Bug 140735 - Editing: The Find operation does not recognize national characters.
Summary: Editing: The Find operation does not recognize national characters.
Status: RESOLVED NOTABUG
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Writer (show other bugs)
Version:
(earliest affected)
7.0.4.2 release
Hardware: All All
: medium normal
Assignee: Not Assigned
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-03-01 13:06 UTC by Jerzy Moruś
Modified: 2021-03-02 23:10 UTC (History)
1 user (show)

See Also:
Crash report or crash signature:


Attachments
Diacritic-sensitive vs. Match character width (9.81 KB, image/png)
2021-03-02 08:26 UTC, Mike Kaganski
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Jerzy Moruś 2021-03-01 13:06:55 UTC
Version: 7.0.4.2 (x64)
Build ID: dcf040e67528d9187c66b2379df5ea4407429775
CPU threads: 4; OS: Windows 10.0 Build 19042; UI render: Skia/Raster; VCL: win
Locale: pl-PL (pl_PL); UI: pl-PL

Writer. The Find operation does not work correctly in this version of the program at my location.
There are 9 additional letters in Polish:
Base letter	Capital letter	code	Small letter	code
	A		Ą	U + 0104	ą	U + 0105
	C		Ć	U + 0106	ć	U + 0107
	E		Ę	U + 0118	ę	U + 0119
	L		Ł	U + 0141	ł	U + 0142
	N		Ń	U + 0143	ń	U + 0144
	O		Ó	U + 00d3	ó	U + 00f3
	S		Ś	U + 015a	ś	U + 015b
	Z		Ź	U + 0179	ź	U + 017a
	Z		Ż	U + 017b	ż	U + 017c



Only the letter "ł" and "Ł" is searched correctly. In other cases, entering a letter, finds that letter and its base letter. Entering the letter code (with regular expression selected), e.g. \u0104, generates the message "Search key not found". Codes \u0141 (Ł) and \u0142 (ł) are recognized correctly.
Comment 1 Mike Kaganski 2021-03-01 19:14:12 UTC
I can't repro using *Find* (Ctrl+F). However, I repro using Replace (Ctrl+H) dialog, with or without regular expressions selected; with all "Other options" unchecked, using Version: 7.1.1.1 (x64) / LibreOffice Community
Build ID: 575c5867c4cc13d7ae78f9ce39a54a52ed38c769
CPU threads: 12; OS: Windows 10.0 Build 19042; UI render: Skia/Raster; VCL: win
Locale: ru-RU (ru_RU); UI: en-US
Calc: CL

I simply copied the text from comment 0, and pasted into a new Writer document as plain text. Copied the characters from Capital letter column, pasted into Find box (in Find toolbar, and in Find and Replace dialog), and pressed "Find All" respective buttons. In Find toolbar, it worked as expected ("Ą" found two places - under "Capital letter" and under "Small letter" - when [ ] Match Case was unselected, and only a single place when it was selected); in Find and Replace dialog, it found 34 places when [ ] Match Case was unselected, and 2 places (in "Base letter" in addition to "Capital letter") when selected.

But it worked as expected in the dialog, as well as in the Find toolbar, in Version: 5.0.0.5 (x64)
Build ID: 1b1a90865e348b492231e1c451437d7a15bb262b
Locale: ru-RU (ru_RU)
Comment 2 Mike Kaganski 2021-03-01 19:19:26 UTC
... but it works fine when [x] Dicaritic-sensitive is checked in the dialog's Other options. So this is not a bug, and it works as expected.
Comment 3 Jerzy Moruś 2021-03-02 08:15:36 UTC
Thank you for resolving the explanation. As an excuse I would like to write that in my language version (PL) the "Diacritic-sensitive" option was described (in Polish) as "Rozróżniaj wielkość znaków" which when translated into English reads "Distinguish character width" and the help itself says:
Distinguish character width (only with Asian languages enabled)
Distinguishes between half-width and full-width characters.
Comment 4 Mike Kaganski 2021-03-02 08:26:25 UTC
Created attachment 170168 [details]
Diacritic-sensitive vs. Match character width

(In reply to Jerzy Moruś from comment #3)
> Thank you for resolving the explanation. As an excuse I would like to write

No need for excuses. It's good to file whatever you consider a bug. Resolving it as NOTABUG with an explanation is a good outcome. It is much better than doubting and not filing at all what could turn out to be a real bug.

> that in my language version (PL) the "Diacritic-sensitive" option was
> described (in Polish) as "Rozróżniaj wielkość znaków" which when translated
> into English reads "Distinguish character width" and the help itself says:
> Distinguish character width (only with Asian languages enabled)
> Distinguishes between half-width and full-width characters.

I suppose it's some confusion here. The "Match character width" option is a different thing: see the attachment. By the way, the "Match character width" is only visible when "Asian" is checked under Options->Language Settings->Languages. The "Diacritic-sensitive" option seems to be missing from help completely.
Comment 5 Mike Kaganski 2021-03-02 08:42:47 UTC
FTR: the related help topic, where the relevant information is missing:

https://help.libreoffice.org/7.1/en-US/text/shared/01/02100000.html
Comment 6 Jerzy Moruś 2021-03-02 08:57:40 UTC
Another observation: In Calc and in the Basic editor (I checked it in them) this option does not need to be checked for everything to work.
Comment 7 Mike Kaganski 2021-03-02 09:06:02 UTC
(In reply to Jerzy Moruś from comment #6)
> Another observation: In Calc and in the Basic editor (I checked it in them)
> this option does not need to be checked for everything to work.

... which would be an own bug, where the option should not be available if does not apply to the module (for unclear reason), or should be implemented properly.
Comment 8 sdc.blanco 2021-03-02 13:26:35 UTC
(In reply to Mike Kaganski from comment #5)
> FTR: the related help topic, where the relevant information is missing:
see bug 103587, where the issue reported here may be resolved, with the result appearing here:

https://help.libreoffice.org/7.2/en-US/text/shared/01/02100000.html

If you (Jerzy, Mike) still see problems in relation to the help for diacritic sensitive, then probably best to report them in bug 103587. For example, it might be helpful to add a Polish example.  Make a concrete suggestion there about what you think the exact text should be. (it will increase the likelihood of relevant further action.)
Comment 9 Jerzy Moruś 2021-03-02 19:20:41 UTC
As colleagues from our Polish Forum report, version 7.1.0.3 already has the correct name of this option.
I would love to add a Polish translation to help, but I don't know how to do it.
Comment 10 sdc.blanco 2021-03-02 23:10:57 UTC
(In reply to Jerzy Moruś from comment #9)
> I would love to add a Polish translation to help, but I don't know how to do
> it.
Let me answer in 3 different ways.

1.  It is possible to make a suggestion for how to translate the current English text.

This link should place you in the right place. You can make a suggestion about how it should be translated.  And I think, sometimes, local adjustments are made. That is, you could add a Polish example in your suggestion.

https://translations.documentfoundation.org/translate/libo_help-master/textshared01/pl/?checksum=e7c1f513ee180d09&sort_by=-priority,position

2.  Or maybe you want to get an account in that system, and contribute to some more translating. 

Look on this page to find the Polish team (and some email addresses). They should be able to give advice about how to get started with translating.

https://wiki.documentfoundation.org/Language_Teams

(You can see here that about 25% of the help is still not translated to Polish).
https://translations.documentfoundation.org/languages/pl/

3. If you give me an example of what might stand on the English page, then I can add it to the English help -- and then -- at some point -- it would be translated in the Polish help.  But probably better to try 1 or 2 first.