| Issue 1360: | UTF-8 decoder is not function of input stream; will break when UTF-8 sequence split across reads | |
| 1 person starred this issue and may be notified of changes. | Back to list |
Affected Version: hterm What steps will reproduce the problem? 1. Run: perl -e '$|=1; print "Here"; print "\342"; print "\200"; print "\231"; print "s Johnny.\n"' 2. Run: perl -e '$|=1; print "Here"; print "\342"; print "\200"; sleep 1; print "\231"; print "s Johnny.\n"' What is the expected output? What do you see instead? Both #1 (no sleep) and #2 (with sleep) should produce the same output: "Here’s Johnny." hterm only correctly decodes #1. With a sleep inserted, #2 prints "Hereâs Johnny." This suggests that hterm will give varying output depending on how, e.g., SSH chooses to divide its octets across TCP segments (and therefore across writes to the pty). Please provide any additional information below. In general it is preferable for the terminal's emulation to be strictly a function of the input octet sequence, without regarding to timing or the chunking of octets across buffers returned by read().
Apr 30, 2012
#1
winst...@gmail.com
May 1, 2012
Wrong project
Status:
Invalid
|
|
| ► Sign in to add a comment |