diff options
| author | Michał Muskała <[email protected]> | 2018-02-03 15:41:11 +0100 |
|---|---|---|
| committer | Michał Muskała <[email protected]> | 2018-04-24 11:57:29 +0200 |
| commit | ad72c0d37ffb214cac874f51ac29fe2cdb47a2a4 (patch) | |
| tree | fcc905efbf7a6e10aa4a0b8358910dc4242c47ff /lib/stdlib | |
| parent | a5dc5af3fe3b1e25f0596d71d3ac36291b0099dc (diff) | |
| download | otp-ad72c0d37ffb214cac874f51ac29fe2cdb47a2a4.tar.gz otp-ad72c0d37ffb214cac874f51ac29fe2cdb47a2a4.tar.bz2 otp-ad72c0d37ffb214cac874f51ac29fe2cdb47a2a4.zip | |
Introduce map_get guard-safe function
Rationale
Today all compound data types except for maps can be deconstructed in guards.
For tuples we have `element/2` and for lists `hd/1` and `tl/1`. Maps are
completely opaque to guards. This means matching on maps can't be
abstracted into macros, which is often done with repetitive guards. It
also means that maps have to be always selected whole from ETS tables,
even when only one field would be enough, which creates a potential
efficiency issue.
This PR introduces an `erlang:map_get/2` guard-safe function that allows
extracting a map field in guard. An alternative to this function would be
to introduce the syntax for extracting a value from a map that was planned
in the original EEP: `Map#{Key}`.
Even outside of guards, since this function is a guard-BIF it is more
efficient than using `maps:get/2` (since it does not need to set up the
stack), and more convenient from pattern matching on the map (compare:
`#{key := Value} = Map, Value` to `map_get(key, Map)`).
Performance considerations
A common concern against adding this function is the notion that "guards
have to be fast" and ideally execute in constant time. While there are
some counterexamples (`length/1`), what is more important is the fact
that adding those functions does not change in any way the time
complexity of pattern matching - it's already possible to match on map
fields today directly in patterns - adding this ability to guards will
niether slow down or speed up the execution, it will only make certain
programs more convenient to write.
This first version is very naive and does not perform any optimizations.
Diffstat (limited to 'lib/stdlib')
| -rw-r--r-- | lib/stdlib/src/erl_internal.erl | 2 | ||||
| -rw-r--r-- | lib/stdlib/src/ms_transform.erl | 3 |
2 files changed, 3 insertions, 2 deletions
diff --git a/lib/stdlib/src/erl_internal.erl b/lib/stdlib/src/erl_internal.erl index 89b97b901e..6d3d5baa23 100644 --- a/lib/stdlib/src/erl_internal.erl +++ b/lib/stdlib/src/erl_internal.erl @@ -76,6 +76,7 @@ guard_bif(floor, 1) -> true; guard_bif(hd, 1) -> true; guard_bif(length, 1) -> true; guard_bif(map_size, 1) -> true; +guard_bif(map_get, 2) -> true; guard_bif(node, 0) -> true; guard_bif(node, 1) -> true; guard_bif(round, 1) -> true; @@ -337,6 +338,7 @@ bif(list_to_tuple, 1) -> true; bif(load_module, 2) -> true; bif(make_ref, 0) -> true; bif(map_size,1) -> true; +bif(map_get,2) -> true; bif(max,2) -> true; bif(min,2) -> true; bif(module_loaded, 1) -> true; diff --git a/lib/stdlib/src/ms_transform.erl b/lib/stdlib/src/ms_transform.erl index 6616e957c0..ec8cfd56c2 100644 --- a/lib/stdlib/src/ms_transform.erl +++ b/lib/stdlib/src/ms_transform.erl @@ -944,6 +944,7 @@ real_guard_function(node,1) -> true; real_guard_function(round,1) -> true; real_guard_function(size,1) -> true; real_guard_function(map_size,1) -> true; +real_guard_function(map_get,2) -> true; real_guard_function(tl,1) -> true; real_guard_function(trunc,1) -> true; real_guard_function(self,0) -> true; @@ -1115,5 +1116,3 @@ normalise_list([H|T]) -> [normalise(H)|normalise_list(T)]; normalise_list([]) -> []. - - |