id,node_id,number,title,user,state,locked,assignee,milestone,comments,created_at,updated_at,closed_at,author_association,pull_request,body,repo,type,active_lock_reason,performed_via_github_app,reactions,draft,state_reason
272391665,MDU6SXNzdWUyNzIzOTE2NjU=,48,Switch to ujson,9599,closed,0,,,4,2017-11-08T23:50:29Z,2019-06-24T06:57:54Z,2019-06-24T06:57:43Z,OWNER,,"ujson is already a dependency of Sanic, and should be quite a bit faster.",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/48/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
274315193,MDU6SXNzdWUyNzQzMTUxOTM=,106,Document how pagination works,9599,closed,0,,,1,2017-11-15T21:44:32Z,2019-06-24T06:42:33Z,2019-06-24T06:42:33Z,OWNER,,I made a start at that in this comment: https://news.ycombinator.com/item?id=15691926,107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/106/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
275125805,MDU6SXNzdWUyNzUxMjU4MDU=,124,Option to open readonly but not immutable,9599,closed,0,,,5,2017-11-19T02:11:03Z,2019-06-24T06:43:46Z,2019-06-24T06:43:46Z,OWNER,,Immutable assumes no other process can modify the file. An option to open reqdonly instead would enable other processes to update the file in place.,107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/124/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
276455748,MDU6SXNzdWUyNzY0NTU3NDg=,146,datasette publish gcloud,9599,closed,0,,,2,2017-11-23T18:55:03Z,2019-06-24T06:48:20Z,2019-06-24T06:48:20Z,OWNER,,"See also #103 

It looks like you can start a Google Cloud VM with a ""docker container"" option - and the Google Cloud Registry is easy to push containers to. So it would be feasible to have `datasette publish gcloud ...` automatically build a container, push it to GCR, then start a new VM instance with it:

https://cloud.google.com/container-registry/docs/pushing-and-pulling
",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/146/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
317714268,MDU6SXNzdWUzMTc3MTQyNjg=,238,External metadata.json,9599,closed,0,,,3,2018-04-25T17:02:30Z,2019-06-24T06:52:55Z,2019-06-24T06:52:45Z,OWNER,,"A frustration I'm having with https://register-of-members-interests.datasettes.com/ is that I keep coming up with new canned queries but I don't want to redeploy the whole thing just to add them to `metadata.json`

Maybe Datasette could optionally take a `--metadata-url` option which causes it to load from a URL instead and occasionally check for updates.",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/238/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
323716411,MDU6SXNzdWUzMjM3MTY0MTE=,267,"Documentation for URL hashing, redirects and cache policy",9599,closed,0,,,3,2018-05-16T17:29:01Z,2019-06-24T06:41:02Z,2019-06-24T06:41:02Z,OWNER,,See my comments on #258 for a starting point,107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/267/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
324188953,MDU6SXNzdWUzMjQxODg5NTM=,272,Port Datasette to ASGI,9599,closed,0,9599,3268330,42,2018-05-17T21:16:32Z,2019-06-24T04:54:15Z,2019-06-24T03:33:06Z,OWNER,,"Datasette doesn't take much advantage of Sanic, and I'm increasingly having to work around parts of it because of idiosyncrasies that are specific to Datasette - caring about the exact order of querystring arguments for example.

Since Datasette is GET-only our needs from a web framework are actually pretty slim.

This becomes more important as I expand the plugins #14 framework. Am I sure I want the plugin ecosystem to depend on a Sanic if I might move away from it in the future?

If Datasette wasn't all about async/await I would use WSGI, but today it makes more sense to use ASGI. I'd like to be confident that switching to ASGI would still give me the excellent performance that Sanic provides.

https://github.com/django/asgiref/blob/master/specs/asgi.rst",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/272/reactions"", ""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
329147284,MDU6SXNzdWUzMjkxNDcyODQ=,305,Add contributor guidelines to docs,9599,closed,0,,,2,2018-06-04T17:25:30Z,2019-06-24T06:40:19Z,2019-06-24T06:40:19Z,OWNER,,https://channels.readthedocs.io/en/latest/contributing.html is a nice example of this done well.,107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/305/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
340730961,MDU6SXNzdWUzNDA3MzA5NjE=,340,Embrace black,9599,closed,0,,,1,2018-07-12T17:32:29Z,2019-06-24T06:50:27Z,2019-06-24T06:50:26Z,OWNER,,"Run [black](https://github.com/ambv/black) against everything. Then set up CI to fail if code doesn't conform to black's style.

Here's how Starlette does this:

* https://github.com/encode/starlette/blob/e3d090b3597167f7b3a4f76e4bb3c0d3e94be61a/.travis.yml#L14
* https://github.com/encode/starlette/blob/e3d090b3597167f7b3a4f76e4bb3c0d3e94be61a/scripts/lint - essentially runs `black starlette tests --check`

And here's an example of a test run that failed:  https://travis-ci.org/encode/starlette/jobs/403172478",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/340/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed
459587155,MDExOlB1bGxSZXF1ZXN0MjkwODk3MTA0,518,Port Datasette from Sanic to ASGI + Uvicorn,9599,closed,0,9599,3268330,12,2019-06-23T15:18:42Z,2019-06-24T13:42:50Z,2019-06-24T03:13:09Z,OWNER,simonw/datasette/pulls/518,"Most of the code here was fleshed out in comments on #272 (Port Datasette to ASGI) - this pull request will track the final pieces:

- [x] Update test harness to more correctly simulate the `raw_path` issue
- [x] Use `raw_path` so table names containing `/` can work correctly
- [x] Bug: JSON not served with correct content-type
- [x] Get ?_trace=1 working again
- [x] Replacement for `@app.listener(""before_server_start"")`
- [x] Bug: `/fixtures/table%2Fwith%2Fslashes.csv?_format=json` downloads as CSV
- [x] Replace Sanic request and response objects with my own classes, so I can remove Sanic dependency
- [x] Final code tidy-up before merging to master",107914493,pull,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/518/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",0,
459714943,MDU6SXNzdWU0NTk3MTQ5NDM=,525,Add section on sqite-utils enable-fts to the search documentation,9599,closed,0,9599,,2,2019-06-24T06:39:16Z,2019-06-24T16:36:35Z,2019-06-24T16:29:43Z,OWNER,,"https://datasette.readthedocs.io/en/stable/full_text_search.html already has a section about csvs-to-sqlite, sqlite-utils is even more relevant.",107914493,issue,,,"{""url"": ""https://api.github.com/repos/simonw/datasette/issues/525/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed