diff options
author | Kartik K. Agaram <vc@akkartik.com> | 2016-10-22 12:16:02 -0700 |
---|---|---|
committer | Kartik K. Agaram <vc@akkartik.com> | 2016-10-22 12:16:02 -0700 |
commit | e939023612da044098e649e1e77cacceb73c564d (patch) | |
tree | 065f3e158bfb78d97c12c0c4b435681e722fe7ba /030container.cc | |
parent | ada5eb55cb185edf30dcac48b25cc485d44677ef (diff) | |
download | mu-e939023612da044098e649e1e77cacceb73c564d.tar.gz |
3553
Tag all transforms as idempotent or not. I'd fallen off this wagon. I might even be getting it wrong. Something a type system should automatically verify.
Diffstat (limited to '030container.cc')
-rw-r--r-- | 030container.cc | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/030container.cc b/030container.cc index 6abbbcc6..03e146ec 100644 --- a/030container.cc +++ b/030container.cc @@ -175,7 +175,7 @@ const type_tree* root_type(const type_tree* t) { //: also store a copy in each reagent in each instruction in each recipe :(after "End Type Modifying Transforms") -Transform.push_back(compute_container_sizes); +Transform.push_back(compute_container_sizes); // idempotent :(code) void compute_container_sizes(recipe_ordinal r) { recipe& caller = get(Recipe, r); @@ -820,7 +820,7 @@ def main [ +mem: storing 34 in location 2 :(after "Transform.push_back(expand_type_abbreviations)") -Transform.push_back(expand_type_abbreviations_in_containers); +Transform.push_back(expand_type_abbreviations_in_containers); // idempotent :(code) // extremely inefficient; we process all types over and over again, once for every single recipe // but it doesn't seem to cause any noticeable slowdown |