check out rdwd

[brry]

This is not an issue, but rather a comment / question.
I am writing a comprehensive R package for DWD data. When I started it, I was not aware of your code, else I might not have done that. Now that I'm on it, I might as well finish it and put it on CRAN.
I have a couple of things you don't seem to have implemented and vice versa. Your idea to automatically merge recent and historical files is quite cool - would you mind if I borrow that?
I'd be happy to get some feedback / testing from you if you find the time and interest...
https://github.com/brry/rdwd#rdwd

[nFrechen]

Hey Berry,
of course you are welcomed to borrow code from our project! Bad though we didn't meet earlier to develop the package together... I admit I didn't do much to make the package known (and especially did not try to get it on CRAN). I see that your package has better basic structure. We started with only daily data download (because back then it was on a different server using http instead of ftp). The hourly data download is sort of a late addition we didn't fit in the basic structure very well...yet.

We had some problems with file encodings and wrong (or rather inconsistent) formatting in some files and had to lengthen our code in some places to cope with them. Did you have the similar problems? I had contact at some point with someone from the open data department of the DWD and suggested to to change some of the things, but not much happened since then...

Did you think about providing translations of the table headers (especially if you put it on CRAN)?

Greetings,
Nanu

[brry]

Thanks! I have also seen some inconsistent column widths in meta files, for example, so rdwd::readDWD can now cope with that. Column name translation (at least in the documentation) is a good idea!