| 1 |
|
| 2 | Known differences
|
| 3 | =================
|
| 4 |
|
| 5 | *As of 1.x release*
|
| 6 |
|
| 7 | - Topics such as cross-origin, content security policy, mixed content, service workers are ignored, given our server-side context.
|
| 8 |
|
| 9 | - Url input must be an absolute url, using either `http` or `https` as scheme.
|
| 10 |
|
| 11 | - On the upside, there are no forbidden headers, and `res.url` contains the final url when following redirects.
|
| 12 |
|
| 13 | - For convenience, `res.body` is a transform stream, so decoding can be handled independently.
|
| 14 |
|
| 15 | - Similarly, `req.body` can either be a string or a readable stream.
|
| 16 |
|
| 17 | - Only support `res.text()` and `res.json()` at the moment, until there are good use-cases for blob.
|
| 18 |
|
| 19 | - There is currently no built-in caching, as server-side caching varies by use-cases.
|