Develop Reference

C# - Compare two .csv files without a key or unique Id

I have a two .csv files that are in the format
File1
ID,Name,Token
123,ABC,345
555,XYZ,777
777,YYY,765
666,UUU,
543,MNO,
File2
ID,Name,Token
777,ABC,345
125,XYZ,999
976,RRR,
I have to compare these and exclude those employees who have same ID or/and same token because these employees are considered same.
So, in my above example, in File2
The employee Id 777 has two matching employees in File 1, the one that have 777 as token number and the one that have 777 as Id.
I have done file comparison before also and every time I have used hashtables/dictionaries where the ID/Token were unique for every employee so it was simple as I only had to compare keys. But here as we can see, one employee in File2 matches with 2 Employees in File1.
I assume to tackle this issue we can use a linear search and do a full scan of File1 for every employee of File2. But I think it would be useless as it would increase processing time a lot because in my file there would be hundreds of rows. For example, if File 2 has 400 rows and File 1 has 300, then for each employee in File2, the application would scan 300 times. This doesn't make sense.
What should I do in this case? Should I fix the data at source and eliminate these duplicates entries there only?
Please let me know if the problem statement doesn't make sense. I will try to explain more clearly. Thanks!
UPDATE- What I tried
I read File1 in a dictionary and then read File2 line by line and compared the ID and token. This works well for those employees that have unique id and token but for duplicates in the data it fails.
if((File2_ID).contains(File1_ID)|| (File2_Token).contains(File1_Token))
{
//do something
}
The issue with this approach is that the application checks employee ID 777 in File2 with the Id 777 in File 1, it thinks it has found it match , it processes the data and moves to the next line in File2. It doesn't consider employee with token number 777 in File1.
UPDATE 2 - I had forgotten to mention that tokens can be blank too.
I was wondering if I try some data cleansing thing. Create a functionality that removes these duplicates, and then add clean file in the application for comparison.

You could use something like the following to lookup by both ID and Token:
FileEntry foundById;
idMap.TryGetValue(File1_ID, out var foundById);
FileEntry foundByToken;
idToken.TryGetValue(File1_Token, out var foundByToken);
if (foundById && foundByToken)
{}
else if (foundById)
{}
else if (foundByToken)
{}

How to get list of .xlsx file in C#

I am developing Windows Form application and I want to have a list of .xlsx files in the folder say -
C:\Equipment\Exchanger\AES\
The folder may contain files like -
00-E-001,
00-E-002,
03-E-005,
04-E-001 and so on..
I don't want to open file dialogs but want to see a combo box showing the FileName (no extension). Based on selection of combobox I want to show the values of certain cells in the various text boxes but that is later part.

You can use Directory.GetFiles to query the full path strings of every *.xlsx file in the folder, then strip it down to just the filename using Select and Path.GetFileName.. It makes sense (to me) to sort them alphabetically, then you just need them in a list so the combo can use them
yourCombo.DataSource = Directory.GetFiles(yourFolderPath, "*.xlsx").Select(Path.GetFileName).OrderBy(n => n.ToLower()).ToList();
If you don't want the extension in the combo (it's a waste of pixels :) ) you can use Path.GetFileNameWithoutExtension
When the time comes to get which file was selected by the user it's:
var fullPath = Path.Combine(yourFolderPath, yourCombo.SelectedItem);
(You'll have to add the extension back on if you used GetFilenameWithoutExtension: yourCombo.SelectedItem + ".xlsx")
I recommend you set your combo's DropDownStyle to DropDownList

First of all, your code should have a look at the folder, which contains needed files.
List<string> xlsxFiles = Directory.GetFiles(yourdirectory, "*.*", SearchOption.AllDirectories)
.Where(file => new string[] { ".xlsx" }
.Contains(Path.GetExtension(file)))
.ToList(); // Looking into directory and filtering files
for(int i = 0; i < xlsxFiles.Count; i++)
{
xlsxFiles[i] = Path.GetFileName(Path.ChangeExtension(xlsxFiles[i], null)); // Removes files' extensions
}
This is the simple way of looking into the directory and filtering files to needed ones.
Then, I suggest having a look into EPPlus library, which allows you to work directly with tables and cells, without manual extraction of the file
Here is the handy tutorial of using EPPlus

ListBox sorting by files creation time

My listBox1 is filled with names of txt files. It is sorting them automatically alphabetically. But I need to sort these files by creation time, from the newest to the oldest. Can someone help me?

you are not giving us much to go on :P
But I am going to give it my best shot to help you. Because I do not know how you have populated the List (by hand, or with code) I cant really guide you there.
But lets say you did it by code. The following snippet I typed might help you find what you need
//array of file names (these need to be paths to files, so if you do not prefix it with ./path/ or anything, then the executable needs to be in the same folder)
//If you are debugging, make sure the paths are correct if you expect them to be in the same folder, it should then be in the debug folder.
//If you included some test files in the solution and want them copied to your debug folder. Click on them, go to the properties pane, and set build action to content, and one of the other properties should read copy of newer instead of never copy
var fileNames = new string[]{"1.txt", "2.txt", "3.txt"};
var fileInfos = fileNames.Select(f => new FileInfo(f));
var orderedFileInfos = fileInfos.OrderBy(f => f.CreationTime);
orderedFileInfos is now a list of type FileInfo ordered by ascending on their creationtime.
This is what you can use to populate your ListBox
I hope i helped you progress in your project.

How to get all files from a folder in sorted order in C#.Net? [duplicate]

Is is possible to get files that is ordered same as in Windows Explorer
I know "natural sort", but it's not what I need, I need to get the file list ordered by the same attribute in Windows Explorer, for example:
If I ordered a directory by the attribute "create date", then I will get a file list as below:
name create date file size
1.txt 2012/1/1 125Kb
2.tab 2012/3/2 15Kb
3.bmp 2013/5/5 26Kb
If my windows explorer order file list with the attribute "file size", the the file list would be:
name create date file size
2.tab 2012/3/2 15Kb
3.bmp 2013/5/5 26Kb
1.txt 2012/1/1 125Kb
Could anyone help?

I think this is going to be a lot more complex than you expect. Folder settings are stored in the registry in two places:
HKCU\Software\Microsoft\Windows\Shell\BagMRU
HKCU\Software\Microsoft\Windows\Shell\Bags
The first path contains a structure which reflects the structure of the file system, and the second path contains details about those items, including a REG_BINARY value called "Sort" which records the sort order used for that folder.
See Willi Balenthin's website for details on the structure, including sample code (in Python)

Here's how to get a list of files sorted by their name:
var path = #"C:\windows"; // obviously change this to whatever you want
var files = System.IO.Directory.GetFiles (path).ToList ();
file.Sort();
And that's it!
Here's how you would do it per your given code sample:
var temperaturePressureSignalFilesList = Directory.GetFiles(TemperaturePressureSignalDirectory, "*.txt", SearchOption.TopDirectoryOnly).ToList();
temperaturePressureSignalFilesList.Sort();

using System.Linq;
DirectoryInfo info = new DirectoryInfo(""); FileInfo[] files =
info.GetFiles().OrderBy(p => p.CreationTime).ToArray(); foreach
(FileInfo file in files) {
// DO Something... }
here is the sample code for get files in directory by creation time.
You can get files by size same way.

I guess you are talking about viewing pane in Windows Explorer (it's essentially a Windows File Manager but also known under different name). Some clarification is needed. You can apply your custom sorting on various columns; moreover, you can have multiple viewing panes (windows) open sorted on different columns. Thus, the problem definition is a bit unclear.
Assuming that you know the sorting order in your viewing panes, then you can use System.IO.DirectoryInfo and derived FileSystemInfo[] objects; the latter has files.OrderBy method.
Hope this will help. My best, Alex

If you want natural sort order, you should either P/Invoke StrCmpLogicalW (http://msdn.microsoft.com/en-us/library/bb759947.aspx) or find a managed natural sort algorithm. There is no built-in natural sort in .NET Framework.

I think you cannot know which is the order in the pane (by size, name or whatever), you must read the list and then sort it the way you want or prompt the user to select a sorting attribute.
As Kenny posted Sorting Directory.GetFiles() here is an approach, anyway I still thinking there is no possibly way to know which is the sorting order that user selected in the viewing pane.

I think you would have to write a shell extension for windows explorer that captures sort events on columns and writes that metadata to disk in some structured way. You may have multiple explorer windows open so might be an idea to apply timestamp or id so you know which explorer window you are dealing with. Then in your app read that metadata to get the sort order and apply accordingly. Not easy but doable.

Output dataset as collection of .csv files, .zip not an option C#

I have produced a solution in a C# web app that runs a report which returns a DataSet containing 9 Datatables.
The tables are linked with many-to-many relationships, and the purpose of the report is to produce a snapshot of the progress on records contained in the tables for anyone point in time.
This dictates that all 9 tables are queried at once, as querying the tables individually at different times will mean that the results dont match up between tables.
I was returning the 9 csv files in a zip file, which worked fine in development. However, in the organisation I am working for a lot of user's machines are very limited and we cannot rely on them having winzip or some other archiving software, and as such I have been told to find an alternative solution.
So far, the problems I have come up against are:
All 9 datatables must be produced from the same query, which rules out giving users the option to run the report against tables individually as the results may not match up if they are run at different times of the day.
I can only return one file at a time to the browser, which stops me returning the collection of csv files unzipped.
I did think about returning the data as separate tabs on an excel workbook, however some of the results have leading zeros, which will be trimmed off when they are loading into the worksheet.
It would be great if anyone knew of an obvious solution I'm missing?

Using DotNetZip, you can zip and unzip files like so:
using Ionic.Zip;
namespace ConsoleApplication23
{
class Program
{
static void Main(string[] args)
{
//Zip your files like so
ZipFile x = new ZipFile();
x.AddFile(#"C:\myFile");
x.AddFile(#"C:\mySecondFile");
x.Save(#"c:\myZipFile.zip");
//Unzip like so
ZipFile y = ZipFile.Read(#"c:\myZipFile.zip");
foreach (ZipEntry e in y)
{
e.Extract(#"c:\test", ExtractExistingFileAction.OverwriteSilently);
}
}
}
}