From ec7a5e8496c4bfdf69caf0ea42ffbb8b2eaef48c Mon Sep 17 00:00:00 2001
From: ado <awesley@gmail.com>
Date: Wed, 6 Dec 2023 18:02:57 -0500
Subject: [PATCH] Image generation

---
 generate.py | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 84 insertions(+), 3 deletions(-)

diff --git a/generate.py b/generate.py
index a6a93dd..5ed371f 100644
--- a/generate.py
+++ b/generate.py
@@ -1,7 +1,13 @@
 from dataclasses import dataclass
 from pathlib import Path
 import re
+from html.parser import HTMLParser
+from PIL import Image
+from os import path
 
+# image optimization ~ 
+INPUT_SOURCE = "./politecafe/input"
+OUTPUT_TARGET = "./politecafe/images"
 
 @dataclass
 class Article:
@@ -11,9 +17,79 @@ class Article:
         return Path(self.path).read_text()
 
 
+class SSGParser(HTMLParser):
+    def __init__(self):
+        super().__init__()
+        self.scales = [0.6, 1]
+        self.output = ""
+
+    def handle_endtag(self, tag):
+        if tag not in ["img", "br"]:
+            self.output += "</" + tag + ">"
+
+    def handle_data(self, data):
+        data = data.replace("<", "&lt;")
+        data = data.replace(">", "&gt;")
+        self.output += data
+
+    def handle_starttag(self, tag, attrs):
+        new_attrs = []
+        self.output += "<" + tag
+        if tag != "img":
+            new_attrs = attrs
+        else:
+            img = None
+            img_path = None
+            new_attrs.append(("sizes", "(min-width: 768px) 25vw, 35vw"))
+            for attr in attrs:
+                if attr[0] != "src":
+                    new_attrs.append(attr)
+                elif attr[0] == "src":
+                    img_path = path.join(INPUT_SOURCE, attr[1])
+                    img = Image.open(img_path)
+                    width = img.size[0]
+                    height = img.size[1]
+
+                    src_set = []
+                    for scale in self.scales:
+                        new_width = int(width * scale)
+                        new_height = int(height * scale)
+                        img_name = Path(path.basename(img_path)).stem
+                        destination_path = path.join(
+                            OUTPUT_TARGET, f"{img_name}{new_width}{new_height}.webp"
+                        )
+                        result = img.resize(size=(new_width, new_height))
+                        # write image
+                        result.save(
+                            fp=destination_path,
+                            format="WEBP",
+                            height=new_height,
+                            width=new_width,
+                        )
+                        # modify attribute
+                        src = path.relpath(destination_path, "politecafe")
+                        src_set.append(f"{src} {new_width}w")
+                    # capture src_set
+                    new_attrs.append(("srcset", ",".join(src_set)))
+        # write attrs to tag
+        for name, value in new_attrs:
+            self.output += ' {}="{}"'.format(name, value)
+
+        self.output += ">"
+
+
 articles = [
     Article(
-        "./input/art-001.html",
+        "./politecafe/input/tech-002.html",
+    ),
+    Article(
+        "./politecafe/input/art-002.html",
+    ),
+    Article(
+        "./politecafe/input/art-001.html",
+    ),
+    Article(
+        "./politecafe/input/tech-001.html",
     ),
 ]
 
@@ -22,6 +98,10 @@ slugs = []
 article_contents = []
 for article in articles:
     contents = article.read_contents()
+    parser = SSGParser()
+    parser.feed(contents)
+    contents = parser.output
+
     id_match = re.search(r'<h2 id="(.*?)"', contents)
     title_match = re.search(r'<h2 id=".*?>(.*?)</h2>', contents)
     if id_match is None:
@@ -36,10 +116,11 @@ for article in articles:
         slugs.append(slug)
 
 # plug into template, write to index.html
-template = Path("./input/template.html").read_text()
+template = Path("./politecafe/input/template.html").read_text()
+
 template = template.replace(
     "<div><!--- Articles ---></div>", "\n".join(article_contents)
 )
 template = template.replace("<div><!--- Slugs ---></div>", "\n".join(slugs))
-with open("./index.html", "w", encoding="utf-8") as index_html:
+with open("./politecafe/index.html", "w", encoding="utf-8") as index_html:
     index_html.write(template)