Add _TSK_BIG_TABLES typedef.

[jeromekelleher]

Closes #1528

[codecov]

Codecov Report

Merging #1531 (59bd85f) into main (53b3179) will decrease coverage by 0.00%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #1531      +/-   ##
==========================================
- Coverage   93.70%   93.70%   -0.01%     
==========================================
  Files          27       27              
  Lines       22760    22759       -1     
  Branches     1076     1076              
==========================================
- Hits        21327    21326       -1     
  Misses       1399     1399              
  Partials       34       34              

Flag	Coverage Δ
c-tests	92.03% <100.00%> (-0.01%)	⬇️
lwt-tests	92.97% <ø> (ø)
python-c-tests	95.29% <ø> (ø)
python-tests	98.80% <ø> (ø)

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

Impacted Files	Coverage Δ
c/tskit/tables.c	90.16% <100.00%> (-0.01%)	⬇️

---

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 53b3179...59bd85f. Read the comment docs.

[benjeffery]

Awesome! Looks good to me - you mentioned you had some concerns about this way of overflow checking, are you happy with this now?

[jeromekelleher]

Looks good to me - you mentioned you had some concerns about this way of overflow checking, are you happy with this now?

I thought some more about it and you're right, it is much better! I got rid of the checks for > 0 since these are unsigned types, but basically it's what you proposed.

I'm slightly worried about the fact we don't have any tests for the boundary behaviour for num_rows though, I tried leaving out the "+1" and all the tests still passed.

[benjeffery]

Yep, added an issue for that #1532

[jeromekelleher]

@Mergifyio rebase

[mergify]

Command rebase: success

Branch has been successfully rebased

[jeromekelleher]

Yep, this is working on CircleCI, just verified it's definitely running in 64 bit mode by checking the total memory used reported in valgrind (much more in 64 bit mode).