[FEATURE] Unitialised resize for deserialisation #160
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
seqan-actions
added
lint
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #160 +/- ##
=======================================
Coverage 99.94% 99.94%
=======================================
Files 44 44
Lines 1687 1702 +15
=======================================
+ Hits 1686 1701 +15
Misses 1 1
☔ View full report in Codecov by Sentry. |
eseiler
force-pushed
the
misc/cereal_voodoo
branch
from
67f26f7 to
10ad88d
Compare
seqan-actions
added
lint
|
|
||
| resize_for_overwrite(vector_size); | ||
|
|
||
| if constexpr (cereal_text_archive<archive_t>) |
There was a problem hiding this comment.
do we ever use a text archive? I thought we cerealize in binary format?
There was a problem hiding this comment.
No, but our cereal tests check for every archive, also text archives. In practice, I hope that a bitvector will never be serialised as json/XML :D
| auto gen = [&dist, &engine]() | ||
| { | ||
| return dist(engine); | ||
| }; | ||
| seqan::hibf::bit_vector vec(size_in_bits); | ||
| std::ranges::generate(vec, gen); |
There was a problem hiding this comment.
| auto gen = [&dist, &engine]() | |
| { | |
| return dist(engine); | |
| }; | |
| seqan::hibf::bit_vector vec(size_in_bits); | |
| std::ranges::generate(vec, gen); | |
| seqan::hibf::bit_vector vec(size_in_bits); | |
| std::ranges::generate(vec, dist(engine)); |
does this not accept a temporary?
There was a problem hiding this comment.
Not like this, dist(engine) returns a single integer, but generate wants a callable which it then calls multiple times
I ran some benchmarks with raptor.
uninit= with PRinit= without PRIndex size: 135.5 GiB (left) and 363.2 GiB (right)