import re
import gradio as gr
import pandas as pd

def parse_text(txt_file):
    # 将上传的文件名作为输入参数
    file_name = txt_file.name
    filename_with_dot = file_name.rsplit('/', 1)[1]
    print(filename_with_dot)
    filename = filename_with_dot.split('.', 1)[0]
    print(filename)
    with open(file_name, "r") as f:
        # 读取上传文件中的内容
        content = f.read()
        # 在这里进行文本解析
        lines = content.split("\n")
        numbered_lines = [(i + 1, line) for i, line in enumerate(lines)]

        # 将解析结果转换为字符串输出
        output_str = ""
        for num, line in numbered_lines:
            output_str += f"{num}. {line}\n"
        filename = filename.split('.', 0)[0]
        return filename, output_str

def process_text(text):
    try:
        sentences = re.split(r'\d+\.', text)
    except Exception:
        try:
            sentences = re.split(r'\d+', text)
        except Exception:
            pass
    sentences = [s.strip() for s in sentences if s.strip()]
    return sentences

def pd_form(sentences):
    rows = []
    for i in range(len(sentences)):
            row_number = i + 1
            description = sentences[i]
            row = {
                "镜号": row_number,
                "画面范例": '',
                "时长": '',
                "景别": '',
                "画面说明": description,
                "歌词": '',
                "备注": '',
            }
            rows.append(row)
    return rows


def to_excel(t_file):
    text_file, name_file = parse_text(t_file)[1], parse_text(t_file)[0]
    output_file_name = name_file + '.xlsx'
    txt = process_text(text_file)
    rows = pd_form(txt)
    writer = pd.ExcelWriter(output_file_name, engine='xlsxwriter')
    df = pd.DataFrame(rows)
    df.to_excel(writer, index=False, sheet_name="Sheet1")
    writer.close()
    return output_file_name


iface = gr.Interface(
    fn=to_excel,
    inputs=gr.inputs.File(label="请上传 txt 格式文件"),
    outputs=gr.File(label="下载 xlsx 格式文档"),
    title="txt - xlsx 分镜格式转换器",
    description="说明： 如果转换不成功，请重试几次")

iface.launch()