app.py

import pandas as pd
import streamlit as st
import matplotlib.pyplot as plt
import matplotlib.font_manager as font_manager
import io
import base64
import os
from datetime import datetime, timedelta
import math
from matplotlib.patches import FancyBboxPatch
from pypinyin import lazy_pinyin, Style
from matplotlib.backends.backend_pdf import PdfPages

def get_font(size=14):
    """Loads the SimHei font for Chinese character support."""
    font_path = "simHei.ttc"
    if not os.path.exists(font_path):
        # Fallback for different OS or font naming
        font_path = "SimHei.ttf"
    if not os.path.exists(font_path):
        # If font is still not found, return None and let matplotlib use its default
        st.warning(f"Font file not found at {font_path}. Chinese characters may not display correctly.")
        return None
    return font_manager.FontProperties(fname=font_path, size=size)

def get_pinyin_abbr(text):
    """获取文本前两个汉字的拼音首字母"""
    if not text:
        return ""
    # 提取前两个汉字
    chars = [c for c in text if '\u4e00' <= c <= '\u9fff']
    if len(chars) < 2:
        chars = chars + [''] * (2 - len(chars))
    else:
        chars = chars[:2]
    # 获取拼音首字母
    pinyin_list = lazy_pinyin(chars, style=Style.FIRST_LETTER)
    return ''.join(pinyin_list).upper()

def process_schedule(file):
    """Processes the uploaded Excel file to extract and clean schedule data."""
    try:
        date_df = pd.read_excel(file, header=None, skiprows=7, nrows=1, usecols=[3])
        date_str = pd.to_datetime(date_df.iloc[0, 0]).strftime('%Y-%m-%d')
        base_date = pd.to_datetime(date_str).date()
    except Exception:
        date_str = datetime.today().strftime('%Y-%m-%d')
        base_date = datetime.today().date()
    
    try:
        df = pd.read_excel(file, header=9, usecols=[1, 2, 4, 5])
        df.columns = ['Hall', 'StartTime', 'EndTime', 'Movie']
        df['Hall'] = df['Hall'].ffill()
        df.dropna(subset=['StartTime', 'EndTime', 'Movie'], inplace=True)
        df['Hall'] = df['Hall'].astype(str).str.extract(r'(\d+号)')
        df['StartTime_dt'] = pd.to_datetime(df['StartTime'], format='%H:%M', errors='coerce').apply(
            lambda t: t.replace(year=base_date.year, month=base_date.month, day=base_date.day) if pd.notnull(t) else t
        )
        df['EndTime_dt'] = pd.to_datetime(df['EndTime'], format='%H:%M', errors='coerce').apply(
            lambda t: t.replace(year=base_date.year, month=base_date.month, day=base_date.day) if pd.notnull(t) else t
        )
        df.loc[df['EndTime_dt'] < df['StartTime_dt'], 'EndTime_dt'] += timedelta(days=1)
        df = df.sort_values(['Hall', 'StartTime_dt'])
        
        merged_rows = []
        for hall, group in df.groupby('Hall'):
            group = group.sort_values('StartTime_dt')
            current = None
            for _, row in group.iterrows():
                if current is None:
                    current = row.copy()
                else:
                    if row['Movie'] == current['Movie']:
                        current['EndTime_dt'] = row['EndTime_dt']
                    else:
                        merged_rows.append(current)
                        current = row.copy()
            if current is not None:
                merged_rows.append(current)
        
        merged_df = pd.DataFrame(merged_rows)
        
        # 将开始时间统一提前10分钟，结束时间统一提前5分钟
        merged_df['StartTime_dt'] = merged_df['StartTime_dt'] - timedelta(minutes=10)
        merged_df['EndTime_dt'] = merged_df['EndTime_dt'] - timedelta(minutes=5)
        
        merged_df['StartTime_str'] = merged_df['StartTime_dt'].dt.strftime('%H:%M')
        merged_df['EndTime_str'] = merged_df['EndTime_dt'].dt.strftime('%H:%M')
        
        return merged_df[['Hall', 'Movie', 'StartTime_str', 'EndTime_str']], date_str
    except Exception as e:
        st.error(f"Error processing schedule data: {e}")
        return None, date_str

def create_print_layout(data, date_str):
    """Generates the print-friendly PDF and PNG layouts from the schedule data."""
    if data is None or data.empty:
        return None
    
    # Create figures for PNG and PDF output with A4 dimensions
    png_fig = plt.figure(figsize=(8.27, 11.69), dpi=300)
    png_ax = png_fig.add_subplot(111)
    png_ax.set_axis_off()
    png_fig.subplots_adjust(left=0.02, right=0.98, top=0.98, bottom=0.02)
    
    pdf_fig = plt.figure(figsize=(8.27, 11.69), dpi=300)
    pdf_ax = pdf_fig.add_subplot(111)
    pdf_ax.set_axis_off()
    pdf_fig.subplots_adjust(left=0.02, right=0.98, top=0.98, bottom=0.02)
    
    def process_figure(fig, ax, is_pdf=False):
        """The core drawing function to render the schedule onto a matplotlib axis."""
        ax.set_ylim(0, 1) # Use a 0-1 coordinate system for consistent positioning

        # Sort halls numerically
        halls = sorted(data['Hall'].unique(), key=lambda h: int(h.replace('号', '')) if h and h.replace('号', '').isdigit() else 0)
        num_halls = len(halls)
        num_movie_rows = len(data)
        
        # 1. Calculate total vertical slots required to fill the page
        num_separators = num_halls - 1 if num_halls > 1 else 0
        total_slots = num_movie_rows + num_separators + 2  # movies + separators + top/bottom padding
        
        if total_slots == 0: return

        # 2. Calculate the height of each slot
        slot_height = 1.0 / total_slots
        
        # 3. Calculate font size to be 90% of the slot height
        figure_height_inches = 11.69
        font_size_points = (slot_height * figure_height_inches) * 0.9 * 72 # (height in inches) * 90% * 72 points/inch
        movie_font_size = font_size_points
        hall_font_size = movie_font_size * 0.8
        
        date_font = get_font(12)
        hall_font = get_font(hall_font_size)
        movie_font = get_font(movie_font_size)
        
        # Draw the date at the top
        ax.text(0.00, 1.00, date_str, fontsize=12, color='#A9A9A9',
                ha='left', va='top', fontproperties=date_font, transform=ax.transAxes, zorder=3)

        # 4. Initialize Y position, starting below the top padding slot
        y_position = 1.0 - slot_height

        # 5. Loop through each hall and its movies to draw the schedule
        for i, hall in enumerate(halls):
            hall_data = data[data['Hall'] == hall]
            movie_count = 1
            
            for _, row in hall_data.iterrows():
                # Draw Hall Number (only for the first movie of the hall)
                if movie_count == 1:
                    hall_num = hall.replace("号", "")
                    hall_text = f"${hall_num}^{{\\#}}$"  # Use LaTeX for superscript '#'
                    ax.text(0.03, y_position, hall_text,
                            fontsize=hall_font_size, fontweight='bold',
                            ha='left', va='top', fontproperties=hall_font,
                            transform=ax.transAxes, zorder=2)

                # Draw Pinyin Abbreviation and Movie Title
                pinyin_abbr = get_pinyin_abbr(row['Movie'])
                ax.text(0.20, y_position, f"{movie_count}. {pinyin_abbr} {row['Movie']}",
                        fontsize=movie_font_size, ha='left', va='top', fontproperties=movie_font,
                        transform=ax.transAxes, zorder=2, clip_on=True)

                # Draw Time Information
                ax.text(0.95, y_position, f"{row['StartTime_str']} - {row['EndTime_str']}",
                        fontsize=movie_font_size, ha='right', va='top', fontproperties=movie_font,
                        transform=ax.transAxes, zorder=2)
                
                y_position -= slot_height # Move down for the next movie
                movie_count += 1
            
            # After a hall's schedule, draw a separator line (if not the last hall)
            if i < num_halls - 1:
                # The line is drawn in the middle of the dedicated separator slot
                line_y = y_position - (slot_height / 2)
                ax.plot([0.03, 0.97], [line_y, line_y], color='black', linewidth=0.8, transform=ax.transAxes, zorder=1)
                
                y_position -= slot_height # Move down past the separator slot

    # Process and render the layout for both figure objects
    process_figure(png_fig, png_ax)
    process_figure(pdf_fig, pdf_ax, is_pdf=True)
    
    # Save PNG to a buffer
    png_buffer = io.BytesIO()
    png_fig.savefig(png_buffer, format='png', bbox_inches='tight', pad_inches=0.05)
    png_buffer.seek(0)
    image_base64 = base64.b64encode(png_buffer.getvalue()).decode()
    plt.close(png_fig)
    
    # Save PDF to a buffer
    pdf_buffer = io.BytesIO()
    with PdfPages(pdf_buffer) as pdf:
        pdf.savefig(pdf_fig, bbox_inches='tight', pad_inches=0.05)
    pdf_buffer.seek(0)
    pdf_base64 = base64.b64encode(pdf_buffer.getvalue()).decode()
    plt.close(pdf_fig)
    
    return {
        'png': f"data:image/png;base64,{image_base64}",
        'pdf': f"data:application/pdf;base64,{pdf_base64}"
    }

def display_pdf(base64_pdf):
    """Generates the HTML to embed and display a PDF in Streamlit."""
    pdf_display = f"""
    <iframe src="{base64_pdf}" width="100%" height="800" type="application/pdf"></iframe>
    """
    return pdf_display

# --- Streamlit App Main ---
st.set_page_config(page_title="LED 屏幕时间表打印", layout="wide")
st.title("LED 屏幕时间表打印")

uploaded_file = st.file_uploader("选择打开【放映时间核对表.xls】文件", accept_multiple_files=False, type=["xls", "xlsx"])

if uploaded_file:
    with st.spinner("文件正在处理中，请稍候..."):
        schedule, date_str = process_schedule(uploaded_file)
        if schedule is not None and not schedule.empty:
            output = create_print_layout(schedule, date_str)
            
            st.success("处理完成！请在下方预览和下载。")
            
            # Create tabs to switch between PDF and PNG views
            tab1, tab2 = st.tabs(["PDF 预览 (推荐打印)", "PNG 预览"])
            
            with tab1:
                st.markdown(display_pdf(output['pdf']), unsafe_allow_html=True)
                st.download_button(
                    label="下载 PDF 文件",
                    data=base64.b64decode(output['pdf'].split(',')[1]),
                    file_name=f"Schedule_{date_str}.pdf",
                    mime="application/pdf"
                )
            
            with tab2:
                st.image(output['png'], use_container_width=True)
                st.download_button(
                    label="下载 PNG 图像",
                    data=base64.b64decode(output['png'].split(',')[1]),
                    file_name=f"Schedule_{date_str}.png",
                    mime="image/png"
                )
        else:
            st.error("无法处理文件，或文件中没有找到有效排期。请检查文件格式或内容是否正确。")