flip doesn't work as expected

[Xeonilian]

Hi Guangchuang,
Today my computer crashed and I reinstall R, Rstudio and all the packages. I found flip did werid things. After flipping sometimes, the branches were not evenly distribute in my own case. In the following case, the tree even overlaped. Please try this example.

set.seed(20180220)
randtree <- rcoal(20)
plot(randtree)
t <- ggtree(randtree, layout = "rectangular") + geom_tiplab() + geom_nodelab()
t$data$label <-  1:39
t
t1 <- flip(t, 33, 22) # no problem
t1
t2 <- flip(t1, 24, 23) # branches overlap
t2
t$data$y
t2$data$y

The sessioninfo for this result is

sessionInfo()
R version 3.4.3 (2017-11-30)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)

Matrix products: default

locale:
[1] LC_COLLATE=Chinese (Simplified)_China.936 LC_CTYPE=Chinese (Simplified)_China.936
[3] LC_MONETARY=Chinese (Simplified)_China.936 LC_NUMERIC=C
[5] LC_TIME=Chinese (Simplified)_China.936

attached base packages:
[1] grid stats graphics grDevices utils datasets methods base

other attached packages:
[1] gtable_0.2.0 reshape2_1.4.3 ggtree_1.10.4 treeio_1.2.1 ggplot2_2.2.1 quantreg_5.35 SparseM_1.77
[8] picante_1.6-2 nlme_3.1-131.1 vegan_2.4-6 lattice_0.20-35 permute_0.9-4 ape_5.0

loaded via a namespace (and not attached):
[1] Rcpp_0.12.15 cluster_2.0.6 magrittr_1.5 MASS_7.3-48 munsell_0.4.3 colorspace_1.3-2
[7] rlang_0.1.6 stringr_1.2.0 plyr_1.8.4 tools_3.4.3 parallel_3.4.3 mgcv_1.8-23
[13] MatrixModels_0.4-1 digest_0.6.15 lazyeval_0.2.1 tibble_1.4.2 Matrix_1.2-12 purrr_0.2.4
[19] tidyr_0.8.0 glue_1.2.0 labeling_0.3 stringi_1.1.6 compiler_3.4.3 pillar_1.1.0
[25] rvcheck_0.0.9 scales_0.5.0 jsonlite_1.5

I tried on my Mac where I have not updated for a while, here flip works as I expected.
I can not say which one is the source, they all pretty old, see sessionInfo below.

R version 3.4.1 (2017-06-30)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS High Sierra 10.13.3

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.4/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] grid stats graphics grDevices utils datasets
[7] methods base

other attached packages:
[1] ggtree_1.8.1 treeio_1.0.2 gtable_0.2.0
[4] ggplot2_2.2.1 reshape2_1.4.2 quantreg_5.33
[7] SparseM_1.77 picante_1.6-2 nlme_3.1-131
[10] vegan_2.4-3 lattice_0.20-35 permute_0.9-4
[13] ape_4.1

loaded via a namespace (and not attached):
[1] Rcpp_0.12.12 cluster_2.0.6 magrittr_1.5
[4] MASS_7.3-47 munsell_0.4.3 colorspace_1.3-2
[7] rlang_0.1.2 stringr_1.2.0 plyr_1.8.4
[10] tools_3.4.1 parallel_3.4.1 mgcv_1.8-18
[13] MatrixModels_0.4-1 lazyeval_0.2.0 tibble_1.3.3
[16] Matrix_1.2-10 tidyr_0.6.3 labeling_0.3
[19] stringi_1.1.5 compiler_3.4.1 rvcheck_0.0.9
[22] scales_0.4.1 jsonlite_1.5

I looked closer at ggtree. In the new version of flip a varieble yy is created. And value of y is changed afterwards.

yy <- df$y[-c(sp1, sp2)]
df$y[-c(sp1, sp2)] <- yy + ((min(sp2.df$y, na.rm = TRUE) - 
    max(yy)) - (min(yy) - max(sp1.df$y, na.rm = TRUE)))/2

I don't quite understand what these line do, but perhaps they are the source.

Actually I'm doing the final check before submission. I really hope it can work, then people get the same figure as in my paper.
Thank you very much for spending time on checking this issue.

Cheers, Xeonilian

[GuangchuangYu]

fixed in v >= 1.10.5.