comparison src/capnproto-git-20161025/doc/roadmap.md @ 133:1ac99bfc383d

Add Cap'n Proto source
author Chris Cannam <cannam@all-day-breakfast.com>
date Tue, 25 Oct 2016 11:17:01 +0100
parents
children
comparison
equal deleted inserted replaced
132:42a73082be24 133:1ac99bfc383d
1 ---
2 layout: page
3 title: Road Map
4 ---
5
6 # Road Map
7
8 Here's what's (hopefully) in store for future versions of Cap'n Proto! Of course, everything here
9 is subject to change.
10
11 ## Near-ish future
12
13 Provisionally, these are probably the things that will be worked on in the next few releases of Cap'n Proto and its C++ reference implementation.
14
15 * **Shared memory RPC:** Zero-copy inter-process communication.
16 * **Three-way introductions (level 3 RPC):** Allow RPC interactions between more than two parties,
17 with new connections formed automatically as needed.
18 * **Fiber-based concurrency:** The C++ runtime's event loop concurrency model will be augmented
19 with support for fibers, which are like threads except that context switches happen only at
20 well-defined points (thus avoiding the need for mutex locking). Fibers essentially provide
21 syntax sugar on top of the event loop model.
22 * **Dynamic schema transmission:** Allow e.g. Python applications to obtain schemas directly from
23 the RPC server so that they need not have a local copy. Great for interactive debugging.
24 * **Improved MSVC support:** Once MSVC improves its support for C++11 language features, we will
25 support Cap'n Proto's reflection and RPC APIs in MSVC. (Currently, only core serialization is
26 supported.)
27 * **Implement encapsulated types:** This will allow you to create a hand-written wrapper around a
28 type which will be automatically injected into the generated code, so that you can provide a
29 nicer interface which encapsulates the type's inner state.
30 * **Implement maps:** Based on encapsulated and parameterized types.
31
32 ## Before version 1.0
33
34 These things absolutely must happen before any 1.0 release. Note that it's not yet decided when
35 a 1.0 release would happen nor how many 0.x releases might precede it.
36
37 * **Expand test coverage:** There are lots of tests now, but some important scenarios, such as
38 handling invalid of invalid input, need better testing.
39 * **Performance review:** Performance is already very good compared to competitors, but at some
40 point we need to break out the profiler and really hone down on the details.
41 * **Security review:** We need a careful security review to make sure malicious input cannot
42 crash an application or corrupt memory.
43
44 ## Wish List
45
46 These are features we'd like to implement some day but haven't decided yet how to prioritize.
47 Some of these features could make their way into Cap'n Proto before version 1.0. Others will
48 certainly come after. If you have opinions on what you'd like to see next,
49 [tell us](https://groups.google.com/group/capnproto)!
50
51 ### Language Features
52
53 * **Inline lists:** Lets you define a field composed of a fixed number of elements of the same
54 type, and have those elements be stored directly within the struct rather than as a separate
55 object. Useful mainly to avoid the need to validate list lengths when the length should always
56 be the same. Also saves a pointer's worth of space.
57 * **Type aliases:** Ability to define a type which is just an alias of some other type, and
58 have it show up as e.g. a `typedef` in languages that support that. (The current `using`
59 keyword is intended only for local use and does not affect code generation.)
60 * **Doc comments:** Harvest doc comments from schema files and use them to generate doc comments
61 on generated code. Also make them available in the compiled schema so that a documentation
62 generator could use them.
63
64 ### C++ API Features
65
66 * **JSON codec:** API for transcoding to JSON format, useful for interacting with legacy
67 infrastructure.
68 * **Snappy integration:** Integrate [Snappy compression](https://code.google.com/p/snappy/) into
69 the API to further reduce bandwidth needs with minimal CPU overhead.
70 * **Annotations API:** For each annotation definition, generate code which assists in extracting
71 that annotation from schema objects in a type-safe way.
72
73 ### Storage
74
75 * **ORM interface:** Define a standard interface for capabilities that represent remotely-stored
76 objects, with get, put, publish, and subscribe methods. Ideally, parameterize this interface
77 on the stored type.
78 * **mmap-friendly mutable storage format:** Define a standard storage format that is friendly
79 to mmap-based use while allowing modification. (With the current serialization format, mmap
80 is only useful for read-only structures.) Possibly based on the ORM interface, updates only
81 possible at the granularity of a whole ORM entry.
82
83 ### Tools
84
85 * **Schema compatibility checker:** Add a `capnp` command which, given two schemas, verifies
86 that the latter is a compatible upgrade from the former. This could be used as a git hook
87 to prevent submission of schema changes that would break wire compatibility.
88 * **RPC debugger:** Add a `capnp` command which sends an RPC from the command line and prints
89 the result. Useful for debugging RPC servers.
90
91 ### Infrastructure
92
93 Note: These are very large projects.
94
95 * **JSON-HTTP proxy:** Develop a web server which can expose a Cap'n Proto RPC backend as a
96 JSON-over-HTTP protocol.
97 * **Database:** A fast storage database based on Cap'n Proto which implements the ORM interface
98 on top of the mmap storage format.