PyCran
Overview 👀
Yet another metadata parser for R source packages and R metadata information
Package: ABACUS
Version: 1.0.0
Depends: R (>= 3.1.0)
Imports: ggplot2 (>= 3.1.0), shiny (>= 1.3.1),
Suggests: rmarkdown (>= 1.13), knitr (>= 1.22)
License: GPL-3
MD5sum: 50c54c4da09307cb95a70aaaa54b9fbd
NeedsCompilation: noFor more see: https://cran.r-project.org/src/contrib/PACKAGES
PyCran lets us parse raw metadata and get it as dictionary, you can:
- Encode metadata dictionary to raw format,
- Decode raw metadata and receive it as dictionary,
- Load from tar archive with R library sources.
Installation 💾
$ pip install pycran
Usage 🚀
Decode
import pycran
raw_metadata = """
Package: ABACUS
Version: 1.0.0
Depends: R (>= 3.1.0)
Imports: ggplot2 (>= 3.1.0), shiny (>= 1.3.1),
Suggests: rmarkdown (>= 1.13), knitr (>= 1.22)
License: GPL-3
MD5sum: 50c54c4da09307cb95a70aaaa54b9fbd
NeedsCompilation: no
"""
assert pycran.decode(raw_metadata) == {
"Package": "ABACUS",
"Version": "1.0.0",
"Depends": "R (>= 3.1.0)",
"Imports": "ggplot2 (>= 3.1.0), shiny (>= 1.3.1),",
"Suggests": "rmarkdown (>= 1.13), knitr (>= 1.22)",
"License": "GPL-3",
"MD5sum": "50c54c4da09307cb95a70aaaa54b9fbd",
"NeedsCompilation": "no",
}Encode
import pycran
metadata = {
"Package": "ABACUS",
"Version": "1.0.0",
"Depends": "R (>= 3.1.0)",
"Imports": "ggplot2 (>= 3.1.0), shiny (>= 1.3.1),",
"Suggests": "rmarkdown (>= 1.13), knitr (>= 1.22)",
"License": "GPL-3",
"MD5sum": "50c54c4da09307cb95a70aaaa54b9fbd",
"NeedsCompilation": "no",
}
expected = """
Package: ABACUS
Version: 1.0.0
Depends: R (>= 3.1.0)
Imports: ggplot2 (>= 3.1.0), shiny (>= 1.3.1),
Suggests: rmarkdown (>= 1.13), knitr (>= 1.22)
License: GPL-3
MD5sum: 50c54c4da09307cb95a70aaaa54b9fbd
NeedsCompilation: no
"""
assert pycran.encode(metadata) == expectedLoad from R source archive
import pycran
# you can pass path to archive
pycran.from_file("PATH/TO/PACKAGE/ABACUS_1.0.0.tar.gz")
# or you can pass tarfile object
import tarfile
pycran.from_file(tarfile.open("PATH/TO/PACKAGE/ABACUS_1.0.0.tar.gz"))Parse raw metadata
In cases when you need to parse metadata for multiple
packages you can pass the data to pycran.parse function
import pycran
# somehow you download the contents of https://cran.r-project.org/src/contrib/PACKAGES
package_list = requests.get(https://cran.r-project.org/src/contrib/PACKAGES).text()
# And parse it as a result you will get a generator which you can iterate
pycran.parse(package_list)Enjoy!
