Created attachment 143589 [details]
Screenshots, Import Lotus Files Dialog

An screenshot doesn't help for test, please attach a sample test file, and set to unconfirmed again.

(In reply to m.a.riosv from comment #2)
The file I tried is a business file so I can not attach it.
I will search for attachable files.

Created attachment 143639 [details]
Screenshots(AOO vs. LibO)

(In reply to m.a.riosv from comment #2)
> An screenshot doesn't help for test, please attach a sample test file, and
> set to unconfirmed again.
I could find the file you can try.

STR:
1. Open a Apche OpenOffice Forum
2. https://forum.openoffice.org/ja/forum/viewtopic.php?f=10&t=651#p2353
3. downloads a bugdoc.zip, and expanded

LibO:
4. Drop a bugdoc.123 file onto LibO icon on Desktop
5. Appear LibO window with "Import Lotus file" Dialog
6. Open a "Character set:" Dropdown
7. Select a Japanese(Windows-932)
8. Garbled characters when opened

AOO:
4. Drop a bugdoc.123 file onto AOO icon on Desktop
5. Appear AOO window with "Import Lotus file" Dialog
6. Open a "Character set:" Dropdown
7. Select a Japanese(ShiftJIS)
8. No-garbled characters when opened

Confirmed,
looks there are missing filters for Japanese, not those available on AOo.

FYI:
https://opengrok.libreoffice.org/xref/core/writerperfect/source/common/WPFTEncodingDialog.cxx?r=5f170e69#81
https://opengrok.libreoffice.org/xref/core/writerperfect/source/common/WPFTEncodingDialog.cxx?r=5f170e69#31
https://github.com/apache/openoffice/blob/c014b5f2b55cff8d4b0c952d5c16d62ecde09ca1/main/svx/source/dialog/txenctab.src#L90

confusion between the bibisect commit and the underlying commit.

the source commit is

commit b0067c89e6b2a4e29465d9da9a731ae30a66dce6
Author: osnola <alonso@loria.fr>
Date:   Sun Apr 2 09:42:59 2017 +0200

    libwps import filter improvements
    
      + some astyle modifications,
      + add .wk4 and .123 to the list of file extensions,
      + add support to open Lotus files protected by a password.
    
    Change-Id: I94d4afffd73f0999ff2b1958704cb3985fcd0cc9

Hello,
yes currently the libwps' filter does not support shift-JIS, one solution is to choose the old filter in the filter's dialog "Lotus-1-2-3" which must still exists:-~

After, it is possible to add support shift-JIS encoding in libwps but I will need a file with contains some SJIS characters and its corresponding screenshot (to check that the conversion is valid).

If I must compare the old and the libwps's filter: clearly the libwps's filter supports less encodings while it must be able to open password protected files, retrieve more styles, ... 

Note: There is also a big problem in the old filter ; in fact, a 123 file can contain (and does contain) many sub streams which can be stored in one consecutive block or decomposed in many sub block. The list of streams is stored at the end of the 123 file, the file begins with the main stream which contains the sheet's data, styles, ... (which explains why the old filter often works) ; but if a chart, a picture is inserted in a spreadsheet, the content of the file will become:
  main Zone (part 1)
  picture Zone
  main Zone (part 2)
  ...
  metadata Zone
  file structure (list of stream)
( so the old parser will begin to parse the main Zone (part 1) and will probably stop somewhere in the picture zone)

Created attachment 143883 [details]
new result

I modified libwps ( see https://sourceforge.net/p/libwps/code/ci/42a5c4e27e3b2c7e4de1bcf045ede9a8737f559d/ ), so now it must be able to use Japanese(Shift-JIS) encoding. I still need to do same for wps v1-v4 filter and the new filter for QuattroPro qpw files.

Notes:
- concerning https://forum.openoffice.org/ja/forum/viewtopic.php?f=10&t=651#p2353, I am
  not sure that the problem pointed in this thread is fixed, ie. when I opened this
  file with OpenOffice and LibreOffice (using the old parser), saving the result seems
  to take forever ; but my versions of OpenOffice and LibreOffice are not the last
  ones...
- concerning libwps, I guess that we will need to use libicu to read more encodings, 
  ...

(In reply to osnola from comment #12)

I chose Japanese(Windows-932) in Win-x86_64@42, and after opening, there was not garbage character.

Version: 6.2.0.0.alpha0+ (x64)
Build ID: 0a1a4ffb4f87adff7fbbbc60202b6a0e42fedd0c
CPU threads: 4; OS: Windows 10.0; UI render: default; 
TinderBox: Win-x86_64@42, Branch:master, Time: 2018-08-08_23:17:46
Locale: ja-JP (ja_JP); Calc: group threaded

Hello,
yes normally, this must be ok in master, ie. I release a new version of libwps which  must allow to parse CP932 and CP950 encodings in its different filters and David integrated with https://gerrit.libreoffice.org/#/c/58704/ .

Note:
CP_037, CP_424, CP_437, CP_500, CP_737, CP_775,
	            DOS_850, CP_852, CP_855, CP_856, CP_857, CP_858, CP_860,
	            CP_861, CP_862, CP_863, CP_864, CP_865, CP_866,
	            CP_869, CP_874, CP_878, CP_875, CP_932, CP_936, CP_950,
	            CP_970, CP_1006, CP_1026,
	            EUC_JP,
	            ISO_2022_CN, ISO_2022_JP, ISO_2022_KR,
	            WIN3_ARABIC, WIN3_BALTIC, WIN3_CEUROPE,
	            WIN3_CYRILLIC, WIN3_GREEK, WIN3_HEBREW, WIN3_TURKISH,
	            WIN3_VIETNAMESE, WIN3_WEUROPE,
	            LICS,

	            MAC_ARABIC, MAC_CELTIC, MAC_CEUROPE, MAC_CROATIAN,
	            MAC_CYRILLIC, MAC_DEVANAGA, MAC_FARSI, MAC_GAELIC,
	            MAC_GREEK, MAC_GUJARATI, MAC_GURMUKHI, MAC_HEBREW,
	            MAC_ICELAND, MAC_INUIT, MAC_ROMAN, MAC_ROMANIAN,
	            MAC_THAI, MAC_TURKISH,

Oops...

Note:
- I also create a branch of libwps to test the integration with libicu  
  (https://sourceforge.net/p/libwps/code/ci/testICU/tree/) 
  which may be able to read: 
                    CP_037, CP_424, CP_437, CP_500, CP_737, CP_775,
	            DOS_850, CP_852, CP_855, CP_856, CP_857, CP_858, CP_860,
	            CP_861, CP_862, CP_863, CP_864, CP_865, CP_866,
	            CP_869, CP_874, CP_878, CP_875, CP_932, CP_936, CP_950,
	            CP_970, CP_1006, CP_1026,
                    CP_1250, CP_1251, CP_1252, CP_1253, CP_1254, 
                    CP_1255, CP_1256, CP_1257, CP_1258,
	            EUC_JP,
	            ISO_2022_CN, ISO_2022_JP, ISO_2022_KR,

	            MAC_ARABIC, MAC_CELTIC, MAC_CEUROPE, MAC_CROATIAN,
	            MAC_CYRILLIC, MAC_DEVANAGA, MAC_FARSI, MAC_GAELIC,
	            MAC_GREEK, MAC_GUJARATI, MAC_GURMUKHI, MAC_HEBREW,
	            MAC_ICELAND, MAC_INUIT, MAC_ROMAN, MAC_ROMANIAN,
	            MAC_THAI, MAC_TURKISH,
   if you see some encodings which are missing, I can add them...

Cannot reproduce with Libreoffice 6.2.0RC1.
No problem with release version(RC1).

Version: 6.2.0.1 (x64)
Build ID: 0412ee99e862f384c1106d0841a950c4cfaa9df1
CPU threads: 4; OS: Windows 10.0; UI render: default; VCL: win; 
Locale: ja-JP (ja_JP); UI-Language: en-US
Calc: threaded

Cannot reproduce with LibO 6.2.0R.
Thanks for fixing this!

(In reply to baffclan from comment #17)
> Cannot reproduce with LibO 6.2.0R.
> Thanks for fixing this!

Thanks for retesting with the latest version.
Setting to RESOLVED WORKSFORME as the commit fixing this issue hasn't been identified.