I have an index, with a transformation:
docs.FeedPosts
.SelectMany(doc => (doc.Labels).DefaultIfEmpty(), (doc, docLabelsItem1) => new {AnnouncementGuid = doc.AnnouncementGuid, CreationDateUtc = doc.CreationWhenAndWhere.Time, FeedOwner = doc.FeedOwner, Key = doc.Key, Labels_Text = doc.Labels
.Select(label => label.Text), SequentialId = ((long)doc.SequentialId), SubjectGuid = doc.SubjectGuid, SubjectId = doc.SubjectId})
(transform)
results
.Select(doc => new {doc = doc, tags = Database.Load(doc.Key)})
.Select(__h__TransparentIdentifier1 => new {AnnouncementGuid = __h__TransparentIdentifier1.tags.AnnouncementGuid, AreCommentsLocked = __h__TransparentIdentifier1.tags.AreCommentsLocked, Author = __h__TransparentIdentifier1.tags.Author, Comments = __h__TransparentIdentifier1.tags.Comments, CreationWhenAndWhere = __h__TransparentIdentifier1.tags.CreationWhenAndWhere, FeedOwner = __h__TransparentIdentifier1.tags.FeedOwner, Key = __h__TransparentIdentifier1.tags.Key, Labels = __h__TransparentIdentifier1.tags.Labels, MessageBody = __h__TransparentIdentifier1.tags.MessageBody, SequentialId = __h__TransparentIdentifier1.tags.SequentialId, SubjectGuid = __h__TransparentIdentifier1.tags.SubjectGuid, SubjectId = __h__TransparentIdentifier1.tags.SubjectId})
Which works for QUERYING the data. But if I then request statistics, I get back the wrong document count! (I Limit the query to 0 results, and request raven statistics).
It seem that because my document looks like this:
{
Labels: [
{ Text: "label 1" }
{ Text: "label 2" }
]
}
Raven generates TWO index entries for this one document - If I look in the raven query tool, the first index contains the actual index data, the second index document is just completely empty.
If I have 3 labels in a document, it generates 3 index results... and my 'count' is 3 times what it should be.
Whats going on?
Thanks
I translated your index to the query syntax, because that is easier to look at:
from doc in docs.FeedPosts
from NOT_USING_THIS in doc.labels
select new
{
AnnouncementGuid = doc.AnnouncementGuid,
CreationDateUtc = doc.CreationWhenAndWhere.Time,
FeedOwner = doc.FeedOwner,
Key = doc.Key,
Labels_Text = doc.Labels.Select(label => label.Text),
SequentialId = ((long)doc.SequentialId),
SubjectGuid = doc.SubjectGuid,
SubjectId = doc.SubjectId}
}
The second from clause is the SelectMany() in your index, whose value you are not using
If you'll remove that and work on top of the root object, you won't have this issue.
The docs for this are:
http://ravendb.net/docs/faq/skipped-results
Related
Is there a way in Erlang to create a select query on ets table, which will get all the elements that contains the searched text?
ets:select(Table,
[{ %% Match spec for select query
{'_', #movie_data{genre = "Drama" ++ '_' , _ = '_'}}, % Match pattern
[], % Guard
['$_'] % Result
}]) ;
This code gives me only the data that started (=prefix) with the required text (text = "Drama"), but the problem is I need also the the results that contain the data, like this example:
#movie_data{genre = "Action, Drama" }
I tried to change the guard to something like that -
{'_', #movie_data{genre = '$1', _='_'}}, [string:str('$1', "Drama") > 0] ...
But the problem is that it isn't a qualified guard expression.
Thanks for the help!!
It's not possible. You need to design your data structure to be searchable by the guard expressions, for example:
-record(movie_data, {genre, name}).
-record(genre, {comedy, drama, action}).
example() ->
Table = ets:new('test', [{keypos,2}]),
ets:insert(Table, #movie_data{name = "Bean",
genre = #genre{comedy = true}}),
ets:insert(Table, #movie_data{name = "Magnolia",
genre = #genre{drama = true}}),
ets:insert(Table, #movie_data{name = "Fight Club",
genre = #genre{drama = true, action = true}}),
ets:select(Table,
[{#movie_data{genre = #genre{drama = true, _ = '_'}, _ = '_'},
[],
['$_']
}]).
How can i dynamically change the selected columns in the generated sql query when using a linq expression?
Its a new session for each time the query is executed.
Even when I set the MapExp as null after first creation an then changing the bool value to false, it still generates the column in the sql query.
The code runs in a wpf application.
System.Linq.Expressions.Expression<Func<Entity, Model>> MapExp = x => new Model
{
Id=xId,
Count= LoadFormulaField ? x.Count: null,
...
};
var result = session.Query<Entity>().Select(MapExp))
Your problem seems to be the ternary-conditional as part of the expression which is causing the "Count" column to always be queried.
One option to avoid this could be:
var query = session.Query<Entity>();
IQueryable<Model> result = null;
if (LoadFormulaField)
{
result = query.Select(x => new Model
{
Id = x.Id,
Count = x.Count,
});
}
else
{
result = query.Select(x => new Model
{
Id = x.Id,
});
}
Which would get a little less ugly if you separate in a couple of methods I think.
I have 2 linq queries. First query does nothing because of unique index and this is OK. But second also does nothing while it should add records . If I bypass first query second query works. Should I refresh entity ? How ?
foreach (var product in productList)
{
cc2nexo_SubiektProduct newproduct = new cc2nexo_SubiektProduct();
newproduct.Name = product.Name;
newproduct.VAT = product.VAT;
newproduct.Id = product.Id;
foreach (var stawkaVAT in myNexo_ExitoEntities.StawkiVat)
{
if (stawkaVAT.Stawka * 100 == tryconvert_dec(newproduct.VAT))
{
newproduct.VAT_Id = stawkaVAT.Id;
}
}
myNexo_ExitoEntities.cc2nexo_SubiektProduct.Add(newproduct);
SurroundWithTryCatchDB(() =>
{
myNexo_ExitoEntities.SaveChanges();
});
}
var orders = (from myorders in myNexo_ExitoEntities.temp_SubiektOrderList
select myorders).ToList();
foreach (var order in orders)
{
cc2nexo_SubiektOrderList neworder = new cc2nexo_SubiektOrderList();
neworder.Data_utworzenia_sprawy = tryconvert_date(order.Data_utworzenia_sprawy);
neworder.Data_modyfikacji_sprawy = tryconvert_date(order.Data_modyfikacji_sprawy);
neworder.Data_umowy = tryconvert_date(order.Data_umowy);
neworder.Id = order.Id;
myNexo_ExitoEntities.cc2nexo_SubiektOrderList.Add(neworder);
SurroundWithTryCatchDB(() =>
{
myNexo_ExitoEntities.SaveChanges();
});
Debug.WriteLine(neworder.LastName);
}
I am receiving an error
Cannot insert duplicate key row in object 'dbo.cc2nexo_SubiektProduct' with unique index 'K_ID'. The duplicate key value is (1).The statement has been terminated
The reason of problem was SurroundWithTryCatchDB procedure not listed in my question. Because first query caused exception due to unique index all next SaveChanges did not work. I have changed my first query to work with unique values and now everything is OK.
I have a basic schema
Post {
Labels: [
{ Text: "Mine" }
{ Text: "Incomplete" }
]
}
And I am querying raven, to ask for all posts with BOTH "Mine" and "Incomplete" labels.
queryable.Where(candidate => candidate.Labels.Any(label => label.Text == "Mine"))
.Where(candidate => candidate.Labels.Any(label => label.Text == "Incomplete"));
This results in a raven query (from Raven server console)
Query: (Labels,Text:Incomplete) AND (Labels,Text:Mine)
Time: 3 ms
Index: Temp/XWrlnFBeq8ENRd2SCCVqUQ==
Results: 0 returned out of 0 total.
Why is this? If I query for JUST containing "Incomplete", I get 1 result.
If I query for JUST containing "Mine", I get the same result - so WHY where I query for them both, I get 0 results?
EDIT:
Ok - so I got a little further. The 'automatically generated index' looks like this
from doc in docs.FeedAnnouncements
from docLabelsItem in ((IEnumerable<dynamic>)doc.Labels).DefaultIfEmpty()
select new { CreationDate = doc.CreationDate, Labels_Text = docLabelsItem.Text }
So, I THINK the query was basically testing the SAME label for 2 different values. Bad.
I changed it to this:
from doc in docs.FeedAnnouncements
from docLabelsItem1 in ((IEnumerable<dynamic>)doc.Labels).DefaultIfEmpty()
from docLabelsItem2 in ((IEnumerable<dynamic>)doc.Labels).DefaultIfEmpty()
select new { CreationDate = doc.CreationDate, Labels1_Text = docLabelsItem1.Text, Labels2_Text = docLabelsItem2.Text }
Now my query (in Raven Studio) Labels1_Text:Mine AND Labels2_Text:Incomplete WORKS!
But, how do I address these phantom fields (Labels1_Text and Labels2_Text) when querying from Linq?
Adam,
You got the reason right. The default index would generate 2 index entries, and your query is executing on a single index entry.
What you want is to either use intersection, or create your own index like this:
from doc in docs.FeedAnnouncements
select new { Labels_Text = doc.Labels.Select(x=>x.Text)}
And that would give you all the label's text in a single index entry, which you can execute a query on.
I have the following query:
val = val.Select(item => new SimpleBill { CTime = item.CTime, Description = item.Description, ID = item.ID,
IDAccount = item.IDAccount, OrderNumber = item.OrderNumber, PSM = item.PSM, Sum = item.Sum,
Type = (BillType)item.Type,
ByteStatus = ((Bill)item).Payments.OrderByDescending(payment => payment.ID).FirstOrDefault().Status,
LastPaymentDate = ((Bill)item).Payments.OrderByDescending(payment => payment.ID).FirstOrDefault().CTime,
LastPaymentSum = ((Bill)item).Payments.OrderByDescending(payment => payment.ID).FirstOrDefault().Sum });
}
Is it possible to avoid repeating the ((Bill)item).Payments.OrderByDescending(payment => payment.ID).FirstOrDefault() part 3 times? I tried turning it into a method and into a delegate - the code compiled in both cases, but produced an exception at runtime.
You can use the let contstruct as follows:
val = from item in val
let lastPayment = ((Bill)item).Payments
.OrderByDescending(payment => payment.ID)
.FirstOrDefault()
select new SimpleBill
{
lastPayment.CTime,
//Rest of fields
}
However, as you may noticed this uses the LINQ Query syntax vs. Method syntax. IIRC let is only available in the former.