search by phash

5 files changed, 24 insertions, 120 deletions

diff --git a/check/app/models/__init__.py b/check/app/models/__init__.py
index 93f5256..e69de29 100644
--- a/check/app/models/__init__.py
+++ b/check/app/models/__init__.py
@@ -1 +0,0 @@
-__init__.py
\ No newline at end of file
diff --git a/check/app/models/sql_factory.py b/check/app/models/sql_factory.py
index 499113d..3ed3af0 100644
--- a/check/app/models/sql_factory.py
+++ b/check/app/models/sql_factory.py
@@ -3,7 +3,7 @@ import glob
 import time
 import pandas as pd
 
-from sqlalchemy import create_engine, Table, Column, String, BigInteger, Integer, DateTime, func
+from sqlalchemy import create_engine, Table, Column, String, BigInteger
 from sqlalchemy.orm import sessionmaker
 from sqlalchemy.ext.declarative import declarative_base
 
@@ -17,12 +17,13 @@ connection_url = "mysql+mysqlconnector://{}:{}@{}/{}?charset=utf8mb4".format(
 )
 
 loaded = False
-Session = None
+engine = create_engine(connection_url, encoding="utf-8", pool_recycle=3600)
+Session = sessionmaker(bind=engine)
 
 Base = declarative_base()
 
-class ImageHashTable(Base):
-  __tablename__ = 'imagehashes'
+class PHashTable(Base):
+  __tablename__ = 'phashes'
   id = Column(Integer, primary_key=True)
   sha256 = Column(String(36, convert_unicode=True), nullable=False)
   phash = Column(BigInteger(blank=True), nullable=False)
@@ -35,116 +36,16 @@ class ImageHashTable(Base):
       'ext': self.ext,
     }
 
-def load_sql_datasets(replace=False, base_model=None):
-  engine = create_engine(connection_url, encoding="utf-8", pool_recycle=3600)
-  # db.set_character_set('utf8')
-  # dbc = db.cursor()
-  # dbc.execute('SET NAMES utf8;')
-  # dbc.execute('SET CHARACTER SET utf8;')
-  # dbc.execute('SET character_set_connection=utf8;')
-  Session = sessionmaker(bind=engine)
+def search_by_phash(phash, threshold=6):
+  connection = engine.connect()
+  cmd = 'SELECT phashes.*, BIT_COUNT(phash ^ :phash) as hamming_distance FROM images_image HAVING hamming_distance < :threshold ORDER BY hamming_distance ASC LIMIT 1'
+  matches = connection.execute(text(cmd), phash=phash, threshold=threshold)
+  return matches
 
-  for fn in glob.iglob(os.path.join(path, "*.csv")):
-    key = os.path.basename(fn).replace(".csv", "")
-    if replace:
-      print('loading dataset {}'.format(fn))
-      df = pd.read_csv(fn)
-      # fix columns that are named "index", a sql reserved word
-      df.reindex_axis(sorted(df.columns), axis=1)
-      columns = [column.name for column in ImageHashTable.__table__.columns]
-      df.columns = columns
-      df.to_sql(name=ImageHashTable.__tablename__, con=engine, if_exists='replace', index=False)
-  return dataset
-
-class SqlDataset:
-  """
-  Bridge between the facial information CSVs connected to the datasets, and MySQL
-  - each dataset should have files that can be loaded into these database models
-  - names will be fixed to work in SQL (index -> id)
-  - we can then have more generic models for fetching this info after doing a FAISS query
-  """
-  def __init__(self, name, engine=None, base_model=None):
-    self.name = name
-    self.tables = {}
-    if base_model is None:
-      self.engine = create_engine(connection_url, encoding="utf-8", pool_recycle=3600)
-      base_model = declarative_base(engine)
-    self.base_model = base_model
-
-  def describe(self):
-    """
-    List the available SQL tables for a given dataset.
-    """
-    return {
-      'name': self.name,
-      'tables': list(self.tables.keys()),
-    }
-
-  def get_identity(self, id):
-    """
-    Get an identity given an ID.
-    """
-    # id += 1
-    file_record_table = self.get_table('file_record')
-    file_record = file_record_table.query.filter(file_record_table.id == id).first()
-
-    if not file_record:
-      return None
-
-    identity_table = self.get_table('identity')
-    identity = identity_table.query.filter(identity_table.id == file_record.identity_id).first()
-
-    if not identity:
-      return None
-
-    return {
-      'file_record': file_record.toJSON(),
-      'identity': identity.toJSON(),
-      'face_roi': self.select('face_roi', id),
-      'face_pose': self.select('face_pose', id),
-    }
-
-  def search_name(self, q):
-    """
-    Find an identity by name.
-    """
-    table = self.get_table('identity')
-    identity_list = table.query.filter(table.fullname.ilike(q)).order_by(table.fullname.desc()).limit(15)
-    return identity_list
-
-  def search_description(self, q):
-    """
-    Find an identity by description.
-    """
-    table = self.get_table('identity')
-    identity_list = table.query.filter(table.description.ilike(q)).order_by(table.description.desc()).limit(15)
-    return identity_list
-
-  def get_file_records_for_identities(self, identity_list):
-    """
-    Given a list of identities, map these to file records.
-    """
-    identities = []
-    file_record_table = self.get_table('file_record')
-    for row in identity_list:
-      file_record = file_record_table.query.filter(file_record_table.identity_id == row.id).first()
-      if file_record:
-        identities.append({
-          'file_record': file_record.toJSON(),
-          'identity': row.toJSON(),
-        })
-    return identities
-
-  def select(self, table, id):
-    """
-    Perform a generic select.
-    """
-    table = self.get_table(table)
-    if not table:
-      return None
-    session = Session()
-    # for obj in session.query(table).filter_by(id=id):
-    #   print(table)
-    obj = session.query(table).filter(table.id == id).first()
-    session.close()
-    return obj.toJSON()
+def add_phash(sha256, phash, ext):
+  rec = PHashTable(
+    sha256=sha256, phash=phash, ext=ext,
+  )
+  session = Session()
+  session.add(rec)
+  session.flush()
diff --git a/check/app/server/__init__.py b/check/app/server/__init__.py
index 93f5256..e69de29 100644
--- a/check/app/server/__init__.py
+++ b/check/app/server/__init__.py
@@ -1 +0,0 @@
-__init__.py
\ No newline at end of file
diff --git a/check/app/server/api.py b/check/app/server/api.py
index 620e604..5c5b4bf 100644
--- a/check/app/server/api.py
+++ b/check/app/server/api.py
@@ -16,5 +16,11 @@ api = Blueprint('api', __name__)
 
 @api.route('/')
 def index():
-  """List the datasets and their fields"""
+  """API status noop"""
+  return jsonify({ 'status': 'ok' })
+
+@api.route('/v1/match/')
+def match():
+  """Search by image"""
+  # get threshold
   return jsonify({ 'status': 'ok' })
diff --git a/check/app/settings/__init__.py b/check/app/settings/__init__.py
index 93f5256..e69de29 100644
--- a/check/app/settings/__init__.py
+++ b/check/app/settings/__init__.py
@@ -1 +0,0 @@
-__init__.py
\ No newline at end of file