Prepare for submission to ropensci

[kauedesousa]

The package is almost ready for submission to ropensci. This is a Milestone proposed to R. Bivand in his course on "Analysing Spatial Data". The package will be also submitted to JOSS. Here some things to do.

To do:

• Write vignette (done!)
• Write paper
• Check consistency with ropensci templates
• Check text and typos

[adamhsparks]

Happy to help with this since I've written several that are rOpenSci packages and reviewed a few as well.

[adamhsparks]

I have done the following so far today to help prepare for submission to rOpenSci:

• Add codemeta.json using {codemetar}

• Precompute the vignette so that installation is faster, see https://ropensci.org/technotes/2019/12/08/precompute-vignettes/

• Tidy up the vignette text

• Checked and corrected spelling using devtools::spell_check()

• Corrected minor issues in DESCRIPTION file such as grammar, author roles and line breaks at 80 chars

• Formatted documentation

[kauedesousa]

Great! I will start to write the paper.

[kauedesousa]

I just committed the first draft of the paper AgrDataSci@e39feee

Tomorrow I will work on how to add the Fig1.png (or .svg) to the document.

[kauedesousa]

Added the fig but still some work to do.

[adamhsparks]

I've made a few very minor edits to paper.Rmd

[kauedesousa]

@adamhsparks, do you think we are ready to send it to ropensci?

[adamhsparks]

I've just pushed a last few minor changes and corrections for the paper and the tests.

Let's do it!

[kauedesousa]

Great! I will submit.

[kauedesousa]

Submitted ropensci/software-review#357

[kauedesousa]

I have started the updates from the first comments/suggestions we got in the rOpenSci review. There is a folder here with the point-to-point response to each comment. I couldn't go further because the server is not working. It generates the ids but do not return the data from the request (in R and using the example provided in the API documentation).

[adamhsparks]

Thanks.

I read through it. A few thoughts to help you.

1. The /man structure is part of using the pre-built vignettes. I'm not aware of any issues with it. CRAN hasn't baulked at it and I just submitted two package updates that use this method in the last month.

2. The tests use skip_on_cran() due to CRAN policies for Internet connectivity. You'll get a nastygram and have the package pulled if the test fails due to this so I just skip them on CRAN. Ideally webmocking would help with this. I tried to use and eventually failed/gave up with nasapower to use VCR, https://docs.ropensci.org/vcr/. But still I'd encourage us to look at implementing this in the tests. I'll try to have another go at it with chirps and see if I can get it to work properly.

[adamhsparks]

I think we're ready to go back to the reviewers now. I don't find anything else outstanding.

[kauedesousa]

Great! I just sent it.