New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[feature] Support OmniSci (MapD) GPU SQL Analytics Engine #4925

Closed
archenroot opened this Issue Jan 6, 2019 · 7 comments

Comments

Projects
None yet
3 participants
@archenroot
Copy link

archenroot commented Jan 6, 2019

Hi,

I would like to use dbeaver for connections to OmniSci platform via JDBC.

Is there any temporary workaround to inject just JDBC and try it or it will require all the data type mapping, DDL/DML language to be mapped into some general model of dbeaver?

Thx for response.

@serge-rider

This comment has been minimized.

Copy link
Member

serge-rider commented Jan 11, 2019

No, only JDBC driver (jar file) + driver configuration is required. n most cases no any data type mappings are required.
If you will succeed in OmniSci configuration - please share driver configuration (URL, class ,etc). We'll include it in one of near DBeaver versions.

@archenroot

This comment has been minimized.

Copy link
Author

archenroot commented Jan 21, 2019

@serge-rider - ok, so connection is established:
image

The config of the driver is straightforward:
image

Here is some sample query:
image

And attached JDBC driver:
mapdjdbc-4.3.0.jar.gz

@archenroot

This comment has been minimized.

Copy link
Author

archenroot commented Jan 21, 2019

And additionally I attach XML config:

<provider id="generic">
                <driver id="B7E68812-B745-F526-374A-D9084F050A77" categories="" custom="true" embedded="false" n$
                        <library type="jar" path="${home}/mapdjdbc-4.3.0.jar" custom="true"/>
                        <parameter name="supports-indexes" value="true"/>
                        <parameter name="omit-single-catalog" value="false"/>
                        <parameter name="all-objects-pattern" value="%"/>
                        <parameter name="quote-reserved-words" value="true"/>
                        <parameter name="omit-single-schema" value="false"/>
                        <parameter name="supports-scroll" value="false"/>
                        <parameter name="omit-catalog" value="false"/>
                        <parameter name="script-delimiter-redefiner" value=";"/>
                        <parameter name="supports-struct-cache" value="true"/>
                        <parameter name="supports-truncate" value="true"/>
                        <parameter name="script-delimiter" value=";"/>
                        <parameter name="legacy-sql-dialect" value="false"/>
                        <parameter name="script-delimiter-after-query" value="false"/>
                        <parameter name="use-search-string-escape" value="false"/>
                        <parameter name="omit-type-cache" value="false"/>
                        <parameter name="split-procedures-and-functions" value="false"/>
                        <parameter name="supports-multiple-results" value="false"/>
                        <parameter name="supports-stored-code" value="true"/>
                        <parameter name="script-delimiter-after-block" value="false"/>
                        <parameter name="supports-references" value="true"/>
                        <parameter name="ddl-drop-column-short" value="false"/>
                        <parameter name="supports-limits" value="true"/>
                        <parameter name="supports-select-count" value="true"/>
                </driver>
        </provider>

@serge-rider serge-rider added this to the 5.3.4 milestone Jan 22, 2019

@archenroot

This comment has been minimized.

Copy link
Author

archenroot commented Jan 22, 2019

@serge-rider - I found one first issue probably can be handled somehow, it does translate timestamp with wrong values:

I run group by command:
SELECT count (*), pickupDate FROM fhv_tripdata group by pickupDate;

image

while in the console of mapd (and how real data looks like) I get:
image

I think its just something with formating, etc. any idea how to handle?

@serge-rider

This comment has been minimized.

Copy link
Member

serge-rider commented Jan 28, 2019

Thanks for details.
However driver descirption xml seems to be corrupted (there is no information about driver class and others).

Do you have some sample DB/script to test it?

Are you aware about any direct link for driver download? Unfortunately they don't have Maven artifact (https://community.omnisci.com/t/jdbc-driver-maven-artifact/613)

@serge-rider

This comment has been minimized.

Copy link
Member

serge-rider commented Jan 30, 2019

@archenroot Are you aware how to obtain jdbc driver by direct link? Or some archive with the driver?
I can put archive you attached somewhere on GitHub but I'm not sure I should. Some official web site would be much better :)

serge-rider added a commit that referenced this issue Jan 30, 2019

@uslss uslss removed the ready for qa label Jan 31, 2019

@uslss uslss closed this Jan 31, 2019

@archenroot

This comment has been minimized.

Copy link
Author

archenroot commented Feb 1, 2019

@serge-rider - I will check with mapd and ask them to deploy to maven, its best way I think.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment