[solved] 794 - Search doesn't find umlauts in utf-8 files

Bugs and other issues or requests which have been resolved.
Post Reply
Message
Author
larsen
Posts: 168
Joined: 12.01.2013, 15:12

[solved] 794 - Search doesn't find umlauts in utf-8 files

#1 Post by larsen » 08.03.2019, 10:29

Steps to reproduce:
- Place the attaced text file into any directory
- Search for "ümlaut" in that directory

The search should find the file, but it returns nothing.
Attachments
test_umlaut.txt
(7 Bytes) Downloaded 550 times
Last edited by larsen on 05.04.2025, 20:27, edited 1 time in total.

User avatar
ralfso
Posts: 816
Joined: 31.10.2007, 18:21
Location: Gifhorn, Germany

Re: 794 - Search doesn't find umlauts in utf-8 files

#2 Post by ralfso » 08.03.2019, 12:48

Hello Larsen,

I don't know how you created this text-file.

If I save this file to my HD and do a search for containing text "ümlauf" it isn't found.
If I open the file in PSPad and only save it once more (also UTF-8), the file is found for containing text "ümlaut"
Regards
Ralf

Win10 (64-bit), Intel i7-2600 (3,4 GHz), 8 GB Ram, 500 GB SSD, 2x2000 GB HD, NVIDIA GeForce 545

larsen
Posts: 168
Joined: 12.01.2013, 15:12

Re: 794 - Search doesn't find umlauts in utf-8 files

#3 Post by larsen » 08.03.2019, 12:58

Saved it with UltraEdit.

Odamn-Ete
Posts: 280
Joined: 28.06.2017, 07:10

Re: 794 - Search doesn't find umlauts in utf-8 files

#4 Post by Odamn-Ete » 08.03.2019, 14:25

When I open your file in Notepadd++ it shows the folowing:

ümlaut

FC search doesn't find ümlaut, because the 'ü' isn't there.

Check your fonts and settings for UltraEdit.

In checking on this I have to confirm that FC Search doesn't find umlauts (ÄÖÜ) in .txt files in cases where the file encoding is utf-8 without BOM.

.txt files with encoding utf-8 with BOM, FC search finds the files (i.e. text in files)

larsen
Posts: 168
Joined: 12.01.2013, 15:12

Re: 794 - Search doesn't find umlauts in utf-8 files

#5 Post by larsen » 02.04.2025, 16:10

Came upon this problem again and noticed my old thread.
FreeCommander XE 2025 Build 927 64-bit donor

UltraEdit doesn't store a BOM. Notepad++ (8.7.8 ) shows UTF-8, but apparently saves the file with the Little-Endian-Notation "FF FE" BOM.

It works in FC when I manually configure the encoding to use for the search dialog.
However, I'd like not to be forced to use the advanced mode (for a cleaner dialog), but "Options > Advanced Mode" is needed for the encoding to be shown.

A bigger problem however is that the encoding needs to be selected every single time the dialog is re-opened.

Same problem described here:
http://www.forum.freecommander.com/view ... af6#p38687

Proposed solution:
Default value should be UTF-8 as this will be helpful for most users. Also, the value should be restored when opening the search dialog.

horst.epp
Posts: 519
Joined: 15.11.2008, 20:18

Re: 794 - Search doesn't find umlauts in utf-8 files

#6 Post by horst.epp » 02.04.2025, 16:59

larsen wrote: 08.03.2019, 10:29 Steps to reproduce:
- Place the attaced text file into any directory
- Search for "ümlaut" in that directory

The search should find the file, but it returns nothing.
It doesn't make sense to add files in this way.
You have to zip them and attach the archive.
Windows 11 Home, Version 24H2 (OS Build 26100.4652)
Everything Version 1.5.0.1396a (x64), Everything Toolbar 1.5.5, Listary Pro 6.3.2.88
FreeCommander XE 2025 Build 933 64-bit donor

larsen
Posts: 168
Joined: 12.01.2013, 15:12

Re: 794 - Search doesn't find umlauts in utf-8 files

#7 Post by larsen » 02.04.2025, 17:28

As you didn't state a reason why this should be necessary, I can only guess that you mean downloading the file would change the encoding or something else.
Doesn't make a difference. Just tested it.

Marek
Author
Author
Posts: 4234
Joined: 10.04.2006, 09:48
Location: Germany
Contact:

Re: 794 - Search doesn't find umlauts in utf-8 files

#8 Post by Marek » 02.04.2025, 21:48

Proposed solution:
Default value should be UTF-8 as this will be helpful for most users. Also, the value should be restored when opening the search dialog.
- Select the encoding and then create new profile for the search
- Set the profile as first in the list
- Activate option "Open first profile on start"

For defining encoding you must define some search text, you can remove the text from profile in the file FreeCommander.find.ini.

larsen
Posts: 168
Joined: 12.01.2013, 15:12

Re: 794 - Search doesn't find umlauts in utf-8 files

#9 Post by larsen » 03.04.2025, 09:59

Thanks for the workaround, but please also change this in the program itself.
Also, I still need to have the advanced mode enabled, which I prefer not to.
Main problem I see is that many users might be surprised or not even notice that the search dialog doesn't return what the user expects.

btw: There seems to be a bug regarding profiles. Although "AdvMode=1" is saved in FreeCommander.find.ini for my profile, the Advanced Mode is not enabled on opening the search dialog if it has been disabled before. But when I use "Apply selected profile (Enter)" the Advance Mode is indeed enabled. So this probably should also be enabled when opening the dialog ("Open first profile on start" is enabled). Hope this makes sense ;-)

horst.epp
Posts: 519
Joined: 15.11.2008, 20:18

Re: 794 - Search doesn't find umlauts in utf-8 files

#10 Post by horst.epp » 03.04.2025, 14:45

larsen wrote: 02.04.2025, 17:28 As you didn't state a reason why this should be necessary, I can only guess that you mean downloading the file would change the encoding or something else.
Doesn't make a difference. Just tested it.
Browsers have many special functions and encoding.
You may get all sorts of changes to file names by simply downloading, depending on the users OS.
Packing as archive allows getting the real binary value of the file name and content you want.
Windows 11 Home, Version 24H2 (OS Build 26100.4652)
Everything Version 1.5.0.1396a (x64), Everything Toolbar 1.5.5, Listary Pro 6.3.2.88
FreeCommander XE 2025 Build 933 64-bit donor

larsen
Posts: 168
Joined: 12.01.2013, 15:12

Re: 794 - Search doesn't find umlauts in utf-8 files

#11 Post by larsen » 05.04.2025, 20:26

Build 928 now has the encoding selection field always visible (the value is saved and restored).

@Marek: Thanks for changing this!

Post Reply

Who is online

Users browsing this forum: Bing [Bot] and 3 guests