Noah Misch <noah@leadboat.com> writes:
> I tried your patches against libedit-28. Wherever a command contains a
> newline, unpatched psql writes the three bytes "\^A" to the history file, and
> patched psql writes the four bytes "\012". Unpatched psql correctly reads
> either form of the history file. Patched psql misinterprets a history file
> created by unpatched psql, placing 0x01 bytes in the recalled command where it
> should have newlines. That's a worrisome compatibility break.
I think you got the test cases backwards, or maybe neglected the aspect
about how unpatched psql will only translate ^J to ^A in the oldest
(or maybe the newest? too pressed for time to recheck right now) history
entry.
The issue is that a patched psql, or a psql with a sufficient old libedit,
will apply ^J -> ^A to all entries when saving, and the reverse when
loading. Without the patch, only the oldest entry gets transformed.
Failure to reverse the encoding in all lines is what creates a
user-visible problem. If we do not fix this, that's what we risk.
We do not escape a problem by refusing to fix it.
regards, tom lane