Python之基于十六进制判断文件类型

核心代码：

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Author  : suk
import struct
from io import BytesIO

# 支持文件类型
# 用16进制字符串的目的是可以知道文件头是多少字节
# 各种文件头的长度不一样，少则2字符，长则8字符
def typeList(types):
    type_dict = {'jpg': ['FFD8FFE000104A464946'],
                 'png': ['89504E470D0A1A0A0000'],
                 'gif': ['47494638396126026F01'],
                 'tif': ['49492A00227105008037'],
                 'bmp': ['424D8E1B030000000000'],
                 'dwg': [''],
                 'html': ['3C21444F435459504520'],
                 'htm': ['3C21646F637479706520'],
                 'css': ['48544D4C207B0D0A0942'],
                 'js': ['696B2E71623D696B2E71'],
                 'rtf': ['7B5C727466315C616E73'],
                 'psd': [''],
                 'eml': ['46726F6D3A203D3F6762'],
                 'wps': ['D0CF11E0A1B11AE10000'],
                 'mdb': ['5374616E64617264204A'],
                 'ps': '[252150532D41646F6265]',
                 'pdf': ['255044462D312E'],
                 'rmvb': ['2E524D46000000120001'],
                 'flv': ['464C5601050000000900'],
                 'mp4': ['00000020667479706D70'],
                 'mp3': [''],
                 'mpg': ['000001BA210001000180'],
                 'wmv': ['3026B2758E66CF11A6D9'],
                 'wav': ['52494646E27807005741'],
                 'avi': ['52494646D07D60074156'],
                 'mid': ['4D546864000000060001'],
                 'zip': ['504B0304140000000800', '504B0304140000080800', '504B03040A0000080000'],
                 'rar': ['526172211A0700CF9073'],
                 'ini': ['235468697320636F6E66'],
                 'jar': ['504B03040A0000000000'],
                 'exe': ['4D5A9000030000000400'],
                 'jsp': ['3C25402070616765206C'],
                 'mf': ['4D616E69666573742D56'],
                 'xml': ['3C3F786D6C2076657273'],
                 'sql': ['494E5345525420494E54'],
                 'java': ['7061636B616765207765'],
                 'bat': ['406563686F206F66660D'],
                 'gz': ['1F8B0800000000000000'],
                 'properties': ['6C6F67346A2E726F6F74'],
                 'class': ['CAFEBABE0000002E0041'],
                 'chm': [''],
                 'mxp': [''],
                 'docx': ['504B0304140006000800', '504B03040A0000000000'],
                 'torrent': ['6431303A637265617465'],
                 'mov': ['6D6F6F76'],
                 'wpd': ['FF575043'],
                 'dbx': ['CFAD12FEC5FD746F'],
                 'pst': ['2142444E'],
                 'qdf': ['AC9EBD8F'],
                 'pwl': ['E3828596'],
                 'ram': ['2E7261FD']
                 }
    ret = {}
    for k_hex, v_prefix in type_dict.items():
        if k_hex in types:
            ret[k_hex] = v_prefix
    return ret

# 字节码转16进制字符串
def bytes2hex(bytes):
    num = len(bytes)
    hexstr = u""
    for i in range(num):
        t = u"%x" % bytes[i]
        if len(t) % 2:
            hexstr += u""
        hexstr += t
    return hexstr.upper()

# 获取文件类型
def file_type(filename):
    binfile = open(filename, 'rb')  # 必需二制字读取
    tl = typeList(types=["jpg", "zip", "docx"])
    ftype = None
    for type_name, hcode_list in tl.items():
        flag = False
        for hcode in hcode_list:
            numOfBytes = int(len(hcode) / 2)  # 需要读多少字节
            binfile.seek(0)  # 每次读取都要回到文件头，不然会一直往后读取
            hbytes = struct.unpack_from("B" * numOfBytes, binfile.read(numOfBytes))  # 一个 "B"表示一个字节
            f_hcode = bytes2hex(hbytes)  # 如果判断不出来，打印出这个值，往字典增加即可
            # print("上传数据流hex", s_hcode, '=', "代码字典hex", hcode)  # 如果判断不出来，打印出这个值，往字典增加即可
            if f_hcode == hcode:
                flag = True
                break
        if flag:
            ftype = type_name
            break
    binfile.close()
    return ftype

# 获取字节流类型
def stream_type(stream, types):
    """
    :param stream:流数据
    :param types:需要判断文件类型,格式:["jpg","jpn"]
    :return:
    """
    tl = typeList(types=types)
    ftype = None
    for type_name, hcode_list in tl.items():
        flag = False
        for hcode in hcode_list:
            numOfBytes = int(len(hcode) / 2)  # 需要读多少字节
            hbytes = struct.unpack_from("B" * numOfBytes, stream[0:numOfBytes])  # 一个 "B"表示一个字节
            s_hcode = bytes2hex(hbytes)
            # print("上传数据流hex", s_hcode, '=', "代码字典hex", hcode)  # 如果判断不出来，打印出这个值，往字典增加即可
            if s_hcode == hcode:
                flag = True
                break
        if flag:
            ftype = type_name
            break
    return ftype

def stream_split(stream, count=3):
    """
    主要处理流是分段获取的数据
    :param stream: 块流
    :param count: 取多少段合成来判断类型,默认三段
    :return:
    """
    block_stream = BytesIO()
    temp = 1
    for block in stream:
        block_stream.write(block)
        if temp == count:
            break
        temp += 1
    return block_stream.getvalue()

is_file_type.py

type_dict字典，根据自己上传的文件，来填写，数据来自互联网。

---

基于Flask的上传示例

@index.route('/upload', methods=['GET', 'POST'])
def upload():

    if request.method == 'GET':
        return render_template('upload.html')

    upload_obj = request.files.get('code_file')

    if not upload_obj:
        return '没有选择文件上传'

    ret = stream_type(stream_split(upload_obj.stream), ["jpg", "png", "pdf"])

    if not ret:
        return '上传失败，文件类型不匹配，类型必须 "jpg" or "png" or "pdf"'

    file_name = upload_obj.filename
    upload_obj.save(os.path.join('files', file_name))

    return '上传文件成功'

upload.html

{% extends 'layout.html' %}
{% block content %}
<h1>上传代码</h1>
<form action="" method="post" enctype="multipart/form-data">
    <input type="file" name="code_file">
    <input type="submit" value="上传"></input>
</form>
{% endblock %}

---

开始上传文件：

上传不在列表中的文件类型

---

上传在列表中的文件类型