issue_comments: 473158506
This data as json
| 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/412#issuecomment-473158506 | https://api.github.com/repos/simonw/datasette/issues/412 | 473158506 | MDEyOklzc3VlQ29tbWVudDQ3MzE1ODUwNg== | 9599 | 2019-03-15T04:53:53Z | 2019-03-15T04:53:53Z | OWNER | I've been thinking about how Datasette instances could query each other for a while - it's a really interesting direction. There are some tricky problems to solve to get this to work. There's a SQLite mechanism called "virtual table functions" which can implement things like this, but it's not supported by Python's https://github.com/coleifer/sqlite-vtfunc is a library that enables this feature. I experimented with using that to implement a function that scrapes HTML content (with an eye to accessing data from other APIs and Datasette instances) a while ago: https://github.com/coleifer/sqlite-vtfunc/issues/6 The bigger challenge is how to get this kind of thing to behave well within a Python 3 async environment. I have some ideas here but they're going to require some very crafty engineering. |
{
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} |
411257981 |