I am using a hashtable to read data from file and make clusters.
Say the data in file is:
umair,i,umair
sajid,mark,i , k , i
The output is like:
[{umair,umair},i]
[sajid,mark,i,i,k]
But my code does not work. Here is the code:
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.IO;
using System.Collections;
namespace readstringfromfile
{
class Program
{
static void Main()
{
/* int i = 0;
foreach (string line in File.ReadAllLines("newfile.txt"))
{
string[] parts = line.Split(',');
开发者_StackOverflow社区 foreach (string part in parts)
{
Console.WriteLine("{0}:{1}", i,part);
}
i++; // For demo only
}*/
Hashtable hashtable = new Hashtable();
using (StreamReader r = new StreamReader("newfile.txt"))
{
string line;
while ((line = r.ReadLine()) != null)
{
string[] records = line.Split(',');
foreach (string record in records)
{
if (hashtable[records] == null)
hashtable[records] = (int)0;
hashtable[records] = (int)hashtable[records] + 1;
Console.WriteLine(hashtable.Keys);
}
/////this portion is not working/////////////////////////////////////
foreach (DictionaryEntry entry in hashtable)
{
for (int i = 0; i < (int)hashtable[records]; i++)
{
Console.WriteLine(entry);
}
}
}
}
}
}
}
You're working with the records
array when inserting into the hashtable (and when reading from it) instead of using the foreach-variable record
. Also, in the final look, you iterate based on records
instead of the current entry.Key
. You're also declaring the hashtable in a too wide scope, causing all rows to be inserted into the same hashtable, instead of one per row.
public static void Main() {
var lines = new[] { "umair,i,umair", "sajid,mark,i,k,i" };
foreach (var line in lines) {
var hashtable = new Hashtable();
var records = line.Split(',');
foreach (var record in records) {
if (hashtable[record] == null)
hashtable[record] = 0;
hashtable[record] = (Int32)hashtable[record] + 1;
}
var str = "";
foreach (DictionaryEntry entry in hashtable) {
var count = (Int32)hashtable[entry.Key];
for (var i = 0; i < count; i++) {
str += entry.Key;
if (i < count - 1)
str += ",";
}
str += ",";
}
// Remove last comma.
str = str.TrimEnd(',');
Console.WriteLine(str);
}
Console.ReadLine();
}
However, you should consider using the generic Dictionary<TKey,TValue> class, and use a StringBuilder if you're building alot of strings.
public static void Main() {
var lines = new[] { "umair,i,umair", "sajid,mark,i,k,i" };
foreach (var line in lines) {
var dictionary = new Dictionary<String, Int32>();
var records = line.Split(',');
foreach (var record in records) {
if (!dictionary.ContainsKey(record))
dictionary.Add(record, 1);
else
dictionary[record]++;
}
var str = "";
foreach (var entry in dictionary) {
for (var i = 0; i < entry.Value; i++) {
str += entry.Key;
if (i < entry.Value - 1)
str += ",";
}
str += ",";
}
// Remove last comma.
str = str.TrimEnd(',');
Console.WriteLine(str);
}
Console.ReadLine();
}
You're attempting to group elements of a sequence. LINQ has a built-in operator for that; it's used as group ... by ... into ...
or the equivalent method .GroupBy(...)
That means you can write your code (excluding File I/O etc.) as:
var lines = new[] { "umair,i,umair", "sajid,mark,i,k,i" };
foreach (var line in lines) {
var groupedRecords =
from record in line.Split(',')
group record by record into recordgroup
from record in recordgroup
select record;
Console.WriteLine(
string.Join(
",", groupedRecords
)
);
}
If you prefer shorter code, the loop be equivalently written as:
foreach (var line in lines)
Console.WriteLine(string.Join(",",
line.Split(',').GroupBy(rec=>rec).SelectMany(grp=>grp)));
both versions will output...
umair,umair,i
sajid,mark,i,i,k
Note that you really shouldn't be using a Hashtable
- that's just a type-unsafe slow version of Dictionary
for almost all purposes. Also, the output example you mention includes []
and {}
characters - but you didn't specify how or whether they're supposed to be included, so I left those out.
A LINQ group is nothing more than a sequence of elements (here, identical strings) with a Key (here a string). Calling GroupBy
thus transforms the sequence of records into a sequence of groups. However, you want to simply concatenate those groups. SelectMany
is such a concatenation: from a sequence of items, it concatenates the "contents" of each item into one large sequence.
精彩评论