Exclude folders in duplicate search

Hans L · Jul 11, 2017#12017-07-11T14:05+00:00

Since Help > Online help still (after two months or so) points to 404, I will ask a simple question here.

Find duplicates > Exclude folders

A tutorial page I found says this:

"Exclude Folders
This filter allows you to specify folders within your find location that should NOT be searched for duplicates. For instance, if you want to exclude all .svn folders, you could add *.svn into this field. If you want to exclude all folders named "bak", then you would simply type bak into this field. Note that this is a filter, so all subfolders within your find locations with names matching the filter will be excluded."

When did folders get extensions? Is this something I have totally missed? I'm sure nothng prevents one from adding an "extension" to a folder, but I have not seen one that I can remember.

I put a particular folder name (incl. path) in this field, and it worked. Then, I put two folder names (incl. paths) separated with a comma; the first was ignored, while the second was 'honored'.

I would be grateful for some guidance.

Regards,

Hans L

Mofi · Jul 11, 2017#22017-07-11T17:39+00:00

Folders don't have an extension. But it is of course possible to give a folder the name bak instead of backup, or MyFolder.bak instead of Backup of MyFolder.

.svn is a special subfolder in folders managed with Subversion which is a software versioning system supported built-in by UEStudio. The folder name starts with a point because on Unix/Linux files and folders which name is starting with a point are hidden files and folders. On Windows there is the hidden attribute to mark a file or folder as hidden. But on Unix/Linux file systems there is no attribute to mark a file or folder as hidden. Windows version of Subversion uses the same folder naming as on Unix/Linux although on Windows file systems there is a hidden attribute also available.

On Windows it is not advisable, but also not forbidden, to name folders with a point in name because the could be easily misinterpreted as file names on not good written code.

For example a command prompt window is opened and in this window following commands are executed:

Code: Select all

md MyTextFiles.txt
echo Summary>Summary.txt
for /F %I in ('dir *.txt /B') do @echo Name: %~nI  Extension: %~xI
del Summary.txt >nul
rd MyTextFiles.txt

The output of for loop is:

Code: Select all

Name: MyTextFiles  Extension: .txt
Name: Summary  Extension: .txt

The folder with name MyTextFiles.txt is also listed because also matched by the pattern *.txt which would not be the case when the folder name would be just MyTextFiles.

The solution is using the for command line:

Code: Select all

for /F %I in ('dir *.txt /A-D /B') do @echo Name: %~nI  Extension: %~xI

The additional dir option /A-D results in ignoring all folder entries having directory attribute set.

A similar problem exists with file names of Unix/Linux files starting with a point for being hidden on Unix/Linux and not having a file extension. A very common example for such a file is .htaccess which exists on many websites at least in root directory of the website.

Again a Windows command line example:

Code: Select all

for /F %I in (".htaccess") do @echo Name: %~nI  Extension: %~xI

The output is:

Code: Select all

Name:   Extension: .htaccess

So Windows command interpreter interprets the file .htaccess as file without a file name having only a file extension according to simply rule everything left of last point in name of file is the file name and last point and everything right of it is the file extension.

Such for Windows unusual file names should be taken into account on processing names of files on Windows. See for example Create a backup copy of files or General file name evaluating functions where you can see how I have done it in a batch file code and in functions for UltraEdit scripts..