1 curl the next few years - perhaps
2 =================================
4 Roadmap of things Daniel Stenberg and Steve Holme want to work on next. It is
5 intended to serve as a guideline for others for information, feedback and
6 possible participation.
11 Improve performance. Measurements and tests have shown that in several cases
12 doing transfers over HTTP/2 can be notably slower than the same transfer done
13 over HTTP/1. Some of that difference can be attributed the inefficient window
14 size handling currently in use but there are probably more to be learned and
15 worked on to optimize this.
20 The standardization process of QUIC has been taken to the IETF and can be
21 followed on the [IETF QUIC Mailing
22 list](https://www.ietf.org/mailman/listinfo/quic). I'd like us to get on the
23 bandwagon. Ideally, this would be done with a separate library/project to
24 handle the binary/framing layer in a similar fashion to how HTTP/2 is
25 implemented. This, to allow other projects to benefit from the work and to
26 thus broaden the interest and chance of others to participate.
31 The new version of the TLS protocol is in the pipeline and will soon start to
32 get used out in the wild. It offers some new interesting features and will
33 need the TLS libraries to adapt and quite likely provide additional or
34 modified APIs. libcurl needs to adapt accordingly.
40 Two cookie drafts have been adopted by the httpwg in IETF and we should
41 support them as the popular browsers will as well:
43 [Deprecate modification of 'secure' cookies from non-secure
44 origins](https://tools.ietf.org/html/draft-ietf-httpbis-cookie-alone-00)
46 [Cookie Prefixes](https://tools.ietf.org/html/draft-ietf-httpbis-cookie-prefixes-00)
48 [Firefox bug report about secure cookies](https://bugzilla.mozilla.org/show_bug.cgi?id=976073)
53 How to find services for specific domains/hosts.
58 To avoid network traffic to/from the proxy getting snooped on. There's a git
59 branch in the public git repository for this that we need to make sure works
60 for all TLS backends and then merge!
65 make sure there's an easy handle passed in to `curl_formadd()`,
66 `curl_formget()` and `curl_formfree()` by adding replacement functions and
67 deprecating the old ones to allow custom mallocs and more
72 Add support for third-party SASL libraries such as Cyrus SASL.
74 SASL authentication in LDAP
75 ---------------------------
79 Simplify the SMTP email
80 -----------------------
82 Simplify the SMTP email interface so that programmers don't have to
83 construct the body of an email that contains all the headers, alternative
84 content, images and attachments - maintain raw interface so that
85 programmers that want to do this can
90 Allow the email protocols to return the capabilities before
91 authenticating. This will allow an application to decide on the best
92 authentication mechanism
97 Allow Windows threading model to be replaced by Win32 pthreads port
102 Implement a dynamic buffer size to allow SFTP to use much larger buffers and
103 possibly allow the size to be customizable by applications. Use less memory
104 when handles are not in use?
109 1. Embed a language interpreter (lua?). For that middle ground where curl
110 isn’t enough and a libcurl binding feels “too much”. Build-time conditional
113 2. Simplify the SMTP command line so that the headers and multi-part content
114 don't have to be constructed before calling curl
119 1. build for windows (considered hard by many users)
121 2. curl -h output (considered overwhelming to users)
123 3. we have > 170 command line options, is there a way to redo things to
124 simplify or improve the situation as we are likely to keep adding
125 features/options in the future too
127 4. docs (considered "bad" by users but how do we make it better?)
131 5. authentication framework (consider merging HTTP and SASL authentication to
132 give one API for protocols to call)
134 6. Perform some of the clean up from the TODO document, removing old
135 definitions and such like that are currently earmarked to be removed years
141 1. makefile.vc files as there is no point in maintaining two sets of Windows
142 makefiles. Note: These are currently being used by the Windows autobuilds