[naver] simplify code + add test

mikf · Mar 5, 2024 · 8a527f6 · 8a527f6
1 parent 009322a
commit 8a527f6
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 11 deletions.
diff --git a/gallery_dl/extractor/naver.py b/gallery_dl/extractor/naver.py
@@ -10,7 +10,6 @@
 
 from .common import GalleryExtractor, Extractor, Message
 from .. import text
-from urllib.parse import unquote
 
 
 class NaverBase():
@@ -63,16 +62,13 @@ def metadata(self, page):
         return data
 
     def images(self, page):
-        return [
-            (unquote(url, encoding="EUC-KR")
-             .replace("://post", "://blog", 1)
-             .partition("?")[0], None)
-            if "\ufffd" in unquote(url)
-            else
-            (url.replace("://post", "://blog", 1)
-             .partition("?")[0], None)
-            for url in text.extract_iter(page, 'data-lazy-src="', '"')
-        ]
+        results = []
+        for url in text.extract_iter(page, 'data-lazy-src="', '"'):
+            url = url.replace("://post", "://blog", 1).partition("?")[0]
+            if "\ufffd" in text.unquote(url):
+                url = text.unquote(url, encoding="EUC-KR")
+            results.append((url, None))
+        return results
 
 
 class NaverBlogExtractor(NaverBase, Extractor):

diff --git a/test/results/naver.py b/test/results/naver.py
@@ -24,6 +24,33 @@
     "#sha1_metadata": "a6e23d19afbee86b37d6e7ad934650c379d2cb1e",
 },
 
+{
+    "#url"     : "https://blog.naver.com/PostView.nhn?blogId=rlfqjxm0&logNo=70161391809",
+    "#comment" : "filenames in EUC-KR encoding (#5126)",
+    "#category": ("", "naver", "post"),
+    "#class"   : naver.NaverPostExtractor,
+    "#urls": (
+        "https://blogfiles.pstatic.net/20130305_23/ping9303_1362411028002Dpz9z_PNG/1_사본.png",
+        "https://blogfiles.pstatic.net/20130305_46/rlfqjxm0_1362473322580x33zi_PNG/오마갓합작.png",
+    ),
+
+    "blog": {
+        "id"  : "rlfqjxm0",
+        "num" : 43030507,
+        "user": "에나",
+    },
+    "post": {
+        "date"       : "dt:2013-03-05 17:48:00",
+        "description": "&amp;nbsp;◈ &amp;nbsp; &amp;nbsp; PROMOTER&amp;nbsp;：핑수 ˚ 아담 EDITOR：핑수 &amp;nbsp; 넵：이크：핑수...",
+        "num"        : 70161391809,
+        "title"      : "[공유] {&#8201;합작}&#8201; OH, MY GOD! ~ 아 또 무슨 종말을 한다 그래~"
+    },
+    "count"    : 2,
+    "num"      : range(1, 2),
+    "filename" : r"re:1_사본|오마갓합작",
+    "extension": "png",
+},
+
 {
     "#url"     : "https://blog.naver.com/gukjung",
     "#category": ("", "naver", "blog"),