html_url,issue_url,id,node_id,user,user_label,created_at,updated_at,author_association,body,reactions,issue,issue_label,performed_via_github_app
https://github.com/dogsheep/twitter-to-sqlite/issues/3#issuecomment-541493242,https://api.github.com/repos/dogsheep/twitter-to-sqlite/issues/3,541493242,MDEyOklzc3VlQ29tbWVudDU0MTQ5MzI0Mg==,9599,simonw,2019-10-14T03:35:36Z,2019-10-14T03:35:36Z,MEMBER,"https://developer.twitter.com/en/docs/tweets/search/api-reference/get-search-tweets
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",488833975,Command for running a search and saving tweets for that search,
https://github.com/simonw/datasette/pull/595#issuecomment-541494080,https://api.github.com/repos/simonw/datasette/issues/595,541494080,MDEyOklzc3VlQ29tbWVudDU0MTQ5NDA4MA==,9599,simonw,2019-10-14T03:43:11Z,2019-10-14T03:43:11Z,OWNER,"Ooh this is tricky... https://travis-ci.org/simonw/datasette/jobs/597216285 failed because it looks like `websockets-8.0.2` isn't compatible with Python 3.5

This is a problem. I need to keep Datasette working with 3.5 for compatibility with Glitch. But obviously I also want it to work with 3.8. 

I wonder if there's a smart solution here. @tomchristie any thoughts?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506300941,bump uvicorn to 0.9.0 to be Python-3.8 friendly,
https://github.com/simonw/datasette/pull/592#issuecomment-541494347,https://api.github.com/repos/simonw/datasette/issues/592,541494347,MDEyOklzc3VlQ29tbWVudDU0MTQ5NDM0Nw==,9599,simonw,2019-10-14T03:45:29Z,2019-10-14T03:45:29Z,OWNER,"This is a really nice implementation. I particularly like how you have the button as hidden until JavaScript runs and makes it visible.

https://github.com/simonw/datasette/pull/592/files#diff-0932659dded7aa3ce2d4c7badb187a9cR6-R8","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",505950145,Offer SQL formatting,
https://github.com/simonw/datasette/pull/590#issuecomment-541495034,https://api.github.com/repos/simonw/datasette/issues/590,541495034,MDEyOklzc3VlQ29tbWVudDU0MTQ5NTAzNA==,9599,simonw,2019-10-14T03:51:32Z,2019-10-14T03:51:32Z,OWNER,"We need some unit tests for this.

I'd be fine with us renaming `extra_database.db` to `extra database.py` in these existing fixtures:

https://github.com/simonw/datasette/blob/af2e6a5cf186a7200d76cb67ac30fa59cc24d84e/tests/fixtures.py#L174-L185

Then we could add an extra `test_database_name_with_spaces(...)` test in `test_api.py` which uses parametrize to exercise a bunch of URLs and check they all return a 200 - something a bit like this but covering a range of URL paths (one for database, table, row and table with filters):

https://github.com/simonw/datasette/blob/90d4f497f9b3f6a5882937c91fddb496ac3e7368/tests/test_api.py#L746-L765","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",505818256,Handle spaces in DB names,
https://github.com/simonw/datasette/pull/591#issuecomment-541495108,https://api.github.com/repos/simonw/datasette/issues/591,541495108,MDEyOklzc3VlQ29tbWVudDU0MTQ5NTEwOA==,9599,simonw,2019-10-14T03:52:09Z,2019-10-14T03:52:09Z,OWNER,Neat solution.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",505837199,Sort databases on homepage by argument order,
https://github.com/simonw/datasette/pull/578#issuecomment-541495682,https://api.github.com/repos/simonw/datasette/issues/578,541495682,MDEyOklzc3VlQ29tbWVudDU0MTQ5NTY4Mg==,9599,simonw,2019-10-14T03:57:17Z,2019-10-14T03:57:17Z,OWNER,"So this is a new Docker feature from April 2019? https://engineering.docker.com/2019/04/multi-arch-images/

> So why buildx?  Let’s first start with the name.  The x stands for experimental.  In the future, when these new build features are stable and made generally available, we will drop the x and integrate these features directly into the existing docker build command.  Note that as buildx is experimental, features and flags are subject to change.

Is there a strong use-case for supporting this while it's still experimental? Users who want Datasette to run on ARM can build their own image pretty easily, so I'm not sure if it's worth depending on an experimental feature for the core Datasette project.

Is there any indication from the Docker team as to when this experimental feature might stop being marked as experimental?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",499954048,Added support for multi arch builds,
https://github.com/simonw/datasette/pull/595#issuecomment-541499978,https://api.github.com/repos/simonw/datasette/issues/595,541499978,MDEyOklzc3VlQ29tbWVudDU0MTQ5OTk3OA==,4312421,stonebig,2019-10-14T04:32:33Z,2019-10-14T04:33:26Z,NONE,"Maybe make the setup rule conditional, so that below python-3.6, it looks for unicorn-0.8 ?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506300941,bump uvicorn to 0.9.0 to be Python-3.8 friendly,
https://github.com/simonw/datasette/pull/590#issuecomment-541562581,https://api.github.com/repos/simonw/datasette/issues/590,541562581,MDEyOklzc3VlQ29tbWVudDU0MTU2MjU4MQ==,2657547,rixx,2019-10-14T08:57:46Z,2019-10-14T08:57:46Z,CONTRIBUTOR,"Ah, thank you – I saw the need for unit tests but wasn't sure what the best way to add one would be.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",505818256,Handle spaces in DB names,
https://github.com/simonw/datasette/pull/590#issuecomment-541587823,https://api.github.com/repos/simonw/datasette/issues/590,541587823,MDEyOklzc3VlQ29tbWVudDU0MTU4NzgyMw==,2657547,rixx,2019-10-14T09:58:23Z,2019-10-14T09:58:23Z,CONTRIBUTOR,Added tests.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",505818256,Handle spaces in DB names,
https://github.com/simonw/datasette/pull/595#issuecomment-541664602,https://api.github.com/repos/simonw/datasette/issues/595,541664602,MDEyOklzc3VlQ29tbWVudDU0MTY2NDYwMg==,647359,tomchristie,2019-10-14T13:03:10Z,2019-10-14T13:03:10Z,NONE,"🤷‍♂️ @stonebig's suggestion would be the best I got too, *if* you want to support 3.5->3.8.

It's either that, or hold off on 3.8 support until you're ready to go to 3.6->3.8.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506300941,bump uvicorn to 0.9.0 to be Python-3.8 friendly,
https://github.com/dogsheep/github-to-sqlite/issues/7#issuecomment-541721437,https://api.github.com/repos/dogsheep/github-to-sqlite/issues/7,541721437,MDEyOklzc3VlQ29tbWVudDU0MTcyMTQzNw==,9599,simonw,2019-10-14T14:44:12Z,2019-10-14T14:44:12Z,MEMBER,Docs: https://github.com/dogsheep/github-to-sqlite/blob/0.5/README.md#retrieving-issue-comments-for-a-repository,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506276893,issue-comments command for importing issue comments,
https://github.com/dogsheep/twitter-to-sqlite/issues/10#issuecomment-541748580,https://api.github.com/repos/dogsheep/twitter-to-sqlite/issues/10,541748580,MDEyOklzc3VlQ29tbWVudDU0MTc0ODU4MA==,9599,simonw,2019-10-14T15:30:44Z,2019-10-14T15:30:44Z,MEMBER,Had several recommendations for https://github.com/tqdm/tqdm which is what goodreads-to-sqlite uses.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",492297930,Rethink progress bars for various commands,
https://github.com/simonw/datasette/pull/578#issuecomment-541837823,https://api.github.com/repos/simonw/datasette/issues/578,541837823,MDEyOklzc3VlQ29tbWVudDU0MTgzNzgyMw==,887095,heussd,2019-10-14T18:19:42Z,2019-10-14T18:19:42Z,NONE,"My use case was: I wanted to use datasette on a Raspberry Pi. `docker pull datasetteproject/datasette` pulled the official image, which then failed to execute because it is not ARM ready. Building my own quite took some time (~60 minutes via Qemu on Intel i5). 

You are right, the build method is quite new and I would not be surprised if the syntax / command will change in future. The outcome however, a Docker multi-architecture manifest, is aligned with Docker's strategy on how to tackle multiple architectures: transparently, on the registry-side.

I just thought it would be nice to have the official image ready for multiple architectures. But I fully understand if the current methods feel too experimental to be mergable...","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",499954048,Added support for multi arch builds,
https://github.com/simonw/datasette/pull/595#issuecomment-541922637,https://api.github.com/repos/simonw/datasette/issues/595,541922637,MDEyOklzc3VlQ29tbWVudDU0MTkyMjYzNw==,9599,simonw,2019-10-14T21:08:19Z,2019-10-14T21:08:19Z,OWNER,There more discussion around this on the Glitch forums here: https://support.glitch.com/t/can-you-upgrade-python-to-latest-version/7980,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506300941,bump uvicorn to 0.9.0 to be Python-3.8 friendly,
https://github.com/simonw/datasette/pull/595#issuecomment-541931047,https://api.github.com/repos/simonw/datasette/issues/595,541931047,MDEyOklzc3VlQ29tbWVudDU0MTkzMTA0Nw==,9599,simonw,2019-10-14T21:25:38Z,2019-10-14T21:25:38Z,OWNER,I like the conditional dependency for the moment - maybe until 3.5 becomes officially unsupported.,"{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",506300941,bump uvicorn to 0.9.0 to be Python-3.8 friendly,