Refine Source|Buffer.indexOf(ByteBuffer) contracts #424
Conversation
core/common/src/ByteStrings.kt
Outdated
| @@ -85,6 +85,8 @@ public fun Source.readByteString(byteCount: Int): ByteString { | |||
| * expands the source's buffer as necessary until [byteString] is found. This reads an unbounded number of | |||
| * bytes into the buffer. Returns `-1` if the stream is exhausted before the requested bytes are found. | |||
| * | |||
| * For empty byte strings this function always return `0`. | |||
There was a problem hiding this comment.
See #423. I'm not sure how to handle empty byte strings yet.
There was a problem hiding this comment.
It seems to me that this is really better than fetching the data, provided that we always ignore what has been fetched.
Maybe it should be explicitly written that in the case of an empty byte string, the data will never be read from the source?
There was a problem hiding this comment.
Maybe it should be explicitly written that in the case of an empty byte string, the data will never be read from the source?
Then, it would be inconsistent with the Buffer's "overload". These two functions already behave differently, but I'd like to keep them as close to each other as possible.
core/common/src/ByteStrings.kt
Outdated
| @@ -85,6 +85,8 @@ public fun Source.readByteString(byteCount: Int): ByteString { | |||
| * expands the source's buffer as necessary until [byteString] is found. This reads an unbounded number of | |||
| * bytes into the buffer. Returns `-1` if the stream is exhausted before the requested bytes are found. | |||
| * | |||
| * For empty byte strings this function always return `0`. | |||
There was a problem hiding this comment.
It seems to me that this is really better than fetching the data, provided that we always ignore what has been fetched.
Maybe it should be explicitly written that in the case of an empty byte string, the data will never be read from the source?
core/common/src/ByteStrings.kt
Outdated
| * @param byteString the sequence of bytes to find within the buffer. | ||
| * @param startIndex the index into the buffer to start searching from. | ||
| * | ||
| * @throws IllegalArgumentException if [startIndex] is negative or if it is greater of equal to buffer's [Buffer.size]. |
There was a problem hiding this comment.
BTW, this behavior is also inconsistent with CharSequence.indexOf, which I'm considering maybe not as a paragon, but as something to be aligned with.
Closes #422