add statistics and plot

nonebot_plugin_memes_api/config.py

@@ -11,7 +11,8 @@ class MemeListImageConfig(BaseModel):
     text_template: str = "{keywords}"
     add_category_icon: bool = True
     label_new_timedelta: timedelta = timedelta(days=30)
-    label_hot_frequency: int = 24
+    label_hot_threshold: int = 21
+    label_hot_days: int = 7
 
 
 class Config(BaseModel):

nonebot_plugin_memes_api/manager.py

@@ -99,7 +99,6 @@ def search(
         meme_names = process.extract(
             meme_name, self.__meme_names.keys(), limit=limit, score_cutoff=score_cutoff
         )
-        logger.debug(meme_names)
         result: dict[str, MemeInfo] = {}
         for name, _, _ in meme_names:
             for meme in self.__meme_names[name]:

nonebot_plugin_memes_api/matchers/__init__.py

@@ -1 +1 @@
-from . import info, manage, search, help, command  # noqa
+from . import info, manage, search, help, command, statistics  # noqa

nonebot_plugin_memes_api/matchers/help.py

@@ -1,5 +1,5 @@
 import hashlib
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, timezone
 from itertools import chain
 
 from nonebot_plugin_alconna import Image, Text, on_alconna
@@ -47,18 +47,21 @@ async def _(user_id: UserId, session: EventSession):
         memes = sorted(memes, key=lambda meme: meme.date_modified, reverse=sort_reverse)
 
     label_new_timedelta = list_image_config.label_new_timedelta
-    label_hot_frequency = list_image_config.label_hot_frequency
+    label_hot_threshold = list_image_config.label_hot_threshold
+    label_hot_days = list_image_config.label_hot_days
 
     meme_generation_keys = await get_meme_generation_keys(
-        session, SessionIdType.GLOBAL, timedelta(days=1)
+        session,
+        SessionIdType.GLOBAL,
+        time_start=datetime.now(timezone.utc) - timedelta(days=label_hot_days),
     )
 
     meme_list: list[MemeKeyWithProperties] = []
     for meme in memes:
         labels = []
         if datetime.now() - meme.date_created < label_new_timedelta:
             labels.append("new")
-        if meme_generation_keys.count(meme.key) >= label_hot_frequency:
+        if meme_generation_keys.count(meme.key) >= label_hot_threshold:
             labels.append("hot")
         disabled = not meme_manager.check(user_id, meme.key)
         meme_list.append(

nonebot_plugin_memes_api/matchers/statistics.py

@@ -0,0 +1,218 @@
+from datetime import datetime, timedelta
+from typing import Any, Optional, Union
+
+from dateutil.relativedelta import relativedelta
+from nonebot.matcher import Matcher
+from nonebot_plugin_alconna import (
+    Alconna,
+    AlconnaQuery,
+    Args,
+    Option,
+    Query,
+    UniMessage,
+    on_alconna,
+    store_true,
+)
+from nonebot_plugin_session import EventSession, SessionIdType
+
+from ..plot import plot_duration_counts, plot_key_and_duration_counts
+from ..recorder import get_meme_generation_records, get_meme_generation_times
+from ..utils import add_timezone
+from .utils import find_meme
+
+statistics_matcher = on_alconna(
+    Alconna(
+        "表情调用统计",
+        Args["meme_name?", str],
+        Option("-g|--global", default=False, action=store_true, help_text="全局统计"),
+        Option("--my", default=False, action=store_true, help_text="我的"),
+        Option(
+            "-t|--type",
+            Args["type", ["day", "week", "month", "year", "24h", "7d", "30d", "1y"]],
+            help_text="统计类型",
+        ),
+    ),
+    aliases={"表情使用统计"},
+    block=True,
+    priority=11,
+    use_cmd_start=True,
+)
+
+
+def wrapper(
+    slot: Union[int, str], content: Optional[str], context: dict[str, Any]
+) -> str:
+    if slot == "my" and content:
+        return "--my"
+    elif slot == "global" and content:
+        return "--global"
+    elif slot == "type" and content:
+        if content in ["日", "24小时", "1天"]:
+            return "--type 24h"
+        elif content in ["本日", "今日"]:
+            return "--type day"
+        elif content in ["周", "一周", "7天"]:
+            return "--type 7d"
+        elif content in ["本周"]:
+            return "--type week"
+        elif content in ["月", "30天"]:
+            return "--type 30d"
+        elif content in ["本月", "月度"]:
+            return "--type month"
+        elif content in ["年", "一年"]:
+            return "--type 1y"
+        elif content in ["本年", "年度"]:
+            return "--type year"
+    return ""
+
+
+pattern_my = r"(?P<my>我的)"
+pattern_type = r"(?P<type>日|24小时|1天|本日|今日|周|一周|7天|本周|月|30天|本月|月度|年|一年|本年|年度)"  # noqa E501
+pattern_global = r"(?P<global>全局)"
+pattern_cmd = r"表情(?:调用|使用)统计"
+
+statistics_matcher.shortcut(
+    rf"{pattern_my}{pattern_cmd}",
+    prefix=True,
+    wrapper=wrapper,
+    arguments=["{my}"],
+).shortcut(
+    rf"{pattern_global}{pattern_cmd}",
+    prefix=True,
+    wrapper=wrapper,
+    arguments=["{global}"],
+).shortcut(
+    rf"{pattern_my}{pattern_global}{pattern_cmd}",
+    prefix=True,
+    wrapper=wrapper,
+    arguments=["{my}", "{global}"],
+).shortcut(
+    rf"{pattern_my}?{pattern_global}?{pattern_type}{pattern_cmd}",
+    prefix=True,
+    wrapper=wrapper,
+    arguments=["{my}", "{global}", "{type}"],
+)
+
+
+@statistics_matcher.handle()
+async def _(
+    matcher: Matcher,
+    session: EventSession,
+    meme_name: Optional[str] = None,
+    query_global: Query[bool] = AlconnaQuery("global.value", False),
+    query_my: Query[bool] = AlconnaQuery("my.value", False),
+    query_type: Query[str] = AlconnaQuery("type", "24h"),
+):
+    meme = await find_meme(matcher, meme_name) if meme_name else None
+
+    is_my = query_my.result
+    is_global = query_global.result
+    type = query_type.result
+
+    if is_my and is_global:
+        id_type = SessionIdType.USER
+    elif is_my:
+        id_type = SessionIdType.GROUP_USER
+    elif is_global:
+        id_type = SessionIdType.GLOBAL
+    else:
+        id_type = SessionIdType.GROUP
+
+    now = datetime.now().astimezone()
+    if type == "24h":
+        start = now - timedelta(days=1)
+        td = timedelta(hours=1)
+        fmt = "%H:%M"
+        humanized = "24小时"
+    elif type == "day":
+        start = now.replace(hour=0, minute=0, second=0, microsecond=0)
+        td = timedelta(hours=1)
+        fmt = "%H:%M"
+        humanized = "本日"
+    elif type == "7d":
+        start = now - timedelta(days=7)
+        td = timedelta(days=1)
+        fmt = "%m/%d"
+        humanized = "7天"
+    elif type == "week":
+        start = now.replace(hour=0, minute=0, second=0, microsecond=0) - timedelta(
+            days=now.weekday()
+        )
+        td = timedelta(days=1)
+        fmt = "%a"
+        humanized = "本周"
+    elif type == "30d":
+        start = now - timedelta(days=30)
+        td = timedelta(days=1)
+        fmt = "%m/%d"
+        humanized = "30天"
+    elif type == "month":
+        start = now.replace(day=1, hour=0, minute=0, second=0, microsecond=0)
+        td = timedelta(days=1)
+        fmt = "%m/%d"
+        humanized = "本月"
+    elif type == "1y":
+        start = now - relativedelta(years=1)
+        td = relativedelta(months=1)
+        fmt = "%y/%m"
+        humanized = "一年"
+    else:
+        start = now.replace(month=1, day=1, hour=0, minute=0, second=0, microsecond=0)
+        td = relativedelta(months=1)
+        fmt = "%b"
+        humanized = "本年"
+
+    if meme:
+        meme_times = await get_meme_generation_times(
+            session, id_type, meme_key=meme.key, time_start=start
+        )
+        meme_keys = [meme.key] * len(meme_times)
+    else:
+        meme_records = await get_meme_generation_records(
+            session, id_type, time_start=start
+        )
+        meme_times = [record.time for record in meme_records]
+        meme_keys = [record.meme_key for record in meme_records]
+
+    if not meme_times:
+        await matcher.finish("暂时没有表情调用记录")
+
+    meme_times = [add_timezone(time) for time in meme_times]
+    meme_times.sort()
+
+    def fmt_time(time: datetime) -> str:
+        if type in ["24h", "7d", "30d", "1y"]:
+            return (time + td).strftime(fmt)
+        return time.strftime(fmt)
+
+    duration_counts: dict[str, int] = {}
+    stop = start + td
+    count = 0
+    key = fmt_time(start)
+    for time in meme_times:
+        while time >= stop:
+            duration_counts[key] = count
+            key = fmt_time(stop)
+            stop += td
+            count = 0
+        count += 1
+    duration_counts[key] = count
+    while stop <= now:
+        key = fmt_time(stop)
+        stop += td
+        duration_counts[key] = 0
+
+    key_counts: dict[str, int] = {}
+    for key in meme_keys:
+        key_counts[key] = key_counts.get(key, 0) + 1
+
+    if meme:
+        title = (
+            f"表情“{meme.key}”{humanized}调用统计"
+            f"（总调用次数为 {key_counts.get(meme.key, 0)}）"
+        )
+        output = await plot_duration_counts(duration_counts, title)
+    else:
+        title = f"{humanized}表情调用统计"
+        output = await plot_key_and_duration_counts(key_counts, duration_counts, title)
+    await UniMessage.image(raw=output).send()

nonebot_plugin_memes_api/plot.py

@@ -0,0 +1,76 @@
+from io import BytesIO
+
+import matplotlib
+from matplotlib import pyplot as plt
+from matplotlib.axes import Axes
+from matplotlib.font_manager import fontManager
+from matplotlib.ticker import MaxNLocator
+from nonebot.utils import run_sync
+
+matplotlib.use("agg")
+fallback_fonts = [
+    "PingFang SC",
+    "Hiragino Sans GB",
+    "Microsoft YaHei",
+    "Source Han Sans SC",
+    "Noto Sans SC",
+    "Noto Sans CJK SC",
+    "WenQuanYi Micro Hei",
+]
+for fontfamily in fallback_fonts.copy():
+    try:
+        fontManager.findfont(fontfamily, fallback_to_default=False)
+    except ValueError:
+        fallback_fonts.remove(fontfamily)
+matplotlib.rcParams["font.family"] = fallback_fonts
+
+
+@run_sync
+def plot_key_and_duration_counts(
+    key_counts: dict[str, int], duration_counts: dict[str, int], title: str
+) -> BytesIO:
+    up_x = list(key_counts.keys())
+    up_y = list(key_counts.values())
+    low_x = list(duration_counts.keys())
+    low_y = list(duration_counts.values())
+    up_height = len(up_x) * 0.3
+    low_height = 3
+    fig_width = 8
+    fig, axs = plt.subplots(
+        nrows=2,
+        figsize=(fig_width, up_height + low_height),
+        height_ratios=[up_height, low_height],
+    )
+    up: Axes = axs[0]
+    up.barh(up_x, up_y, height=0.6)
+    up.xaxis.set_major_locator(MaxNLocator(integer=True))
+    low: Axes = axs[1]
+    low.plot(low_x, low_y, marker="o")
+    if len(low_x) > 24:
+        low.set_xticks(low_x[::3])
+    elif len(low_x) > 12:
+        low.set_xticks(low_x[::2])
+    low.yaxis.set_major_locator(MaxNLocator(integer=True))
+    fig.suptitle(title)
+    fig.tight_layout()
+    output = BytesIO()
+    fig.savefig(output)
+    return output
+
+
+@run_sync
+def plot_duration_counts(duration_counts: dict[str, int], title: str) -> BytesIO:
+    x = list(duration_counts.keys())
+    y = list(duration_counts.values())
+    fig, ax = plt.subplots(figsize=(6, 4))
+    ax.plot(x, y, marker="o")
+    if len(x) > 24:
+        ax.set_xticks(x[::3])
+    elif len(x) > 12:
+        ax.set_xticks(x[::2])
+    ax.yaxis.set_major_locator(MaxNLocator(integer=True))
+    fig.suptitle(title)
+    fig.tight_layout()
+    output = BytesIO()
+    fig.savefig(output)
+    return output