RFC: npm fund depth

[fishcharlie]

Formatted RFC

What / Why

This RFC proposes adding a depth flag to the npm fund command.

References

• [FEATURE] npm fund --depth=0 cli#1345

[fishcharlie]

This is my first time submitting an RFC to npm 🎉. 🤞 that I did it correctly. However if I didn't, please let me know and I'll make any corrections needed.

I look forward to discussing this RFC and answering any questions that arise with the community!

[ljharb]

what happens if one of your direct deps has no funding info, but it depends on dozens of packages that do declare funding info?

[fishcharlie]

@ljharb In this case, if you set depth=0 I believe it should not output the sub-dependencies funding information. It should be strictly limited to printing funding information of your direct dependencies.

I wish there was an existing example of that behavior that we could emulate, however I can't think of a similar example for this case.

Do you think I should clarify that more in the document?

[ruyadorno]

hi @fishcharlie, first of all thanks for the contribution 😊 I'm going to leave here the feedback we got from the last time we discussed it (as we went over it again during this week's npm OpenRFC call).

In most of npm subcommands, we're shifting the focus away from the --depth flag in favor of using the --all flag. The --all flag is now supported in the npm ls as of npm7 beta as well as in npm outdated (and maybe some other cmds I don't recall at the moment).

There were a bunch of discussions when we first set to remove support to --depth in npm outdated in a previous RFC that you can check in case you want to dig more on the rationale there: #133

An important change @isaacs noticed is that we'd still want to preserve the default behavior of npm fund printing funding info of your entire install tree, so the RFC should also mention that the all config would now need a different default value (current is false). This new default value of all should now be null allowing the subcommands to override it with their expected default values (which will be false for ls and outdated and true for fund).

Next action item here now is to tweak the RFC to reflect these proposed changes 😄