Add Ignore names option to Synchronize Dirs

[mmm]

TC currently does not support search for copies of the same data scattered all over the disk space if filenames differ.

Proposal:
Please consider implementing Ignore name option in Synchronize Dirs.

Rationale
Hunting for copies created by CTRL+C and CTRL+V is quite a common task whenever you want to get rid of redundant data or simply need to free up disk space.

The list of typical filename discrepancies include:
Copy and paste "myfile" vs "myfile – copy (2)" // by far the most common use case
space "myfile" vs "my file" // edited manually
underscore "myfile" vs "my_file" // edited manually
automated padding "myfile" vs "myfile0002" // made by automation machines



[Christian Ghisler, author] 2022-03-04
Unfortunately you cannot do this job with my sync tool.
Why? There wouldn't be a 1:1 relationship between files on the left and those one on the right if names were ignored.

[mmm]
Yes, I know. Having said that, I am convinced that a large number of TC users would greatly benefit from the option that is currently missing.



I would like to invite comments and thoughts from broader audience.

[Hacker]

mmm,

TC currently does not support search for copies of the same data scattered all over the disk space if filenames differ.

Would Duplicate search fit your needs? It searches for copies of the same data scattered all over the disk space and filenames do not matter.

[Christian Ghisler, author] 2022-03-04
Unfortunately you cannot do this job with my sync tool.
Why? There wouldn't be a 1:1 relationship between files on the left and those one on the right if names were ignored.

[mmm]
Yes, I know. Having said that, I am convinced that a large number of TC users would greatly benefit from the option that is currently missing.

How would you suggest to implement it? Currently it looks like the author says it's not possible, and you say "I know, but please implement it anyway". Without some idea of how to make it work it will be difficult to implement.

Roman

[mmm]

Roman,
I will be more than happy to discuss with you more details of "what" I would like to accomplish with the new tool; whether it be a search tool for duplications or a true sync tool.

With all due respect I am going to refrain from commenting on "how" to implement the algorithm as I would consider it totally inappropriate, not to mention that I do not find myself qualified enough for mentoring TC developers.

Hope you understand,
mmm

[mmm]

Would Duplicate search fit your needs? It searches for copies of the same data scattered all over the disk space and filenames do not matter.

I do not think your statement is 100% correct. If am not mistaken, current implementation of "search for duplicates" does not help when you are looking for duplicates of files located in a selected folder (left pane) and you want to investigate potential copies located in a target data structure that is not within the "source" folder (right pane).

Put another, 'search for duplicates' does the job within a single data structure only; i.e. from the selected folder down the hierarchy.

Yet another, you cannot find copies of local files located on a remote server.

Please let me know if I am wrong.

[Hacker]

mmm,

'search for duplicates' does the job within a single data structure only; i.e. from the selected folder down the hierarchy.

That is not correct, you can add as many folders as you like under "Search in".

With all due respect I am going to refrain from commenting on "how" to implement the algorithm as I would consider it totally inappropriate, not to mention that I do not find myself qualified enough for mentoring TC developers.

Actually, it's the complete opposite - it's the best way to get a suggestion implemented - to present a workable concept / idea.

Roman

[mmm]

Oh my goodness. I did not know I could search in multiple folders.
Many thanks for your educating me!

When I fully test the entire search process I will let you know about the outcome.

Thanks and regards,
mmm

[nsp]

Oh my goodness. I did not know I could search in multiple folders.

Just my 2¢ :
I usually first search for folders (for a whole list or projects to cleanup) -> feed to listbox (select ALL)-> search for duplicates in selected file/folder ->feed to listbox -> use specific numpad[+]

If you have to search many times in same list of folder, you can save/edit the list into a file and use loadsearch to fead the listbox and then select all then do your searchs....

[mmm]

Very nice. Thanks.
This kind of feedback is highly appreciated. Always.

Just out of curiosity - did you ever consider submitting a request for automating this process in TC in order to eliminate/minimize manual work ? Do you think it is worth implementing, considering number of potential users and the number of steps to follow?

[nsp]

Just out of curiosity - did you ever consider submitting a request for automating this process in TC in order to eliminate/minimize manual work ? Do you think it is worth implementing, considering number of potential users and the number of steps to follow?

You can do it yourself in suggestion forum..
//Now it is a bit late for 10.5 as during R.C version Christian usually do not add features (he just do BugFix).
I use PowerPro to automate my work and in many case i find a way to complement TC with scripts.

[mmm]

mmm
Just out of curiosity - did you ever consider submitting a request for automating this process in TC in order to eliminate/minimize manual work ? Do you think it is worth implementing, considering number of potential users and the number of steps to follow?

nsp
You can do it yourself in suggestion forum..

Well, this conversation has just circled.

---------------------------------------
Let's go second round then:
---------------------------------------

Much as I try I cannot figure out how to hunt for copies scattered all over disk space and select redundant files for removal.

Synchronize Dirs is absolutely fascinating tool designed for this type of job; however, there is one major constraint - it is restricted to files with the same names only. As far as I know you cannot find copies created by means of Copy and Paste on Windows; e.g. "myfile - copy (53).xlsx".

Does TC file manager support disk cleanup?
If not, I would like to encourage TC team to give it a thought.
I cannot stress more that I am looking for a robust built-in solution with graphical interface and minimum user interaction.

[Hacker]

mmm,
So the main question is what does not suit you about duplicate search?

Roman

[mmm]

Search part - GUI for selecting data to search for aka source files aka golden files (left pane)
GUI for selecting data structure to search in (right pane)

Selection and deletion part - GUI for picking duplications/copies to be deleted


Put simply, I am missing the incredibly smooth and intuitive workflow implemented in Synchronize Dirs.

[Hacker]

mmm,

Search part - GUI for selecting data to search for aka source files aka golden files (left pane)
GUI for selecting data structure to search in (right pane)

True, if you treat a group of files as "originals" and another group as "possible duplicates", I don't think there is really a good way in TC to do that.

Selection and deletion part - GUI for picking duplications/copies to be deleted

Well, after the search is complete, if you click Feed to listbox and press Num+, there is a dialog which allows you to select duplicates based on various criteria (it is different from the usual Num+ selection dialog).

Personally I am using specialized tools for duplicate search that offer an option to keep a database of file hashes to speed up consecutive searches.

Roman

[mmm]

Personally I am using specialized tools for duplicate search that offer an option to keep a database of file hashes to speed up consecutive searches.

How sad.

I hope my suggestion is still valid regardless your personal preference.

[Hacker]

mmm,

How sad.

I have learned to live with this deep sadness

I hope my suggestion is still valid regardless your personal preference.

Of course. My preference is irrelevant. Just please remember, the more detailed your suggestion / concept is, the better chance of it being implemented.

Roman