All checksums have been double-checked against existing RMD160 and
SHA512 hashes
The following distfiles were unfetchable (possibly fetched
conditionally?):
./mail/qmail/distinfo netqmail-1.05-TAI-leapsecs.patch
### GMime 3.2.7
* Added some configure logic to auto-detect the system shift-jis charset alias. (issue #81)
* Fixed tests/Makefile.am to exit with a non-negative value (issue #82)
* Fixed logic to skip expired or revoked gpg subkeys when looking for the correct subkey to
use for signing or encrypting. (issue #88)
* Fixed a regression introduced into 3.2.6 as part of the header parsder rewrite that lost
the ability to warn about invalid headers for non-toplevel MIME parts. (issue #89)
* Fixed S/MIME to always set GPGME_KEYLIST_MODE_VALIDATE when looking up certificates
as this is needed in order to correctly populate the GMimeCertificates (issue #90)
### GMime 3.2.6
* Added methods to get 64-bit timestamps for the creation-date and expiration-date of
GMimeCertificates and GMimeSignatures. (issue #68)
* Fixed the vala abstract public constructor for GMimeObject. (issue #76)
* Fixed a bug in GMimeParser's header parser. (issue #78)
* Added build files for Visual Studio 2017 thanks to Vitaliy Didik.
pkglint --only "https instead of http" -r -F
With manual adjustments afterwards since pkglint 19.4.4 fixed a few
indentations in unrelated lines.
This mainly affects projects hosted at SourceForce, as well as
freedesktop.org, CTAN and GNU.
### GMime 3.2.5
* Modified GMimeParser to prevent stack overflows when parsing deeply nested messages.
GMimeParser now has a limit on how deep multipart and/or message/rfc822 MIME part
nesting is allowed to go before the parser will take action to prevent a stack
overflow. If the max level is reached at a message/rfc822 part, then that part
will be consumed by the parser as a generic GMimePart rather than a
GMimeMessagePart. Likewise, if the max level is reached at any type of multipart,
then the content of said multipart will be packed into the GMimeMultipart's
preface and not parsed any further.
* g_mime_multipart_foreach has been rewritten to avoid recursion, thereby avoiding potential
stack overflows.
* The gmime-port-2-6-to-3-0.sh script has been fixed to use proper sed syntax.
Changelog:
### GMime 3.2.4
* Replaced calls to g_memmove with memmove.
It seems that the latest versions of glib have deprecated g_memmove
in favor of having developers use libc's memmove() function directly.
This change reduces the number of compiler warnings during the gmime
build process.
* Added a new GMIME_DECRYPT_NO_VERIFY flag that disables signature verification.
For cases where it is not necessary to verify the signatures (or it is known
that there are no signatures), making use of this flag can significantly
improve the performance of decrypting OpenPGP content.
* Modified GMimeParser to not set the OpenPGP state for base64/uuencoded content.
While the GMimeParser is parsing a MIME message (or other MIME entity), it will
normally attempt to identify OpenPGP markers in the content of GMimeParts.
However, when the content is base64 or uu-encoded, the parser is unable to
accurately detect these markers and so in previous versions, it was falsely
claiming that such MIME parts had no OpenPGP content even though it was possible,
after decoding their content, that they did in fact have OpenPGP content.
For more details about this bug, see issue #60.
* Added reporting of RFC 5322 addr-spec syntax violations to the GMimeParserWarning API.
* Fixed a stack underflow error in the uudecode.c sample.
* Improved Vala bindings.
- Use correct symbol prefixes which avoid loads of cname attributes.
- Unhide FilterBest.charset() method which conflicts with charset field.
- Object.write_to_stream conflicts with function pointer with the same name
but with a different signature.
Fixes https://gitlab.gnome.org/GNOME/gmime/issues/2
Fixes https://gitlab.gnome.org/GNOME/gmime/issues/3
### GMime 3.2.3
* Fixed GMimeFilterBasic for uudecode.
Don't allow the outbuf to ever get set to NULL which could happen
if the begin-line had not yet been found (and thus
g_mime_filter_set_size() had never been called to allocate the
outbuf buffer).
* Fixed a bug in g_mime_uuencode_step().
* Modified GMimeParser to work around broken mailers that send base64
encoded message/rfc822 parts.
Fixes https://gitlab.gnome.org/GNOME/gmime/issues/1
* Fixed a bug in g_mime_quoted_encode_close() where it would incorrectly
end the quoted-printable output with a line containing only "=\n" even
when it is not needed.
* Improved g_mime_content_encoding_from_string(). This function no
longer requires the input string to be an exact match for "7bit",
"8bit", "base64", etc. It can now handle whitespace before and
after the value. In other words, it is now easy to use this
function on raw header values before any whitespace trimming
has been done.
* Really, really fixed the packaging to include the Vala build files.
### GMime 3.2.2
* Fixed packaging to include Vala files.
### GMime 3.2.1
* Fixed GMimeParser to recognize the message/global mime-type
(a UTF-8 version of message/rfc822). (issue #50)
* Updated GMime to use libidn2 instead of the older libidn
library. (issue #48)
* Fixed address quoting logic and IDN2 encoding.
The rules for quoting address names should use 'specials'
instead of 'tspecials' and when encoding domain names via
IDN2, check if the encoded domain matches the original
domain name (other than case). If they match, prefer the
non-encoded domain name since the user may have used
uppercase characters to enhance readability of the domain
name.
* Added GMIME_DECRYPT_ENABLE_ONLINE_CERTIFICATE_CHECKS and
GMIME_DECRYPT_ENABLE_KEYSERVER_LOOKUPS as possible flags to
pass to g_mime_crypto_context_decrypt(). Also added
GMIME_VERIFY_ENABLE_ONLINE_CERTIFICATE_CHECKS and
GMIME_VERIFY_ENABLE_KEYSERVER_LOOKUPS as possible flags to
pass to g_mime_crypto_context_verify().
Clients that wish to enable online certificate and/or
keyserver lookups now need to explicitly enable this
functionality.
These changes are designed to make it more difficult
for clients to be susceptible to Efail privacy exploits.
Specifically, it is meant to address the privacy concerns
regarding CRL and OCSP status check backchannels.
For more information about Efail, see https://www.efail.de/
* Fixed g_mime_message_write_to_stream() to prioritize message
headers over body headers (even when they have an offset of -1).
(issue #46)
* The GMimeParser can now warn about a number of RFC-compliance
issues that it finds when parsing messages.
* Fixed GMimeTextPart to make sure that the GMimeFilterCharset is
non-null before trying to use it. This can happen if the charset
specified in the Content-Type header is unsupported by the
iconv library.
GMime is a C library which may be used for the creation and parsing
of messages using the Multipurpose Internet Mail Extension (MIME),
as defined by numerous IETF specifications.
GMime features an extremely robust high-performance parser designed
to be able to preserve byte-for-byte information allowing developers
to re-seralize the parsed messages back to a stream exactly as the
parser found them. It also features integrated GnuPG and S/MIME
v3.2 support.
Built on top of GObject (the object system used by the GNOME
desktop), many developers should find its API design and memory
management very familiar.
This package contains v3 of the gmime API.
