fix non-quoted db&catalog issues

[ThomasHepworth]

Type of PR

• BUG
• FEAT
• MAINT
• DOC

Is your Pull Request linked to an existing Issue or Pull Request?

See #1556

Give a brief description for the solution you have provided

Previously, we weren't quoting catalog and database names which was resulting in unusual catalog and database combinations causing SQL errors.

I've simply added quotes to self.splink_data_store (the only location both catalog and db are used) and it appears to be resolving the issue.

To test, you can try creating a random database named 1111 and setting that as your target db from within the linker object:

# Create a database called '1111'
spark.sql("CREATE DATABASE IF NOT EXISTS `1111`")

# Run our actual link job
linker = SparkLinker(
    df_spark,
    settings_dict,
    database="1111"
)

PR Checklist

• Added documentation for changes
• Added feature to example notebooks or tutorial (if appropriate)
• Added tests (if appropriate)
• Made changes based off the latest version of Splink
• Run the linter

[ThomasHepworth]

CI tests are all failing still... I've fixed this here.

[github-actions]

Test: test_2_rounds_1k_duckdb

Percentage change: -30.0%

	date	time	stats_mean	stats_min	commit_info_branch	commit_info_id	machine_info_cpu_brand_raw	machine_info_cpu_hz_actual_friendly	commit_hash
849	2022-07-12	18:40:05	1.89098	1.87463	splink3	c334bb9	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz	2.7934 GHz	c334bb9
1986	2023-08-23	11:35:18	1.32291	1.3118	(detached head)	b87da07	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz	2.7934 GHz	b87da07

Test: test_2_rounds_1k_sqlite

Percentage change: -27.9%

	date	time	stats_mean	stats_min	commit_info_branch	commit_info_id	machine_info_cpu_brand_raw	machine_info_cpu_hz_actual_friendly	commit_hash
851	2022-07-12	18:40:05	4.32179	4.25898	splink3	c334bb9	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz	2.7934 GHz	c334bb9
1988	2023-08-23	11:35:18	3.12197	3.07046	(detached head)	b87da07	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz	2.7934 GHz	b87da07

Click here for vega lite time series charts

[espenfl]

@ThomasHepworth That is great. Thanks a lot. I also started on a PR and was stuck trying to add a proper test for it. I think it would be nice to also add this as part of the PR.

[ThomasHepworth]

@espenfl ah, sorry, I wasn't aware you were also working on a fix!

I'm happy to close this PR and let you polish up a fix and a proper test if you'd like to make a contribution?

Our contribution guide is available here. Please add myself and Robin as reviewers once you're done, if you're up for it.

[espenfl]

@ThomasHepworth No problem. That happens. No way for you to know this as I was a bit slow submitting a draft. I learned some new things and we will get it fixed. That is all that matters. Please let this PR be open. Greatly appreciated. I was actually trying to make a new catalog name which involved figuring out how to set this in the Spark config. Easier and better to just do it with a new database as you do it here as the names are combined anyway in the issue that we want to solve.

[espenfl]

@ThomasHepworth Can I contribute with something to get this one merged?

[ThomasHepworth]

#1558 (comment)

👋 apologies for the delayed reply, I was on leave last week.

Please do. I think it may in fact be easier if you submit a pull request as then I'll be able to QA and merge it personally.

I'll close this PR once you've opened another.

[espenfl]

@ThomasHepworth Unfortunately I am unable to get this done in the next two weeks so maybe we can just leave this as is and get it merged when the dev team find it suitable? Thanks.

[ADBond]

All seems to be working okay, nice one 👍