str(row[ci]).ljust(widths[ci])[: widths[ci]] for ci in range(ncols)]
+        print(ve + ve.join(cells) + ve)
+    print(bot)
     print(_("fmt.rows_count", n=len(rows)))
 
 
-def summarize_val(v, max_len=120) -> str:
+def summarize_val(v: Any, max_len: int = 120) -> str:
     """Обрезает длинное значение."""
     s = str(v)
     if len(s) > max_len:
@@ -45,7 +81,7 @@ def summarize_val(v, max_len=120) -> str:
     return s
 
 
-def format_part_to_md(part, full=False):
+def format_part_to_md(part: dict, full: bool = False) -> str:
     """Преобразует часть сообщения в строку Markdown.
 
     Args:
@@ -149,7 +185,7 @@ def format_part_to_md(part, full=False):
     return ""
 
 
-def collect_snapshot(project_dir):
+def collect_snapshot(project_dir: str) -> dict[str, Any]:
     """Собирает статистику по проекту для log.md.
 
     Returns:
@@ -182,7 +218,14 @@ def collect_snapshot(project_dir):
     return snapshot
 
 
-def update_log(output_dir, title, note, dialog_filename, snapshot, now_str) -> None:
+def update_log(
+    output_dir: str,
+    title: str,
+    note: str | None,
+    dialog_filename: str | None,
+    snapshot: dict[str, Any],
+    now_str: str,
+) -> None:
     """Создаёт или дополняет log.md в директории проекта."""
     log_path = os.path.join(output_dir, "log.md")
     heading = f"{title}: {note}" if note else title