1.0 alpha released

Includes the following updates:

• Simplified/reworked parser
• Support for RFC2231 header parameters
• Use Psr streams and decorators instead of built-in stream_filters and extensions (zbateson/stream-decorators and GuzzleHttp streams)
• Ability to use other character sets
• Dynamically generate streams when saving messages -- only generate a dynamic stream on modification to a message and simplify writing

In addition much code has been refactored and cleaned to be more maintainable, testable and stable.

Fix for quoted-printable decoding in streams

Decoded streams weren't correctly buffered, so streams that happened to have a quoted-printable escape sequence split between two read operations would result in incorrect decoding. #72

0.4.9

• Return attachment part from addAttachmentPart/addAttachmentPartFromFile (#65)
• Fix composer install warning on Windows (#61)

Fix adding attachments and signed part detection

Fixes:

• addAttachmentPart() on multipart/alternative results in empty multipart #59
• detection of signature part in a multipart/signed message #60

Support commonly misspelled 'x-uuencode' encoding

Small release to address #56 to support additional spellings of the 'x-uuencode' transfer encoding.

Fix for line breaks in quoted header values

Fixes an issue with new lines in quoted headers (#55)

Fix for decoding mime headers

• Fix for half-width Kana (#11)
• Additional charsets supported in mime-decoding (same as supported for body)

Signature verification, bug fixes

• Added method to get original signed part (Message::getOriginalMessageStringForSignatureVerification) #47
• Fixes for HHVM 3.15 and up #46
• Fix for problem with mime-encoded headers #45
• Slightly better writing (normalized line-endings)
• Some additional tests

Fix for non-mime messages

This fixes a critical issue with 0.4.3 (#40) where non-mime messages aren't handled properly.

Fix inline content handling (BROKEN)

This release is broken (#40) please update to 0.4.3.

• Allow underscore in mime-encoded header part #36
• Better inline content handling #29 #34
  • Part filtering using the new PartFilter
  • More dynamic mime parts (fewer references to parts)
  • Added removeAllHtmlParts, removeAllTextParts, getTextPartCount, getHtmlPartCount
  • getTextPart, getHtmlPart, getTextStream, getHtmlStream, getTextContent and getHtmlContent all now optionally take an integer index to get inline content parts other than the first one
  • getAllParts, getPartCount, getPart, getChild, getChildParts, getChildCount all optionally take a PartFilter to filter out returned parts
• Added more test code
• Code cleanup
• Fixed documentation