Cleaning algoritm for cleaning text

Question

0.00/5 (No votes)

See more:

In the end I will finish this code. But I want to check how you will make it?
On the moment, i am stuck and I need to refresh myself. A pause.
The Problem enunciate like this:
I have many images that contain -pre and -fullview in the end of their names.
Some are the same image, some are diferent...it's a mess.
The -pre images are at low rezolution, but -fullview are max rezolution.
My program must sort -pre from -ful (i did it already with 2 lists).
then, erase -pre, IF image is the SAME name for -pre and -full.
Keep in mind there are also unique images that have pre but not full. Also, only -full images that dont have -pre to be clear out.
How would you deal with this ... shingle ?
Thank you.

Aham...I looked over the stack element, but... i never worked seriously with it, so its hard for me to implement it at the moment. I mention about it, because maybe that is what i miss to do my life easier. You tell me.

What I have tried:

List<string> ImagesName = new List<string>();
ImagesName.AddRange(new string[] { "a.jpg", "a1-pre.jpg", "a1-fullview.jpg", "a2-pre.png", "a2-pre.png", "a3-fullview.jpg", "a3-fullview.jpg", "a4-fullview.jpg" });


//pre, fullview image duplicate cleaning
List<string> Duplicates = new List<string>();
string pre = "-pre.", ful = "-fullview."; bool isX = true;
string link = "", LinkWithNoPre = "", dupelink = "";
List<string> linkWithPre = new List<string>();
List<string> linkWithFul = new List<string>();
for (int i = 0; i < ImagesName.Count; i++)
{
    if (ImagesName[i].Contains(pre)) linkWithPre.Add(ImagesName[i]);
}
for (int i = 0; i < ImagesName.Count; i++)
{
    if (ImagesName[i].Contains(ful)) linkWithFul.Add(ImagesName[i]);
}

for (int k = 0; k < ImagesName.Count; k++)
{
    link = ImagesName[k];
    if (link.Contains(pre))
    {
        //clears the "-pre.jpg" from name
        int hu = link.IndexOf(pre);
        LinkWithNoPre = link.Remove(hu, link.Length - hu);


        for (int i = 0; i < linkWithPre.Count; i++)
        {
            link = linkWithPre[i];
            if (link.Contains(LinkWithNoPre))
            {

            }
        }
    }
}

Posted 7-Feb-19 2:53am

_Q12_

Updated 7-Feb-19 5:07am

Add a Solution

3 solutions

Add a Solution

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Richard MacCutchan · Answer 1 · 2019-02-07T03:04:00

Solution 1

So you need to remove pre files that have fullview versions.
Start by enumerating all the fullview names. For each name look to see if there is a pre version, and if so delete it. Or am I missing something?

Posted 7-Feb-19 3:04am

Richard MacCutchan

Comments

_Q12_ 7-Feb-19 10:24am

I like your solution . I was thinking a bit more complicated.
So, for each fullversion, if pre exist, delete it. Very nice.
---
And if are 2 identical in a group of 3 files (1 pre and 2 identical full) ?
I suppose 2 identical lists to compare the first with the second, and make the 3d list with the results. Oh boy.

Richard MacCutchan 7-Feb-19 10:35am

You cannot have two files with identical names in the same folder.

_Q12_ 7-Feb-19 10:56am

heh, thats true. :)

phil.o · Answer 2 · 2019-02-07T03:07:00

You could create a dictionary accepting a string as key (the filename without the -pre or -ful suffix) associated to a tuple of boolean values indicating whether the filename exists with -pre, -ful, or both suffixes.
You could then iterate through the dictionary and act accordingly.

C#

var dic = new Dictionary<string, Tuple<bool, bool>>;
// Here you populate the dictionary with file informations
foreach (string name in dic.Keys) {
  var entry = dic[name];
  if (entry.Item1 && !entry.Item2) {
    // Here the case where the file only exists with the -pre suffix
  }
  else if (!entry.Item1 && entry.Item2) {
    // Here the case where the file only exists with the -ful suffix
  }
  else if (entry.Item1 && entry.Item2) {
    // Here the case where both suffixes are present
  }

MadMyche · Answer 3 · 2019-02-07T04:43:00

Solution 3

I really think the key to this would be to enumerate the actual files via System.IO

1. Use the base-name of the file (a, a1, a2)
2. Check and see if they exist, same with the -pre and -ful variants
3. Compare the base, pre, & full variants to see if they are the same file.
4. If they match you rules, delete what is not needed.

Now there are a couple of ways you could compare them:
1. Use system.io to get basic properties like file-size.
2. Open the files in streamreader and check the MD5 of the contents
3. Open the file and...
3a. Use system.drawing and GDI to check dimensions
-OR-
3b. Use EXIF properties to check dimensions

Posted 7-Feb-19 4:43am

MadMyche

Comments

_Q12_ 7-Feb-19 11:08am

Nice idea also ! Nice approach.
The thing is...i avoid to mention in the start, but i get those images as links actually. In the end, they will be downloaded and physically be on my hdd. But your solution can be well put in practice, After I down them all. Yes, its very nice, especially, checking for dimensions in GDI and System.IO (wich i can do); the other variants i must learn to do them (MD5 stream and Exif).
Pretty sweet.

MadMyche 7-Feb-19 11:47am

Yea, the links throws a wrench into it; these would all require them to be local. Unless the list has more info like file-size etc.
For MD5, there is a good example at https://stackoverflow.com/questions/10520048/calculate-md5-checksum-for-a-file

Cleaning algoritm for cleaning text

3 solutions

Solution 1

Solution 2

Solution 3

Add your solution here

Preview 0

Cleaning algoritm for cleaning text

3 solutions

Solution 1

Solution 2

Solution 3

Add your solution here

Preview 0

Existing Members

...or Join us