refactoring and thumbnailer

distribusi/cli.py

@@ -13,76 +13,71 @@ def build_argparser():
     distribusi works by traversing the file system and directory hierarchy to
     automatically list all the files in the directory, detect the file types
     and providing them with relevant html classes and tags for easy styling.
-    """)
-
-    parser.add_argument(
-        '-d', '--directory', help="Select which directory to distribute", default="."
+    """
     )
 
     parser.add_argument(
-        '-s', '--style', help="Select a CSS style sheet to include"
+        "-d", "--directory", help="Select which directory to distribute", default="."
     )
 
+    parser.add_argument("-s", "--style", help="Select a CSS style sheet to include")
+
     parser.add_argument(
-        '-v', '--verbose', help="Print verbose debug output", action="store_true"
+        "-v", "--verbose", help="Print verbose debug output", action="store_true"
     )
 
     parser.add_argument(
-        '-t',
-        '--thumbnail',
+        "-t",
+        "--thumbnail",
         help="Generate 450x450 thumbnails for images",
         action="store_true",
     )
 
     parser.add_argument(
-        '-n',
-        '--no-template',
+        "-n",
+        "--no-template",
         help="Don't use the template to output html",
         action="store_true",
     )
 
     parser.add_argument(
-        '-nf',
-        '--no-filenames',
-        help="Don't add file names to listing",
+        "-c",
+        "--captions",
+        help="Adds image captions based on EXIF metadata",
         action="store_true",
     )
 
     parser.add_argument(
-        '-c',
-        '--captions',
-        help="Adds image captions based on EXIF metadata, requires 'exiftool'",
+        "-r",
+        "--remove-index",
+        help="Recursively removes all instances of index.html that have been previously made by distribusi",
         action="store_true",
     )
 
     parser.add_argument(
-        '-r',
-        '--remove-index',
-        help="Recursively removes all instances of index.html that have been previously made by distribusi",
-        action="store_true")
-
-    parser.add_argument(
-        '-e',
-        '--exclude-directory',
+        "-e",
+        "--exclude-directory",
         help="Exclude one or multiple directories from indexing",
         nargs="*",
-        metavar='DIR')
+        metavar="DIR",
+    )
 
     parser.add_argument(
-        '-f',
-        '--force',
+        "-f",
+        "--force",
         help="Force whether distribusi overwrites or removes instances of index.html not generated by distribusi, use at own risk!",
-        action="store_true")
+        action="store_true",
+    )
 
     parser.add_argument(
-        '--no-hidden',
-        help="Exclude hidden directories",
-        action="store_true")
+        "--no-hidden", help="Exclude hidden directories", action="store_true"
+    )
 
     parser.add_argument(
-        '--menu-with-index',
+        "--menu-with-index",
         help="Append index.html to menu items to aid navigation",
-        action="store_true")
+        action="store_true",
+    )
 
     return parser
 

distribusi/distribusi.py

@@ -1,11 +1,9 @@
 import base64
 import os
-import subprocess
-from io import BytesIO
 
 import magic
 from PIL import Image
-
+from exif import Image as ExifImage
 from distribusi.page_template import html_footer, html_head
 from distribusi.mappings import CODE_TYPES, FILE_TYPES, SUB_TYPES
 
@@ -14,62 +12,49 @@ MIME_TYPE = magic.Magic(mime=True)
 
 def caption(image):
     try:
-        process = subprocess.Popen(
-            ['exiftool', '-Comment', image], stdout=subprocess.PIPE)
-        out, err = process.communicate()
-    except Exception as e:
-        print(e)
-        print('Do you have exiftool installed?')
-    try:
-        caption = out.decode("utf-8").split(": ", 1)[1]
+        with open(image, "rb") as image_file:
+            exif_image = ExifImage(image_file)
+            caption = exif_image.communication
     except Exception as e:
-        caption = ''
         print(e)
+        caption = ""
     return caption
 
 
-def thumbnail(image, name, args):
+def thumbnail(full_path_image, name, args):
+    if full_path_image.endswith("_thumbnail.jpg"):
+        return
     try:
         size = (450, 450)
-        im = Image.open(image)
-        im.thumbnail(size)
-        
-        if (im.mode == 'RGBA'):
-            bg = Image.new('RGBA', im.size, (255,255,255))
-            composite = Image.alpha_composite(bg, im)
-            im=composite.convert('RGB')
-        
-        output = BytesIO()
-        im.save(output, format='JPEG')
-        im_data = output.getvalue()
-        data_url = base64.b64encode(im_data).decode()
-        if args.captions:
-            cap = caption(image)
-        else:
-            cap = name
-        return (
-            "<figure><a href='{}'><img class='thumbnail' src='data:image/jpg;base64,{}'></a><figcaption>{}</figcaption></figure>"
-        ).format(name, data_url, cap)
+        thumbnail_image = Image.open(full_path_image)
+        thumbnail_image.thumbnail(size)
+
+        if thumbnail_image.mode == "RGBA":
+            bg = Image.new("RGBA", im.size, (255, 255, 255))
+            composite = Image.alpha_composite(bg, thumbnail_image)
+            thumbnail_image = composite.convert("RGB")
+
+        image_filename_no_ext = os.path.splitext(full_path_image)[0]
+        thumbnail_filename = f"{image_filename_no_ext}_thumbnail.jpg"
+        thumbnail_image.save(thumbnail_filename, format="JPEG")
+        return os.path.basename(thumbnail_filename)
     except Exception as e:
-        print('Thumbnailer:', e)
-        return "<figure><a href='{}'><img src='{}'></a><figcaption>{}</figcaption></figure>".format(name, name, name)
+        print("Thumbnailer:", e)
+        return
 
 
-def div(args, type_, subtype, tag, name):
-    id_name = name.split('.')[0].replace(' ', '_')
-    if args.no_filenames:
-        filename = ''
-    else:
-        filename = '<span class="filename">{}</span>'.format(name)
+def format_div(args, type_, subtype, tag, name):
+    id_name = name.split(".")[0].replace(" ", "_")
+    filename = f'<span class="filename">{name}</span>'
 
-    if 'image' in type_:
+    if "image" in type_:
         html = '<div id="{}" class="{}">{}</div>'
-    elif 'pdf' in subtype:
-        html = '<div id="{}" class="{}">{}' + filename + '</div>'
-    elif 'dir' in type_ or 'html' in subtype or 'unkown-file' in subtype:
+    elif "pdf" in subtype:
+        html = '<div id="{}" class="{}">{}' + filename + "</div>"
+    elif "dir" in type_ or "html" in subtype or "unkown-file" in subtype:
         html = '<div id="{}" class="{}">{}</div>'
     else:
-        html = '<div id="{}" class="{}">{}' + filename + '</div>'
+        html = '<div id="{}" class="{}">{}' + filename + "</div>"
 
     return html.format(id_name, subtype, tag)
 
@@ -80,111 +65,141 @@ def check_distribusi_index(args, index):
     """
 
     if not args.force:
-        with open(index, 'r') as f:
+        with open(index, "r") as f:
             if '<meta name="generator" content="distribusi" />' in f.read():
                 return True
             else:
                 if args.verbose:
-                    print(index, 'not generated by distribusi, skipping')
+                    print(index, "not generated by distribusi, skipping")
                 return False
     elif args.force:
         return True
 
+
 def write_index(args, index, html, html_head, html_footer):
-    with open(index, 'w') as f:
+    with open(index, "w") as index_file:
         if not args.no_template:
             if args.style:
-                fs = open(args.style, "r")
-                style = fs.read()
+                file_style = open(args.style, "r")
+                style = file_style.read()
                 styled_html_head = html_head % style
             else:
-                styled_html_head = html_head % ''
-            f.write(styled_html_head)
+                styled_html_head = html_head % ""
+            index_file.write(styled_html_head)
 
         for line in html:
-            f.write(line + '\n')
+            index_file.write(line + "\n")
 
         if not args.no_template:
-            f.write(html_footer)
+            index_file.write(html_footer)
+
+
+def handle_text_files(name, full_path):
+    if name.endswith(".html") or subtype == "html":
+        subtype = "html"
+        # what types of text files to expand
+        tag = '<section id="{}">{}</section>'.format(
+            name, open(full_path).read()
+        )
+    elif subtype in CODE_TYPES or name.endswith(".txt"):
+        # if the plain text is code,
+        # which types do we wrap in pre-tags?
+        tag = "<pre>" + open(full_path).read() + "</pre>"
+    else:
+        subtype = subtype + " unkown-file"
+        tag = "<a href='{}'>{}</a>"
+        # a = FILE_TYPES[type_]
+    return subtype, tag
+
+
+def handle_image_files(name, full_path, args):
+    if args.thumbnail:
+        caption = ""
+        thumbnail_filename = thumbnail(full_path, name, args)
+        if thumbnail_filename is None:
+            return
+        if args.captions:
+            caption = caption(full_path)
+        return f"<figure><a href='{name}'><img class='thumbnail' src='{thumbnail_filename}'></a><figcaption>{caption}</figcaption></figure>"
+
+    return FILE_TYPES[type_].format(name, caption)
+
+
+def remove_index_html(root, files):
+    index = os.path.join(root, "index.html")
+    if "index.html" in files:
+        try:
+            if check_distribusi_index(args, index):
+                if args.verbose:
+                    print("Removing index.html from", root)
+                os.remove(index)
+        except Exception as e:
+            print(e)
+    return
+
+
+def remove_hidden_files_and_folders(dirs, files):
+    dirs = list(filter(lambda d: not d.startswith("."), dirs))
+    files = list(filter(lambda f: not f.startswith("."), files))
+    return dirs, files
 
 
 def distribusify(args, directory):  # noqa
     for root, dirs, files in os.walk(directory):
+        html = []
 
         if args.exclude_directory:
             if args.verbose:
-                print('Excluding directory:', ", ".join(args.exclude_directory))
+                print("Excluding directory:", ", ".join(args.exclude_directory))
             dirs[:] = [d for d in dirs if d not in args.exclude_directory]
 
         if args.no_hidden:
-            dirs = list(filter(lambda d: not d.startswith('.'), dirs))
-            files = list(filter(lambda f: not f.startswith('.'), files))
-
-        dirs.sort()
-        files.sort()
+            dirs, files = remove_hidden_files_and_folders(dirs, files)
 
-        if not args.remove_index:
-            html = []
+        if args.remove_index:
+            remove_index_html(root, files)
 
         if args.verbose:
-                print('Generating directory listing for', root)
+            print("Generating directory listing for", root)
 
         for name in sorted(files):
+            if "index.html" in name:
+                continue
+
+            if name.endswith("_thumbnail.jpg"):
+                continue
 
-                if 'index.html' not in name:
             full_path = os.path.join(root, name)
             mime = MIME_TYPE.from_file(full_path)
-                    # example: MIME plain/text becomes 'type' plain 'subtype' text
-                    type_, subtype = mime.split('/')
-
+            type_, subtype = mime.split("/")
             caption = name
 
             if args.verbose:
-                        print('Found', name, 'as', mime)
-
+                print("Found", name, "as", mime)
             if type_ in FILE_TYPES:
-                        
-                        a = FILE_TYPES[type_].format(name, caption)
-
-                        # expansion for different kind of text files
-                        if type_ == 'text':
-                            if name.endswith('.html') or subtype == 'html':
-                                subtype = 'html'
-                                # what types of text files to expand
-                                a = '<section id="{}">{}</section>'.format(name, open(full_path).read())
-                            elif subtype in CODE_TYPES or name.endswith('.txt'):
-                                # if the plain text is code,
-                                # which types do we wrap in pre-tags?
-                                a = "<pre>" + open(full_path).read() + "</pre>"
-                            else:
-                                subtype = subtype+' unkown-file'
-                                a = "<a href='{}'>{}</a>"
-                                # a = FILE_TYPES[type_]
-
-                        if type_ == 'image':
-                            if args.thumbnail:
-                                a = thumbnail(full_path, name, args)
-                            if args.no_filenames:
-                                caption = ""
-                                a = FILE_TYPES[type_].format(name, caption)
-                            if args.captions:
-                                caption = caption(full_path)
-                                a = FILE_TYPES[type_].format(name, caption)
+                match type_:
+                    case "text":
+                        subtype, tag = handle_text_files(name, full_path)
+                    case "image":
+                        tag = handle_image_files(name, full_path, args)
+                        if tag is None:
+                            continue
+                    case _:
+                        tag = FILE_TYPES[type_].format(name, caption)
 
             if subtype in SUB_TYPES:
-                        a = SUB_TYPES[subtype]
+                tag = SUB_TYPES[subtype]
 
             if type_ not in FILE_TYPES and subtype not in SUB_TYPES:
                 # catch exceptions not yet defined in FILE_TYPES or SUB_TYPES
-                        a = "<a href='{}'>{}</a>"
+                tag = "<a href='{}'>{}</a>"
                 if args.verbose:
-                            message = 'not in list of file types, adding as plain href: \n'
+                    message = "not in list of file types, adding as plain href: \n"
                     print(type_, subtype, message, name)
-                            subtype = subtype + ' unkown-file'
+                    subtype = subtype + " unkown-file"
 
-                    a = a.replace('{}', name)
-
-                    html.append(div(args, type_, subtype, a, name))
+            tag = tag.replace("{}", name)
+            html.append(format_div(args, type_, subtype, tag, name))
 
         if root != directory:
             if args.menu_with_index:
@@ -192,28 +207,17 @@ def distribusify(args, directory):  # noqa
             else:
                 html.append('<a href="../">../</a>')
 
-            for name in dirs:
+        for name in sorted(dirs):
             if args.menu_with_index:
-                    a = "<a href='{}/index.html'>{}</a>".replace('{}', name)
+                tag = "<a href='{}/index.html'>{}</a>".replace("{}", name)
             else:
-                    a = "<a href='{}'>{}/</a>".replace('{}', name)
+                tag = "<a href='{}'>{}/</a>".replace("{}", name)
 
-                html.insert(0, div(args, 'dir', 'dir', a, 'folder'))
+            html.insert(0, format_div(args, "dir", "dir", tag, "folder"))
 
-            index = os.path.join(root, 'index.html')
+        index = os.path.join(root, "index.html")
         if os.path.exists(index):
             if check_distribusi_index(args, index):
                 write_index(args, index, html, html_head, html_footer)
         elif not os.path.exists(index):
             write_index(args, index, html, html_head, html_footer)
-
-        if args.remove_index:
-            index = os.path.join(root, 'index.html')
-            if 'index.html' in files:
-                try:
-                    if check_distribusi_index(args, index):
-                        if args.verbose:
-                            print('Removing index.html from', root)
-                        os.remove(index)
-                except Exception as e:
-                    print(e)

distribusi/mappings.py

@@ -1,14 +1,15 @@
-CODE_TYPES = ['x-c', 'x-shellscript', 'x-python']
+CODE_TYPES = ["x-c", "x-shellscript", "x-python"]
 
 FILE_TYPES = {
-    'image': '<figure><img class="image" src="{}"><figcaption>{}</figcaption></figure>',
-    'text': '<a href="{}" class="text">{}</a>',
-    'video': ('<video controls>' '<source src="{}"></video>'),
-    'audio': ('<audio controls class="audio">' '<source src="{}"></audio>'),
+    "image": '<figure><img class="image" src="{}"><figcaption>{}</figcaption></figure>',
+    "text": '<a href="{}" class="text">{}</a>',
+    "video": ("<video controls>" '<source src="{}"></video>'),
+    "audio": ('<audio controls class="audio">' '<source src="{}"></audio>'),
 }
 
 SUB_TYPES = {
-        'pdf': (
+    "pdf": (
         '<object data="{}" class="pdf" type="application/pdf">'
-            '<embed src="{}" type="application/pdf" /></object>')
+        '<embed src="{}" type="application/pdf" /></object>'
+    )
 }

setup.py

@@ -1,28 +1,28 @@
 from setuptools import find_packages, setup
 
-dependencies = ['pillow >= 10.3.0', 'python-magic >= 0.4.15, < 1.0']
+dependencies = ["pillow >= 10.3.0", "python-magic >= 0.4.15, < 1.0", "exif >= 1.6.0"]
 
-with open('README.md', 'r') as handle:
+with open("README.md", "r") as handle:
     long_description = handle.read()
 
 setup(
-    name='distribusi',
-    version='0.0.10',
-    url='https://git.vvvvvvaria.org/varia/distribusi',
-    license='GPLv3',
-    author='Varia',
+    name="distribusi",
+    version="0.0.11",
+    url="https://git.vvvvvvaria.org/varia/distribusi",
+    license="GPLv3",
+    author="Varia",
     description=(
-        'Distribusi is a content management system for '
-        'the web that produces static pages based on '
-        'the file system.'
+        "Distribusi is a content management system for "
+        "the web that produces static pages based on "
+        "the file system."
     ),
     long_description=long_description,
-    long_description_content_type='text/markdown',
-    packages=find_packages(exclude=['tests']),
+    long_description_content_type="text/markdown",
+    packages=find_packages(exclude=["tests"]),
     include_package_data=True,
     zip_safe=False,
-    platforms='any',
+    platforms="any",
     install_requires=dependencies,
-    entry_points={'console_scripts': ['distribusi = distribusi.cli:cli_entrypoint']},
-    classifiers=['Programming Language :: Python :: 3', 'Environment :: Console'],
+    entry_points={"console_scripts": ["distribusi = distribusi.cli:cli_entrypoint"]},
+    classifiers=["Programming Language :: Python :: 3", "Environment :: Console"],
 )