Feat: search and download images

anixarttierlist/api.py

@@ -0,0 +1,48 @@
+import logging
+
+import requests
+
+BASE_URL = "https://shikimori.one/api/graphql"
+
+
+def get_image_url(name: str) -> str:
+    variables = {"name": name, "limit": 1}
+    query = """
+    query($name: String, $limit: Int!) {
+        animes(search: $name, limit: $limit, kind: "!special") {
+            id
+            name
+            russian
+            japanese
+            english
+            poster { id originalUrl mainUrl }
+        }
+    }
+    """
+    headers = {'User-Agent': 'AnixartTierList'}
+    response = requests.post(
+        BASE_URL, json={'query': query, 'variables': variables}, headers=headers)
+    if response.status_code != 200:
+        logging.error(f'Не удалось получить ссылку на изображение для: {name}')
+        return ""
+    data = response.json()
+    pic_url = data['data']['animes'][0]['poster']['originalUrl']
+    return pic_url
+
+
+def download_image(url: str, name: str, save_folder: str = './animes'):
+    try:
+        img_data = requests.get(url).content
+        with open(f"{save_folder}/{name}.jpeg", 'wb') as handler:
+            handler.write(img_data)
+    except Exception as e:
+        print(e)
+        logging.error(f'Не удалось скачать изображение для: \'{
+                      name}\' по ссылке: {url}')
+
+
+if __name__ == '__main__':
+    url = get_image_url(
+        "Kimi no koto ga daidaidaidaidai suki na 100-nin no Kanojo")
+    download_image(
+        url, "Kimi no koto ga daidaidaidaidai suki na 100-nin no Kanojo")

anixarttierlist/group.py

@@ -5,7 +5,7 @@ from sklearn.feature_extraction.text import CountVectorizer
 from sklearn.metrics.pairwise import cosine_similarity
 
 
-def find_first_substring(strings: List[str]):
+def find_first_substring(strings: List[str]) -> str:
     if not strings:
         return ""
     if len(strings) == 1:
@@ -20,12 +20,12 @@ def find_first_substring(strings: List[str]):
     return ""
 
 
-def group_by_common_part(strings, n=3):
+def group_by_common_part(strings, n=3) -> List[str]:
     vectorizer = CountVectorizer(analyzer='char_wb', ngram_range=(n, n))
     X = vectorizer.fit_transform(strings)
     similarity_matrix = cosine_similarity(X)
 
-    similarity_treshold = .5
+    similarity_threshold = .5
 
     groups = []
     assigned = [False] * len(strings)
@@ -38,12 +38,12 @@ def group_by_common_part(strings, n=3):
         assigned[i] = True
 
         for j in range(i+1, len(strings)):
-            if assigned[j] or similarity_matrix[i, j] < similarity_treshold:
+            if assigned[j] or similarity_matrix[i, j] < similarity_threshold:
                 continue
             group.append(strings[j])
             assigned[j] = True
 
         common_part = find_first_substring(group)
-        groups.append(common_part)
+        groups.append((group, common_part))
 
     return groups