Polish characters displaying ? on reports

building reports
1

Hi,
We’ve had issues with Polish special characters not displaying correctly on reports. They are represented as ?.
I’ve checked the SAP support site and implemented the support for multibyte characters though the solutions are for Oracle DB. The universe/connection is using MySQL and I’m not sure how to proceed at this point. I’ve installed the Arial Unicode MS font which from my research supports Polish characters.

Can anyone help me out on this one?

2

It may be the viewer/user computer that needs the font. The report may have the correct unicode bytes.

3

hi, I’d check:

  • how are data stored in MySQL (if in utf?),
  • how is set the db driver (for additional params),
  • if an universe is used check parameter unicode_strings (YES),
  • and which font is used for that field in report.
    BR Tomas
4
  1. According to the developer, its stored UTF8 in DB.
  2. Default, havent changed anything on the driver side. I just checked now, clicked Details and under the connection tab, Character set is blank.
  3. Yes, Unicode_Strings in Universe is set to YES
  4. Arial Unicode MS is the font used.

I’ll try changing the character set on the drivers side to see if it would help things out.

5

Its displaying ? on mine and I have the Arial Unicode MS font installed on my machine.

6

Is it possible to save the document locally and open it with a hex editor?
On a unix system I’d use the “od” command or the “hexdump” command. There are several freeware tools that can do the same on Windows.

Either your system does not have the correct font (or some setting is preventing its use) or the document does not contain the right characters.

I think it’s easier to prove the document correct or incorrect than to mess with Windows settings unnecessarily.

Are you reading the report in your browser? In MS Word? As a PDF? That also may affect your computers ability to display correctly. If the report structure makes it hard to see internally perhaps make your title something like “XXXXXXX?XXXXXXX” where there are simple repeated non-extended characters wrapped around the target character.

I do recall some difficulty with multi-byte settings on our old RS6000 running AIX 4.1 - I believe our solution was to not use them. But that was several years ago.

7

Do the characters display correctly when querying in MySQL Workbench? Just checking because this happened with us a few years ago and it turned out the database wasn’t holding them correctly.

Check your my.cfg file contains the following:

[client]default-character-set=utf8
[mysql]default-character-set=utf8

[mysqld]collation-server = utf8_unicode_ci
init-connect='SET NAMES utf8'character-set-server = utf8
8

Is this the config file for MySQL itself or the cfg file on the dataaccess folder on the BO installation folder?

9

An update, I checked the data contained within the table in MySQL and the Polish special characters are appearing correctly. When I check the actual field, the collation is “utf8mb4_unicode_ci”. Now I’m not sure what that means. On the ODBC driver side, I have set the connection to utf8. When i refresh the report now, instead of “?” its now showing characters but its different from what is shown on the database. Should i match the character set on the ODBC connection to what is on the DB? I’m seeing utf8mb4 on the ODBC settings but I’m not sure if that is the one I should have selected.

10

Apologies, it’s been a while so I cannot remember which it was.

In terms of collation itself, here’s a good reference: What is Collation in Databases? | Database.Guide

From that, you should be able to figure out what you need - if not, your DBA should be able to help.