[ODBC-272] MariaDB ODBC corrupts imports. Created: 2020-02-07  Updated: 2020-04-06

Status: Open
Project: MariaDB Connector/ODBC
Component/s: None
Affects Version/s: 3.1.6
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Fabio Valeri Assignee: Lawrin Novitsky
Resolution: Unresolved Votes: 0
Labels: corruption, excel, odbc
Environment:

10.4.8-MariaDB
Windows 10 Pro Workstation, OS Version: 10.0.18362 N/A Build 18362
MSOffice 365
R version 3.6.1 (2019-07-05)

Attachments: Zip Archive labor_limits_201911_20200207.zip    

 Description   

When using MariaDB-odbc to import data from MariaDB into R or Excel the imported table is corrupted. This is not the case if I use MySQL-odbc. The kind of corruption is that certain columns get inserts with 0 and shifts all rows of only these columns downwards.

In Stackoverflow is a detailled description of this problem created in 31.01.2019 with no response until this bug-report.

I loaded up the table which I used to show the bug.

 Comments   
Comment by Lawrin Novitsky [ 2020-04-05 ]

Thank you for your report.
Importing to Excel, have you experienced the same corruption as with R?
I've just tried to import into Office365 Excel spreadsheet, and I can't see any corruption. The small difference is that I used current repository version, which is 3.1.6 + few fixes. But I doubt any of those fixes could affect importing to Office.

I'll install and try R as well. I just don't have it at the moment.

Comment by Fabio Valeri [ 2020-04-06 ]

Using MariaODBC-driver to import the table via MSquery I got the following messages: string data, right truncated. Diese Datei konnte nicht gelesen werden. (These data couldn't be read). With Mysql-ODBC I could import the table.

Comment by Lawrin Novitsky [ 2020-04-06 ]

I've just tried to load data via query and it also worked fine for me - no errors, no corruption, the data is the same as in the MariaDB table. The only thing I noticed was that in the preview it ordered the data by group0, analyse(and maybe other) columns. Thus the order of rows was a different a bit. But data was correct.
Could you elaborate a bit what do you do? And could you please turn on ODBC tracing and attach the trace here? And maybe separately the trace of importing of the table into R. That might work faster.

Generated at Thu Feb 08 03:27:29 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.