Age | Commit message (Collapse) | Author |
|
non-empty challenge in a smtp server.
|
|
|
|
to user's request.
|
|
to c-client, which in turn tries an empty password instead of
cancelling the connection.
|
|
parameter to make sure the user understands that this is used when the
STARTLS command is needed and not when a TLS encryption is needed.
|
|
is being saved.
|
|
attempts to check for new mail on a locked stream that is being used
for a save operation. Reported by Carlos E.R.
|
|
connection is closed while the post is being composed, but later the
post is attempted to be sent. In some instances, the stream may be
open and closed during this time, but Alpine does not realize the
connection was closed and attempts the post without a netstream,
producing a crash.
|
|
|
|
has been disabled during the compilation of the Openssl library.
|
|
|
|
|
|
|
|
|
|
keystroke sequence of commands is active. Based on a report from Holger
Trapp.
|
|
|
|
problems exist, so I can go back to the exact bits that caused the
problem.
|
|
of Openssl that do different things. I need to understand them
better. In the mean time, we go back to the original code.
|
|
consistent in what it signs. Problem reported by Björn Krellner.
|
|
version 3.0.
|
|
user agrees to save a certificate of another user, use the saved
certificate immediately to verify the smime message. Reported by
Stefan Mueller.
|
|
submitted in 2009, but resubmitted by Erich Eckner on 2019.
|
|
fails to add a trailing line to signed text. Reported by Stefan Mueller.
|
|
and not the new calls.
|
|
was being allocated. Based on a report by Erich Eckner.
|
|
|
|
Reported by Matt Roberds to the Debian bug system at
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=886370.
|
|
Gmail.
|
|
"mechanic" in comp.mail.pine.
* Configure script modifications to set the ldap-dir value correctly.
* Fix some typos in documentation.
|
|
that fix the following startup crashes:
* Crash when Alpine started with empty url fragment;
* Crash when Alpine started with option last-time-prune-questioned
in wrong format;
* Crash when Alpine started with printf formatting characters from
command line; and
* Crash when Alpine started with an extremely long command line
option.
* Crash when Alpine is started with the wrong piped input when
opening a folder from the command line. Reported by Mateusz
Kocielski from LogicalTrust.
|
|
a. Add a new variable: encryption-protocol-range, which can be
used to specify the minimum and maximum versions of the TLS
protocol that Alpine will attempt to use to encrypt its
communication with the server.
b. Add support for the Server Name Identification (SNI) extension
needed for TLSv1.3.
c. Remove the DTLS code. It was not being used.
|
|
emails, the filenames will be displayed in RFC1522 representation,
instead of in decoded form. Reported and patched by Wang Kang.
|
|
for big amounts of time while displaying messages in the screen.
Reset sleep time to 5 seconds in case it finds it needs to sleep
more than 5 seconds or a negative amount of time.
|
|
because of that, Alpine sets their encoding to quoted-printable, in
contradiction to RFC 2045. We handle this by keeping the encoding
reported by the server. Based on a report by Holger Trapp.
|
|
Content-Transfer-Encoding: QUOTED-PRINTABLE, Alpine will stop
processing that message. Later this causes Alpine to crash because
when it displays messages, it assumes that both header and body parts
are processed. Reported by Mark Crispin in 2010, in the Alpine-info
list (message with subject "crash bug in
alpine/mailpart.c:format_msg_att()") with no example, and reported
now by Holger Trapp, with an example.
|
|
then forwarding the message will set the first part to be flowed, and
sent that way even when the option Do Not Send Flowed Text is
enabled.
To avoid this issue, we remove any flowed text parameters from the
body of the message, so we create a "remove_parameter" function, and
then we set the flowed parameter only when the user requested.
Reported by Holger Trapp.
|
|
a failure using an SMTP server. This was due to c-client sending a
RSET command after the failure and Alpine reporting the reply of this
command.
|
|
S/MIME,
and then reenabled it. This crash was due to a double free of memory. To avoid
this in the future, we created a function free_x509_store(), which whenever
called, will only free memory once. A similar crash would occur when one
attempted to enter the S/MIME configuration screen if S/MIME was turned off.
In this case, Alpine would try to dereference a null pointer.
|
|
version 1.3.
|
|
signing first and encrypting second, which makes Alpine fail to parse
the signed data after encryption. Reported by Holger Trapp.
|
|
LibreSSL in connecting to external servers. This complements the
changes to support S/MIME. In particular, we add support for
validation of certificates by using C:\\libressl\ssl\certs as the
place to save CA certificates. In order to help users, some
certificates are distributed. TODO: Kerberos port, w2k.
|
|
we need to allocate memory for storing the resolved path. Reported by
Fabian Schmidt.
|
|
avoid memory leaks when trying to get the opening text a second time,
and get the actual opening text for html messages that are part of a
multipart/* text. In addition, Alpine caches this text, so that operations
such as resizing, or marking a message deleted, does not make Alpine
recompute this text.
|
|
character in UTF-8, but 0xfff8 is undefined.
|
|
coming from the configure script and imap/Makefile. Remove the double
definition from imap/Makefile. Reported and patched by Wang Kang.
|
|
|
|
is used. The revert to using code for the Windows operating system.
Reported by Andrew Ho.
|
|
multipart/mixed message that contain a signed part.
|
|
multipart/signed, Alpine will include the text of the original
message in a reply message, instead of including a multipart
attachment. Suggested by Barry Landy.
|
|
part. In this case, these attachments are not included in a reply,
even if the option "Include Attachments in Reply" is enabled. We
include these attachments in a reply too in this case (experimental.)
|