CRAN submission.

[grabear]

We can use this issue to discuss anything related to CRAN submission including any feedback that we might get.

#79 is also going to contain some commits so that we can submit to CRAN.

[grabear]

@sdhutchins
I've done a lot of work this afternoon to get the regular build check to work (e.g. devtools::check())

After I get a few minor things fixed I'm going to run R CMD check --as-cran, and then try running our package through Rhub or Winbuilder .

[grabear]

@sdhutchins Any hot tips on how to prevent this:

Warning: replacing previous import 'scales::alpha' by 'microbiome::alpha' when loading 'MicrobiomeR'

or this:

Package suggested but not available for checking: 'covr'

[sdhutchins]

@sdhutchins Any hot tips on how to prevent this:

Warning: replacing previous import 'scales::alpha' by 'microbiome::alpha' when loading 'MicrobiomeR'

or this:

Package suggested but not available for checking: 'covr'

One way with scales is to @import scales instead of @importFrom scales alpha

[sdhutchins]

For covr, I'm not sure. I don't think covr is on cran.

[grabear]

This is an increadibly annoying Note that occurs when using NSE:

> checking R code for possible problems ... NOTE

 convert_proportions: no visible binding for global variable 'SampleID'
  convert_proportions: no visible binding for global variable 'Abundance'
  convert_proportions: no visible global function definition for ':='
  convert_proportions: no visible binding for global variable
    'TreatmentGroup'
  correlation_data: no visible binding for global variable 'treatment_1'
  correlation_data: no visible binding for global variable 'treatment_2'
  correlation_data: no visible binding for global variable
    'wilcox_p_value'
  correlation_data: no visible binding for global variable 'Significance'
  correlation_data: no visible binding for global variable
    'log2_mean_ratio'
  correlation_data: no visible binding for global variable 'label'
  correlation_data: no visible binding for global variable '.'
  correlation_plot: no visible binding for global variable 'mean_treat2'
  correlation_plot: no visible binding for global variable 'mean_treat1'
  correlation_plot: no visible binding for global variable 'Abundance'
  correlation_plot: no visible binding for global variable
    'color_wilcox_p_value'
  correlation_plot: no visible binding for global variable 'rank_label'
  heat_tree_parameters: no visible binding for global variable
    'log2_mean_ratio'
  heat_tree_parameters: no visible binding for global variable
    'wilcox_p_value'
  heat_tree_plots: no visible binding for global variable 'n_supertaxa'
  melt_taxmap: no visible binding for global variable 'X.SampleID'
  melt_taxmap: no visible binding for global variable 'otu_id'
  otu_id_filter: no visible binding for global variable 'otu_id'
  otu_prevalence_filter: no visible binding for global variable
    'n_samples'
  permanova: no visible global function definition for 'as.formula'
  permanova: no visible global function definition for 'anova'
  permanova: no visible global function definition for 'coefficients'
  sample_id_filter: no visible binding for global variable 'sample_id'
  stacked_barplot: no visible global function definition for ':='
  stacked_barplot: no visible binding for global variable 'SampleID'
  stacked_barplot: no visible binding for global variable
    'Relative.Abundance'
  taxa_prevalence_filter: no visible binding for global variable
    'n_samples'
  top_coefficients_barplot: no visible global function definition for
    'par'
  top_coefficients_barplot: no visible global function definition for
    'barplot'
  transformer: no visible binding for global variable '.'
  transposer: no visible binding for global variable '.'
  Undefined global functions or variables:
    . := Abundance Relative.Abundance SampleID Significance
    TreatmentGroup X.SampleID anova as.formula barplot coefficients
    color_wilcox_p_value label log2_mean_ratio mean_treat1 mean_treat2
    n_samples n_supertaxa otu_id par rank_label sample_id treatment_1
    treatment_2 wilcox_p_value
  Consider adding
    importFrom("graphics", "barplot", "par")
    importFrom("stats", "anova", "as.formula", "coefficients")
  to your NAMESPACE file.

Solutions include:

1. https://dplyr.tidyverse.org/articles/programming.html#different-data-sets

If this function is in a package, using .data also prevents R CMD check from giving a NOTE about undefined global variables (provided that you’ve also imported rlang::.data with @importFrom rlang .data).

2. Writing R Packages: no visible binding for global variable STAT545-UBC/Discussion#451 (comment)

quiets concerns of R CMD check re: the .'s that appear in pipelines
if(getRversion() >= "2.15.1") utils::globalVariables(c("."))

@sdhutchins probably going to try the first one. It seems more elegant (even though it will take longer..). However if that doesnt work we can just do number 2. What do you think?

[grabear]

For covr, I'm not sure. I don't think covr is on cran.

Should it be in the Remotes: field in the DESCRIPTION file?

[sdhutchins]

Maybe?
If it's only in Suggests, I'm not sure.

[grabear]

I tried, and I don't think it worked.

[grabear]

I'm going to try putting it in the Imports field.