[gmime] Created tag 2.99.0

[Rico Tzschichholz <ricotz src gnome org>]

The unsigned tag '2.99.0' was created.

Tagger: Jeffrey Stedfast <fejj gnome org>
Date: 1491670840 -0400

    Tagged for 2.99.0

Changes since the last tag '2.6.23':

Gaute Hope (3):
      GMimeFilterHtml: add option to enclose citation in blockquotes
      bq: correctly calculate citation depth when no newline between quoted and unquoted line, close quotes 
at beginning of line
      blockquote: catch '>   >' type quotes and handle empty lines

Javier Jardón (1):
      autogen.sh: Honor the NOCONFIGURE variable

Jeffrey Stedfast (359):
      Bumped version to 2.99.0
      Updated TODO
      Redesigned the way header values get cached by the parser to save original formatting.
      Replaced GMimeHeaderIter with an indexed API
      Updated TODO
      Updated TODO
      Fixed g_mime_stream_write_to_stream() to return gint64
      Fixed g_mime_message_get_reply_to/sender to return InternetAddressList
      Updated TODO
      Don't reformat the header when we have the raw (unfolded) value available
      Use the correct virtual write_to_stream() method for the body
      Fixed GMimeMultipart's write_to_stream() logic
      Fixed writing out a multipart's preface
      Track whether or not the end boundary should be written for multiparts
      Replaced GMimeRecipientType with GMimeAddressType and simplified code
      Updated PORTING to document changes from 2.6 to 3.0
      Privatized the GMimePkcs7Context structure and updated configure
      Fixed GMimeHeaderList to properly emit changed events
      Renamed GMime-2.6.metadata to GMime-3.0.metadata
      Moved get/set_always_trust to GMimeCryptoContext
      Properly implemented always_trust for GMimePkcs7Context
      Fixed the build
      code cleanup
      Reimplemented GMimeGpgContext using GpgMe
      Updated PORTING
      Fixed the GMimeGpgContext to mostly work
      Introduced the beginnings of GMimeParserOptions
      bumped min req'd version of GpgME to 1.6.0 (will prob up it to 1.8.0 later)
      GMIME_ENABLE_RFC2047_WORKAROUNDS is no more. Use GMimeParserOptions instead.
      updated jwz.mbox output summary to match new defaults
      code cleanup
      Added charset parameter to g_mime_utils_header_encode_*()
      Use gpgme's sig->summary instead of our own status + errors
      Updated PORTING
      Fixed examples
      Got rid of the need for a msgid mutex lock
      code cleanup
      Assume g_mime_charset_map_init() has been called
      Updated TODO
      Rewrote address parser
      Fixed parsing of group addresses like "name:" (e.g. empty w/ no ';')
      Changed expectation in address parser unit test
      Removed more broken addresses not worth supporting fro unit tests
      Improved address parser: allow unquoted ,'s and .'s
      Added g_mime_references_copy and registered GMimeReferences as a boxed type
      Fixed #includes for gmime-common.c and gmime-multipart.c
      Merge pull request #1 from gauteh/blockquote_html
      Fixed indenting
      Updated README to use markdown
      Renamed README -> README.md
      Add back NOCONFIGURE test to autogen.sh
      Merge branch 'master' into gnome-master
      Implemented a bunch of GMimeApplicationPkcs7Mime
      Modified g_mime_crypto_context_sign() to take a 'detach' argument
      Added a registry for GMimeCryptoContext's
      We no longer need to pass a GMimeCryptoContext to some methods
      Fixed mutex locking
      Get/set the export-session-key flag on the gpgme_ctx_t
      gpgme expects "1", not "true"
      Implemented session-keys upport for GMimePkcs7Context
      Merge decrypt and decrypt_session
      Replace get/set_always_trust() and get/set_retrieve_session_key() with flags
      fixed the build
      Updated TODO
      Updated PORTING
      Updated PORTING
      Fixed the build when crypto is disabled
      Updated docs
      Updated docs
      Implemented g_mime_application_pkcs7_mime_verify()
      fixed the build
      Don't require a crypto ctx for any of the application/pkcs7-mime API's
      Add more pkcs7 tests and automate running
      Added g_mime_references_get_type()
      Improved error reporting for PKCS7 context
      Started working on fixing pkcs7 tests
      Improved error reporting in GPG context
      Use the binary version of smime.crt instead
      Import the CA certificate in the pkcs7 tests (needed for encryption)
      No need for g_mime_crypto_context_verify() to take a 'digest' argument
      Don't bother setting up sign+encrypt for pkcs7 tests
      Check gpgme_add_signers() for errors
      Fixed test-pkcs7 to build
      Share common utility functions between gpg and pkcs7 ctx's
      Added gmime-gpgme-utils.[c,h]
      Added CI with Travis CI
      Attempt to fix the Travis CI build
      Updated .travis.yml
      Modified package dependency list for Travis CI
      More updates to Travis CI package dependencies
      automake1.15 is apparently not a valid package
      Allow automake 1.11 for use with Travis CI
      Try autoconf 2.54 instead of 2.69?
      Find out what version of gpgme is installed on Travis CI
      Fixed test-mime unit tests
      Added --enable-coverage to configure.ac
      Try enabling coveralls
      Added coveralls badge
      Fixed jwz.mbox expected output
      Ignore coverage of examples and generators
      Automated message/partial unit tests
      Updated README.md
      Updated README.md
      Updated README.md
      Updated copyright
      Renamed g_mime_stream_[file,fs]_new_for_path() to _open()
      Updated test-partial to use a FILE stream
      Update README.md
      Update README.md
      Update README.md
      Update HACKING
      Update README.md
      Update README.md
      Removed support for C#/Mono bindings
      Updated README.md
      Removed more mono cruft
      Updated TODO
      Fixed clang compile warnings
      Cleaned up gpg testing
      Use gpgme_op_decrypt() for decrypting pkcs7 instead of decrypt_verify()
      Disable CRL checks in gpgsm.conf for unit tests
      Flush some streams in the unit tests
      Clear gpgme signers in error conditions
      Updated .gitignore
      Added a GMimeTextPart for convenience
      Updated README.md
      Updated REAMDE.md
      Fixed autogen.sh NOCONFIGURE check
      Use more accessor methods in unit tests to increase coverage
      Fixed another autogen.sh NOCONFIGURE check
      Updated .travis.yml
      Minor doc fix
      Removed unneeded overrides
      Updated gmime-param.c to respect GMimeParserOptions
      Updated configure.ac
      Updated README.md
      Simplified g_mime_multipart_signed_verify()
      GpgME's OpenCMS ctx does not support passphrase_cb's
      Removed debugging cruft from testsuite.c
      Uncommented code for session_key support
      Simplify gpgme -> gmime hash_algo conversion logic
      Removed debug output from test-pkcs7.c
      Fixed pkcs7 verify logic
      Use g_set_error_literal() when appropriate
      Fixed gpg verify logic (and simplified pkcs7 verify logic)
      Fixed the build
      Share code between gpg and pkcs7 contexts
      We can't automate pkcs7 or smime tests (yet)
      Fixed the build
      Lower gpgme dependency to 1.2.0 and see what fails on the bots
      Temporarily force the verbosity of test-pgpmime to 4
      g_mime_crypto_context_import_keys() now returns # of keys imported
      Reset streams back to NULL
      Is CI failing due to gpgme 1.2.0's gpgme_op_decrypt_verify()?
      Revert previous commit (didn't work)
      code cleanup
      Removed ChangeLog file
      Removed debugging printf
      Added some debug printfs for the CI bots
      Removed debugging printf
      Disable some test-pgpmime logic for old gpgme versions
      Don't override verbosity anymore
      Renamed g_mime_part_[get,set]_content_object() to [get,set]_content()
      Updated docs
      Updated docs
      Updated docs
      Updated docs
      Updated docs
      Added encapsulated pgp signing test
      Fixed GpgMe passphrase callback function to not close the pipe
      Re-enable some PGP/MIME tests to see if things work on the bots now
      code cleanup
      code cleanup
      Added unit tests for application/pkcs7-mime
      Improved debugging output for unit tests
      base64 encode application/pkcs7-mime content
      fixed test-smime to pass 100%
      Updated g_mime_text_part_set_text() to set an appropriate transfer encoding
      Updated README.md
      Updated README.md
      Make GMimeMultipartSigned/Encrypted easier to use
      Fixed g_mime_multipart_signed_sign()
      Updated README.md
      code cleanup
      Renamed g_mime_header_list_[get,set]() to g_mime_header_list_[get,set]_value()
      Updated docs
      Updated TODO
      Fixed parse_broken_date() to always return 0 on garbage input
      Revert the g_mime_header_list_get/set rename
      Modified GMimeHeaderList's changed event to pass useful args
      Rearchitected the way GMimeObject's process added/removed headers
      Added more message/partial unit tests
      block/unblock changed event in g_mime_object_set_content_id()
      Update TODO
      Replace g_mime_parser_get/set_scan_from() with get/set_format()
      Added support for the MMDF format
      Updated docs
      Fixed MMDF support
      Fixed header_parse() logic for invalid cases to properly reset state
      Dropped GMimeDigestAlgo argument from sign() and encrypt() methods
      code cleanup
      Modified GMimeParser to auto-detect OpenPGP markers
      Test more GMimeParser API's
      Dropped g_mime_parser_construct_*_with_options(). Just roll it into regular method.
      Disable PGP tests if crypto is not enabled.
      Updated docs
      Added support for inline-PGP
      Fixed test-pgpmime tests
      Updated docs
      Improved GMimeParser's OpenPGP marker scanning logic
      Fixed formatting
      Use g_set_error_literal where appropriate
      Don't set the passphrase cb until we need it
      Renamed GMimeCertificateTrust to GMimeTrust and sync'd values with gpgme
      Protect against dereferencing a NULL subkey
      code formatting
      Added NO_COMPRESS and SYMMETRIC options for GMimeEncryptFlags
      minor code cleanup
      Renamed various _destroy() functions to _free()
      Renamed InternetAddress:priv to :changed
      Stubbed out a GMimeParamList
      Fixed the build
      okay, really fix the build this time
      Fixed unit tests and examples
      Renamed 'priv' member of GMimeContentType/Disposition to 'changed'
      Updated code to use the new GMimeParamList instead of a linked-list
      Fixed the build
      Updated docs
      Updated docs
      Improved GMimeParamList parser/encoder
      verbosify the rfc2184 tests to see why travis ci is breaking
      Don't depend on an uninitialized variable!
      get rid of debugging spew
      added more debug spew. why art thou failing on the bots!?!?
      Fixed for real this time.. I hope?
      Updated TODO
      Implemented g_mime_content_type_encode()
      Fixed the build
      Fixed the unit tests
      Modified GMimeHeader[List] to subclass GObject
      Updated gmime.c
      Converted GMimeParam and GMimeParamList into GObjects
      Updated README.md
      Updated README.md
      Fixed g_mime_header_set_value()
      Removed dead code
      Updated InternetAddress and GMimeParam to capture the charset used in rfc2047 encoded-words
      Added GError arg to g_mime_stream_[file,fs]_open()
      Added GMimeFormatOptions
      Fixed GMimePart's encode() method to only pick the best encoding for the content
      Proliferate GMimeFormatOptions throughout the API
      Implemented support for hiding headers using GMimeFormatOptions
      If GMimeParam:method is DEFAULT, use GMimeFormatOptions:method
      Added GMimeStreamFilter:owner
      Make use of GMimeFormatOptions:newline
      Updated g_mime_param_list_encode() to use GMimeFormatOptions:newline
      Updated docs
      Always use accessor functions for GMimeParserOptions
      No need to special-case NULL anymore
      Updated TODO
      Slight parser optimization and additional check
      Another minor optimization to GMimeParser
      Drop support for non-GLib regex
      Make parser_scan_content() less hacky by using a GMimeStreamNull
      Disable GMimeStreamNull counting newlines by default
      Split GMimeFilterCRLF into Unix2Dos and Dos2Unix
      Updated test-mbox.c
      Fixed GMimeParser to handle header names with trailing whitespace
      Updated docs
      Updated docs
      Updated docs
      Updated docs
      Rewrote GMimeFilterStrip to handle DOS line endings
      Use the FormatOptions when verifying multipart/signed signatures
      minor parser fixups
      Renamed g_mime_parser_get_from() to g_mime_parser_get_mbox_marker()
      Use GChecksum instead of a custom MD5Context
      Dropped support for GMIME_STREAM_BUFFER_MODE_CACHE_READ
      Updated VS2010 gmime.vcxproj
      Removed caching of iconv descriptors; not really needed anymore
      Avoid breaking strict aliasing rules in GMimeParser
      Reorganized GMimeParser members
      Get rid of dead code
      Fixed parser logic
      synced up with latest manual modifications
      regenerated charset map
      generate the charset map on Linux instead
      Added porting script from 2.6 to 3.0
      Update gmime.doap
      Replace GMimeFilterMd5 with GMimeFilterChecksum
      Added GMimeFilterSmtpData and deleted GMimeFilterCRLF
      Made g_mime_header_get_raw_[name,value] public
      GMimeHeader::value is now the unfolded and decoded version
      Removed g_mime_message_[get,set]_date_as_string()
      Updated docs
      Directly set the raw values when InternetAddress/ContentType/Disposition change
      Specified Content-Id and Disposition-Notification-To formatters
      Updated docs
      Added description to gmime.doap and removed MAINTAINERS
      Fixed GMimeMessage address header sync and added unit tests
      Updated TODO
      Migrated to using GDateTime instead of time_t and an int tz_offset
      Updated TODO
      Updated test-mime.c
      Rewrote GMimeReferences to use a GPtrArray
      Updated TODO
      Added gmime-references.h to gmime.h
      Fixed the build
      Fixed the build for --disable-crypto
      Fixed argument validation
      Added unit tests for GMimeReferences
      Added more unit testing for GMimeReferences
      Removed test[1,2,3].eml
      Added new set of message/partial messages for the unit tests
      Added some unit tests for the header formatters
      Added support for GPGME_ENCRYPT_THROW_KEYIDS
      Fixed up documentation comments
      doc and g-ir-scanner fixes
      updated docs
      Register GMimeParserOptions as a boxed type
      Register GMimeFormatOptions as a boxed type
      Updated docs
      Added m4/gpgme.m4
      Updated gmime.spec.in
      Added (nullable) attribute for Parser and FormatOptions
      more annotations
      more attributes
      skip introspecting g_mime_crypto_context_register
      Added skip attribute to a few more methods
      Updated docs
      Fixed typo
      Fixed more typos
      call g_mime_[parser,format]_options_get_type()
      Save (munged) From lines at the start of message/rfc822 parts
      Updated TODO
      Use g_object_new instead of g_object_newv
      Re-use the ensure_newline field instead of needing message->compliance
      Removed gmtime_r and timezone checks (no longer needed)
      Updated docs
      minor code cleanup
      Don't use uninitialized gpgme_error_t
      Make sure to always initialize inend when unfolding a header
      Removed dead code
      Just use openpgp_state_t
      Properly handle fseek/lseek errors
      O_RDONLY has a value of 0, so use O_WRONLY instead
      Fixed error check for open()
      Make sure to always close fin
      Fixed some unit test buglets
      Privatize the GMimeParserOptions and GMimeFormatOptions structs
      Added libidn support
      minor fixups
      Updated docs
      Added README.md to EXTRA_DIST
      Added GPGME and Libidn info to README.md
      Added Documentation: yes/no to configure output
      Added internet_address_mailbox_get_idn_addr() to header
      minor fixups
      Fixed test-partial.c to load the path passed on cmd-line

M. Dietrich (1):
      Better find citation cut depth