- Upgrade tests to be compatible with OCaml 5.2 (@kit-ty-kate, #99)
- Be able to set field parsers when we parse an entire email (@dinosaure, #89)
- Be able to wrap a multipart email into a part (@dinosaure, #90)
- Don't use RNG to produce the example to be compatible with OCaml 5 (@dinosaure, #93)
- Upgrade to cmdliner.1.1.0 (@hannesm, #94)
- Use a smaller email to test isomorphism (@dinosaure, 931c6ad)
- Lint dependencies (@dinosaure, #95)
- Delete
rresult
dependency (@dinosaure, @hannesm, #96) - Delete
fmt
dependency (@dinosaure, #97)
- Replace deprecated functions of
fmt
library (@dinosaure, #86) - Fix OPAM constraints (@kit-ty-kate, @dinosaure, #85)
- Clarify the semantic about the underlying quoted-printable encoder (@dinosaure, #84)
- Be able to pass the seed to generate the boundary (@dinosaure, #83)
- Let the user to decide if it wants to encode or not the given stream
according to the
Content-Transfer-Encoding
(@dinosaure, #82) - Add
Content_type.Type.of_string
(@dinosaure, #81) - Add
Header.add_unless_exists
(@dinosaure, #81) - Add a regression test about #77 (@lyrm, @dinosaure, #78)
- Fix bug on the Base64 decoder (@clecat, @lyrm, @dinosaure, #78)
- Don't try to split quoted-printable and add useless spaces to respect isomorphism (@dinosaure, @lyrm, #77)
- Fix how we escape
\b
in content-type parameters (@dinosaure, #73) - Add
Header.{length,to_list,of_list,to_list_with_location}
(@dinosaure, #72) - Enable
fws
token in quoted-string values (as a parameter of the content-type) (@lyrm, @dinosaure, #71) - Handle correctly an empty part with
angstrom
(@dinosaure, #68) - Generate a corpus of a million emails where we ensure that
mrmime
assumes a kind of isomorphism, see https://github.com/mirage/hamlet for more details (@lyrm, @dinosaure, #67) - Fix some comparison functions (on IANA values and field-name) (@lyrm, @dinosaure, #66)
- Simplify the
Mail.t
type (@lyrm, @dinosaure, #63) - Fix pattern-matching failure when decoder is closed (@dinosaure, #62)
- Keep the order of fields into a header when we replace one (@dinosaure, #61)
- Ensure stable memory with lwt when we parse contents (@dinosaure, #60)
- Fix how we parse quoted-printable contents (@dinosaure, #59)
- Add a space in front of any field's values (@dinosaure, #56)
- Ensure to emit only lines (@dinosaure, #55)
- Fix how we parse Content-Transfer-Encoding (@dinosaure, #54)
- Better pretty-print multiple domains on mailbox (@dinosaure, #53)
- Unlock the arbitrary limit on the internal buffer to decode the header (@dinosaure, #51)
- Delete unused phrases field value (@dinosaure, #50)
- Fix fuzzer (@dinosaure, #49)
-
Return the zone of the date (#41, @dinosaure)
-
Add
Content_type.to_string
(#43, @dinosaure) -
Be resilient on date (accept nano-seconds) (#44, @dinosaure)
-
Add a simple example to craft an email with an attachment (#47, @dinosaure)
-
Drop the support of OCaml 4.08.0 (#47, @dinosaure)
-
mrmime.prettym
is not a part of the distribution breaking changes We decided to splitmrmime
andprettym
mostly because this part (prettym
) is used by some others packages and they should not follow the release cycle ofmrmime
when they are not about email stuffs. The package is available here:
- Add
Header.message_id
(#39, @dinosaure, @hannesm) - Update README.md (#40, @dinosaure)
- Update the way to generate quoted-printable contents
According
pecu.0.5
(#38, @dinosaure)
- Fix unstructured values (@dinosaure, #34)
- Fix
Mrmime.Hd
decoder about the separation between the header and the body (@dinosaure, #35) - ocamlformat.0.15.0 pass (@dinosaure, #36)
- Use emile.1.0 (@dinosaure)
- Fix OPAM file, add
dune
constraint (@kit-ty-kate) - Delete dead-code (@seliopou, #29)
- Move to
angstrom.0.14.0
(@dinosaure, #30) - Use
bigarray-overlap.0.2.0
(@dinosaure, #31)
- New lightweight version of
mrmime
- Use
unstrctrd
- Add some accessors/converters
- First release