reply code.  Maybe this used to be on the same line as the BBS name?

Make resume only perform any actions if curses is suspended.
This may fix the Syncterm OS X raw on exit in -iC mode.
(But may not either)

May fix OS X stack clobbering issue.

Fixed spelling: supplimental -> supplemental.  My CVS client says the server does not support rename, so I had to remove and re-add with the fixed spelling.

never maintened or used since. So... remove from CVS.

Replacing the whole separate localized English dictionaries with a general English dictionary and smaller supplimental ones containing localized spellings of words used in those regions

Updated parsing for dictionary filenames (to appear in the list for user preferences): Dictionary filenames can now include "-supplimental" after the localization (such as dictionary_en-US-supplimental.txt" if it's a smaller dictionary only with localized spellings of words.  Those dictionaries will include "(Supplimental)" at the end of their name in the user preferences.  Dictionary filenames without a localization after the language (such as dictionary_en.txt) will include "(General)" at the end of their name in the user preferences.  Dictionaries with just the localization without "-supplimental" will just have the language name and localization as the language name in the user preferences (i.e., dictionary_en-US.txt will be "English (US)".

global "PlainAuthOnly=true"

"CRAM-MD5 authentication required" instead of "MD5 required"

and require_crypt are set to false as well.

Log free disk space errors to sbbsecho.log (as well as print to stdout).

(defaults to false). If you want to disable all CRAM-MD5 based authentication
for both inbound and outbound BinkIT/BinkP connections/sessions (e.g. for
trouble-shooting password mismatch issues), then set this to true.

"QNET-HTTP 1.3 invoked with options: ..."

Made a couple updates to the function that loads a dictionary file. It seems at least slightly faster now.

- don't advertise support for encryption (by sending an "OPT CRYPT" message)
- don't complain (send "Encryption requires CRAM-MD5" M_ERR) if the remote
  doesn't support CRAM-MD5

mismatch, it's an authentication scheme mismatch).

one of the configured linked-nodes matching the remote address(es) is set for
CRAM-MD5 passwords only and a plain-text password was provided by the remote.
Previously, the function would log a warning:
"CRAM-MD5 required (and not provided) by <addr>" but succeeed (the actual
password value was not check) and it woudl flag the session as "non-secure"
(sending M_OK non-secure), looking like this on the remote side:
  Authentication successful: non-secure
and any received files would go into the configured non-secure inbound.
This may explain why some sysops sometimes get unexplained files in their
non-secure inbound (looking for "non-secure" in the logs should confirm).
Now, you should still see (on the answering side) the log message:
"CRAM-MD5 required (and not provided) ...", but it'll also send a M_ERR
Passowrd mismatch error to the remote, thus ending the session.

rechocfg.c (true). No effect for any sbbsecho.ini which already has this
key with a value (and they all should).

do so incorrectly. This is a correct fix (check the nocrypt property, not
just that the linked node object is != false).

Replaced dictionary_english.txt with some English dictionaries for various regions with standard language tags.

Added support for some common standard language tags and parsing them from the dictionary filenames.  Updated the version to 1.65.