Re: [fetch] Proposed backward-incompatible change: requiring cloning all requests/responses (#61)

@domenic 

I agree with you basically regarding correspondence between `getReader()` and `releaseLock()`.

But, having `fetch()`, `cache.put()`, etc. mark a Request/Response used even if they're originally null/empty (impossible to make any change on the body) is not consistent with the `offset` based approach. The former is to mark `X` used when passed to `F()` which drains `X`'s contents if any, regardless of whether any change has actually happened on `X` by `F()` or not. The latter is to mark `X` used based on whether any change has happened to `X` by `F()` or not.

---
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fetch/issues/61#issuecomment-115108928

Received on Thursday, 25 June 2015 05:26:53 UTC