Merge photo highlights by name and capture year

photo-upload/README.md

@@ -179,8 +179,11 @@ photos. When none of these date sources is available, the service uses the uploa
 date. Re-sharing the same images, caption, and
 highlight with dates replaces the earlier undated upload. Captions are plain
 text and optional.
-An optional highlight name (up to 200 characters) adds one link to **Highlights**,
-pointing to the first photo in the batch. Leave it blank for an ordinary upload.
+An optional highlight name (up to 200 characters) adds a link to **Highlights**.
+If that name already exists in the same capture year, publishing merges the new
+photos into that highlight and keeps one link at its newest photo. A batch spanning
+years gets one link per year. Photo dates and gallery order stay unchanged. Leave
+the name blank for an ordinary upload.
 
 Gallery images are progressive JPEGs with a maximum edge of 1600 pixels and a
 300 KiB size limit. The uploader tries moderate JPEG quality levels, then reduces

photo-upload/app.py

@@ -177,6 +177,8 @@ def photo_entry(job, date, index, caption, highlight):
                f'date: "{item_date.isoformat()}"', "categories: art"]
     if index == 0 and highlight:
         content.append(f"anchor: {json.dumps(highlight)}")
+    if highlight:
+        content.append(f"highlight_name: {json.dumps(highlight)}")
     content.extend(["---", ""])
     if index == 0 and not highlight:
         content.extend([f'<div id="photo-{job}"></div>', ""])

photo-upload/highlights.py

@@ -0,0 +1,56 @@
+"""Keep one gallery highlight link per name and capture year."""
+
+from datetime import datetime, timezone
+import json
+from pathlib import Path
+import re
+
+
+def field(content, name):
+    match = re.search(rf"^{re.escape(name)}: (.+)$", content.split("---", 2)[1], re.MULTILINE)
+    if not match:
+        return None
+    value = match.group(1).strip()
+    return json.loads(value) if value.startswith('"') else value
+
+
+def normalize_highlights(art):
+    """Place each link on the newest photo in its named, dated group."""
+    groups = {}
+    for path in Path(art).glob("*.md"):
+        content = path.read_text()
+        name = field(content, "highlight_name") or field(content, "anchor")
+        date = field(content, "date")
+        if not name or not date:
+            continue
+        taken = datetime.fromisoformat(str(date))
+        if taken.tzinfo is None:
+            taken = taken.replace(tzinfo=timezone.utc)
+        groups.setdefault((name, taken.year), []).append((taken, path, content))
+
+    changed = []
+    for (name, _), entries in groups.items():
+        # An old single anchor is already the only link; a newer named photo
+        # moves the link without changing any photo's capture date or position.
+        newest = max(entries, key=lambda entry: (entry[0], entry[1].name))[1]
+        if len(entries) == 1 and field(entries[0][2], "anchor") == name:
+            continue
+        for _, path, content in entries:
+            old = content
+            had_anchor = field(content, "anchor") is not None
+            if (path == newest and had_anchor and
+                    all(field(other[2], "anchor") is None for other in entries if other[1] != path)):
+                continue
+            if had_anchor:
+                content = re.sub(r"^anchor: .+\n", "", content, count=1, flags=re.MULTILINE)
+            if path == newest:
+                content = content.replace("categories: art\n", "categories: art\n" +
+                                          f"anchor: {json.dumps(name, ensure_ascii=False)}\n", 1)
+            elif had_anchor:
+                title = field(content, "title")
+                if title and title.startswith("photo-") and f'id="{title}"' not in content:
+                    content = content.replace("---\n\n", f'---\n\n<div id="{title}"></div>\n\n', 1)
+            if content != old:
+                path.write_text(content)
+                changed.append(path)
+    return changed

photo-upload/publisher.py

@@ -11,6 +11,7 @@ import sys
 import time
 
 from app import atomic_json
+from highlights import normalize_highlights
 from posts import publish_post
 
 log = logging.getLogger("publisher")
@@ -117,6 +118,8 @@ class Publisher:
                     self.git("add", "--", str(entry))
                 shutil.copytree(submission / "images", self.repo / images, dirs_exist_ok=True)
                 self.git("add", "--", str(images))
+            for path in normalize_highlights(self.repo / "_art"):
+                self.git("add", "--", str(path.relative_to(self.repo)))
             if self.git("diff", "--cached", "--name-only"):
                 self.git("commit", "-m", f"Publish {len(jobs)} submission(s)")
             self.git("push", "origin", f"HEAD:refs/heads/{self.branch}")

photo-upload/tests/test_upload.py

@@ -379,6 +379,56 @@ def test_publisher_names_photo_entry_from_capture_date(publisher, service):
     assert f'_art/2025-04-18-photo-{response.json["id"]}-00.md' in paths
 
 
+def test_publisher_merges_highlights_by_name_and_capture_year(publisher):
+    run, remote, client = publisher
+    first = client.post("/api/photos", headers=AUTH, data={
+        "photos": [(io.BytesIO(photo("red")), "first.jpg")],
+        "highlight": "Trip", "photo_dates": "2026-03-01T12:00:00+00:00",
+    }).json["id"]
+    assert run().returncode == 0
+    second = client.post("/api/photos", headers=AUTH, data={
+        "photos": [(io.BytesIO(photo("blue")), "second.jpg")],
+        "highlight": "Trip", "photo_dates": "2026-06-01T12:00:00+00:00",
+    }).json["id"]
+    assert run().returncode == 0
+    files = git(remote, "ls-tree", "-r", "--name-only", "master").splitlines()
+    entries = [git(remote, "show", f"master:{path}") for path in files if path.startswith("_art/")]
+    assert sum('anchor: "Trip"' in entry for entry in entries) == 1
+    assert f'anchor: "Trip"' in next(entry for entry in entries if f"photo-{second}" in entry)
+    assert f'<div id="photo-{first}"></div>' in next(
+        entry for entry in entries if f'title: "photo-{first}"' in entry)
+    old_year = client.post("/api/photos", headers=AUTH, data={
+        "photos": [(io.BytesIO(photo("green")), "old-year.jpg")],
+        "highlight": "Trip", "photo_dates": "2025-12-31T12:00:00+00:00",
+    }).json["id"]
+    assert run().returncode == 0
+    files = git(remote, "ls-tree", "-r", "--name-only", "master").splitlines()
+    entries = [git(remote, "show", f"master:{path}") for path in files if path.startswith("_art/")]
+    assert sum('anchor: "Trip"' in entry for entry in entries) == 2
+    assert 'anchor: "Trip"' in next(entry for entry in entries if f"photo-{old_year}" in entry)
+    commit = git(remote, "rev-parse", "master")
+    assert run().returncode == 0
+    assert git(remote, "rev-parse", "master") == commit
+
+
+def test_publisher_splits_a_highlight_across_capture_years(publisher):
+    run, remote, client = publisher
+    response = client.post("/api/photos", headers=AUTH, data={
+        "photos": [(io.BytesIO(photo("red")), "old.jpg"),
+                   (io.BytesIO(photo("blue")), "new.jpg")],
+        "highlight": "Trip",
+        "photo_dates": "2025-12-31T12:00:00+00:00\n2026-01-01T12:00:00+00:00",
+    })
+    assert response.status_code == 202
+    result = run()
+    assert result.returncode == 0, result.stderr
+    job = response.json["id"]
+    files = git(remote, "ls-tree", "-r", "--name-only", "master").splitlines()
+    entries = [git(remote, "show", f"master:{path}") for path in files if f"photo-{job}" in path]
+    assert len(entries) == 2
+    assert all('anchor: "Trip"' in entry for entry in entries)
+
+
 def test_publisher_replaces_an_undated_version_of_the_same_photos(publisher, service):
     run, remote, client = publisher
     raw = photo()