| Unexpected behavior when combining "Ignore Case" and "Ignore Composition" [message #1755] |
Sun, 07 January 2024 11:59  |
Atlas
Messages: 140
Registered: August 2009
|
Senior Member |
|
|
Search 1
1. Apply advanced filter -> File Name -> Contain any of the string -> Apply "Multiple Strings" + "Ignore Composition" -> Search for a single character "ℕ" in file name
2. Result: This seems to work as expected and I see files that contain both "ℕ" and "N" (normal capital letter N) in their file names.
Search 2
1. Apply advanced filter -> File Name -> Contain any of the string -> Apply "Multiple Strings" + "Ignore Composition" + "Ignore Case" -> Search for a single character "ℕ" in file name
2. Result: This seems to only return files with " N " (uppercase N as a single word with spaces around the letter) or " n " (lower case N as single word with spaces around the letter) in their names, but no file with "ℕ".
This is confusing, because Search 2 is exactly the same as Search 1, except that Search 2 adds "Ignore Case", and this should give me more results because upper and lower case are treated the same. Thus, Search 1 results should be a subset of Search 2 results. Instead, Search 2 return fewer results, and it does two strange things: (1) It doesn't return any result with "ℕ", which it should since I use "Ignore Composition" (2) It return file names with the letter "N" or "n" in their names, but only when they occur as whole words with spaces around the letters.
Am I missing something here? Please do a quick check and let me know.
[Updated on: Mon, 08 January 2024 09:45] Report message to a moderator |
|
|
|
| Re: Unexpected behavior when combining "Ignore Case" and "Ignore Composition" [message #1790 is a reply to message #1755] |
Tue, 19 March 2024 12:14  |
FoxTrot Engineering
Messages: 406
Registered: April 2020
|
Senior Member |
|
|
There are actually two issues here:
- a bug when using "Ignore Composition" and "Ignore Case" at the same time when when performing [Apply advanced filter]; this only happens for characters that are decomposed to upper case letters
- searching for "ℕ" in a filter may not find "N", because FoxTrot considers that ℕ should be indexed as an independent word, even when there is no space or punctuation to separate it from the adjacent characters, and this interferes with how filters are applied in this situation
Both issues will be fixed in the next release
Jérôme - FoxTrot Engineering
|
|
|
|
] 
FoxTrot Search Forum
Search
Help
Members
Register
Login
Home
