Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Already on GitHub? Sign in to your account

Binary data get() issue #48

Closed
aslushnikov opened this Issue Apr 30, 2012 · 2 comments

Comments

Projects
None yet
2 participants

I've got a pdf file with the size of 125884 bytes https://github.com/aslushnikov/node-memcached/blob/master/issues/48/1335827866.pdf. Once I stored it in the memcached, it almost always comes back corrupted: the data lacks 3 or 4 bytes, but sometimes it happens that it is returned full.

I guess the problem is that you check the end of the message from memcached with the \r\n sequence, but in memcached documentation it is strongly recommended not to do this way.

Hope it helps, and thank you for your work,
Andrey

Owner

3rd-Eden commented Jun 21, 2012

I'm currently rewriting the ASCII parser so I hopefully will resolve this issue with it. Instead of checking for \r\n sequences it will be scanning each char to determine to responses.

No clue when this will be done though.

Owner

3rd-Eden commented Sep 13, 2012

Fixed in d5a7f55

@3rd-Eden 3rd-Eden closed this Sep 13, 2012

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment