Spaces:
Running
Running
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 a specific font file, falling back to a common name if not found.""" | |
font_path = "simHei.ttc" | |
if not os.path.exists(font_path): | |
font_path = "SimHei.ttf" | |
return font_manager.FontProperties(fname=font_path, size=size) | |
def get_pinyin_abbr(text): | |
"""Gets the first letter of the Pinyin for the first two Chinese characters of a text.""" | |
if not text: | |
return "" | |
# Extract the first two Chinese characters | |
chars = [c for c in text if '\u4e00' <= c <= '\u9fff'] | |
chars = chars[:2] | |
# Get the first letter of the pinyin for each character | |
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 the movie schedule.""" | |
try: | |
# Attempt to read the date from a specific cell | |
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: | |
# Fallback to the current date if reading fails | |
date_str = datetime.today().strftime('%Y-%m-%d') | |
base_date = datetime.today().date() | |
try: | |
# Read the main schedule data | |
df = pd.read_excel(file, header=9, usecols=[1, 2, 4, 5]) | |
df.columns = ['Hall', 'StartTime', 'EndTime', 'Movie'] | |
# Clean and format the data | |
df['Hall'] = df['Hall'].ffill() | |
df.dropna(subset=['StartTime', 'EndTime', 'Movie'], inplace=True) | |
df['Hall'] = df['Hall'].astype(str).str.extract(r'(\d+号)') | |
# Convert time strings to datetime objects | |
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 | |
) | |
# Handle overnight screenings | |
df.loc[df['EndTime_dt'] < df['StartTime_dt'], 'EndTime_dt'] += timedelta(days=1) | |
df = df.sort_values(['Hall', 'StartTime_dt']) | |
# Merge consecutive screenings of the same movie | |
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'] # Extend the end time | |
else: | |
merged_rows.append(current) | |
current = row.copy() | |
if current is not None: | |
merged_rows.append(current) | |
merged_df = pd.DataFrame(merged_rows) | |
# Adjust times as per requirement | |
merged_df['StartTime_dt'] = merged_df['StartTime_dt'] - timedelta(minutes=10) | |
merged_df['EndTime_dt'] = merged_df['EndTime_dt'] - timedelta(minutes=5) | |
# Format times back to strings for display | |
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: | |
# Return None if any processing error occurs | |
return None, date_str | |
def create_print_layout(data, date_str): | |
"""Creates PNG and PDF layouts of the schedule, maximizing content size.""" | |
if data is None or data.empty: | |
return None | |
# --- Figure Setup --- | |
# A4 paper size in inches, 300 dpi for good quality | |
fig_width, fig_height = 8.27, 11.69 | |
dpi = 300 | |
# Create separate figures for PNG and PDF | |
png_fig = plt.figure(figsize=(fig_width, fig_height), dpi=dpi) | |
png_ax = png_fig.add_subplot(111) | |
png_ax.set_axis_off() | |
pdf_fig = plt.figure(figsize=(fig_width, fig_height), dpi=dpi) | |
pdf_ax = pdf_fig.add_subplot(111) | |
pdf_ax.set_axis_off() | |
# Set tight margins to maximize the content area on the page. | |
margin = 0.03 | |
png_fig.subplots_adjust(left=margin, right=1-margin, top=1-margin, bottom=margin) | |
pdf_fig.subplots_adjust(left=margin, right=1-margin, top=1-margin, bottom=margin) | |
# --- Drawing Logic --- | |
def process_figure(fig, ax): | |
# --- Dynamic Sizing --- | |
halls = sorted(data['Hall'].unique(), key=lambda h: int(h.replace('号','')) if h else 0) | |
# Calculate total lines needed: one for each movie + a gap between each hall's schedule | |
num_movie_lines = len(data) | |
num_hall_breaks = len(halls) - 1 | |
total_lines = num_movie_lines + num_hall_breaks | |
if total_lines == 0: return | |
# Calculate the height available for content based on margins | |
available_height = 1 - (2 * margin) | |
# Calculate the height for each line to perfectly fill the page | |
line_height = available_height / total_lines | |
# Dynamically set font size based on line height. | |
# This makes the text as large as possible while still fitting. | |
# (72 points per inch, 0.65 is a factor for vertical padding) | |
font_size_pt = line_height * fig_height * 72 * 0.65 | |
hall_font = get_font(font_size_pt) | |
movie_font = get_font(font_size_pt) | |
date_font = get_font(font_size_pt * 0.7) | |
# --- Content Drawing --- | |
ax.text(0.01, 0.99, date_str, | |
fontsize=date_font.get_size(), color='#A9A9A9', | |
ha='left', va='top', fontproperties=date_font, | |
transform=ax.transAxes, zorder=2) | |
y_position = 0.99 # Start drawing from the top | |
for hall in halls: | |
hall_data = data[data['Hall'] == hall].sort_values('StartTime_str') | |
y_block_top = y_position | |
hall_num_str = hall.replace("号", "") | |
hall_display_text = f"${hall_num_str}^{{\\#}}$" | |
# Center text vertically within its line height for better appearance | |
text_v_align_offset = line_height * 0.5 | |
for i, (_, row) in enumerate(hall_data.iterrows()): | |
if i == 0: | |
ax.text(0.05, y_position - text_v_align_offset, hall_display_text, | |
fontproperties=hall_font, fontweight='bold', | |
ha='center', va='center', transform=ax.transAxes, zorder=2) | |
pinyin_abbr = get_pinyin_abbr(row['Movie']) | |
movie_display_text = f"{i+1}. {pinyin_abbr} {row['Movie']}" | |
ax.text(0.15, y_position - text_v_align_offset, movie_display_text, | |
fontproperties=movie_font, ha='left', va='center', | |
transform=ax.transAxes, zorder=2, clip_on=True) | |
time_display_text = f"{row['StartTime_str']} - {row['EndTime_str']}" | |
ax.text(0.99, y_position - text_v_align_offset, time_display_text, | |
fontproperties=movie_font, ha='right', va='center', | |
transform=ax.transAxes, zorder=2) | |
y_position -= line_height | |
# Add a decorative rounded box around each hall's schedule | |
y_block_bottom = y_position + line_height | |
rect_height = y_block_top - y_block_bottom | |
rect = FancyBboxPatch((0, y_block_bottom), 1, rect_height, | |
boxstyle="round,pad=0.01,rounding_size=0.02", | |
edgecolor='gray', facecolor='white', | |
linewidth=0.8, zorder=1, transform=ax.transAxes) | |
ax.add_patch(rect) | |
if hall != halls[-1]: | |
y_position -= line_height # Add gap for the next block | |
# Render the layout on both figures | |
process_figure(png_fig, png_ax) | |
process_figure(pdf_fig, pdf_ax) | |
# --- File Saving --- | |
# Save PNG to a memory 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 memory 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 --- | |
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"]) | |
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: | |
output = create_print_layout(schedule, date_str) | |
# Use tabs for PDF and PNG previews | |
tab1, tab2 = st.tabs(["PDF 预览", "PNG 预览"]) | |
with tab1: | |
st.markdown(display_pdf(output['pdf']), unsafe_allow_html=True) | |
with tab2: | |
st.image(output['png'], use_container_width=True) | |
else: | |
st.error("无法处理文件,请检查文件格式或内容是否正确。") |