Add a template_suffix parameter to BigQuery Table.insert_data

[barakcoh]

Enable the use of template tables by adding the template_suffix parameter.
Documentation: https://cloud.google.com/bigquery/streaming-data-into-bigquery#template-tables

Would love tips on how you feel this should be tested.

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed, please reply here (e.g. I signed it!) and we'll verify. Thanks.

---

• If you've already signed a CLA, it's possible we don't have your GitHub username or you're using a different email address. Check your existing CLA data and verify that your email is set on your git commits.
• If you signed the CLA as a corporation, please let us know the company's name.

[barakcoh]

I signed it!

[tseaver]

We need to add a unit test which covers the case that template_suffix is passed to insert_data.

[tseaver]

@jgeewax can you figure out the CLA issue?

[barakcoh]

@jgeewax @tseaver sorry, I misunderstood the CLA process and thought I signed it when I filled out the form but then got the DocuSign. we're reviewing it internally and will sign it soon.

[barakcoh]

@tseaver re coverage, i can update existing tests with template_suffix='xyz' and I guess the unit tests will pass as it'll have no real effect and so will the coverage.

don't know about system tests as I haven't gone through the steps to run them locally. frankly, it's quite a hassle (at least for such a small change)

[tseaver]

@barakcoh if you can add the template_suffix parameter to just one of the tests for insert_data (e.g., to test_insert_data_with_bound_client) and fix the assertion for the posted data, that will be sufficient.

[tseaver]

It would actually fit better in test_insert_data_with_alternate_client, which already passes the other optional fields.

[barakcoh]

@tseaver done. also rebased the latest upstream (not sure if you prefer it over a merge).

[tseaver]

LGTM as soon as the CLAbot says we're good.

[liorr]

I signed it

[tseaver]

@jgeewax can you verify, or kick the bot to check again?

[jgeewax]

@barakcoh Looks like the e-mail used in these commits (ba....@wonder....com) hasn't signed the CLA. Can you confirm the e-mail address you signed with?

Some options...

• re-sign with the address on the commits (as listed on the patch) CLA signing link
• amend the commit author e-mail for the commits you pushed to the e-mail you signed with (since they weren't merged in yet, this seems fine to me, @tseaver tell me if that's not OK)

---

Sorry about this. I promise it's not that we don't want your commits ... it's just one of those things we need before we can accept outside contributions.

[barakcoh]

@jgeewax the commits are under my email (as you've noted above) but CLA was eventually signed by my boss @liorr ([email protected]), I just tried signing it again using the link you provided but got It looks like you've already signed this CLA..

Do we have to submit all patches using just the email that signed the CLA? Seems kind of a turn off.. hope we can whitelist several emails / GH users under the CLA.

[jgeewax]

@barakcoh Seems weird that it says you've already signed... I can't find your GitHub username or the e-mail attached the commits in this PR, nor @liorr 's username. Is there a specific company that signed the corporate CLA so I can check on that?

The link should show a list of agreements you've signed, that looks like this:

Do we have to submit all patches using just the email that signed the CLA? Seems kind of a turn off.. hope we can whitelist several emails / GH users under the CLA.

Any e-mail that is set as the "author" of a commit in a PR needs to have signed the CLA (otherwise, we don't necessarily know if those commits are yours...). So if you author commits as [email protected] and [email protected] and send a single pull request, both of those e-mails must have signed the CLA -- it's OK to have multiple in a single PR though.

If you have a ton of people and want to do a blanket signature, there's a way to sign "for the company" (here) and you just maintain a Google Group with all the e-mails that are "authorized".

[barakcoh]

thanks @jgeewax, here's what I have:

1. Corporate CLA covering @Wondermall w/ Google Group - https://groups.google.com/forum/#!managemembers/google-conrtibutors-wondermall/members/active
2. I'm in the group with my @Wondermall email account: bar...@wond...com
3. All commits are using my @Wondermall email account

what am I missing here?

I've also asked the signed (@liorr) to join the group but that's pending.

[googlebot]

CLAs look good, thanks!

[barakcoh]

oh thank god! thanks so much @jgeewax !

[jgeewax]

:) as much as I'd love to take credit for it... I think the googlebot just caught up somehow... Sorry for the delay !

[barakcoh]

@jgeewax I figured that much, but still appreciative of the hand-holding. looks like the bot caught on exactly as I submitted my a comment.

@tseaver we good to merge?

[tseaver]

@dhermes ships that cross. I was re-running the tests locally prior to merging.

[dhermes]

My bad. No lingering issues?

[tseaver]

Nope.