Python解析Excel图表Chart的信息实战指南

import xml.etree.ElementTree as ET
from zipfile import ZipFile
import io
import urllib.request

def get_chat_info(direct_link):
    result = {}
    res = {}

    try:
        # 下载 Excel 文件
        file = urllib.request.urlopen(direct_link).read()
        archive = ZipFile(io.BytesIO(file))

        try:
            # 读取图表 XML
            data = archive.read('xl/charts/chart1.xml')
            res['code'] = 200
            res['msg'] = "获取图表信息成功"

            tree = ET.parse(io.BytesIO(data))
            root = tree.getroot()

            # 命名空间
            ns = {
                'c': 'http://schemas.openxmlformats.org/drawingml/2006/chart',
                'a': 'http://schemas.openxmlformats.org/drawingml/2006/main'
            }

            # ================== 图表标题 ==================
            title_element = root.find('.//c:title/c:tx/c:rich', ns)
            if title_element is not None:
                title_text = ""
                for t in title_element.iter('{http://schemas.openxmlformats.org/drawingml/2006/main}t'):
                    title_text += t.text
                result['title'] = title_text

            title_ax_element = root.find('.//c:chart/c:title/c:tx', ns)
            if title_ax_element is not None:
                tx_pr = title_ax_element.find('.//a:defRPr', ns)
                if tx_pr is not None:
                    latin = tx_pr.find('.//a:latin', ns)
                    result['title_font'] = latin.get('typeface') if latin is not None else "no"
                    result['title_size'] = tx_pr.get('sz') or "no"

            # ================== 系列名称 ==================
            ser_elements = root.findall('.//c:chart/c:plotArea/*/c:ser', ns)
            series_name = ""
            for ser in ser_elements:
                v = ser.find('.//c:v', ns)
                if v is not None:
                    series_name = v.text
            result['series_name'] = series_name

            # ================== X / Y 数据 ==================
            x_values, y_values = [], []

            for num_ref in root.findall('.//c:numRef', ns):
                for v in num_ref.findall('.//c:v', ns):
                    y_values.append(v.text)

            for pt in root.findall('.//c:cat/c:strRef/c:strCache/c:pt', ns):
                v = pt.find('.//c:v', ns)
                if v is not None:
                    x_values.append(v.text)

            result['x_values'] = x_values
            result['y_values'] = y_values

            # ================== 数值轴 ==================
            val_ax = root.find('.//c:valAx', ns)
            if val_ax is not None:
                tx_pr = val_ax.find('.//c:txPr/a:p/a:pPr/a:defRPr', ns)
                if tx_pr is not None:
                    latin = tx_pr.find('.//a:latin', ns)
                    result['valAx_font'] = latin.get('typeface') if latin is not None else "no"
                    result['valAx_size'] = tx_pr.get('sz') or "no"

            # ================== 分类轴 ==================
            cat_ax = root.find('.//c:catAx', ns)
            if cat_ax is not None:
                tx_pr = cat_ax.find('.//c:txPr/a:p/a:pPr/a:defRPr', ns)
                if tx_pr is not None:
                    latin = tx_pr.find('.//a:latin', ns)
                    result['catAx_font'] = latin.get('typeface') if latin is not None else "no"
                    result['catAx_size'] = tx_pr.get('sz') or "no"

            # ================== 系列字体 ==================
            ser = root.find('.//c:chart/c:plotArea/*/c:ser', ns)
            if ser is not None:
                tx_pr = ser.find('.//a:defRPr', ns)
                if tx_pr is not None:
                    latin = tx_pr.find('.//a:latin', ns)
                    result['ser_font'] = latin.get('typeface') if latin is not None else "no"
                    result['ser_size'] = tx_pr.get('sz') or "no"

            res['data'] = result

        except:
            res['code'] = 404
            res['msg'] = "未找到图表信息"

    except:
        res['code'] = 500
        res['msg'] = "未获取excel信息"

    return res

aa = get_chat_info("http://192.168.31.161:8080/555.xlsx")
print(aa)

{
  "code": 200,
  "msg": "获取图表信息成功",
  "data": {
    "title": "各季度采购合理性折线图",
    "title_font": "宋体",
    "title_size": "1200",
    "series_name": "采购合理性",
    "x_values": [
      "2018-1", "2018-2", "2018-3", "2018-4",
      "Jan-19", "2019-2", "2019-3", "2019-4",
      "2020-1", "2020-2", "2020-3", "2020-4",
      "2021-1", "2021-2", "2021-3", "2021-4",
      "2022-1", "2022-2", "2022-3", "2022-4"
    ],
    "y_values": [
      "0.99", "0.92", "0.91", "0.37", "0.85",
      "0.97", "0.8", "0.88", "0.67", "0.91",
      "0.76", "0.75", "0.99", "0.95", "0.89",
      "0.83", "0.44", "0.75", "0.94", "0.41"
    ],
    "valAx_font": "宋体",
    "valAx_size": "1000",
    "catAx_font": "宋体",
    "catAx_size": "1000",
    "ser_font": "宋体",
    "ser_size": "1000"
  }
}