This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
Context ----- This bug is related to bug 6822: the root of the bug is a mix between raw and decoded forms within the moki representation. Description ----- In the moki representation of a typical HTTP exchange, the list of HTTP headers appears in raw form. The problem is that it is actually a mix between raw and decoded form. Typically, end of lines in an HTTP header is CR LF, but end of lines in XML text entities is only LF, and so the CR is escaped and appears at the end of the line, while the LF is taken as a regular end-of-line character. Example of a typical header line in the moki: Content-Type: text/html
 Possible solutions ----- rawHeaders could be: 1. the list of HTTP header fields, one per line, in which we should simply replace CR LF by LF 2. the raw HTTP header, but in that case we need to encode the text as base64. I guess the main problem with 2. is that the resulting string won't be human-readable anymore.
Removed the offending 
 character. The reference moki representations of the whole test suite need to be re-validated, but other recent and more substantives changes in the Checker require that anyway.