pyproject

src/auto_archiver/formatters/html_formatter.py

@@ -0,0 +1,78 @@
+from __future__ import annotations
+from dataclasses import dataclass
+from abc import abstractmethod
+import mimetypes
+from jinja2 import Environment, FileSystemLoader
+import uuid, os, pathlib
+
+from ..core import Metadata
+from ..core import Media
+from . import Formatter
+
+
+@dataclass
+class HtmlFormatter(Formatter):
+    name = "html_formatter"
+
+    def __init__(self, config: dict) -> None:
+        # without this STEP.__init__ is not called
+        super().__init__(config)
+        self.environment = Environment(loader=FileSystemLoader(os.path.join(pathlib.Path(__file__).parent.resolve(), "templates/")))
+        self.environment.filters.update({
+            'is_list': is_list_jinja,
+            'is_video': is_video_jinja,
+            'is_image': is_image_jinja,
+            'is_audio': is_audio_jinja,
+            'is_media': is_media_jinja,
+        })
+        self.template = self.environment.get_template("html_template.html")
+
+    @staticmethod
+    def configs() -> dict:
+        return {
+            "detect_thumbnails": {"default": True, "help": "if true will group by thumbnails generated by thumbnail enricher by id 'thumbnail_00'"},
+
+        }
+
+    def format(self, item: Metadata) -> Media:
+        media = item.media
+        # thumbnails
+        # TODO: thumbnails_media work per media, gah
+        # if self.detect_thumbnails:
+
+        content = self.template.render(
+            url=item.get_url(),
+            title=item.get_title(),
+            media=item.media,
+            metadata=item.get_clean_metadata()
+        )
+        html_path = os.path.join(item.get_tmp_dir(), f"formatted{str(uuid.uuid4())}.html")
+        with open(html_path, mode="w", encoding="utf-8") as outf:
+            outf.write(content)
+        return Media(filename=html_path)
+
+
+# JINJA helper filters
+
+
+def is_list_jinja(v) -> bool:
+    return isinstance(v, list)
+
+
+def is_video_jinja(s: str) -> bool:
+    m = mimetypes.guess_type(s)[0]
+    return "video" in (m or "")
+
+
+def is_image_jinja(s: str) -> bool:
+    m = mimetypes.guess_type(s)[0]
+    return "image" in (m or "")
+
+
+def is_audio_jinja(s: str) -> bool:
+    m = mimetypes.guess_type(s)[0]
+    return "audio" in (m or "")
+
+
+def is_media_jinja(v) -> bool:
+    return isinstance(v, Media)