fix: 부스 UI 변경 대응

.github/workflows/develop-build.yml

@@ -31,7 +31,7 @@ jobs:
         with:
           registry: ${{ env.REGISTRY }}
           username: ${{ github.actor }}
-          password: ${{ secrets.GH_TOKEN }}
+          password: ${{ secrets.GITHUB_TOKEN }}
 
       # =======================
       # booth-checker
@@ -92,4 +92,4 @@ jobs:
         with:
           subject-name: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME_BOOTH_DISCORD }}
           subject-digest: ${{ steps.push_booth_discord.outputs.digest }}
-          push-to-registry: true
+          push-to-registry: true

.github/workflows/latest-build.yml

@@ -31,7 +31,7 @@ jobs:
         with:
           registry: ${{ env.REGISTRY }}
           username: ${{ github.actor }}
-          password: ${{ secrets.GH_TOKEN }}
+          password: ${{ secrets.GITHUB_TOKEN }}
 
       # =======================
       # booth-checker
@@ -104,4 +104,4 @@ jobs:
       - name: Publish release
         uses: ncipollo/release-action@v1
         with:
-          generateReleaseNotes: true
+          generateReleaseNotes: true

booth_checker/__main__.py

@@ -16,6 +16,11 @@
 from operator import length_hint
 from unitypackage_extractor.extractor import extractPackage
 
+try:
+    from .fbx_diff import build_fbx_path_list, calculate_fbx_diff
+except ImportError:
+    from fbx_diff import build_fbx_path_list, calculate_fbx_diff
+
 from shared import *
 import booth
 import booth_sql
@@ -254,7 +259,6 @@ def generate_changelog_and_summary(item_data, download_url_list, version_json):
 
     return changelog_html_path, s3_object_url, summary_result, diff_found, None
 
-
 def generate_fbx_changelog_and_summary(item_data, download_url_list, version_json):
     """Generates changelog information for FBX-only tracking."""
     previous_fbx = version_json.get('fbx-files', {}) or {}
@@ -269,46 +273,13 @@ def generate_fbx_changelog_and_summary(item_data, download_url_list, version_jso
             logger.error(f'An error occurred while parsing {filename}: {e}')
             logger.debug(traceback.format_exc())
 
-    previous_hashes = {file_hash for file_hash in previous_fbx.values()}
-    current_hashes = {file_hash for file_hash in current_fbx.values()}
-
-    added = []
-    changed = []
-    deleted = []
-
-    previous_remaining = dict(previous_fbx)
-    current_remaining = dict(current_fbx)
-
-    for name in set(previous_fbx.keys()) & set(current_fbx.keys()):
-        old_hash = previous_fbx[name]
-        new_hash = current_fbx[name]
-        if old_hash != new_hash:
-            changed.append(name)
-        previous_remaining.pop(name, None)
-        current_remaining.pop(name, None)
-
-    for name, new_hash in current_remaining.items():
-        if new_hash in previous_hashes:
-            continue
-        added.append(name)
-
-    for name, old_hash in previous_remaining.items():
-        if old_hash in current_hashes:
-            continue
-        deleted.append(name)
+    added_entries, changed_entries, deleted_entries = calculate_fbx_diff(previous_fbx, current_fbx)
 
-    if not added and not changed and not deleted:
+    if not added_entries and not changed_entries and not deleted_entries:
         logger.info('No FBX hash differences detected; skipping changelog generation.')
         return None, None, None, False, current_fbx
 
-    path_list = []
-    for name in sorted(added):
-        path_list.append({'line_str': name, 'status': 1})
-    for name in sorted(changed):
-        path_list.append({'line_str': name, 'status': 3})
-    for name in sorted(deleted):
-        path_list.append({'line_str': name, 'status': 2})
-
+    path_list = build_fbx_path_list(added_entries, changed_entries, deleted_entries)
     tree = build_tree(path_list)
     html_list_items = tree_to_html(tree) if item_data["changelog_show"] else ''
     summary_data = files_list(tree)
@@ -874,15 +845,6 @@ def strftime_now():
     while True:
         logger.info("BoothChecker cycle started")
 
-        # BOOTH Heartbeat check once per cycle
-        try:
-            logger.info('Checking BOOTH heartbeat')
-            requests.get("https://booth.pm", timeout=10)
-        except requests.RequestException as e:
-            logger.error(f'BOOTH heartbeat failed: {e}. Skipping this cycle.')
-            sleep(refresh_interval)
-            continue
-
         # Recreate temporary folders
         recreate_folder("./download")
         recreate_folder("./process")

booth_checker/booth.py

@@ -9,14 +9,17 @@ def _extract_download_info(div, link_selector, filename_selector):
     if not download_link or not filename_div:
         return None
 
-    href = download_link.get("data-href")
-    filename = filename_div.get_text()
+    href = download_link.get("data-href") or download_link.get("href")
+    filename = filename_div.get_text(strip=True)
 
-    if not href:
+    if not href or not filename:
         return None
 
-    href = re.sub(r'[^0-9]', '', href)
-    return [href, filename]
+    match = re.search(r'/downloadables/(\d+)', href)
+    if not match:
+        return None
+
+    return [match.group(1), filename]
 
 def _crawling_base(url, cookie, selectors, shortlist, thumblist, product_only_filter=None):
     response = requests.get(url=url, cookies=cookie)
@@ -72,9 +75,19 @@ def crawling(order_num, product_only, cookie, shortlist=None, thumblist=None):
         'product_info_selector': 'a',
         'product_info_index': 1,
         'thumb_selector': 'img',
-        'download_item_selector': 'div.legacy-list-item__center, div[data-test="downloadable"]',
-        'download_link_selector': 'a.nav-reverse, div.js-download-button',
-        'filename_selector': 'div.flex-\\[1\\] b'
+        'download_item_selector': (
+            'div.legacy-list-item__center, '
+            'div.mt-16.desktop\\:flex.desktop\\:justify-between.desktop\\:items-center'
+        ),
+        'download_link_selector': (
+            'div.js-download-button[data-test="downloadable"][data-href*="/downloadables/"], '
+            'a.nav-reverse[href*="/downloadables/"]'
+        ),
+        'filename_selector': (
+            'div.min-w-0.u-text-wrap b, '
+            'div.min-w-0.break-words.whitespace-pre-line, '
+            'div.flex-\\[1\\] b'
+        )
     }
     return _crawling_base(url, cookie, selectors, shortlist, thumblist, product_only_filter=product_only)
 
@@ -85,9 +98,9 @@ def crawling_gift(order_num, cookie, shortlist=None, thumblist=None):
         'product_div_class': 'rounded-16 bg-white p-40 mobile:px-16 mobile:pt-24 mobile:pb-40 mobile:rounded-none',
         'product_info_selector': 'div.mt-24.text-left a',
         'thumb_selector': 'img',
-        'download_item_selector': 'div.w-full.text-left, div[data-test="downloadable"]',
-        'download_link_selector': 'a.no-underline.flex.items-center.flex.gap-4, div.js-download-button',
-        'filename_selector': "div[class='min-w-0 break-words whitespace-pre-line']"
+        'download_item_selector': 'div.mt-16.desktop\\:flex.desktop\\:justify-between.desktop\\:items-center',
+        'download_link_selector': 'div.js-download-button[data-test="downloadable"][data-href*="/downloadables/"]',
+        'filename_selector': 'div.min-w-0.break-words.whitespace-pre-line, div.min-w-0.u-text-wrap b'
     }
     return _crawling_base(url, cookie, selectors, shortlist, thumblist)
 

booth_checker/fbx_diff.py

@@ -0,0 +1,167 @@
+import os
+from collections import Counter, defaultdict
+
+
+def _normalize_fbx_entries(fbx_records):
+    entries = []
+    for path_str, file_hash in sorted(fbx_records.items()):
+        entries.append(
+            {
+                "path": path_str,
+                "basename": os.path.basename(path_str),
+                "hash": file_hash,
+            }
+        )
+    return entries
+
+
+def _remove_exact_name_hash_matches(previous_entries, current_entries):
+    previous_by_key = defaultdict(list)
+    current_by_key = defaultdict(list)
+
+    for entry in previous_entries:
+        previous_by_key[(entry["basename"], entry["hash"])].append(entry)
+    for entry in current_entries:
+        current_by_key[(entry["basename"], entry["hash"])].append(entry)
+
+    remaining_previous = []
+    remaining_current = []
+
+    for key in sorted(set(previous_by_key) | set(current_by_key)):
+        previous_list = previous_by_key.get(key, [])
+        current_list = current_by_key.get(key, [])
+        shared_count = min(len(previous_list), len(current_list))
+
+        remaining_previous.extend(previous_list[shared_count:])
+        remaining_current.extend(current_list[shared_count:])
+
+    return remaining_previous, remaining_current
+
+
+def calculate_fbx_diff(previous_fbx, current_fbx):
+    previous_entries = _normalize_fbx_entries(previous_fbx)
+    current_entries = _normalize_fbx_entries(current_fbx)
+
+    remaining_previous, remaining_current = _remove_exact_name_hash_matches(
+        previous_entries, current_entries
+    )
+
+    previous_by_name = defaultdict(list)
+    current_by_name = defaultdict(list)
+
+    for entry in remaining_previous:
+        previous_by_name[entry["basename"]].append(entry)
+    for entry in remaining_current:
+        current_by_name[entry["basename"]].append(entry)
+
+    added = []
+    changed = []
+    deleted = []
+
+    for name in sorted(set(previous_by_name) | set(current_by_name)):
+        previous_list = sorted(
+            previous_by_name.get(name, []),
+            key=lambda entry: (entry["hash"], entry["path"]),
+        )
+        current_list = sorted(
+            current_by_name.get(name, []),
+            key=lambda entry: (entry["hash"], entry["path"]),
+        )
+
+        shared_count = min(len(previous_list), len(current_list))
+        for index in range(shared_count):
+            changed.append(
+                {
+                    "basename": name,
+                    "previous_hash": previous_list[index]["hash"],
+                    "current_hash": current_list[index]["hash"],
+                    "previous_path": previous_list[index]["path"],
+                    "current_path": current_list[index]["path"],
+                }
+            )
+
+        deleted.extend(previous_list[shared_count:])
+        added.extend(current_list[shared_count:])
+
+    return added, changed, deleted
+
+
+def _short_hash(file_hash):
+    return file_hash[:8]
+
+
+def _append_path_context(label, basename, basename_counts, previous_path=None, current_path=None):
+    is_ambiguous_name = basename_counts[basename] > 1
+    path_changed = previous_path and current_path and previous_path != current_path
+
+    if not is_ambiguous_name and not path_changed:
+        return label
+
+    if previous_path and current_path:
+        if previous_path == current_path:
+            return f"{label} {{{current_path}}}"
+        return f"{label} {{from {previous_path} -> {current_path}}}"
+
+    path_value = current_path or previous_path
+    return f"{label} {{{path_value}}}"
+
+
+def build_fbx_path_list(added_entries, changed_entries, deleted_entries):
+    basename_counts = Counter()
+
+    for entry in added_entries:
+        basename_counts[entry["basename"]] += 1
+    for entry in changed_entries:
+        basename_counts[entry["basename"]] += 1
+    for entry in deleted_entries:
+        basename_counts[entry["basename"]] += 1
+
+    path_list = []
+
+    for entry in added_entries:
+        label = _append_path_context(
+            entry["basename"],
+            entry["basename"],
+            basename_counts,
+            current_path=entry["path"],
+        )
+        path_list.append(
+            {
+                "line_str": f"{label} [new {_short_hash(entry['hash'])}]",
+                "status": 1,
+            }
+        )
+
+    for entry in changed_entries:
+        label = _append_path_context(
+            entry["basename"],
+            entry["basename"],
+            basename_counts,
+            previous_path=entry["previous_path"],
+            current_path=entry["current_path"],
+        )
+        path_list.append(
+            {
+                "line_str": (
+                    f"{label} "
+                    f"[{_short_hash(entry['previous_hash'])} -> {_short_hash(entry['current_hash'])}]"
+                ),
+                "status": 3,
+            }
+        )
+
+    for entry in deleted_entries:
+        label = _append_path_context(
+            entry["basename"],
+            entry["basename"],
+            basename_counts,
+            previous_path=entry["path"],
+        )
+        path_list.append(
+            {
+                "line_str": f"{label} [old {_short_hash(entry['hash'])}]",
+                "status": 2,
+            }
+        )
+
+    return path_list