src/images.py

-import subprocess
-import json
-import os
-
-import cairo
-
-import gi
-gi.require_version('GdkPixbuf', '2.0')
-from gi.repository import GdkPixbuf
-
-from . import abstract
-
-
-class PNGParser(abstract.AbstractParser):
-    mimetypes = {'image/png', }
-    meta_whitelist = {'SourceFile', 'ExifToolVersion', 'FileName',
-            'Directory', 'FileSize', 'FileModifyDate', 'FileAccessDate',
-            "FileInodeChangeDate", 'FilePermissions', 'FileType',
-            'FileTypeExtension', 'MIMEType', 'ImageWidth', 'BitDepth', 'ColorType',
-            'Compression', 'Filter', 'Interlace', 'BackgroundColor', 'ImageSize',
-            'Megapixels', 'ImageHeight'}
-
-    def get_meta(self):
-        out = subprocess.check_output(['exiftool', '-json', self.filename])
-        meta = json.loads(out.decode('utf-8'))[0]
-        for key in self.meta_whitelist:
-            meta.pop(key, None)
-        return meta
-
-    def remove_all(self):
-        surface = cairo.ImageSurface.create_from_png(self.filename)
-        surface.write_to_png(self.output_filename)
-        return True
-
-
-class GdkPixbufAbstractParser(abstract.AbstractParser):
-    """ GdkPixbuf can handle a lot of surfaces, so we're rending images on it,
-        this has the side-effect of removing metadata completely.
-    """
-    def get_meta(self):
-        out = subprocess.check_output(['exiftool', '-json', self.filename])
-        meta = json.loads(out.decode('utf-8'))[0]
-        for key in self.meta_whitelist:
-            meta.pop(key, None)
-        return meta
-
-    def remove_all(self):
-        _, extension = os.path.splitext(self.filename)
-        pixbuf = GdkPixbuf.Pixbuf.new_from_file(self.filename)
-        if extension == '.jpg':
-            extension = '.jpeg'
-        pixbuf.savev(self.output_filename, extension[1:], [], [])
-        return True
-
-
-class JPGParser(GdkPixbufAbstractParser):
-    mimetypes = {'image/jpeg'}
-    meta_whitelist = {'SourceFile', 'ExifToolVersion', 'FileName',
-            'Directory', 'FileSize', 'FileModifyDate', 'FileAccessDate',
-            "FileInodeChangeDate", 'FilePermissions', 'FileType',
-            'FileTypeExtension', 'MIMEType', 'ImageWidth',
-            'ImageSize', 'BitsPerSample', 'ColorComponents', 'EncodingProcess',
-            'JFIFVersion', 'ResolutionUnit', 'XResolution', 'YCbCrSubSampling',
-            'YResolution', 'Megapixels', 'ImageHeight'}
-
-
-class TiffParser(GdkPixbufAbstractParser):
-    mimetypes = {'image/tiff'}
-    meta_whitelist = {'Compression', 'ExifByteOrder', 'ExtraSamples',
-            'FillOrder', 'PhotometricInterpretation', 'PlanarConfiguration',
-            'RowsPerStrip', 'SamplesPerPixel', 'StripByteCounts',
-            'StripOffsets', 'BitsPerSample', 'Directory', 'ExifToolVersion',
-            'FileAccessDate', 'FileInodeChangeDate', 'FileModifyDate',
-            'FileName', 'FilePermissions', 'FileSize', 'FileType',
-            'FileTypeExtension', 'ImageHeight', 'ImageSize', 'ImageWidth',
-            'MIMEType', 'Megapixels', 'SourceFile'}
-
-
-class BMPParser(GdkPixbufAbstractParser):
-    mimetypes = {'image/x-ms-bmp'}
-    meta_whitelist = {'SourceFile', 'ExifToolVersion', 'FileName', 'Directory',
-            'FileSize', 'FileModifyDate', 'FileAccessDate',
-            'FileInodeChangeDate', 'FilePermissions', 'FileType',
-            'FileTypeExtension', 'MIMEType', 'BMPVersion', 'ImageWidth',
-            'ImageHeight', 'Planes', 'BitDepth', 'Compression', 'ImageLength',
-            'PixelsPerMeterX', 'PixelsPerMeterY', 'NumColors',
-            'NumImportantColors', 'RedMask', 'GreenMask', 'BlueMask',
-            'AlphaMask', 'ColorSpace', 'RedEndpoint', 'GreenEndpoint',
-            'BlueEndpoint', 'GammaRed', 'GammaGreen', 'GammaBlue', 'ImageSize',
-            'Megapixels'}
-

src/parser_factory.py

-import mimetypes
-import importlib
-import pkgutil
-
-from . import abstract
-
-from typing import TypeVar
-
-T = TypeVar('T', bound='abstract.AbstractParser')
-
-# This loads every parser in a dynamic way
-for module_loader, name, ispkg in pkgutil.walk_packages('.src'):
-    if not name.startswith('src.'):
-        continue
-    elif name == 'src.abstract':
-        continue
-    importlib.import_module(name)
-
-
-def _get_parsers() -> list:
-    """ Get all our parsers!"""
-    def __get_parsers(cls):
-        return cls.__subclasses__() + \
-            [g for s in cls.__subclasses__() for g in __get_parsers(s)]
-    return __get_parsers(abstract.AbstractParser)
-
-
-def get_parser(filename: str) -> (T, str):
-    mtype, _ = mimetypes.guess_type(filename)
-
-    for c in _get_parsers():
-        if mtype in c.mimetypes:
-            return c(filename), mtype
-    return None, mtype

tests/data/dirty.css

+/**
+ * This is my super css framework
+ * version: 1.0
+ * author : jvoisin
+ */
+
+body {
+	color: red;
+	background-color: blue;
+}
+
+.underline {
+	text-decoration: underline; /* underline is cool */	
+}