skip latest pages

they probably dont have good ground truth to begin with

main.py

@@ -157,13 +157,13 @@ class Danbooru(Booru):
     title = "Danbooru"
     base_url = "https://danbooru.donmai.us"
 
-    async def posts(self, tag_query: str, limit):
+    async def posts(self, tag_query: str, limit, page: int):
         log.info("%s: submit %r", self.title, tag_query)
         async with self.limiter:
             log.info("%s: submit upstream %r", self.title, tag_query)
             async with self.session.get(
                 f"{self.base_url}/posts.json",
-                params={"tags": tag_query, "limit": limit},
+                params={"tags": tag_query, "limit": limit, "page": page},
             ) as resp:
                 assert resp.status == 200
                 rjson = await resp.json()
@@ -190,7 +190,12 @@ async def download_images(ctx):
     try:
         limit = int(sys.argv[3])
     except IndexError:
-        limit = 10
+        limit = 30
+
+    try:
+        pageskip = int(sys.argv[4])
+    except IndexError:
+        pageskip = 150
 
     danbooru = Danbooru(
         ctx.session,
@@ -198,7 +203,7 @@ async def download_images(ctx):
         AsyncLimiter(1, 3),
     )
 
-    posts = await danbooru.posts(tagquery, limit)
+    posts = await danbooru.posts(tagquery, limit, pageskip)
     for post in posts:
         if "md5" not in post:
             continue