Python mimetypes: File Type Detection

import mimetypes
 
# From filename
mime_type, encoding = mimetypes.guess_type('document.pdf')
print(mime_type)   # application/pdf
print(encoding)    # None
 
# Compressed files
mime_type, encoding = mimetypes.guess_type('data.txt.gz')
print(mime_type)   # text/plain
print(encoding)    # gzip

import mimetypes
 
files = ['image.jpg', 'script.js', 'style.css', 'data.json', 
         'page.html', 'doc.docx', 'video.mp4', 'song.mp3']
 
for filename in files:
    mime, _ = mimetypes.guess_type(filename)
    print(f"{filename}: {mime}")
 
# image.jpg: image/jpeg
# script.js: text/javascript
# style.css: text/css
# data.json: application/json
# page.html: text/html
# doc.docx: application/vnd.openxmlformats-officedocument.wordprocessingml.document
# video.mp4: video/mp4
# song.mp3: audio/mpeg

import mimetypes
 
# Single extension
ext = mimetypes.guess_extension('image/jpeg')
print(ext)  # .jpg
 
ext = mimetypes.guess_extension('text/html')
print(ext)  # .html
 
ext = mimetypes.guess_extension('application/json')
print(ext)  # .json

import mimetypes
 
# Get all known extensions for a type
exts = mimetypes.guess_all_extensions('image/jpeg')
print(exts)  # ['.jpe', '.jpeg', '.jpg']
 
exts = mimetypes.guess_all_extensions('text/plain')
print(exts)  # ['.txt', '.c', '.h', ...]

import mimetypes
 
# Add custom type
mimetypes.add_type('application/x-custom', '.custom')
 
# Now it works
mime, _ = mimetypes.guess_type('file.custom')
print(mime)  # application/x-custom

import mimetypes
 
# Initialize with system files (usually done automatically)
mimetypes.init()
 
# Add custom files
mimetypes.init(files=['/etc/mime.types', 'my-types.txt'])

import mimetypes
 
def get_type_category(filename):
    """Get broad category of file type."""
    mime, _ = mimetypes.guess_type(filename)
    
    if mime is None:
        return 'unknown'
    
    main_type = mime.split('/')[0]
    return main_type
 
files = ['photo.png', 'song.mp3', 'movie.mp4', 'doc.txt', 'app.exe']
for f in files:
    print(f"{f}: {get_type_category(f)}")
 
# photo.png: image
# song.mp3: audio
# movie.mp4: video
# doc.txt: text
# app.exe: application

import mimetypes
 
ALLOWED_TYPES = {'image/jpeg', 'image/png', 'image/gif'}
 
def is_allowed_upload(filename):
    """Check if file type is allowed."""
    mime, _ = mimetypes.guess_type(filename)
    return mime in ALLOWED_TYPES
 
print(is_allowed_upload('photo.jpg'))   # True
print(is_allowed_upload('script.js'))   # False
print(is_allowed_upload('image.png'))   # True

import mimetypes
 
def get_content_type(filename):
    """Get Content-Type header value."""
    mime, encoding = mimetypes.guess_type(filename)
    
    if mime is None:
        return 'application/octet-stream'
    
    # Add charset for text types
    if mime.startswith('text/'):
        return f'{mime}; charset=utf-8'
    
    return mime
 
print(get_content_type('page.html'))  # text/html; charset=utf-8
print(get_content_type('image.png'))  # image/png
print(get_content_type('unknown.xyz'))  # application/octet-stream

import mimetypes
from http.server import HTTPServer, SimpleHTTPRequestHandler
 
class ContentTypeHandler(SimpleHTTPRequestHandler):
    def guess_type(self, path):
        """Override to use mimetypes."""
        mime, _ = mimetypes.guess_type(path)
        if mime:
            return mime
        return 'application/octet-stream'

import mimetypes
 
url = 'https://example.com/images/photo.jpg?size=large'
mime, _ = mimetypes.guess_type(url)
print(mime)  # image/jpeg

import mimetypes
 
# Strict mode only uses official IANA types
mime, _ = mimetypes.guess_type('file.mp3', strict=True)
print(mime)  # audio/mpeg
 
# Non-strict includes platform-specific types
mime, _ = mimetypes.guess_type('file.mp3', strict=False)
print(mime)  # audio/mpeg

import mimetypes
 
# All extension mappings
print(mimetypes.types_map)  # {'.js': 'text/javascript', ...}
 
# Common types
print(mimetypes.common_types)  # Core types only

import mimetypes
 
def is_image(filename):
    """Check if file is an image."""
    mime, _ = mimetypes.guess_type(filename)
    return mime is not None and mime.startswith('image/')
 
def get_image_type(filename):
    """Get specific image format."""
    mime, _ = mimetypes.guess_type(filename)
    if mime and mime.startswith('image/'):
        return mime.split('/')[1]  # jpeg, png, gif, etc.
    return None
 
print(is_image('photo.jpg'))  # True
print(get_image_type('photo.jpg'))  # jpeg

import mimetypes
 
def get_attachment_params(filename):
    """Get MIME parts for email attachment."""
    mime, _ = mimetypes.guess_type(filename)
    
    if mime is None:
        return 'application', 'octet-stream'
    
    maintype, subtype = mime.split('/')
    return maintype, subtype
 
maintype, subtype = get_attachment_params('report.pdf')
print(f"maintype={maintype}, subtype={subtype}")
# maintype=application, subtype=pdf

import mimetypes
 
# Based on extension, not content!
mime, _ = mimetypes.guess_type('malware.jpg')  # Still returns image/jpeg
 
# For content-based detection, use python-magic:
# import magic
# magic.from_file('file.jpg', mime=True)