#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import os
import sys
import json
import pdfplumber
import pandas
from pprint import pprint


class invoice():
    """ 发票 """

    def _parse_pdfs(self, pdfFullpath):
        """ 分析 """
        info = {}
        with pdfplumber.open(pdfFullpath) as pdf:
            page = pdf.pages[0]

            inf = self._extrace_from_words(page.extract_words())
            info.update(inf)

            inf = self._extrace_from_table(page.extract_tables()[0])
            info.update(inf)

        return info

    def _extrace_from_words(self, words):
        """ 从单词中提取 """
        info = {}

        lines = {}
        for word in words:
            top = int(word['top'])
            bottom = int(word['bottom'])
            pos = (top + bottom) // 2
            text = word['text']
            if pos not in lines:
                lines[pos] = [text]
            else:
                lines[pos].append(text)

        lines_pack = []
        last_pos = None
        for pos in sorted(lines):
            arr = lines[pos]

            if len(lines_pack) > 0 and pos - last_pos <= 10:
                lines_pack[-1] += arr
                continue

            lines_pack.append(arr)
            last_pos = pos
            # continue

        for pack in lines_pack:
            for idx, line in enumerate(pack):
                if '电子普通发票' in line:
                    info['标题'] = line
                    continue

                if '发票代码:' in line:
                    info['发票代码'] = line.split(':')[1]
                    continue

                if '发票号码:' in line:
                    info['发票号码'] = line.split(':')[1]
                    continue

                if '开票日期:' in line:
                    year = line.split(':')[1]
                    month = [ln for ln in pack if ln.isdigit()][0]
                    day = [ln[:2] for ln in pack if '日' in ln][0]
                    info['开票日期'] = f'{year}-{month}-{day}'
                    continue

                if '机器编号:' in line:
                    info['机器编号'] = [ln for ln in pack if ln.isdigit()
                                        and len(ln) > 10][0]
                    continue

                if '码:' in line:
                    c1 = pack[idx].split(':')[1]
                    c2 = pack[idx + 1]
                    c3 = pack[idx + 2]
                    c4 = pack[idx + 3]
                    info['校验码'] = f'{c1} {c2} {c3} {c4}'
                    continue

                if '收款人:' in line:
                    info['收款人'] = line.split(':')[1]
                    continue

                if '开票人:' in line:
                    info['开票人'] = line.split(':')[1]
                    continue

        return info

    def _extrace_from_table(self, table):
        """ 从表中提取 """
        info = {}
        if len(table) != 4:
            return None

        # 购买方
        for cell in table[0]:
            if not cell:
                continue

            lines = cell.splitlines()
            for line in lines:
                if '名        称:' in line:
                    info['购买方名称'] = line.split(':')[1]
                    continue

                if len(line) == 18 and line.isalnum():
                    info['购买方税号'] = line
                    continue

                if len(line) == 27:
                    if '密码' not in info:
                        info['密码'] = []
                    info['密码'].append(line)
                    continue

        # 详细
        for cell in table[1]:
            if not cell:
                continue

            lines = cell.splitlines()
            for line in lines:
                if '货物或应税劳务、服务名称' in line:
                    info['商品'] = lines[1:-1]
                    break

                if '金  额' in line:
                    info['总金额'] = lines[-1][1:]
                    break

                if '税  额' in line:
                    info['总税额'] = lines[-1][1:]
                    break

        # 合计
        for cell in table[2]:
            if not cell:
                continue

            lines = cell.splitlines()
            for line in lines:
                if '¥' in line:
                    info['总计'] = line[1:]

        # 销售方
        for cell in table[3]:
            if not cell:
                continue

            lines = cell.splitlines()
            for line in lines:
                if '名        称:' in line:
                    info['销售方名称'] = line.split(':')[1]
                    continue

                if len(line) == 18 and line.isalnum():
                    info['销售方税号'] = line
                    continue

        return info