From 92a4e77f9b5ef0693605ae50b7072809b140214e Mon Sep 17 00:00:00 2001
From: Will Thompson <will@willthompson.co.uk>
Date: Tue, 28 Jun 2016 15:59:16 +0100
Subject: [PATCH 1/3] chunkupload: typo: uplaod -> upload

---
 static/js/chunkupload.js | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/static/js/chunkupload.js b/static/js/chunkupload.js
index ba83a096..2d37d807 100644
--- a/static/js/chunkupload.js
+++ b/static/js/chunkupload.js
@@ -89,7 +89,7 @@ pandora.chunkupload = function(options) {
             }
         }, false);
         request.addEventListener('error', function (evt) {
-            that.status = 'uplaod failed';
+            that.status = 'upload failed';
             that.progress = -1;
             that.responseText = evt.target.responseText;
             done();
@@ -100,7 +100,7 @@ pandora.chunkupload = function(options) {
             done();
         }, false);
         var formData = new FormData();
-        
+
         Object.keys(options.data).forEach(function(key) {
             formData.append(key, options.data[key]);
         });
@@ -176,7 +176,7 @@ pandora.chunkupload = function(options) {
                 // failed to upload, try again in 5 second
                 retries++;
                 if (maxRetry > 0 && retries > maxRetry) {
-                    that.status = 'uplaod failed';
+                    that.status = 'upload failed';
                     that.progress = -1;
                     done();
                 } else {
@@ -195,7 +195,7 @@ pandora.chunkupload = function(options) {
             // failed to upload, try again in 3 second
             retries++;
             if (maxRetry > 0 && retries > maxRetry) {
-                that.status = 'uplaod failed';
+                that.status = 'upload failed';
                 that.progress = -1;
                 done();
             } else {

From 8d25e3be78e751f65375ae6cbe1a8001b83f3f32 Mon Sep 17 00:00:00 2001
From: Will Thompson <will@willthompson.co.uk>
Date: Tue, 28 Jun 2016 13:11:09 +0000
Subject: [PATCH 2/3] findDocuments: improve entity query performance

When I implemented this in 9a4c24c, there were not many rows in
entity_documentproperties in the database here. Now that there are,
computing the document_document -> entity_documentproperties ->
entity_entity join and then filtering is really, really slow. Postgres
seems to materialize the whole join and then scan it.

If we get a set of matching document IDs for the entity query in a
subquery, and then just filter with IN on that, things are much faster:
scan entity_entity; in a nested loop, get the document_ids via
entity_documentproperties; hash this set; and then scan
document_document.

Searching for a single character, this brings the query from ~1.1s to
~400ms. Searching for a full word, ~800ms to 120ms

This condition is getting really ugly -- I am sorry!

References #2935
---
 pandora/document/managers.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/pandora/document/managers.py b/pandora/document/managers.py
index 2772d63c..f9d26b11 100644
--- a/pandora/document/managers.py
+++ b/pandora/document/managers.py
@@ -35,14 +35,18 @@ def parseCondition(condition, user, item=None):
 
 
 def buildCondition(k, op, v):
+    import entity.models
     if k == 'id':
         v = ox.fromAZ(v)
         return Q(**{k: v})
     if isinstance(v, bool): #featured and public flag
         key = k
     elif k == 'entity':
-        entity_key, v = entity.managers.namePredicate(op, v)
-        key = 'entities__' + entity_key
+        entity_key, entity_v = entity.managers.namePredicate(op, v)
+        key = 'id__in'
+        v = entity.models.DocumentProperties.objects.filter(**{
+            'entity__' + entity_key: entity_v
+        }).values_list('document_id', flat=True)
     else:
         key = "%s%s" % (k, {
             '==': '__iexact',

From 09ebbc9cc6ccf03d3b24c842042ac575749b3abc Mon Sep 17 00:00:00 2001
From: Will Thompson <will@willthompson.co.uk>
Date: Tue, 28 Jun 2016 12:36:07 +0000
Subject: [PATCH 3/3] findDocuments: improve performance of positions queries

References #2935
---
 pandora/document/views.py | 30 ++++++++++++++++++++++++++++--
 1 file changed, 28 insertions(+), 2 deletions(-)

diff --git a/pandora/document/views.py b/pandora/document/views.py
index 66a01e59..ab739f62 100644
--- a/pandora/document/views.py
+++ b/pandora/document/views.py
@@ -165,6 +165,33 @@ def parse_query(data, user):
     return query
 
 
+def get_positions(qs, query_positions):
+    '''
+    qs: a QuerySet
+    query_positions: a list of AZ ids
+
+    TODO: merge this with item.utils.get_positions. The win is to fetch
+    only the integer IDs and convert the (smaller) set of query_positions to
+    ints, rather than fetch all keys for everything in qs (expected to be many
+    orders of magnitude larger), ignore most of it, and convert those ids to
+    strings.
+
+    Returns:
+        {
+            i: index of i in qs
+            for i in query_positions
+        }
+    '''
+    ids = list(qs.values_list('id', flat=True))
+    ret = {}
+    for i in query_positions:
+        try:
+            ret[i] = ids.index(ox.fromAZ(i))
+        except:
+            pass
+    return ret
+
+
 def findDocuments(request, data):
     '''
     Finds documents for a given query
@@ -198,8 +225,7 @@ def findDocuments(request, data):
         #FIXME: actually implement position requests
         response['data']['position'] = 0
     elif 'positions' in data:
-        ids = [i.get_id() for i in qs]
-        response['data']['positions'] = utils.get_positions(ids, query['positions'])
+        response['data']['positions'] = get_positions(qs, query['positions'])
     else:
         r = qs.aggregate(
             Sum('size')