examples: extend radix_tree_complex_value example by igchor · Pull Request #1051 · pmem/libpmemobj-cpp

igchor · 2021-03-05T13:22:31Z

to show that creating city_info on dram is not possible and show
example with a structure wrapped in pmem::obj::p<>

This change is

codecov · 2021-03-05T15:35:04Z

Codecov Report

Merging #1051 (ee95c83) into master (7cc2f38) will decrease coverage by 0.04%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1051      +/-   ##
==========================================
- Coverage   94.27%   94.23%   -0.05%     
==========================================
  Files          48       48              
  Lines        4720     4720              
==========================================
- Hits         4450     4448       -2     
- Misses        270      272       +2

Flag	Coverage Δ
tests_clang_debug_cpp17	`93.66% <ø> (-0.07%)`	⬇️
tests_gcc_debug	`92.03% <ø> (-0.08%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...libpmemobj++/detail/enumerable_thread_specific.hpp	`98.91% <0.00%> (-1.09%)`	⬇️
...ude/libpmemobj++/container/concurrent_hash_map.hpp	`94.35% <0.00%> (-0.48%)`	⬇️
include/libpmemobj++/condition_variable.hpp	`78.57% <0.00%> (+4.76%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7cc2f38...ee95c83. Read the comment docs.

lukaszstolarczuk

Reviewable status: 0 of 1 files reviewed, 7 unresolved discussions (waiting on @igchor)

examples/radix_tree/radix_tree_complex_value.cpp, line 43 at r1 (raw file):

	/* If some fields are often updated at the same time, it might
	 * be better to put the inside a structure and wrap the whole structure

to put them?

examples/radix_tree/radix_tree_complex_value.cpp, line 47 at r1 (raw file):

	 * (which means better performance).
	 *
	 * Is date structure is only used in city_info, year, month and day do

Is -> If

examples/radix_tree/radix_tree_complex_value.cpp, line 48 at r1 (raw file):

	 *
	 * Is date structure is only used in city_info, year, month and day do
	 * not have to be wrapped in pmem::obj::p<>. However, it recommended to

it's

examples/radix_tree/radix_tree_complex_value.cpp, line 49 at r1 (raw file):

	 * Is date structure is only used in city_info, year, month and day do
	 * not have to be wrapped in pmem::obj::p<>. However, it recommended to
	 * do so, if the date can be used in some other context.

I'm, not sure what you mean in this comment - we should wrap year/month/date if we use the date outside of the city_info structure...?

examples/radix_tree/radix_tree_complex_value.cpp, line 82 at r1 (raw file):

		/* WRONG: cannot create city_info on stack (DRAM) since
		 * pmem::obj::string can only be placed on pmem:

you could add info to be completely clear `can't create the structure on stack (since ...) and pass it to radix_tree
or maybe
just extend the info above, that these 3 args you pass to try_emplace are passed to the city_info ctor

examples/radix_tree/radix_tree_complex_value.cpp, line 96 at r1 (raw file):

	 * be updated or none of them (even in case of failure).
	 *
	 * This code will result in two snaphots.

perhaps will result in -> will make

examples/radix_tree/radix_tree_complex_value.cpp, line 105 at r1 (raw file):

	/* Update "Gdańsk" latest_update_date field.
	 *
	 * This code will result in only a single snaphot.

a single -> one

to show that creating city_info on dram is not possible and show example with a structure wrapped in pmem::obj::p<>

igchor

Reviewable status: 0 of 1 files reviewed, 7 unresolved discussions (waiting on @lukaszstolarczuk)

examples/radix_tree/radix_tree_complex_value.cpp, line 43 at r1 (raw file):