Add data migration that renames the image urls in the database for th…

…e VPN Resource Center contentful pages, so that they match the renamed files just added to the codebase

bedrock/contentful/migrations/0007_data_switch_to_local_images.py

@@ -0,0 +1,128 @@
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at https://mozilla.org/MPL/2.0/.
+
+# Generated by Django 3.2.23 on 2024-02-08 14:44
+
+import re
+import sys
+from collections import defaultdict
+
+from django.db import migrations
+
+from bedrock.contentful.constants import CONTENT_TYPE_PAGE_RESOURCE_CENTER
+
+# Renames image urls like the following to be local paths:
+#   "https://images.ctfassets.net/w5er3c7zdgmd/7cTC9dTpbRjoeNn3qwNbv2/63b89449a26f88b9de617f0ff0fa5acf/image2.png?w=688"
+#   "https://images.ctfassets.net/w5er3c7zdgmd/7cTC9dTpbRjoeNn3qwNbv2/63b89449a26f88b9de617f0ff0fa5acf/image2.png?w=1376 1.5x"
+#
+# Note that some of them appear in srcset attributes, hence have the density multiplier,
+# and some have width querystrigs, both of which must be catered for in the renaming.
+#
+# The actual files have been downloaded and added to the Bedrock codebase in the
+# same changeset as this data migration will be immediately available upon deployment
+
+contentful_cdn_images_pattern = re.compile(
+    r"\"https:\/\/images\.ctfassets\.net\/"  # CDN hostname
+    r"\S*"  # any path without a space
+    r"(?:\s1\.5x){0,1}"  # but allow a space at the end IFF it's before '1.5x' (in a non-capturing group)
+    r"\""  # and ending in a quote
+)
+
+
+def _print(args):
+    sys.stdout.write(args)
+    sys.stdout.write("\n")
+
+
+def _url_to_local_filename(url):
+    # Temporarily shelve any density info picked up from the src attribute. 1.5x is the only one used.
+    dpr_multiplier = " 1.5x"
+
+    if dpr_multiplier in url:
+        has_dpr = True
+        url = url.replace(dpr_multiplier, "")
+    else:
+        has_dpr = False
+
+    base_filename = url.split("/")[-1]
+    parts = base_filename.split("?")
+    if len(parts) == 1:
+        filename = parts[0]
+    else:
+        split_filename = parts[0].split(".")
+        if len(split_filename) == 2:
+            name, suffix = split_filename
+        else:
+            # filename has multiple . in it
+            name = ".".join(split_filename[:-1])
+            suffix = split_filename[-1]
+
+        filename = f"{name}_{''.join(parts[1:])}.{suffix}"
+        filename = filename.replace("=", "")
+
+    if has_dpr:
+        filename = f"{filename}{dpr_multiplier}"
+    return filename
+
+
+def switch_image_urls_to_local(apps, schema_editor):
+    ContentfulEntry = apps.get_model("contentful", "ContentfulEntry")
+
+    resource_center_pages = ContentfulEntry.objects.filter(
+        content_type=CONTENT_TYPE_PAGE_RESOURCE_CENTER,
+    )
+    for page in resource_center_pages:
+        # Some safety checks. If these fail, the migration breaks and it gets rolled back
+
+        if page.slug != "unknown":
+            _print(f"Checking page {page.contentful_id} {page.slug} [{page.locale}]")
+        # First the "SEO" images, which are also used on the listing page
+        assert "info" in page.data
+
+        if "seo" in page.data["info"] and "image" in page.data["info"]["seo"]:
+            _print(f"Updating SEO image URL for {page.contentful_id}. Slug: {page.slug}")
+
+            image_url = page.data["info"]["seo"]["image"]
+            updated_image_url = f"/media/img/products/vpn/resource-center/{_url_to_local_filename(image_url)}"
+            page.data["info"]["seo"]["image"] = updated_image_url
+
+            _print("Saving SEO info changes\n")
+            page.save()
+
+        # Now check for images in the body of the page
+        assert len(page.data["entries"]) == 1
+        assert "body" in page.data["entries"][0]
+
+        body = page.data["entries"][0]["body"]
+        old_to_new = defaultdict(str)
+        matches = re.findall(contentful_cdn_images_pattern, body)
+
+        if matches:
+            _print(f"Found {len(matches)} URLs to update")
+            for link in matches:
+                old_to_new[link] = f'"/media/img/products/vpn/resource-center/{_url_to_local_filename(link[1:-1])}"'
+
+            for old, new in old_to_new.items():
+                body = body.replace(old, new)
+
+            page.data["entries"][0]["body"] = body
+            _print("Saving page changes\n")
+            page.save()
+        else:
+            if page.slug != "unknown":
+                _print("No changes needed to page body\n")
+
+    _print("All done.")
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("contentful", "0006_mark-en-us-as-localisation-complete"),
+    ]
+    operations = [
+        migrations.RunPython(
+            switch_image_urls_to_local,
+            migrations.RunPython.noop,
+        )
+    ]