.

.gitignore

@@ -2,3 +2,5 @@
 data/recs.json
 
 backup.sql
+
+app/data/

app/application/__init__.py

@@ -5,12 +5,12 @@ import os
 from json import dumps
 import application.endpoints as endpoints
 import application.config as config
-from application.db import Session 
 
 app = Flask(__name__)
 api = Api(app, version='1', contact={"name":""}, license={"name":"Online Dienst Dokumentation"}, api_spec_url='/api/swagger')
 
 api.add_resource(endpoints.RecipeList,'/api/v1/recipe/')
+api.add_resource(endpoints.Images,'/api/v1/images/<string:id>')
 
 @app.route("/")
 def index():

app/application/db2.py

@@ -10,7 +10,7 @@ import time
 engine = db.create_engine('mysql+mysqldb://root@server/fs2?charset=utf8mb4', echo=False, encoding="utf8", pool_size=1000, max_overflow=0)
 
 Base = declarative_base()
-Session = sessionmaker(bind=engine)
+Session = sessionmaker(bind=engine, autoflush=False)
 
 # https://docs.sqlalchemy.org/en/13/orm/basic_relationships.html#association-object
 
@@ -20,7 +20,7 @@ class Recipe(Base):
     name = Column('name', Text)
     instructions = Column('instructions', Text)
     url = Column('url', Text)
-    img = Column('img', LargeBinary)
+    img = Column('img', LargeBinary(length=(2**32)-1))
     imgURL = Column('imgURL', Text)
     ingredient = relationship("RecIngred", back_populates="recipe")
 
@@ -30,25 +30,26 @@ class RecIngred(Base):
     ingredient_amount = Column('ingredient_amount', Text)
     ingredient_amount_mu = Column('ingredient_amount_mu', Text)    # measurement unit
 
-    recipe_id = Column(Integer, ForeignKey('recipe.recipe_id'), primary_key=True)
-    ingredient_name = Column(String(50), ForeignKey('ingredient.name'), primary_key=True)
-
     recipe = relationship("Recipe", back_populates="ingredient")
     ingredient = relationship("Ingredient", back_populates="recipe")
 
+    recipe_id = Column(Integer, ForeignKey('recipe.recipe_id'), primary_key=True)
+    ingredient_name = Column(String(200), ForeignKey('ingredient.name'), primary_key=True)
+
+
 class Ingredient(Base):
     __tablename__ = "ingredient"
-    name = Column('name', String(50), primary_key=True)
+    name = Column('name', String(200), primary_key=True)
     recipe = relationship("RecIngred", back_populates="ingredient")
     trunks = relationship("IngredTrunk", back_populates="ingredient")
 
 class IngredTrunk(Base):
     __tablename__ = 'ingredtrunk'
-    ingredient_name = Column(String(50), ForeignKey('ingredient.name'), primary_key=True)
-    trunk_name = Column(String(50), ForeignKey('ingredient.name'), primary_key=True)
+    ingredient_name = Column(String(200), ForeignKey('ingredient.name'), primary_key=True)
+    trunk_name = Column(String(50), ForeignKey('trunk.name'), primary_key=True)
 
-    ingredient = relationship("Ingredient", back_populates="trunk")
-    trunk = relationship("Trunk", back_populates="ingredient")
+    ingredient = relationship("Ingredient", back_populates="trunks")
+    trunk = relationship("Trunk", back_populates="ingredients")
 
 class Trunk(Base):
     __tablename__ = "trunk"

app/application/endpoints.py

@@ -5,9 +5,9 @@ import requests
 import application.config as config
 import json
 import base64
-from application.db import Session, Recipe, Ingredient
-import search
-import migrate
+from application.db2 import Session, Recipe
+import application.search as search
+import background.migrate as migrate
 import time
 
 class RecipeList(Resource):
@@ -20,15 +20,15 @@ class RecipeList(Resource):
         args = parser.parse_args()
         ingreds = args["ingred"]
 
-        ingreds = [migrate.stemWord(ingred)[0] for ingred in ingreds + search.defaultArr] 
+        ingreds = [migrate.stem(ingred)[0] for ingred in ingreds] 
 
         start = time.time()
-        indx = search.fastes(ingreds)
+        indx = search.search2(ingreds)
         end = time.time()
         print("get recipes",end - start, "\n")  
 
-        start = time.time()
-        recs = search.getRecDict(indx, ingreds)
+        #start = time.time()
+        recs = search.getRecDict2(indx, ingreds)
         end = time.time()
         print("calc overlay",end - start, "\n")  
 
@@ -36,5 +36,15 @@ class RecipeList(Resource):
         g.session.close()
         return flask.make_response(flask.jsonify({'data': recs}), 200)
 
+class Images(Resource):
+    def get(self, id = None):
+        if id is None:
+            flask.make_response(flask.jsonify({'error': "No ID supplied"}), 401)
+        
+        session = Session()
+        image = session.query(Recipe.img).filter(Recipe.recipe_id == id).first()[0]
+        image = base64.b64decode(image.encode())
+        return flask.Response(image,  mimetype='image/png')
+
 
 

app/application/search.py

@@ -0,0 +1,89 @@
+
+import application.db2 as db
+from flask import g
+import nltk as nltk
+from nltk.corpus import stopwords
+import time
+import heapq
+from collections import Counter
+import background.migrate
+
+def search2(inputArr):
+    indx = {}
+    dbSession = db.Session()
+    for inpu in inputArr:
+        x = dbSession.query(db.Trunk.name, db.Recipe.recipe_id).filter(db.Trunk.name == inpu).join(db.IngredTrunk).join(
+            db.Ingredient).join(db.RecIngred).join(db.Recipe).all()
+
+        indx[inpu] = [str(y[1]) for y in x]
+
+    return(indx)
+
+
+def stemInput(inputArr):
+    inputArr2 = []
+
+    snowball = nltk.SnowballStemmer(language='german')
+    stopset = set(stopwords.words('german'))
+    for word in inputArr:
+        if word in stopset:
+            continue
+        inputArr2.append(snowball.stem(word))
+    return inputArr2
+#
+
+def getRecDict2(indx, inputArr):
+    dbSession = db.Session()
+
+    outDict = {}
+    # 2d to 1d
+    indx = sum(indx.values(), [])
+    k = Counter(indx)
+    indx = k.most_common(1000)
+    indx = dict(indx)
+
+    ingred = [x for x in dbSession.query(db.Recipe.recipe_id, db.IngredTrunk.trunk_name, db.IngredTrunk.ingredient_name ).filter(db.Recipe.recipe_id.in_(indx.keys())).join(db.RecIngred).join(db.Ingredient).join(db.IngredTrunk).all()]
+    ingredDict = {}
+    for k,v, i in ingred:
+        if k not in ingredDict:
+            ingredDict[k] = {}
+        if i not in ingredDict[k]:
+            ingredDict[k][i] = []
+
+        ingredDict[k][i].append(v)
+    inputArr += defaultArr
+    for key, value in ingredDict.items():
+        overlay = calcOverlay2(inputArr, value)
+        while overlay in outDict.keys():
+            overlay -= 0.0001
+        outDict[overlay] = int(key)
+
+    outDict2 = {}
+    for key in heapq.nlargest(20, outDict.keys()):
+        key2 = outDict[key]
+        rec = dbSession.query(db.Recipe).filter(db.Recipe.recipe_id == key2).first()
+        outDict2[key] = (key2, rec.name, rec.url,  [r[0] + ": " + r[1] for r in dbSession.query(db.Ingredient.name,
+                                                                                                db.RecIngred.ingredient_amount).join(db.RecIngred).join(db.Recipe).filter(db.Recipe.recipe_id == key2).all()], rec.img.decode('utf-8'))
+    return outDict2
+
+def stem(l1):
+    snowball = nltk.SnowballStemmer(language='german')
+    stopset = set(stopwords.words('german'))
+    stopset |= set("(),")
+    l1 = [snowball.stem(l) for l in l1]
+    return l1
+
+def calcOverlay2(l1, l2):
+    counter = 0
+    for ll in l2.values():
+        for l in ll:
+            if l in l1:
+                counter += 1
+                break
+
+    counter = counter / len(l2)
+    return counter
+
+# it is assumed that everyone has this
+defaultArr = ["Wasser", "salz", "pfeffer"]
+

app/application/static/main.js

@@ -58,7 +58,7 @@ function renderRecipeList(data){
             )
             
             recString = `
-                <a href="${data1[2]}"> 
+                <a href="${data1[2]}" target="_blank"> 
                     <div class="card text-white bg-primary mb-3" style="max-width: 100%">
                         <div class="card-body recipe-container">
                             <div class="row">

app/background/migrate.py

@@ -0,0 +1,103 @@
+import json
+import cv2
+import base64
+import nltk as nltk
+from nltk.corpus import stopwords
+#import db as db1
+#import db2 as db2
+
+def stemWord(word):
+    try:
+        arr = []
+        stopset = set(stopwords.words('german'))
+        stopset |= set("(),")
+        snowball = nltk.SnowballStemmer(language='german')
+        for token in nltk.word_tokenize(word): 
+            if token in stopset or len(token) < 4:
+                continue
+            stemmed = snowball.stem(token)
+            arr.append(stemmed)
+        if len(arr) == 0:
+            arr.append("")
+        return arr
+    except:
+        return [""]
+
+#migrate('./data/recs.json')
+
+def migrateRecsDb1ToDb2():
+
+    session1 = db1.Session()
+    session2 = db2.Session()
+ 
+
+    count = 0
+    length = session1.query(db1.Recipe).count()
+    for r1 in list(session1.query(db1.Recipe).all())[int(length/2):]:
+        try:
+            if not bool(session2.query(db2.Recipe).filter(db2.Recipe.name == r1.name).first()):
+                r2 = db2.Recipe(name=r1.name, instructions=r1.instructions, url=r1.url, img=r1.img)
+                
+                for ingred in r1.ingredient:
+                    ri2 = db2.RecIngred()
+                    ingredient2 = session2.query(db2.Ingredient).filter(db2.Ingredient.name == ingred.name).first()
+                    if ingredient2 is None:
+                        ingredient2 = db2.Ingredient(name=ingred.name)
+                    ri2.ingredient_amount = ingred.ingredient_amount
+                    ri2.ingredient = ingredient2
+                    r2.ingredient.append(ri2)
+                
+                session2.add(r2)
+                session2.commit()
+            
+        except:
+            session1 = db1.Session()
+            session2 = db2.Session()
+
+        count+=1
+        print(count/length)
+
+def TrunkDb2():
+    session2 = db2.Session()
+ 
+    count = 0
+    length = session2.query(db2.Ingredient).count()
+    for i2 in session2.query(db2.Ingredient).all():
+        try:
+            for trunk1 in stem(i2.name):
+
+                ri2 = db2.IngredTrunk()
+                trunk = session2.query(db2.Trunk).filter(db2.Trunk.name == trunk1).first()
+                if trunk is None:
+                    trunk = db2.Trunk(name=trunk1)
+                
+                if session2.query(db2.IngredTrunk).filter(db2.IngredTrunk.ingredient_name == i2.name, db2.IngredTrunk.trunk_name == trunk1).first() is None:
+                    ri2.trunk = trunk
+                    i2.trunks.append(ri2)
+            
+                session2.commit()
+            
+        except Exception as e:
+            print(e)
+            session2 = db2.Session()
+
+        count+=1
+        print(count/length)
+
+
+def stem(l1):
+    '''Tokenize and stem word, result is 1d list'''
+    snowball = nltk.SnowballStemmer(language='german')
+    stopset = set(stopwords.words('german'))
+    stopset |= set("(),")
+    l2 = []
+
+    for token in nltk.word_tokenize(l1): 
+        token = snowball.stem(token)
+        if token in stopset or not token.isalpha() or len(token) < 2:
+            continue
+        l2.append(token)
+
+    return l2
+#migrateDb1ToDb2()
+#TrunkDb2()

app/run.py

@@ -0,0 +1,10 @@
+from application import app
+from application.search import defaultArr, stem
+import nltk
+
+nltk.download('stopwords')
+nltk.download('punkt')
+defaultArr = stem(defaultArr)
+app.run(host="0.0.0.0", port='5001', debug=False, threaded=True)
+
+

application/db.py

@@ -1,85 +0,0 @@
-import sqlalchemy as db
-from sqlalchemy import Column, String, Integer, Numeric, Table, DateTime, ARRAY, ForeignKey, create_engine, LargeBinary, Enum, Text
-from sqlalchemy.orm import sessionmaker, relationship, column_property
-from datetime import datetime
-from sqlalchemy.ext.declarative import declarative_base
-import enum
-from flask import Flask
-import time
-
-engine = db.create_engine('mysql+mysqldb://root@server/fs?charset=utf8mb4', echo=False, encoding="utf8", pool_size=1000, max_overflow=0)
-
-Base = declarative_base()
-Session = sessionmaker(bind=engine)
-
-# https://docs.sqlalchemy.org/en/13/orm/basic_relationships.html#association-object
-
-
-class Recipe(Base):
-    __tablename__ = "recipe"
-    recipe_id = Column('recipe_id', Integer, primary_key=True, autoincrement=True)
-    name = Column('name', Text)
-    instructions = Column('instructions', Text)
-    url = Column('url', Text)
-    img = Column('img', LargeBinary(length=(2**32)-1))
-    ingredient = relationship("Ingredient", backref="recipe")
-    trunk = relationship("Trunk", backref="recipe")
-
-    def ingredients(self):
-        l = []
-        for i in self.ingredient:
-            l.append(i.ingredients())
-        return l
-
-    def ingredientDict(self):
-        l = {}
-        for i in self.ingredient:
-            l[i.ingredients()] = [i.ingredient_amount, i.ingredient_amount_mu]
-        return l
-
-    def serialize(self):
-        ingredients = []
-
-        if self.img is not None:
-            img = self.img.decode('utf-8')
-        else:
-            img = None
-
-        data = {
-            "recipe_id": self.recipe_id,
-            "name":self.name,
-            "instructions":self.instructions,
-            "url": self.url,
-            "img": img,
-            "ingredients": self.ingredients()  
-        }
-        return data
-
-class Ingredient(Base):
-    __tablename__ = "ingredient"
-    ingredient_id = Column('ingredient_id', Integer,  primary_key=True, autoincrement=True)
-    name = Column('name', Text)
-    ingredient_amount = Column('ingredient_amount', Text)
-    ingredient_amount_mu = Column('ingredient_amount_mu', Text)    # measurement unit
-
-    recipe_id = Column(Integer, ForeignKey('recipe.recipe_id'))
-    
-
-class Trunk(Base):
-    __tablename__ = "trunk"
-    trunk_id = Column('trunk_id', Integer,  primary_key=True, autoincrement=True)
-    name = Column('name', Text)
-
-    recipe_id = Column(Integer, ForeignKey('recipe.recipe_id'))
-
-def initDB(counter):
-    try:
-        Base.metadata.create_all(engine)
-    except Exception as e:
-        print(e)
-        counter += 1
-        if counter < 13:
-            time.sleep(5)
-            initDB(counter)
-
-initDB(0)

ghostdriver.log

@@ -1,27 +0,0 @@
-[INFO  - 2020-04-20T20:11:22.708Z] GhostDriver - Main - running on port 63975
-[INFO  - 2020-04-20T20:11:24.918Z] Session [1c4df520-8343-11ea-a8a6-19d2d02d716b] - page.settings - {"XSSAuditingEnabled":false,"javascriptCanCloseWindows":true,"javascriptCanOpenWindows":true,"javascriptEnabled":true,"loadImages":true,"localToRemoteUrlAccessEnabled":false,"userAgent":"Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/538.1 (KHTML, like Gecko) PhantomJS/2.1.1 Safari/538.1","webSecurityEnabled":true}
-[INFO  - 2020-04-20T20:11:24.918Z] Session [1c4df520-8343-11ea-a8a6-19d2d02d716b] - page.customHeaders:  - {}
-[INFO  - 2020-04-20T20:11:24.918Z] Session [1c4df520-8343-11ea-a8a6-19d2d02d716b] - Session.negotiatedCapabilities - {"browserName":"phantomjs","version":"2.1.1","driverName":"ghostdriver","driverVersion":"1.2.0","platform":"windows-10-32bit","javascriptEnabled":true,"takesScreenshot":true,"handlesAlerts":false,"databaseEnabled":false,"locationContextEnabled":false,"applicationCacheEnabled":false,"browserConnectionEnabled":false,"cssSelectorsEnabled":true,"webStorageEnabled":false,"rotatable":false,"acceptSslCerts":false,"nativeEvents":true,"proxy":{"proxyType":"direct"}}
-[INFO  - 2020-04-20T20:11:24.918Z] SessionManagerReqHand - _postNewSessionCommand - New Session Created: 1c4df520-8343-11ea-a8a6-19d2d02d716b
-[ERROR - 2020-04-20T20:11:26.927Z] Session [1c4df520-8343-11ea-a8a6-19d2d02d716b] - page.onError - msg: Unhandled promise rejection TypeError: undefined is not an object (evaluating 'w.set')
-
-  phantomjs://platform/console++.js:263 in error
-[ERROR - 2020-04-20T20:11:26.927Z] Session [1c4df520-8343-11ea-a8a6-19d2d02d716b] - page.onError - stack:
-  (anonymous function) (https://www.chefkoch.de/rezepte/207581086939665/Nannie-s-schnelle-Kuchen.html:8626)
-
-  phantomjs://platform/console++.js:263 in error
-[ERROR - 2020-04-20T20:11:28.646Z] WebElementLocator - _handleLocateCommand - Element(s) NOT Found: GAVE UP. Search Stop Time: 15874134[INFO  - 2020-04-20T20:12:30.427Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:13:35.766Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:17:30.429Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:18:35.767Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:22:30.429Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:23:35.768Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:27:30.429Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:28:35.771Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:32:30.439Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:33:35.774Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:37:30.439Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:38:35.774Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:42:30.439Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:43:35.784Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:47:30.439Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:48:35.787Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:52:30.439Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:53:35.788Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T20:57:30.440Z] SessionManagerReqHand - [INFO  - 2020-04-20T20:58:35.788Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:02:30.440Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:03:35.789Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:07:30.445Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:08:35.789Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:12:30.457Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:13:35.788Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:17:30.457Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:18:35.798Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:22:30.459Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:23:35.803Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:27:30.460Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-OW
-O  - 2020-04-20T21:22:12.188Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:27:12.188Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-ns - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:07:14.184Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:12:14.184Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:17:14.190Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:22:14.192Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:27:14.196Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-242Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:19:25.242Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:24:25.249Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-[INFO  - 2020-04-20T21:27:31.799Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW
-179Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:11:22.721Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:13:54.176Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:16:22.724Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:18:54.177Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:21:22.725Z] SessionManagerReqHand - _cleanupWindowlessSes[INFO  - 2020-04-20T21:23:54.181Z] SessionManagerReqHand - [INFO  - 2020-04-20T21:26:22.728Z] SessionManagerReqHand - _cleanupWindowlessSessions - Asynchronous Sessions clean-up phase starting NOW

migrate.py

@@ -1,52 +0,0 @@
-import json
-import cv2
-import base64
-import nltk as nltk
-from nltk.corpus import stopwords
-from application.db import Session, Recipe, Ingredient, Trunk
-
-def stemWord(word):
-    try:
-        arr = []
-        stopset = set(stopwords.words('german'))
-        stopset |= set("(),")
-        snowball = nltk.SnowballStemmer(language='german')
-        for token in nltk.word_tokenize(word): 
-            if token in stopset or len(token) < 4:
-                continue
-            stemmed = snowball.stem(token)
-            arr.append(stemmed)
-        if len(arr) == 0:
-            arr.append("")
-        return arr
-    except:
-        return [""]
-
-def migrate(path):
-    recs = ""
-    with open(path, encoding="utf-8") as file:
-        recs = json.load(file)
-    
-    dbSession = Session()
-    counter = 0
-    leng = len(recs)
-    for key, value in recs.items():
-        name=key
-        resString=value[0]
-        link=value[2]
-        img=value[3].encode()
-
-        r = Recipe(name=name, instructions=resString, url=link, img=img)
-
-        for x, y in value[1].items():
-            a = Ingredient(name=x, ingredient_amount=y)
-            r.ingredient.append(a)
-            for x in stemWord(a.name):
-                t = Trunk(name=x)
-                r.trunk.append(t)
-        dbSession.add(r)
-        dbSession.commit()
-        counter+=1
-        print(counter/leng)
-
-#migrate('./data/recs.json')

run.py

@@ -1,11 +0,0 @@
-from application import app
-import nltk
-from search import defaultArr, stem
-
-
-nltk.download('stopwords')
-nltk.download('punkt')
-delattr = stem(defaultArr)
-app.run(host="0.0.0.0", port='5001', debug=True, threaded=True)
-
-

search.py

@@ -1,102 +0,0 @@
-
-from application.db import Session, Recipe, Ingredient, Trunk
-from flask import g
-import nltk as nltk
-from nltk.corpus import stopwords
-import time
-import heapq
-from collections import Counter 
-
-
-
-def fastes(inputArr):
-    indx = {}
-    dbSession = g.session
-    for inpu in inputArr:
-        ids = [] 
-        for recipe_id in dbSession.query(Trunk.recipe_id).filter(Trunk.name == inpu).all():           
-            if str(recipe_id[0]) not in indx:
-                indx[str(recipe_id[0])] = 0
-
-            indx[str(recipe_id[0])] += 1
-    return(indx) 
-    
-def stemInput(inputArr):
-    inputArr2 = []
-
-    snowball = nltk.SnowballStemmer(language='german')
-    stopset = set(stopwords.words('german'))
-    for word in inputArr:
-        if word in stopset:
-             continue
-        inputArr2.append(snowball.stem(word))
-    return inputArr2
-#
-
-def getRecDict(indx, inputArr):
-    dbSession = g.session
-    
-    outDict = {}
-    k = Counter(indx) 
-    # Finding 1000 highest values TODO: this is not correct
-    indx = k.most_common(1000)  
-    indx = dict(indx)
-    for key, value in indx.items():
-        ingred = [x[0] for x in dbSession.query(Trunk.name).filter(Trunk.recipe_id==int(key)).all()]
-        outDict[calcOverlay(inputArr, ingred)] = int(key)
-        
-    outDict2 = {}
-    for key in heapq.nlargest(10, outDict.keys()):
-        key2 = outDict[key]
-        rec = dbSession.query(Recipe).filter(Recipe.recipe_id==key2).first()
-        outDict2[key] = (key2, rec.name, rec.url,  [r[0] + ": " + r[1] for r in dbSession.query(Ingredient.name, Ingredient.ingredient_amount).filter(Ingredient.recipe_id==key2).all()], rec.img.decode('utf-8'))
-    return outDict2
-
-def printDict(indx, inputArr):
-    outDict = getRecDict(indx, inputArr)
-    for key, value in sorted(outDict.items()):
-        if key >= 0.3:
-           
-            print(key, value[0], value[1])
-            for xx in value[2]:
-                print("\t", xx[0])
-    
-
-def stem(l1):
-    snowball = nltk.SnowballStemmer(language='german')
-    stopset = set(stopwords.words('german'))
-    stopset |= set("(),")
-
-    l1 =  [snowball.stem(l) for l in l1]
-    return l1
-
-def calcOverlay(l1, l2):
-    counter = 0
-    for l in l1:
-        if l not in defaultArr:
-            if l in l2:
-                #print(l)
-                counter +=1
-    counter = counter / len(l2)                 
-    return counter
-
-
-#inputArr = ["reis", "tofu", "bohnen", "kichererbsen", "hackfleisch"]
-defaultArr = ["Wasser", "salz", "pfeffer"] # it is assumed that everyone has this
-#inputArr += defaultArr
-maxMissing = 10
-#
-#stemmed = stemInput(inputArr)
-#
-#start = time.time()
-#indx = faster(stemmed)  
-#end = time.time()
-#printDict(indx)
-#print("\n", end - start, "\n")  
-#
-#
-#start = time.time()
-#indx = fastes(stemmed)
-#end = time.time()
-#printDict(indx)
-#print("\n", end - start, "\n")