From 6f070bc1f9b070d1d6e9e64e5f83006215955b8e Mon Sep 17 00:00:00 2001
From: Lucas Thurston <lucas@castironcoding.com>
Date: Wed, 29 Mar 2023 15:22:41 -0700
Subject: [PATCH] [WIP] Implement EMS fetcher

---
 metadata_fetcher/fetchers/ems_fetcher.py | 70 ++++++++++++++++++++++++
 metadata_fetcher/lambda_function.py      |  3 +
 2 files changed, 73 insertions(+)
 create mode 100644 metadata_fetcher/fetchers/ems_fetcher.py

diff --git a/metadata_fetcher/fetchers/ems_fetcher.py b/metadata_fetcher/fetchers/ems_fetcher.py
new file mode 100644
index 000000000..bf7dab81b
--- /dev/null
+++ b/metadata_fetcher/fetchers/ems_fetcher.py
@@ -0,0 +1,70 @@
+import json
+from xml.etree import ElementTree
+from .Fetcher import Fetcher
+
+
+class EmsFetcher(Fetcher):
+    def __init__(self, params):
+        super(EmsFetcher, self).__init__(params)
+
+        # If `next_url` is a param, we know that this is not
+        # the fetch of the first page, so skip setting those
+        # attributes
+        if "next_url" in params:
+            for key in params:
+                setattr(self, key, params[key])
+            return
+
+        self.base_url = params.get("harvest_data").get("url")
+        self.original_url = self.get_current_url()
+        self.next_url = self.original_url
+        self.docs_total = 123
+
+    def get_current_url(self):
+        query_params = f"/search/*/objects/xml?filter=approved%3Atrue&page={self.write_page}"
+        return f"{self.base_url}{query_params}"
+
+    def build_fetch_request(self):
+        request = {"url": self.next_url}
+
+        print(
+            f"[{self.collection_id}]: Fetching page {self.write_page} "
+            f"at {request.get('url')}")
+
+        return request
+
+    def get_text_from_response(self, response):
+        return response.content
+
+    def check_page(self, http_resp):
+        """
+        TODO: review other fetchers, do what they do
+        """
+        hits = 345
+
+        print(
+            f"[{self.collection_id}]: Fetched page {self.write_page} "
+            f"at {http_resp.url} with {hits} hits"
+        )
+
+        return True
+
+    def increment(self, http_resp):
+        super(EmsFetcher, self).increment(http_resp)
+        tree = ElementTree.fromstring(http_resp.content.encode('utf-8'))
+        self.docs_total = len(tree.findall("objects/object"))
+        self.next_url = self.get_current_url() if self.docs_total > 0 else None
+
+    def json(self):
+        current_state = {
+            "harvest_type": self.harvest_type,
+            "collection_id": self.collection_id,
+            "next_url": self.next_url,
+            "write_page": self.write_page,
+            "base_url": self.base_url
+        }
+
+        if not self.next_url:
+            current_state.update({"finished": True})
+
+        return json.dumps(current_state)
diff --git a/metadata_fetcher/lambda_function.py b/metadata_fetcher/lambda_function.py
index 9f4b5cef6..d55b40a59 100644
--- a/metadata_fetcher/lambda_function.py
+++ b/metadata_fetcher/lambda_function.py
@@ -8,6 +8,9 @@
 
 
 def import_fetcher(harvest_type):
+    if harvest_type == "emuseum":
+        harvest_type = "ems"
+
     fetcher_module = importlib.import_module(
         f"fetchers.{harvest_type}_fetcher", package="metadata_fetcher")
     fetcher_module_words = harvest_type.split('_')