sql >> Database teknologi >  >> RDS >> Oracle

LINQ Brug like i stedet for (( NVL(INSTR(x, y), 0) ) =1)

Da jeg kiggede lidt omkring, fandt jeg LIKE-operator i LINQ som var en som har et par gode eksempler på, hvordan du kunne gøre dette. Jeg har testet den nedenfor, der var fra linket ovenfor

Her er en udvidelse til brug af Synes godt om med lambda, som blev indsendt af adobrzyc

    public static class LinqEx
    {
        private static readonly MethodInfo ContainsMethod = typeof(string).GetMethod("Contains");
        private static readonly MethodInfo StartsWithMethod = typeof(string).GetMethod("StartsWith", new[] { typeof(string) });
        private static readonly MethodInfo EndsWithMethod = typeof(string).GetMethod("EndsWith", new[] { typeof(string) });

        public static Expression<Func<TSource, bool>> LikeExpression<TSource, TMember>(Expression<Func<TSource, TMember>> property, string value)
        {
            var param = Expression.Parameter(typeof(TSource), "t");
            var propertyInfo = GetPropertyInfo(property);
            var member = Expression.Property(param, propertyInfo.Name);

            var startWith = value.StartsWith("%");
            var endsWith = value.EndsWith("%");

            if (startWith)
                value = value.Remove(0, 1);

            if (endsWith)
                value = value.Remove(value.Length - 1, 1);

            var constant = Expression.Constant(value);
            Expression exp;

            if (endsWith && startWith)
            {
                exp = Expression.Call(member, ContainsMethod, constant);
            }
            else if (startWith)
            {
                exp = Expression.Call(member, EndsWithMethod, constant);
            }
            else if (endsWith)
            {
                exp = Expression.Call(member, StartsWithMethod, constant);
            }
            else
            {
                exp = Expression.Equal(member, constant);
            }

            return Expression.Lambda<Func<TSource, bool>>(exp, param);
        }

        public static IQueryable<TSource> Like<TSource, TMember>(this IQueryable<TSource> source, Expression<Func<TSource, TMember>> parameter, string value)
        {
            return source.Where(LikeExpression(parameter, value));
        }

        private static PropertyInfo GetPropertyInfo(Expression expression)
        {
            var lambda = expression as LambdaExpression;
            if (lambda == null)
                throw new ArgumentNullException("expression");

            MemberExpression memberExpr = null;

            switch (lambda.Body.NodeType)
            {
                case ExpressionType.Convert:
                    memberExpr = ((UnaryExpression)lambda.Body).Operand as MemberExpression;
                    break;
                case ExpressionType.MemberAccess:
                    memberExpr = lambda.Body as MemberExpression;
                    break;
            }

            if (memberExpr == null)
                throw new InvalidOperationException("Specified expression is invalid. Unable to determine property info from expression.");


            var output = memberExpr.Member as PropertyInfo;

            if (output == null)
                throw new InvalidOperationException("Specified expression is invalid. Unable to determine property info from expression.");

            return output;
        }
    }

For at bruge det, skal du blot tilføje Like-funktionen, hvor du vil placere Indeholder-funktionerne. Du kan se nedenfor for et eksempel

            using (CustomerEntities customerContext = new CustomerEntities())
            {
                IQueryable<Customer> customer = customerContext.Customer.Like(x => x.psn, "%1%");    
            }

Dette vil oprette en sql-forespørgsel, der ser sådan ud.

SELECT 
[Extent1].[psn] AS [psn]
FROM [dbo].[Customer] AS [Extent1]
WHERE [Extent1].[psn] LIKE '%1%'



  1. Oracle konvertering af UNIX tidsstempel til tidsstempel med tidszone

  2. Skift standardskema for bruger

  3. 25 Microsoft Access-genveje for at spare tid i tabeller i dataarkvisning

  4. Vanskeligheder ved at skrive lagret proc til at hente data til Jqgrid-paginering