When data is retrieved from cassandra, this 4 bytes is present, and to make the content identical to what I insert, I will need to strip away the content substr($data, 4).
It seems likely to us that the correct behaviour should be to not prefix the data with the length, at least when we insert blobs with other (non-php) libraries the content is not prefixed.
I can send a pull request, but I wanted to ask about it first. Looking forwards to hear others opinions / experiences with this.
We're having issues with inserting and retrieving blobs.
When blobs are inserted, the data size is added as a prefix in form of an 4 byte integer value is. This can be found here:
When data is retrieved from cassandra, this 4 bytes is present, and to make the content identical to what I insert, I will need to strip away the content
substr($data, 4)
.It seems likely to us that the correct behaviour should be to not prefix the data with the length, at least when we insert blobs with other (non-php) libraries the content is not prefixed.
I can send a pull request, but I wanted to ask about it first. Looking forwards to hear others opinions / experiences with this.