Characters turned into question marks

I have wrote a document by Arabic language Arial font, after I finished I saved the documnet in doxc extension to open it later on other device, but when I opened it by Libre again I noteced all my written word have been turned to just ? question marks , What is the solve of this problem, and how can I get back (recover) my written document? help please

EDIT: hg.docx

First rule. Always save original in native ods format, do a Save As to export to foreign format.

Can you glean anything from this question?

unforetunatly I can not found any solution for this…

Share the document.

You can attach a document clicking on the edit link under your question. Scroll to the bottom. Type twice Enter. Use the paperclip tool.

hg.docx

No Never I am always using Windows 10 last edition. but later I noticed that my non unicode language is setted to ENGlish not Arabic, I think this was the main reason. But there is an important point that though of this setting MS Word doesn’t make this problem (I frequently using it, I never faced by this problem)
My story began when I opened the a blank document by Libre and continued to writing after finishing I saved it with docx ext. and when I reopened it I saw that question marks took place instead of really written letters,
I quess that Libre should solve this fatal error seperatly from (irrespective of)whta is the language settings of user’s windows. Thank you very much.

  1. Do not answer your own question unless it is a real answer. Modify or comment the original one.
  2. The file your shared is not a DOCX file, but a plain text file. Apparently, it was saved in a wrong encoding (meaning that the encoding does not contain Arabic), and there is no way to restore the text.
  3. As already said by Earnest Al, keep your work in ODF and only after you have finished, export to a foreign format.

Are you really sure you saved as .docx? It looks rather that you somehow saved as .txt. .txt and .docx (OOXML variant) follow one another in the menu. It is easy to click on the wrong one. You then get a popup dialog asking you to choose the encoding. You can of course select one of the Arabic candidates but they are legacy alphabets.

Perhaps you tried to correct the error changing the file extension. But this does not work because .docx uses a completely different representation including the formatting.

If you reopen a .txt with legacy encoding, Writer or Word discover a non-Unicode file they can’t interpret. The legacy alphabet results in faulty UTF-8 sequence (for example ISO-8859-6 for Arabic uses bytes 0xA0-0xFE for characters while in Unicode 0x80-0xBF are extension bytes, 0xC0-0xF7 are initial bytes and 0xF8-0xFF are illegal).

When such a file is later saved, all 0x80-0xFF bytes are changed to ?.

No bug, only user error.

The attached file probably comes from a Mac (linebreak is CR+LF). Apart from a few numbers, it consists exclusively of U+003F QUESION MARK and some spaces. As @gabix pointed out, it is neither an .odt not a .docx document. It was probably a plain text one but it underwent encoding conversion to ASCII, losing all non-ASCII characters. There is some math inside but everything is damaged beyond repair.

Put your hand on the original file and save it as an .odt document. Avoid errors when you get it.

Nothing else can be done.

To show the community your question has been answered, click the ✓ next to the correct answer, and “upvote” by clicking on the ^ arrow of any helpful answers. These are the mechanisms for communicating the quality of the Q&A on this site. Thanks!

In case you need clarification, edit your question (not an answer which is reserved for solutions) or comment the relevant answer.

text line-break is CR+LF

That actually suggests that the source is Windows.

The 3 line-ending options for the main OS are:-

  1. Windows: CR + LF (2 bytes: U+000D + U+000A)
  2. Mac (legacy): CR (1 byte: U+000D)
  3. Linux + Mac (osx): LF (1 byte: U+000A)

(Wikipedia)