Missing files in the data for replication

[pachevalier]

The following files are not available in the repo and there is no direct link to download them :

• emploi_sal_2015.csv'
• ICIO2018Z.Rda (international data)
• ICIO2018Z.Rdata
• ICIO2018X.Rdata
• ICIO2018HFCE.Rdata
• ICIO2018NPISH.Rdata
• ICIO2018GGFC.Rdata

Would it be possible to add them in the repo or to add download url ?

[FLimousin]

hello !
the file emploi_sal_2015.csv has been renamed "eqtp_2015.csv", so it is available in the repository

For ICIO data, it is more complicated as files are quite heavy, so i can not add them in the repository. I don't know yet where we will put them, so i do not have a direct link for download. If you give me an email adress, i can use wetransfer or something like that to transfer them to you

[pachevalier]

Thank you very much for your answer.

For large files, you may have a look at services Github Large Files.

But the best would be to give the url of the source file and publish the script which transform the data in the good format.

[FLimousin]

I ll look into github Large files.
The URL for the source is given in both the video and the written tutorial. And the conversion from the OECD CSV file to our R tables is not entirely scripted...

[pachevalier]

Ok, good practices would also be

• to put the raw data in a "data-raw" directory
• avoid absolute paths to files. Prefer using relative path. Rstudio projects or here library can help
• Use Rmarkdown for tutorials or vignette
• Reorganize the code with setup (ie libraries and functions), data import (load the data), analyse, etc (See this example https://gist.github.com/pachevalier/7789677fe8b994c7cc09ee4868b141af)

This would improve readability and reusability of the source code

[FLimousin]

OK !
thank you for your advice !