github
html_url | issue_url | id | node_id | user | created_at | updated_at | author_association | body | reactions | issue | performed_via_github_app |
---|---|---|---|---|---|---|---|---|---|---|---|
https://github.com/simonw/datasette/issues/1213#issuecomment-771965281 | https://api.github.com/repos/simonw/datasette/issues/1213 | 771965281 | MDEyOklzc3VlQ29tbWVudDc3MTk2NTI4MQ== | 9599 | 2021-02-02T20:37:08Z | 2021-02-02T20:39:24Z | OWNER | Starlette's gzip middleware implementation is here: https://github.com/encode/starlette/blob/0.14.2/starlette/middleware/gzip.py | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799663959 | |
https://github.com/simonw/datasette/issues/1213#issuecomment-771968177 | https://api.github.com/repos/simonw/datasette/issues/1213 | 771968177 | MDEyOklzc3VlQ29tbWVudDc3MTk2ODE3Nw== | 9599 | 2021-02-02T20:41:13Z | 2021-02-02T20:41:13Z | OWNER | Starlette accumulates the full response body in a `body` variable and then does this: ```python elif message_type == "http.response.body": # Remaining body in streaming GZip response. body = message.get("body", b"") more_body = message.get("more_body", False) self.gzip_file.write(body) if not more_body: self.gzip_file.close() message["body"] = self.gzip_buffer.getvalue() self.gzip_buffer.seek(0) self.gzip_buffer.truncate() await self.send(message) ``` | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799663959 | |
https://github.com/simonw/datasette/issues/1213#issuecomment-771968675 | https://api.github.com/repos/simonw/datasette/issues/1213 | 771968675 | MDEyOklzc3VlQ29tbWVudDc3MTk2ODY3NQ== | 9599 | 2021-02-02T20:41:55Z | 2021-02-02T20:41:55Z | OWNER | So maybe I could a special response header which ASGI middleware can pick up that means "Don't attempt to gzip this, just stream it through". | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799663959 | |
https://github.com/simonw/datasette/issues/1212#issuecomment-771975941 | https://api.github.com/repos/simonw/datasette/issues/1212 | 771975941 | MDEyOklzc3VlQ29tbWVudDc3MTk3NTk0MQ== | 9599 | 2021-02-02T20:52:36Z | 2021-02-02T20:52:36Z | OWNER | 37 minutes, wow! They're a little slow for me (4-5 minutes perhaps) but not nearly that bad. Thanks for running that profile. I think you're right: figuring out how to use more session scopes would definitely help. The `:memory:` idea is interesting too. The new `memory_name=` feature added in #1151 (released in Datasette 0.54) could help a lot here, since it allows Datasette instances to share the same in-memory database across multiple HTTP requests and connections. Note that `memory_name=` also persists within test runs themselves, independently of any `scope=` options on the fixtures. That might actually help us here! I'd be delighted if you explored this issue further, especially the option of using `memory_name=` for the fixtures databases used by the tests. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 797651831 | |
https://github.com/simonw/datasette/issues/1212#issuecomment-771976561 | https://api.github.com/repos/simonw/datasette/issues/1212 | 771976561 | MDEyOklzc3VlQ29tbWVudDc3MTk3NjU2MQ== | 9599 | 2021-02-02T20:53:27Z | 2021-02-02T20:53:27Z | OWNER | It would be great if we could get `python-xdist` to run too - I tried it in the past and gave up when I ran into those race conditions, but I've not done any further digging to see if there's a way to fix that. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 797651831 | |
https://github.com/simonw/datasette/issues/1214#issuecomment-771992025 | https://api.github.com/repos/simonw/datasette/issues/1214 | 771992025 | MDEyOklzc3VlQ29tbWVudDc3MTk5MjAyNQ== | 9599 | 2021-02-02T21:14:16Z | 2021-02-02T21:14:16Z | OWNER | As a result, navigating to https://github-to-sqlite.dogsheep.net/github/labels?_search=help and clearing out the `_search` field then submitting the form does NOT clear the search term. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799693777 | |
https://github.com/simonw/datasette/issues/1214#issuecomment-771992628 | https://api.github.com/repos/simonw/datasette/issues/1214 | 771992628 | MDEyOklzc3VlQ29tbWVudDc3MTk5MjYyOA== | 9599 | 2021-02-02T21:15:18Z | 2021-02-02T21:15:18Z | OWNER | The cause of this bug is form fields which begin with `_` but ARE displayed as form inputs on the page - hence should not be duplicated in an `<input type="hidden">` element. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799693777 | |
https://github.com/simonw/datasette/issues/1214#issuecomment-772001787 | https://api.github.com/repos/simonw/datasette/issues/1214 | 772001787 | MDEyOklzc3VlQ29tbWVudDc3MjAwMTc4Nw== | 9599 | 2021-02-02T21:28:53Z | 2021-02-02T21:28:53Z | OWNER | Fix is now live on https://latest.datasette.io/fixtures/searchable?_search=terry - clearing "terry" and re-submitting the form now works as expected. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 799693777 | |
https://github.com/simonw/datasette/issues/1212#issuecomment-772007663 | https://api.github.com/repos/simonw/datasette/issues/1212 | 772007663 | MDEyOklzc3VlQ29tbWVudDc3MjAwNzY2Mw== | 4488943 | 2021-02-02T21:36:56Z | 2021-02-02T21:36:56Z | CONTRIBUTOR | How do you get 4-5 minutes? I run my tests in WSL 2, so may be i need to try a real linux VM. | {"total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0} | 797651831 |