Open nanxstats opened 2 years ago
agree 👍
I think if we export read_pkglite()
, it's reasonable to also create and export a new function write_pkglite()
by decoupling the middle part from unpack()
. This means that you can read it from the txt, work on the list, and write it back.
Update: oops, unpack()
only writes to package structures so it's irrelevant. We only need to create the logic to write the list to txt. A typical user flow can be pack()
-> read_pkglite()
-> modify list or just get some data -> write_pkglite()
or not -> unpack()
.
To show this visually:
We should extend the format.Rmd
vignette to explain this (APIs for working with the txt) and include the above chart.
It feels like it would generate more benefits than harms if we make
read_pkglite()
an exported function, because people might need to parsepkglite.txt
directly sometimes. This would avoid the use of:::
.What do you think? @elong0527