search

randrescastaneda / joyn

joyn provides a set of tools to analyze the quality of merging (i.e., joining) data frames. It is a JOY to join with joyn
https://randrescastaneda.github.io/joyn/
Other
9 stars 4 forks source link

Optimize #64

Closed randrescastaneda closed 4 months ago

randrescastaneda commented 5 months ago

Hey guys (@zander-prinsloo and @RossanaTat),

I have made substantial changes in {joyn} following Sebastian's suggestions. Now, the ".joyn" variable is created in collapse::join but it is modified down the road to account for those observations in left that have been updated with values in right. To make the whole process more efficient and faster, this variable is now a factor by default.

Also, the freq_table() has been improved dramatically. It is way faster than the previous version.

I had to skip three check because they were checking matching types (e.g., "m:1", "1:1" and friends) in the joyn_workhorse function. We always merge "m:m" in collapse::join and then filter.

If you guys agree. We can merge and move to version 2.0.1. Keep in mind that this version depends on collapse 2.0.15, which is not available in CRAN yet, so we can't submit now.

image

SebKrantz commented 5 months ago

Great work! I can submit 2.0.15 end of the month. 2.0.14 was submitted on 2024-05-24 and CRAN requires authors to wait a month unless there are immanent issues.

SebKrantz commented 4 months ago

collapse 2.0.15 has made it to CRAN. Sorry for the delay. There are large changes going on in R's C API to which developers need to react.

randrescastaneda commented 4 months ago

collapse 2.0.15 has made it to CRAN. Sorry for the delay. There are large changes going on in R's C API to which developers need to react.

@SebKrantz, Fantastic. Thanks!. joyn 0.2.2 is now in CRAN and depends on collapse 2.0.15. Once again, thank you for your suggestions and great work with collapse. It has become the main package in our team.

Best