I'm trying to write a simple helper function that connects to Oracle database using Oracle.ManagedDataAccess.Core library, and returns the rows.
Returned object should survive when connection to database is closed. Something similar to the following. Not sure what the will be.
public <sometype> GetOracleResults(string connectionString, string cmdText, List<OracleParameter> oracleParameters)
{
<sometype> results = null;
try
{
using (OracleConnection oracleConnection = new OracleConnection(connectionString))
{
oracleConnection.Open();
using (OracleCommand oracleCommand = new OracleCommand(cmdText, oracleConnection))
{
foreach (var param in oracleParameters)
{
oracleCommand.Parameters.Add(param);
}
OracleDataReader oracleDataReader = oracleCommand.ExecuteReader();
if(oracleDataReader.HasRows)
{
results = new <sometype>();
while (oracleDataReader.Read())
{
//loop through the reader and add results
return results;
}
}
}
}
}
catch (Exception)
{
//todo
throw;
}
}
Does it have to be strongly typed? If not, a coworker showed me sticking the data in a List of Dictionary items, E.G.:
var columns = new List<string>();
for (int i = 0; i < myReader.FieldCount; i++)
{
columns.Add(myReader.GetName(i).Trim());
}
while (myReader.Read() && results.Count < 200)
{
Dictionary<string, string> row = new Dictionary<string, string>();
foreach (var column in columns)
{
row.Add(column, Convert.ToString(myReader[column], CultureInfo.InvariantCulture).Trim());
}
results.Add(row);
}
I suppose if you know the schema so that the generic type coming into the method had property names matching the column names, then you could use reflection to match up the column results with the C# object.
I can't work out how to get the Type of a LINQ result. My set contains both strings and bools, so I run into trouble when I try to act on the rows. I attached an incredibly rough workout using try/catch (for a laugh), but it hurts my soul and would much rather know the proper method in obtaining the Type.
private AppointmentInfoClass UpdateDataContext(DataSet phaseDataSet) {
var phaseCollection = new AppointmentInfoClass();
var Type = phaseCollection.GetType();
var properties = Type.GetProperties();
var result = from DataRow myRow in DataBindings.CompanyAppsDataSet.Tables[0].Rows
where (int)myRow["AppointmentID"] == ApptID
select myRow;
var k = 0;
foreach (DataRow row in phaseDataSet.Tables[0].Rows) {
string header;
header = row.Field<string>("Header");
foreach (var field in result) {
try {
properties[k].SetValue(phaseCollection, field.Field<string>(header));
}
catch (Exception) {
properties[k].SetValue(phaseCollection, field.Field<bool>(header).ToString());
}
}
k++;
}
return phaseCollection;
}
It will return the type you have written instead of Type
string s = field.Field<string>("ColumnName");
bool b = field.Field<bool>("ColumnName");
I can't seem to get passed this error message using this code.
System.Data.DuplicateNameException: A column named 'Url' already belongs to this DataTable.
foreach (var user in users.Users)
{
using (DataTable table = new DataTable())
{
table.Clear();
for (int i = 0; i < users.Users[0].GetType().GetProperties().Count(); i++)
{
table.Columns.Add(users.Users[0].GetType().GetProperties()[i].Name);
Console.WriteLine(table.Columns.Add(users.Users[0].GetType().GetProperties()[i].Name));
}
}
}
Any ideas?
UPDATE
var properties = users.Users[0].GetType().GetProperties();
for (int i = 0; i < properties.Count(); i++)
{
table.Columns.Add(properties[i].Name, typeof(String));
}
Because I didn't know what the type was to use the foreach. This:
var properties = users.Users[0].GetType().GetProperties();
foreach (PropertyInformation prop in properties)
{
table.Columns.Add(properties.);
Console.WriteLine(prop.Name);
}
gave me Cannot convert type 'System.Reflection.PropertyInfo' to 'System.Configuration.PropertyInformation'
Anyway I can figure out how to get the correct type in there? #barrick
The error won't lie. In the Console.WriteLine() call, you're adding the column again, which will cause the problem.
Try:
var properties = users.Users[0].GetType().GetProperties();
foreach (PropertyInfo prop in properties)
{
table.Columns.Add(prop.Name);
Console.WriteLine(prop.Name);
}
I am having difficulty with the following method:
public override List<Team> Search(Dictionary<string, string> prms,
int pageSize, int page, out int results)
{
List<Team> t = null;
//Team t = null;
var tresults = new List<Team>();
using (SqlConnection conn = DB.GetSqlConnection())
{
using (SqlCommand cmd = conn.CreateCommand())
{
cmd.CommandText = #"SearchForTeam";
cmd.CommandType = System.Data.CommandType.StoredProcedure;
foreach (var key in prms.Keys)
{
cmd.Parameters.Add(key, prms[key]);
}
SqlDataReader reader
= cmd.ExecuteReader(CommandBehavior.CloseConnection);
while (reader.Read())
{
var temp = Load(reader);
if (t == null)
{
t = temp;
}
else
{
t.CityHistory.Add(temp.CityHistory[0]);
}
}
}
}
results = 0;
return t;
}
The error lies mainly with the if and else statement where the temp in the if block is claiming that it "cannot implicitly convert type DataLayer.Team to System.Collections.GenericList"
EDIT:
Here is my load method:
public Team Load(SqlDataReader reader)
{
var team = new Team()
{
TeamID = Int32.Parse(reader["TeamID"].ToString()),
TeamName = reader["TeamName"].ToString()
};
team.CityHistory.Add(
new TeamCity(
Int32.Parse(reader["TeamCitiesID"].ToString()),
team.TeamID,
Int32.Parse(reader["CityID"].ToString()),
reader["CityName"].ToString(),
Int32.Parse(reader["YearStart"].ToString()),
Int32.Parse(reader["YearEnd"].ToString())
)
);
return team;
}
t is defined as List<Team>, yet you later say t.CityHistory. CityHistory is clearly not a property of List<>. I'd guess it's a property of Team, but since you never show us that we can't say.
Show us the definition of Team and the method signature of Load() and perhaps we can give an answer.
UPDATE (from OP's update)
Now, I'm going to assume that you are getting multiple rows, one for each City, with the team info repeating. So, which you want is:
var temp = Load(reader);
// remove t definition above
var t = tresults.FirstOrDefault(team=> team.TeamId == temp.TeamId);
if (t == null)
{
t = temp;
tresults.Add(t);
}
else
t.CityHistory.Add(temp.CityHistory[0]);
(Updated again, based on Steve's comment)
You have to wrap temp into a List first to assign it to t:
if (t == null) {
t = new List<Team>() { temp }
} else {
t.add(temp);
}
VERY LONG COMMENT:
Consider the following refactoring of your method:
private IEnumerable<Team> LoadTeams()
{
using (SqlConnection conn = DB.GetSqlConnection())
{
using (SqlCommand cmd = conn.CreateCommand())
{
cmd.CommandText = #"SearchForTeam";
cmd.CommandType = System.Data.CommandType.StoredProcedure;
foreach (var key in prms.Keys)
{
cmd.Parameters.Add(key, prms[key]);
}
SqlDataReader reader
= cmd.ExecuteReader(CommandBehavior.CloseConnection);
while (reader.Read())
{
yield return Load(reader);
}
}
}
}
public override List<Team> Search(Dictionary<string, string> prms,
int pageSize, int page, out int results)
{
List<Team> searchResult = new List<Team>;
//Team t = null;
var tresults = new List<Team>();
foreach(Team team in LoadTeams())
{
if (team .....)
searchResult.Add(team);
}
results = 0;
return searchResult;
}
Take into account this NEEDED initialization:
List<Team> searchResult = new List<Team>;
And ask yourself: What should the following excerpt do? :
if (team .....)
searchResult.Add(team);
P.S.: Also the line
results = 0;
should probably look like:
results = searchResult.Count;
I'm having to guess a little here, but I assume that your Load() method is returning a data type 'DataLayer.Team', which sounds like its one Team, and you're trying to assign it to 't', which is a list of teams.
Try:
t.Add(temp)
or
t.Add(temp as Team).
It doesn't help that you're using 'var' declarations all the time.
In MVC 4 and EF 5 i want to run dynamic query.
var returndata = Context.Database.SqlQuery(Type, strsql, null);
i don't know, how many fields it will return and name. Out of this result i want to make table structure that will display on view.
Question : What should i passed as Type?
my query return below result:
Field 1, Field 2, Field 3, Field 4, Field 5
Row1...
Row2..
Appreciate any suggestion.
You could use a raw SQL query because EF doesn't support that:
private static IEnumerable<object[]> Read(DbDataReader reader)
{
while (reader.Read())
{
var values = new List<object>();
for (int i = 0; i < reader.FieldCount; i++)
{
values.Add(reader.GetValue(i));
}
yield return values.ToArray();
}
}
and then:
public ActionResult Index()
{
using (var ctx = new UsersContext())
using (var cmd = ctx.Database.Connection.CreateCommand())
{
ctx.Database.Connection.Open();
cmd.CommandText = "SELECT * FROM UserProfile";
using (var reader = cmd.ExecuteReader())
{
var model = Read(reader).ToList();
return View(model);
}
}
}
and finally in your view:
#model IEnumerable<object[]>
<table>
<tbody>
#foreach (var row in Model)
{
<tr>
#foreach (var column in row)
{
<td>#column</td>
}
</tr>
}
</tbody>
</table>
This method loads data from SQL select (with parameters) to the list of rows, where each row is the dictionary of columns (the key is the column name).
private static List<Dictionary<string, object>> LoadData(string sqlSelect, params object[] sqlParameters)
{
var table = new List<Dictionary<string, object>>();
using (var ctx = new DbEntities())
{
ctx.Database.Connection.Open();
using (var cmd = ctx.Database.Connection.CreateCommand())
{
cmd.CommandText = sqlSelect;
foreach (var param in sqlParameters)
cmd.Parameters.Add(param);
using (var reader = cmd.ExecuteReader())
{
while (reader.Read())
{
var row = new Dictionary<string, object>();
for (int i = 0; i < reader.FieldCount; i++)
row[reader.GetName(i)] = reader[i];
table.Add(row);
}
}
}
}
return table;
}
Finally i made is using TypeBuilder option suggested by "Mortalus" and ExpandoObject object. It has little performance overhead right now.
Take Typebuilder code from "Mortalus" answer then i made code according to my requirement as below.
List<Dictionary<string, object>> expandolist = new List<Dictionary<string, object>>();
foreach (var item in returndata)
{
IDictionary<string, object> expando = new ExpandoObject();
foreach (PropertyDescriptor propertyDescriptor in TypeDescriptor.GetProperties(item))
{
var obj = propertyDescriptor.GetValue(item);
expando.Add(propertyDescriptor.Name, obj);
}
expandolist.Add(new Dictionary<string, object>(expando));
}
return expandolist;
so now, I have "Dictionary" object from dynamic object. and using it you can work easily at design time rather then wait until runtime using "dynamic" object.
I have recently stumbled upon this example:
http://www.markzhou.com/blog/post/2011/06/02/Use-dynamic-type-in-Entity-Framework-41-SqlQuery()-method.aspx
I haven't had the time to test it myself but it seems that it is possible with some additional work to construct the dynamic type.
In short you would want to do something like this:
TypeBuilder builder = Program.CreateTypeBuilder(
"MyDynamicAssembly", "MyModule", "MyType");
Program.CreateAutoImplementedProperty(builder, "name", typeof(string));
Program.CreateAutoImplementedProperty(builder, "type", typeof(string));
Program.CreateAutoImplementedProperty(builder, "id", typeof(int));
Type resultType = builder.CreateType();
dynamic queryResult = context.Database.SqlQuery(
resultType, "SELECT * FROM sys.sysobjects");
Where TypeBuilder is described in details in the post I have attached.
Without knowing anything about the type that is returned, I think you might be out of luck.
If you know what patterns it might fall under, you could use some try { } catch () { }'s on interfaces that match those parameters on your otherwise dynamic query, but that seems like it might be a bit painful.
Unfortunately, EF won't materialize objects unless it knows their Type.
If this is really necessary for you, I think your best bet would be to fall back to ADO.NET and DataTable.
Similarly post by Darin Dimitrov, but it returns DataTable
public DataTable QueryToTable(Entities db, string queryText, SqlParameter[] parametes)
{
using ( DbDataAdapter adapter = new SqlDataAdapter())
{
adapter.SelectCommand = db.Database.Connection.CreateCommand();
adapter.SelectCommand.CommandText = queryText;
if (parametes != null)
adapter.SelectCommand.Parameters.AddRange(parametes);
DataTable table = new DataTable();
adapter.Fill(table);
return table;
}
}
Use
SqlParameter[] parametes = new[]
{
new SqlParameter("date_from", dateFrom)
};
DataTable tab = QueryToTable(new Entities(),
"Select * From SomeTable Where ADate >= #date_from", parametes);
Example for MS SQL Server
Adding to Petr VobornÃk's answer, dynamic query, I add dynamic insert of ResultSet, my application takes the dynamic query of all tables of the entire database, a chunk at a time and then inserts the dynamic results into a remote database, using Always Encrypted (omitted here). Passing a sb command and parameter object.
public void StoreData(DbContext dbContext, Dictionary<string, string> columnInfo, List<Dictionary<string, object>> multiInsertObj, string tableName)
{
_ctx = dbContext;
_columnInfo = columnInfo;
var sb = new StringBuilder();
sb.Append(BuildSqlCommand(tableName, columnInfo, multiInsertObj.Count));
ExecuteSqlCommand(sb, GetParamsObject(columnInfo, multiInsertObj));
}
private static StringBuilder BuildSqlCommand(string tableName, Dictionary<string, string> variableInfo, int variableCount)
{
//Build sql command
var sb = new StringBuilder();
sb.Append("INSERT INTO dbo." + tableName + "(");
foreach (var variable in variableInfo)
{
sb.Append(variable.Key);
sb.Append(", ");
}
sb.Append("SystemNumber, ");
sb.Remove(sb.Length - 2, 2).Append(") VALUES ");
for (var i = 0; i < variableCount; i++)
{
sb.Append("(");
foreach (var name in variableInfo.Keys)
{
sb.Append("#" + name + "_" + i + ",");
}
sb.Append("#SystemNumber" + "_" + i + ",");
sb.Remove(sb.Length - 1, 1).Append("),");
}
sb.Remove(sb.Length - 1, 1);
return sb;
}
private static object[] GetParamsObject(Dictionary<string, string> columnInfo, List<Dictionary<string, object>> multiInsertObj)
{
var variableCount = multiInsertObj.Count;
var rowCount = multiInsertObj[0].Keys.Count;
var objectLength = (rowCount + 1) * variableCount;
var variableDataTypes = columnInfo.Values.ToList();
var paramObj = new object[objectLength];
var j = 0;
var i = 0;
foreach (var row in multiInsertObj)
{
var k = 0;
foreach (var data in row)
{
var sb = new StringBuilder();
sb.Append("#");
sb.Append(data.Key);
sb.Append("_" + i);
paramObj[j] = new SqlParameter(sb.ToString(), SetSqlDataType(variableDataTypes[k])) { Direction = Input, Value = data.Value };
j++;
k++;
}
paramObj[j] = new SqlParameter(("#SystemNumber" + "_" + i), SetSqlDataType("int")) { Direction = Input, Value = _systemNumber };
i++;
j++;
}
return paramObj;
}
private static void ExecuteSqlCommand(StringBuilder sb, params object[] sqlParameters)
{
using (_ctx)
{
_ctx.Database.Connection.Open();
using (var cmd = _ctx.Database.Connection.CreateCommand())
{
cmd.CommandText = sb.ToString();
foreach (var param in sqlParameters)
cmd.Parameters.Add(param);
try
{
cmd.ExecuteNonQuery();
}
catch (Exception e)
{
Console.WriteLine(e);
throw;
}
}
}
}
These approaches seemed a bit complicated to me, at least in my situation. All I needed was to return a DataTable so it could be previewed. Meaning every value could be a string.
So I created two classes:
public class DynamicResultModel
{
public DynamicResultColumnModel headers = new();
public List<DynamicResultColumnModel> rows = new();
}
public class DynamicResultColumnModel
{
public List<string> columns = new();
}
Then created a simple helper class to basically serialize and deserialize a DataTable into and out of a DynamicResultModel
public static DynamicResultModel DataTableToDynamic(DataTable dtData)
{
DynamicResultModel result = new();
DynamicResultColumnModel headers = new();
foreach (DataColumn col in dtData.Columns)
{
result.headers.columns.Add(col.ColumnName);
}
foreach (DataRow row in dtData.Rows)
{
DynamicResultColumnModel rowData = new();
foreach (var item in row.ItemArray)
{
rowData.columns.Add(item.ToString());
}
result.rows.Add(rowData);
}
return result;
}
The only caveat was that I could not return a DynamicResultModel from my controller, I had to serialize it into a string first. Regardless, this worked for me and saved me a lot of time.
So my API endpoint definition was
Task<ApiResponse<string>>
Instead of
Task<ApiResponse<DynamicResponseModel>>