This project has stagnated, but perhaps can provide a foundation for future work. In particular, I beyond any of the TODO/FIXMEs listed, some thoughts. See seanmonstar/reqwest#14 for some background/history. The work that @SergioBenitez did in https://github.com/alexcrichton/cookie-rs and related answered/solved some of the questions in the above issue, esp. w.r.t. efficiency of cookie handling. At the same time, it raised some questions for me in terms of a good direction to take this code in:
- The Cookie work largely involved decreasing/eliminating allocations with some more efficient string indexing, but I believe much of the logic in
user_agent
necessarily involves some canonicalization/normalization that would obviate this benefit. So it would make sense for this project to evolve in a way that e.g. a project likereqwest
supportinguser_agent
doesn't have a negative impact when not needed. - cookie-rs has a CookieJar implementation, which handles storing cookies as well as some security details, but does not aim to implement path/domain matching, etc. Does it fit into this picture anywhere? Should
CookieJar
be updated to e.g. separate out concerns like security, session management, etc.? - Does it make more sense for
user_agent
to be something that wraps a thing like areqwest
/hyper
/curl
/etc. client, or should things likereqwest
to "own" a session?
- error-chain
- env_logger
- change get_with, etc. to not take a closure for "prepare_and_send"?
- are re-directs handled correctly?
- target stable instead of nightly
- utilize publicsuffix.org per 4.1.2.3 http://tools.ietf.org/html/rfc6265#section-4.1.2
- implement created/last-accessed timestamps
- make serde dependency optional
- reqwest example
This project is licensed and distributed under the terms of both the MIT license and Apache License (Version 2.0).
See LICENSE-APACHE and LICENSE-MIT