Software Development - { Coding && Architecture }

In a previous posting I talked about how to read Excel files in .NET. I talked about the IMX parameter and how important it is to set it to “1” in order to interpret all data as strings. If the parameter is something else .NET will try to make an assumption based on the first rows in the file, if these rows are integers then is the column treated as an integer, and if a row further down contains non-numeric characters then will that data be null . After discovering this parameters I implemented a correct connection string (one of several information sources is this one ) and it worked. But as I discovered today it does not work all the time, still not all columns are treated as strings. The solution (this time…) is to treat the first row in the file as a header row and then map the headers with the imported data if needed. When using this solution must the HDR -parameter in the connection string be set to No . Assuming the file has a header row with text, then will .NET treat all f

For some time I have been using the Load() and Get() methods of NHibernate Session object without actually knowing the difference between the two methods. I kind of simply assumed the difference was architectural, but when I later ran into problems I had to examine the methods more in detail and discovered that the difference is pretty fundamental… In short the Load() and Get() methods find objects based on it’s id (i.e. the primary key).  But what happens if the object is not found? The Load() methods throws an exception and the Get() methods returns a null object. That is quite a big difference! The NHibernate documentation is very accurate in that the Load() method should not be used if you are not sure if an object with the specific id exist, and hence it is a pretty big error if the object does not exist.