From 4d0addb87b416f0ac03ed5175a09671d5bebe8fa Mon Sep 17 00:00:00 2001
From: Damien De Paoli <ddp@depaoli.id.au>
Date: Thu, 16 Sep 2021 19:25:23 +1000
Subject: [PATCH] removing extra session.commits, and optimsing run_ai_on -
 dont reprocess faces if no new refimgs since last scan -- so we had set
 last_ai_scan on FILE as well

---
 pa_job_manager.py | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/pa_job_manager.py b/pa_job_manager.py
index 9c0bced..a528d82 100644
--- a/pa_job_manager.py
+++ b/pa_job_manager.py
@@ -19,7 +19,7 @@ DEBUG=1
 
 ### SQLALCHEMY IMPORTS ###
 from sqlalchemy.ext.declarative import declarative_base
-from sqlalchemy import Column, Integer, String, Sequence, Float, ForeignKey, DateTime, LargeBinary, Boolean
+from sqlalchemy import Column, Integer, String, Sequence, Float, ForeignKey, DateTime, LargeBinary, Boolean, func
 from sqlalchemy.exc import SQLAlchemyError
 from sqlalchemy.orm import relationship
 from sqlalchemy import create_engine
@@ -193,6 +193,7 @@ class File(Base):
     day = Column(Integer)
     woy = Column(Integer)
     last_hash_date = Column(Float)
+    last_ai_scan = Column(Float)
     faces = Column( LargeBinary )
     faces_created_on = Column(Float)
 
@@ -754,7 +755,6 @@ def JobForceScan(job):
     ProcessStorageDirs(job)
     FinishJob(job, "Completed (forced remove and recreation of all file data)")
     MessageToFE( job.id, "success", "Completed (forced remove and recreation of all file data)" )
-    session.commit()
     return
 
 ##############################################################################
@@ -1674,7 +1674,6 @@ def MatchRefimgToFace( refimg_id, face_id, face_dist ):
     session.query(FaceRefimgLink).filter(FaceRefimgLink.face_id==face_id).delete()
     rfl = FaceRefimgLink( refimg_id = refimg_id, face_id = face_id, face_distance=face_dist )
     session.add(rfl)
-    session.commit()
     return
 
 ####################################################################################################################################
@@ -1764,11 +1763,18 @@ def ScanFileForPerson( job, e, force=False ):
         for locn, face in zip( face_locations, unknown_encodings ):
             AddFaceToFile( locn, face, e.id, model.id )
         file_h.faces_created_on = time.time()
-    session.commit()
 
     faces = session.query(Face).join(FaceFileLink).filter(FaceFileLink.file_eid==e.id).all()
     # if there are no faces for this file, then dont go any futher
     if not faces:
+        file_h.last_ai_scan = time.time()
+        return
+
+    ri_newest = session.query(func.max(Refimg.created_on)).first()[0]
+    
+    # ri_newest has to exist, no ris and we dont process files but, if we have never scanned before, then last_ai_scan will be None
+    # if last_ai_scan is newer than the most recent refimg created, no need to look again, we have checked those refimgs in the last scan, just skip this file
+    if file_h.last_ai_scan and file_h.last_ai_scan > ri_newest:
         return
 
     dist={}
@@ -1784,6 +1790,7 @@ def ScanFileForPerson( job, e, force=False ):
 
     # record matches in DB...
     ProcessFaceMatches( job, dist, threshold, e, name )
+    file_h.last_ai_scan = time.time()
     return