Search webservice (#2)

* Deployable script for windows file service

* Deploy!

* Server side work

conjurer_librarian/conjurer_librarian.py

@@ -12,9 +12,8 @@ from habanero import Crossref
 from waitress import serve
 from queue import Queue
 import search_bot
+import scrape_bot
 import requests
-from typing import Callable
-from uuid import UUID
 NETRC_FILE = r"C:\Users\Activcom.pl\.netrc"
 
 HOST_ADDRESS = "192.168.1.192"
@@ -22,7 +21,7 @@ PORT_ADDRESS = 5001
 MAIN_BOT_ADDRESS = "http://192.168.1.191:5000"
 SEND_RESULTS = "/conjurer"
 BDSM_UUID_TEST = '96b7f85a-1142-4908-8986-62a2ea25a147'
-
+MAX_CR_RESULTS = 150000
 ENCODING = "utf-8"
 app = Flask(__name__)
 
@@ -37,13 +36,15 @@ class Librarian(object):
         self.cr = Crossref(mailto=authTokens[0])
         self.query = query
         self.uuid = str(uuid)
-        self.limit = 3000
+        self.limit = MAX_CR_RESULTS
         self.fetched = 0
         self.hit = 0
         self.total = 0
         self.app = app
         self.logger = app.logger
         self.live_results = []
+        self.final_result = []
+        self.not_in_db = []
         self.search_result_from_cr = {}
         self.done = False
 
@@ -62,8 +63,7 @@ class Librarian(object):
             self.fetched += len(result['message']['items'])
             self.logger.info (self.total)
             self.logger.info(self.fetched)
-
-            time.sleep(1)
+            time.sleep(0.1)
         self.app.logger.info("CROSSREF DONE")
         #TODO: add to sql with UUID
         #sql row - title, doi, author, uuid, last retrieved, is_available
@@ -142,20 +142,20 @@ class Librarian(object):
             temp.append([item["DOI"], item])
 
         refined_result = await self.check_if_exists(temp)
-        temp = []
+
         for item in refined_result:
             if item["exists"]:
-                temp.append(item["data"])
-        refined_result = temp
-        self.hit = len(refined_result)
-        return refined_result
+                self.final_result.append(item["data"])
+            else:
+                self.not_in_db.append(item["data"])
+        self.hit = len(self.final_result)
+        return self.final_result
     #TODO: DOłożyć sprawdzenie czy w rafinowanym pliku już nie mamy częściowego wyniku
     async def answer_query(self):
         #TODO: DOłożyć sprawdzenie czy w rafinowanym pliku już nie mamy częściowego wyniku tutaj bo zanim w ogóle otworzymy nierafinowany
         #Search sql for UUID
         #SQL SELECT * FROM RESULTS WHERE UUID = uuid
         self.logger.info(f"Search started {self.uuid}")
-        uuid_found = False
         database = None
         query = self.query
         with open("cr_results.json", "r+", encoding="utf-8") as cr_file:
@@ -165,7 +165,7 @@ class Librarian(object):
                 database =  json.load(cr_file)
             except JSONDecodeError:
                 pass
-
+            #TODO: Zwracać i final_result i not_in_db. Not in db_do pliku - i do ręcznego przeglądania potem.
             if database:
                 for item in database.keys():
                     if item in self.uuid:
@@ -174,6 +174,10 @@ class Librarian(object):
                         for item in self.result:
                             temp[item['DOI']] = {"Title" : item['title'], "type": item['type']}
                         self.result = temp
+                        temp2 = {}
+                        for item in self.not_in_db:
+                            temp2[item['DOI']] = {"Title" : item['title'], "type": item['type']}
+                        self.not_in_db = temp2
                         return self.result
                 answer = await self.search_crossref(query=query)
                 database.update(answer)
@@ -194,6 +198,10 @@ class Librarian(object):
             for item in self.result:
                 temp[item['DOI']] = {"Title" : item['title'], "type": item['type']}
             self.result = temp
+            temp2 = {}
+            for item in self.not_in_db:
+                temp2[item['DOI']] = {"Title" : item['title'], "type": item['type']}
+                self.not_in_db = temp2
             return self.result
 
 #============================= FLASK INTERNALS===============================
@@ -221,11 +229,25 @@ class BackgroundTaskSearch(threading.Thread):
     async def _run(self):
         while True:
             database = None
+            ndb_database = None
             librarian = librarian_queue.get()
             self.app.logger.info("STARTED")
             result = await librarian.answer_query()
             result = {librarian.uuid: result}
             self.app.logger.info("Saving to file")
+            with open("not_in_db.json", "r+", encoding="utf-8") as ndb_file:
+                try:
+                    ndb_database =  json.load(ndb_file)
+                except JSONDecodeError:
+                    pass
+                if ndb_database:
+                    ndb_database.update(librarian.not_in_db)
+                else:
+                    ndb_database = librarian.not_in_db
+                ndb_file.truncate(0)
+                ndb_file.seek(0)
+                json.dump(ndb_database, ndb_file)
+
             with open("s_results.json", "r+", encoding="utf-8") as s_file:
                 try:
                     database =  json.load(s_file)
@@ -291,14 +313,15 @@ async def get_partial():
 
 #=======================================MAIN===================================================
 if __name__ == "__main__":
-    logger = logging.getLogger('conjurer_musician')
+    logger = app.logger
     logger.setLevel(logging.DEBUG)
     threads = []
-    #threads.append(threading.Thread(target=waitress_run))
-    threads.append(threading.Thread(target=flask_debug))
+    threads.append(threading.Thread(target=waitress_run))
+    #threads.append(threading.Thread(target=flask_debug))
     bgtask = BackgroundTaskSearch()
     bgtask.app = app
     threads.append(bgtask)
+    threads.append(threading.Thread(target = scrape_bot.scraper, args=(logger,)))
     i = 0
     for worker in threads:
         try: