Serverside work of search

communication_subroutine.py

@@ -0,0 +1,76 @@
+import logging
+import threading
+import json
+from queue import Queue, Empty
+from flask import Flask, jsonify, request
+from waitress import serve
+
+HOST_ADDRESS = "192.168.1.191"
+PORT_ADDRESS = 5000
+OUT_COMM_Q = Queue()
+IN_COMM_Q = Queue()
+internal_q = Queue()
+app = Flask(__name__)
+
+@app.route("/conjurer", methods=["POST"])
+def answer_external_command():
+    record = json.loads(request.data)
+    app.logger.info(record)
+    app.logger.info("DATA RECEIVED")
+
+    
+    while True:
+        try:
+            q_data = internal_q.get(block=False)
+        except Empty:
+            break
+    if q_data.uuid in record:
+        pass
+
+    return jsonify("SUCCESS")
+
+@app.route("/conjurer", methods=["GET"])
+def check_alive():
+    return jsonify("ALIVE")
+
+def flask_debug(_logger):
+    """
+    The `flask_debug` function starts a Flask application in debug mode without using the reloader.
+    Do not use for production for fucks sake.
+    """
+    _logger.info("Attempt debug")
+    # trunk-ignore(bandit/B201)
+    app.run(debug=True, use_reloader=False, host=HOST_ADDRESS, port=PORT_ADDRESS)
+
+
+def waitress_run(_logger):
+    """
+    The `waitress_run` function serves the `app` on host "0.0.0.0"
+    and port 5000 using the Waitress WSGI server.
+    """
+    _logger.info("Attempt waitress")
+    serve(app, host=HOST_ADDRESS, port=PORT_ADDRESS)
+
+def scan_queue(_logger):
+    while True:
+        data = OUT_COMM_Q.get()
+        _logger.info(data)
+        internal_q.put(data)
+        query_list.append(data)
+
+def comm_subroutine(logger):
+    #logger.setLevel(logging.DEBUG)
+    logger.info("Started")
+    threads = []
+    threads.append(threading.Thread(target=flask_debug, args=(logger,)))
+    #threads.append(threading.Thread(target=waitress_run, args=(logger,)))
+    threads.append(threading.Thread(target=scan_queue,args=(logger,)))
+
+    for worker in threads:
+        worker.start()
+    for worker in threads:
+        worker.join()
+
+if __name__ == "__main__":
+    logging_client = logging.getLogger(__name__)
+    comm_subroutine(logging_client)

conjurer_librarian/conjurer_librarian.py

@@ -5,72 +5,103 @@ import time
 from flask import Flask, jsonify, request
 import netrc
 import json
+import asyncio
 from json.decoder import JSONDecodeError
-from platform import uname
-from sys import platform
 from urllib.request import urlopen
 from habanero import Crossref
 from waitress import serve
+from queue import Queue
+import search_bot
+import requests
+from typing import Callable
+from uuid import UUID
+NETRC_FILE = r"C:\Users\Activcom.pl\.netrc"
 
-if platform in ("linux", "linux2"):
-    SEPARATOR_FILE_PATH = "/"
-    if "microsoft-standard" in uname().release:
-        NETRC_FILE = "/home/mtuszowski/.netrc"
-    else:
-        NETRC_FILE = "/home/pi/.netrc"
-elif platform == "win32":
-    NETRC_FILE = "C:\\Users\\mtusz\\.netrc"
-
-#HOST_ADDRESS = "192.168.1.191"
-HOST_ADDRESS = "0.0.0.0"
+HOST_ADDRESS = "192.168.1.192"
 PORT_ADDRESS = 5001
+MAIN_BOT_ADDRESS = "http://192.168.1.191:5000"
+SEND_RESULTS = "/conjurer"
+BDSM_UUID_TEST = '96b7f85a-1142-4908-8986-62a2ea25a147'
+
+ENCODING = "utf-8"
+app = Flask(__name__)
+
+librarian_queue = Queue()
+librarian_list = []
 
 class Librarian(object):
 
-    def __init__(self) -> None:
+    def __init__(self, app, query, uuid) -> None:
         netrc_mod = netrc.netrc(NETRC_FILE)
         authTokens = netrc_mod.authenticators("crossref")
         self.cr = Crossref(mailto=authTokens[0])
-        self.limit = 1000
+        self.query = query
+        self.uuid = str(uuid)
+        self.limit = 3000
         self.fetched = 0
         self.hit = 0
         self.total = 0
+        self.app = app
+        self.logger = app.logger
+        self.live_results = []
+        self.search_result_from_cr = {}
+        self.done = False
 
-    def search_crossref(self, query, uuid):
-        search_result_from_cr = self.cr.works(query = query, limit = self.limit)
-        self.total = search_result_from_cr['message']['total-results']
-        self.fetched = len(search_result_from_cr['message']['items'])
-        #add to sql with UUID
+    async def search_crossref(self, query):
+        self.app.logger.info("STARTED SEARCH")
+        result = self.cr.works(query = query, limit = 1000)
+        self.search_result_from_cr.update(result)
+        self.total = result['message']['total-results']
+        self.fetched += len(result['message']['items'])
+        self.logger.info (self.total)
+        self.logger.info(self.fetched)
+        while self.total > self.fetched and self.limit > self.fetched:
+            result = self.cr.works(query = query, limit = 1000)
+            self.search_result_from_cr.update(result)
+            self.total = result['message']['total-results']
+            self.fetched += len(result['message']['items'])
+            self.logger.info (self.total)
+            self.logger.info(self.fetched)
+
+            time.sleep(1)
+        self.app.logger.info("CROSSREF DONE")
+        #TODO: add to sql with UUID
         #sql row - title, doi, author, uuid, last retrieved, is_available
-        with open("results.json", "r+", encoding="UTF-8") as data_file:
+        
+        with open("rr_results.json", "r+", encoding="utf-8") as data_file:
             # First we load existing data into a dict.
             try:
                 file_data = json.load(data_file)
             except JSONDecodeError:
-                file_data = []
-            # Join new_data with file_data inside emp_details
+                file_data = {}
             summarized_results = []
-            for item in search_result_from_cr['message']['items']:
+            self.app.logger.info("REFINE: Removing all derived works from the list")
+            for item in self.search_result_from_cr['message']['items']:
                 container = "NO"
                 if "container-title" in item:
                     container = "YES"
-                summarized_results.append({"DOI" : item['DOI'], "title": item['title'] if 'title' in item else None,"type": item['type'] if 'type' in item else None, "container": container})
-            result = {"UUID": uuid, "total_results": search_result_from_cr['message']['total-results'], "on_page" : 1, "summary":summarized_results, "results":search_result_from_cr['message']['items'],  }
-            file_data.append(result)
-            print("Saving")
+                else:
+                    summarized_results.append({"DOI" : item['DOI'], "title": item['title'] if 'title' in item else None,"type": item['type'] if 'type' in item else None, "container": container})
+            result = {self.uuid : {"total_results": self.search_result_from_cr['message']['total-results'], "on_page" : 1, "summary":summarized_results, "results":self.search_result_from_cr['message']['items']}}
+            if file_data:
+                file_data.update(result)
+            else:
+                file_data = result
+            
+            self.app.logger.info("REFINE: Dumping to file")
+            data_file.truncate(0)
             data_file.seek(0)
-            # convert back to json.
             json.dump(file_data, data_file, indent=4)
             return result
 
+    async def check_dois_in_local_db(self, doi):
+        self.app.logger.info("REFINE: Running search in the backend app")
+        coro = asyncio.to_thread(search_bot.search_for_doi, doi, self.live_results, self.app.logger)
+        result = await coro
+        return result
 
-    def check_if_exists(self,page_url):
-
-        
-
-        return None
-
-        in_scihub_db = True
+    async def check_if_exists_brute_force(self, page_url):
+        self.app.logger.error("REFINE: Brute force search!")
         if not page_url.startswith(("http:", "https:")):
             raise ValueError("URL must start with 'http:' or 'https:'")
         # trunk-ignore(bandit/B310)
@@ -79,70 +110,86 @@ class Librarian(object):
             text = data.decode("utf-8")
             for line in text.splitlines():
                 if re.match(r".*Unfortunately, Sci-Hub doesn't have the requested document.*", line):
-                    return None
-        if in_scihub_db:
-            return text
-        return None
-
-    def download(self, page_url):
-        if response := self.check_if_exists(page_url):
-            for line in response.splitlines():
+                    return False
                 if m := re.match(r".*<embed type=\"application.pdf\"\s*src=\"(.*\.pdf)",line):
-                    return "https:" + str(m.group(1))
-        return None
+                    direct_download_link = "https:" + str(m.group(1))
+                    self.app.logger.info (direct_download_link)
+                    return True
+        return False
 
-    def provide_sci_hub_link(self, page_url):
-        if self.check_if_exists(page_url):
-            return page_url
-        return None
+    async def check_if_exists(self, doi, brute_force = False):
+        result = {}
+        result = await self.check_dois_in_local_db(doi)
+        if brute_force and not result:
+            for item in doi:
+                item_link = "https://sci-hub.se/" + item[0]
+                tmp = {"DOI" : item["DOI"], "exists" : await self.check_if_exists_brute_force(item_link), "data": item}
+                result.update(tmp)
+        return result
 
-    def refine_query(self, unrefined_reqult):
-        print("Refine result")
+    async def refine_query(self, unrefined_reqult, brute_force = False):
+        temp = []
         refined_result = []
-        for item in unrefined_reqult["summary"]:
-            if item["container"] == "NO" and item["title"]:
-                item_link = "https://sci-hub.se/" + str(item['DOI'])
-                if self.check_if_exists(item_link):
-                    refined_result.append(item)
-                    print ("HIT")
-                    print(item["title"])
-                time.sleep(0.01)
-        print("Existing hits")
+        for key in unrefined_reqult:
+            self.logger.info("KEY:")
+            self.logger.info(key)
+
+        for item in unrefined_reqult[self.uuid]["summary"]:
+            if item["container"] == "NO" and item["title"]:                
+                refined_result.append(item)
+
+        for item in refined_result:
+            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
-
-    def answer_query(self, query, uuid):
+    #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
-        print("Answer result")
+        self.logger.info(f"Search started {self.uuid}")
         uuid_found = False
         database = None
-        with open("results.json", "r", encoding="UTF-8") as new_file:
+        query = self.query
+        with open("cr_results.json", "r+", encoding="utf-8") as cr_file:
             #tu logika komunikacji z SQL i cała logika związana z wyszukaniem
             #ale tymczasowo plik
             try:
-                database =  json.load(new_file)
+                database =  json.load(cr_file)
             except JSONDecodeError:
                 pass
-        if database:
-            for item in database:
-                if item["UUID"] == uuid:
-                    print (item["UUID"])
-                    uuid_found = True
-                    return self.refine_query(item)
 
-        if not uuid_found:
-            answer = self.search_crossref(query=query, uuid=uuid)
-            return self.refine_query(answer)
+            if database:
+                for item in database.keys():
+                    if item in self.uuid:
+                        self.result = await self.refine_query(database)
+                        return self.result
+                answer = await self.search_crossref(query=query)
+                database.update(answer)
+                cr_file.truncate(0)
+                cr_file.seek(0)
+                json.dump(database,cr_file)
+                self.result = await self.refine_query(answer)
+                return self.result
+            else:
+                answer = await self.search_crossref(query=query)
+                database = answer
+                cr_file.truncate(0)
+                cr_file.seek(0)
+                json.dump(database,cr_file)
+                self.result = await self.refine_query(answer)
+                self.app.logger.info("Refined")
+                self.done = True
+                return self.result
 
-#example not exists
-#sample_non_existing_page_url = "https://sci-hub.se/10.4324/9781003006237"
-#sample usage to download
-#sample_existing_page_url = "https://sci-hub.se/10.1057/9781137435026"
-#print(cl.download(sample_existing_page_url))
-#cl.search_crossref("BDSM")
-
-app = Flask(__name__)
+#============================= FLASK INTERNALS===============================
 def flask_debug():
     """
     The `flask_debug` function starts a Flask application in debug mode without using the reloader.
@@ -158,35 +205,101 @@ def waitress_run():
     """
     serve(app, host=HOST_ADDRESS, port=PORT_ADDRESS)
 
+class BackgroundTaskSearch(threading.Thread):
+    def run(self):
+        loop = asyncio.new_event_loop()  # loop = asyncio.get_event_loop()
+        loop.run_until_complete(self._run())
+        loop.close()
+
+    async def _run(self):
+        while True:
+            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("s_results.json", "r+", encoding="utf-8") as s_file:
+                try:
+                    database =  json.load(s_file)
+                except JSONDecodeError:
+                    pass
+                if database:
+                    database.extend(result)
+                else:
+                    database = result
+                self.app.logger.info("DUMPING DATA")
+                s_file.truncate(0)
+                s_file.seek(0)
+                json.dump(database, s_file)
+            self.app.logger.info("FINISHED")
+
+            self.app.logger.info(result)
+            coroutine = asyncio.to_thread(
+            requests.post,
+            f"{MAIN_BOT_ADDRESS}{SEND_RESULTS}",
+            json=result,
+            timeout=360,
+    )       
+            self.app.logger.info("SENT")
+            _ = await coroutine
+            self.app.logger.info("SEND CONFIRMED")
+            time.sleep(1)
+            
+#==================================SERVER ROUTES==========================================
 @app.route("/query", methods=["POST"])
-def query_database():
+async def query_database():
     record = json.loads(request.data)
     app.logger.info(record)
     app.logger.info(record["query"])
     app.logger.info(record["UUID"])
-    cl = Librarian()
-    answer_data = cl.answer_query(record["query"], record["UUID"])
-    app.logger.info(answer_data)
-    summary = (cl.hit,cl.fetched, cl.total, record["UUID"])
-    answer_data = (summary,answer_data)
+    uuid = record["UUID"]
+    #TODO: TESTING ONLY======================
+    uuid = UUID('{12345678-1234-5678-1234-567812345678}')
+
+    cl = Librarian(app, record["query"], uuid)
+    librarian_queue.put(cl)
+    librarian_list.append(cl)
+    answer_data = (record["query"], record["UUID"], librarian_queue.qsize())
     return_data = (
         jsonify(isError=False, message="Success", statusCode=200, data=answer_data),
         200,
     )
     return return_data
 
+@app.route("/get_partial_result", methods=["POST"])
+async def get_partial():
+    record = json.loads(request.data)
+    app.logger.info(record)
+    app.logger.info(record["UUID"])
+    for lib in librarian_list:
+        if lib.uuid == record["UUID"]:
+            answer_data = lib.live_results
+            break
+    return_data = (
+        jsonify(isError=False, message="Success", statusCode=200, data=answer_data),
+        200,
+    )
+    return return_data
+
+#=======================================MAIN===================================================
 if __name__ == "__main__":
     logger = logging.getLogger('conjurer_musician')
     logger.setLevel(logging.DEBUG)
     threads = []
     #threads.append(threading.Thread(target=waitress_run))
     threads.append(threading.Thread(target=flask_debug))
-
+    bgtask = BackgroundTaskSearch()
+    bgtask.app = app
+    threads.append(bgtask)
+    i = 0
     for worker in threads:
-        worker.start()
+        try:
+            logger.info(f"App number: {i}")
+            i+=1
+            worker.start()
+        except RuntimeError as e:
+            logger.error("Exploded")
+            print(str(e))
     for worker in threads:
         worker.join()
-
-#1. robi liste wyników i ją wyświetla - ale tylko tytuły niezależnych dzieł - jeśli jest coś w containerze to trza pogrupować
-#2. Tworzy przyciski pozwalające sprawdzić czy jest dostępne - i jeśli jest zwraca link
-#3. Jest też przycisk "more" - wyszukujemy po 100, wyświetlamy po 20. Po 100 jak jeszcze to dopiero odpytujemy crossref przyciski strzałek przesuwają po obecnej porcji danych

conjurer_librarian/search_bot.py

@@ -0,0 +1,106 @@
+from queue import Queue, Empty
+from threading import Thread
+q = Queue()
+
+#TODO: TO w klasie jako parametry wybierane czy robimy pelen run czy testowy
+#MAXTHREADS = 3
+MAXTHREADS = 1
+
+#DATABASE_PATH = r'C:\\Database\\chunks_1\\'
+DATABASE_PATH = r'C:\\Database\\chunks\\'
+
+ENCODING = "utf-8"
+CHUNk = "_chunk.txt"
+_sentinel = object()
+result_list = []
+
+#TODO: Wpiemdolić to wszystko w klasę z loggerem przysłanym z góry
+
+def producer(out_q, control_q, filename):
+    with open(DATABASE_PATH + filename, "r+", encoding=ENCODING) as operated_file:
+        lines = operated_file.readlines()
+        for line in lines:
+            out_q.put(line)
+            try:
+                check = control_q.get(block=False)
+            except Empty:
+                check = False
+                pass
+            if check is _sentinel:
+                print("TERM signal received")
+                control_q.put(check)
+                break
+    print(f"Worker finished: {filename}")
+    out_q.put(_sentinel)
+ 
+#IMPORTANT!!! ONLY ONE CONSUMER THREAD AS WE ARE NOT PUTTING SENTINELS BACK
+#TODO: Change doi to tuple list and only put term when all doi are found => (doi, True)
+def consumer(in_q, control_q, doi, live_results, logger):
+    sentinels = 0
+    for item in doi:
+        result_list.append({"DOI" : item[0], "exists" : False, "data" : item[1]})
+    while True:
+        #if in_q.empty():
+        #    print("Queue empty!")
+        #if in_q.full():
+        #    print("Queue full!")
+        done_check = True
+        data = in_q.get()
+        if data is _sentinel:
+            logger.info("Worker finished")
+            sentinels += 1
+        else:
+            for item in result_list:
+                if item["DOI"] in data:
+                    logger.info(data)
+                    logger.info("HIT")
+                    item["exists"] = True
+                    live_results.append(item)
+                done_check = done_check and item["exists"]
+            if done_check:
+                control_q.put(_sentinel)
+        if sentinels >= MAXTHREADS:
+            logger.info("All workers finished")
+            break
+
+
+def search_for_doi(doi, live_results, logger):
+    threads = []
+    work_q = Queue(maxsize=200000)
+    control_q = Queue()
+    t_cons = Thread(target = consumer, args = (work_q, control_q, doi, live_results, logger))
+    logger.info("Consumer thread created")
+    threads.append(t_cons)
+    for i in range(0,MAXTHREADS):
+        #filename = "test" + str(i) + CHUNk
+        filename = "40" + CHUNk
+        #filename = str(i) + CHUNk
+        logger.info(f"Creating worker thread no: {i}")
+        threads.append(Thread(target = producer, args = (work_q, control_q, filename)))
+    for worker in threads:
+        worker.start()
+    for worker in threads:
+        worker.join()
+    return result_list
+
+
+
+if __name__ == "__main__":
+    import logging
+    logger = logging.getLogger()
+    logger.setLevel("DEBUG")
+    h1 = logging.StreamHandler()
+    logger.addHandler(h1)
+    logger.info("TEST RUN")
+    live_result = []
+    logger.info(search_for_doi([
+        ("10.1002/9781118786352.wbieg0998.pub2","DATA"),
+        ("10.1002/j.2050-0416.2002.tb00563.x","DATA"),
+        ("10.1111/j.1365-2958.1994.tb00448.x","DATA"),
+        ("10.2165/00128415-200309690-00017","DATA"),
+        ("10.5772/48313","DATA"),
+        ("10.15803/ijnc.7.2_419","DATA"),
+        ("10.1051/0004-6361/201321596e","DATA"),
+        ("10.2307/40835941","DATA"),
+        ], live_result, logger))
+    logger.info(live_result)

file_webservice/conjurer_musician.py

@@ -269,7 +269,6 @@ if __name__ == "__main__":
     threads = []
     threads.append(threading.Thread(target=waitress_run))
     threads.append(threading.Thread(target=thread_rescan))
-    #threads.append(threading.Thread(target=flask_debug))
 
     for worker in threads:
         worker.start()

requirements_bot.txt

@@ -10,4 +10,6 @@ PyPDF2
 requests
 spotipy
 tiktoken
-PyNaCl
+PyNaCl
+flask
+waitress

s_results.json

@@ -0,0 +1 @@
+{"12345678-1234-5678-1234-567812345678": [{"DOI": "10.1057/9781137435026", "title": ["Queer BDSM Intimacies"], "type": "book", "container": "NO"}]}

utils/cut_large_file.py

@@ -39,9 +39,12 @@ def smart_write(lines, file):
     :return: The function `smart_write` returns the size of the encoded bytes of the input `lines` in
     bytes.
     """
-    encoded_bytes = lines.encode("utf-8")
-    size_in_bytes = len(encoded_bytes)
-    file.write(lines)
+    try:
+        encoded_bytes = lines.encode("utf-8")
+        size_in_bytes = len(encoded_bytes)
+        file.write(lines)
+    except:
+        return 0
 
     return size_in_bytes
 
@@ -65,8 +68,9 @@ def cut_my_life_into_pieces(source, output_folder, output_file):
     with open(source, encoding="utf-8") as infile:
         output_size = 0
         output_no = 0
-
+        lineno = 0
         for line in infile:
+            lineno += 1
             if output_size > FILE_SIZE_LIMIT:
                 output_no += 1
                 output_size = 0
@@ -76,7 +80,12 @@ def cut_my_life_into_pieces(source, output_folder, output_file):
                 "a",
                 encoding="utf-8",
             ) as op_file:
-                output_size += smart_write(line, op_file)
+                tmp =  smart_write(line, op_file)
+                if tmp > 0:
+                    output_size += smart_write(line, op_file)
+                else:
+                    print (f'Error at {lineno}')
+                    print(line)
 
 if __name__ == "__main__":
     cut_my_life_into_pieces(SOURCE,OUTPUT_FOLDER,OUTPUT_FILE_SCHEMA)