-
-
Notifications
You must be signed in to change notification settings - Fork 31.4k
build: do not build doc in source tarball #17100
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Conversation
0486a84
to
9a82b27
Compare
9a82b27
to
b88ae43
Compare
@joyeecheung what are the steps to verify this works? |
@joyeecheung any chance this whole thing could be refactored out of the main Makefile? @MylesBorins, one possible option is to use a new test file written by @joyeecheung to validate the output of the doctool - |
@MylesBorins My way is:
Also if you are on Mac, make sure you are testing it with GNU Make v4.x (the default one is 3.x, which does not really have a problem with the previous configuration) |
@refack Sorry, I am not sure I am following, by "the whole thing" do you mean "make doc-only"? But I am pretty sure there are people actually reading its output e.g. people working on doc tools, writing docs, or just wanting to read the latest docs.. |
I'm not saying remove it, I'm suggesting moving it to it's own Makefile (possibly AFAICT |
@refack I think we can refactor it out to another Makefile (still nice to avoid rebuilding unmodified docs), but probably in another PR. |
BTW both #16661 and its previous implementation could result in concurrent |
Landed in 289fcb0, thanks! |
opting to not land this on v6.x. Please feel free to change the labels and open a backport |
Checklist
make -j4 test
(UNIX), orvcbuild test
(Windows) passesAffected core subsystem(s)
build
Fixes: #16650
This is the real fix for #16650 , which makes sure that GNU make v4.x won't build the docs if the source is extracted from the source tarball (i.e. doc/api contains built docs) (see the theory in #16650 (comment))
Also during the investigation of this issue I think I have a better idea about #17043, I'll open a separate PR with a more robust
available-node
and try to fix all the$(NODE)
usage there, hopefully fixing the makefile regression. For this PR the current implementation is enough.cc @nodejs/build