add the `skip_duplicates` option to `Relation.insert1`

[dimitri-yatsenko]

We often insert values that may already exist. The most common example is the filling of lookup tables. To suppress duplicate entry errors, we use the ignore_errors flag. It does the job but it also suppress other errors such as incorrect attribute names. I propose adding the skip_duplicates attribute to Relation.insert.

[fabiansinz]

Agreed.

[eywalker]

Yep, sounds like a good idea.

On Tue, Aug 25, 2015 at 2:47 PM, Fabian Sinz notifications@github.com
wrote:

Agreed.

—
Reply to this email directly or view it on GitHub
#148 (comment)
.

[dimitri-yatsenko]

It does not appear that there is a solution on the MySQL site, so we have to check for duplicates explicitly.
http://stackoverflow.com/questions/548541/insert-ignore-vs-insert-on-duplicate-key-update

So insert1 should check if the new entry is a duplicate first. I am debating whether a transaction is necessary to include both the check and the insert, but it's probably not worth the trouble.

[fabiansinz]

solved by #149