Override reported text encoding for qinghai

qinghai/scrape.py

@@ -38,7 +38,7 @@ def main():
         print(f"Downloading {link[0]} ({i}/{len(links)})")
 
         text = download_article_text(link[0])
-        with open(f"articles-qinghai/{link[1]}_{i}.txt") as f:
+        with open(f"articles-qinghai/{link[1]}_{i}.txt", "w+") as f:
             f.write(text)
 
 
@@ -61,7 +61,9 @@ def get_article_links(page: int) -> List[Tuple[str, str]]:
 
 def download_article_text(link: str) -> str:
     """Get the text of an article from its link."""
-    soup = BeautifulSoup(requests.get(link).text, "html.parser")
+    request = requests.get(link)
+    request.encoding = "gbk"  # The website responds with the wrong encoding
+    soup = BeautifulSoup(request.text, "html.parser")
     return soup.find(class_="page_text").get_text().strip()