rustdoc: Don't generate empty files for stripped items #34045
Conversation
We need to traverse stripped modules to generate redirect pages, but we shouldn't generate anything else for them. This now renders the file contents to a Vec before writing it to a file in one go. I think that's probably a better strategy anyway.
|
r? @cmr (rust_highfive has picked a reviewer for you, use r? to override) |
|
Hum... Strange. It doesn't prevent the file creation for |
|
It should do, do you have an example where it doesn't? Files for What currently happens is empty files are created anyway which I believe is what the problem in #34025 is. |
|
Your solution might be better than mine then. I close my PR in favor of yours. Great job. 😃 |
|
@bors r+ thanks @GuillaumeGomez @ollie27 |
|
📌 Commit cfb4ad2 has been approved by |
|
@BenTheElder: No need, bors will test and merge on its own. |
|
Oh cool :-) On Fri, Jun 3, 2016 at 3:41 PM, Guillaume Gomez notifications@github.com
|
|
I ran linkchecker after this PR and it looks like I uncovered another rustdoc bug, now getting:
I've fixed that in #34068 but assuming bors is gated on linkchecker then this PR won't pass until that is merged. Should I add that commit to this PR if that's easier? |
|
⌛ Testing commit cfb4ad2 with merge da75ed2... |
|
💔 Test failed - auto-linux-64-opt-rustbuild |
|
@ollie27 If you have the patience let's just wait for the other to land. |
|
@bors: retry |
rustdoc: Don't generate empty files for stripped items We need to traverse stripped modules to generate redirect pages, but we shouldn't generate anything else for them. This now renders the file contents to a Vec before writing it to a file in one go. I think that's probably a better strategy anyway. Fixes: #34025
We need to traverse stripped modules to generate redirect pages, but we shouldn't generate
anything else for them.
This now renders the file contents to a Vec before writing it to a file in one go. I think
that's probably a better strategy anyway.
Fixes: #34025