Thanks for the excellent tool, John.
Could recutils be treated more like YAML, with 'sub-values'?
Extending sample2.rec:
%rec: Book
%doc: A book in my personal collection.
Title: GNU Emacs Manual
Author: Richard M. Stallman
Topic: Ediff
Topic: ERC
Topic: ERT
Converting this to YAML, the expected result would be:
$ mlr --irecutils --oyaml cat sample2.rec
- Title: GNU Emacs Manual
Author: Richard M. Stallman
Topic:
- Ediff
- ERC
- ERT
But the actual result looks like the output of rec2csv:
- Title: GNU Emacs Manual
Author: Richard M. Stallman
Topic: Ediff
Topic_2: ERC
Topic_3: ERT
Miller can sort-of parse it with rename already:
$ mlr --irecutils --oyaml \
rename -r '"Topic$",Topic_1'\
then rename -r '"_",.' \
then cat book.rec
...but this does not feel tidy, and requires stating the field's name up-front.
Fully rendered Recfiles would be great, as it's the only plain-text and git-friendly relational database.
Thanks for the excellent tool, John.
Could recutils be treated more like YAML, with 'sub-values'?
Extending
sample2.rec:Converting this to YAML, the expected result would be:
$ mlr --irecutils --oyaml cat sample2.rec - Title: GNU Emacs Manual Author: Richard M. Stallman Topic: - Ediff - ERC - ERTBut the actual result looks like the output of
rec2csv:Miller can sort-of parse it with
renamealready:...but this does not feel tidy, and requires stating the field's name up-front.
Fully rendered Recfiles would be great, as it's the only plain-text and git-friendly relational database.