First some more general remarks on your code not yet identified:
temp = inputFile1.ReadLine();
if (temp == null)
{
endOfFile1 = true;
}
The most reasonable way to tell if there is a new line or not usually comes into this code:
string line;
while ((line = stream.ReadLine()) != null)
Saves space, is readable, is concise. When I first started programming I always had those issues when reading line from files, from the moment I saw that I don't want another thing.
pathInputFile1 We normally suggest to use descriptive names but I admit that file1 is plenty for me.
You are having some trouble because you are doing three different tasks on your algorithm. I can identify at least three steps to achieve a solution for your problem:
- Read the integers from the files
- Merge the integers
- Write the merged integers to a new file
The major problem here is that we may not want to use additional memory, because of that we need to implement all algorithms on a lazily fashion.
public static IEnumerable<int> ReadIntegers(string file)
{
using (var stream = File.OpenText(file))
{
string line;
while ((line = stream.ReadLine()) != null)
{
yield return int.Parse(line);
}
}
}
public static IEnumerable<int> UnionSorted(IEnumerable<int> first, IEnumerable<int> second)
{
var it1 = first.GetEnumerator();
var it2 = second.GetEnumerator();
while (it1.MoveNext())
{
if (!it2.MoveNext()) break;
if (it1.Current > it2.Current)
{
yield return it2.Current;
}
else
{
yield return it1.Current;
}
}
while (it1.MoveNext())
{
yield return it1.Current;
}
while (it2.MoveNext())
{
yield return it2.Current;
}
}
public static void MergeFiles(string file1, string file2, string ouputFile)
{
var merged = UnionSorted(ReadIntegers(file1), ReadIntegers(file2));
using (var writer = File.AppendText(ouputFile))
{
foreach (var number in merged)
{
writer.WriteLine(number);
}
}
}
