(Chunked) HTTP binary message body and CRLFs -


I can not seem to get a more accurate answer on the following question (mostly reading Googling and HTTP / 1.1 specs):

When 'encoded' transfer encoding is used, then the server needs to write both sizes of size in bytes and after that the flywheel data ends with the CRLF. Is not this binary data sending "CRLF-modifier" and the method is somewhat unnecessary? What if there is 0x0 A in the data, then somewhere in 0x0 D (i.e. this is actually part of the data)? Is the client obviously expected to follow Chuck's size, which is given at the top of that part or is the first error in the CRLF which is found in the data? My understanding is to take part of the size given by the server only till now, proceed to the next line, then read the amount of bytes actually from within the following data (CRLF or a CRL inside), then that CRLF will follow the data Avoid and repeat this procedure until there is no other section ... am I right? What does CRRF mean after each data check? Readability?

A selected consumer does not scan the message body for the CRLF pair, before it is specified number of bytes Reads, and then reads two bytes to confirm that they are CR and LF. If they are not, the message body is made ill, and either the shape is incorrectly specified or the data is otherwise corrupted.

The following CRLF is a belt-end-breather assurance (copy, <, em> chunked transfer coding ), but it also works to maintain the rule that at the beginning of the field line Starts.


Comments

Popular posts from this blog

oracle - The fastest way to check if some records in a database table? -

php - multilevel menu with multilevel array -

jQuery UI: Datepicker month format -