-
Notifications
You must be signed in to change notification settings - Fork 4.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
common/http: Remove chromium_url #14583
Conversation
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, if this can nuke source/common/chromium_url
then this is a fairly clean approach (relatively, but the shim could still work..).
/retest |
Retrying Azure Pipelines: |
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Nuking |
|
||
// Canonicalizes a host that requires IDN conversion. Returns true on success | ||
bool DoIDNHost(const gurl_base::char16* src, int src_len, CanonOutput* output) { | ||
- int original_output_len = output->length(); // So we can rewind below. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Coverage and fuzz_coverage resolves more symbols from url
lib, hence it is needed to include url_canon_host.cc
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I feel the shimmed one is safer.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@yanavlasov @dio thinking about this vs. the shim PR, I think ultimately, if we can upstream Chromium to give us an IDN-free optional build, that would be the ideal, since then we don't need to own a patch permanently. Shimming reduces the size of the patch, but in all cases with a patch we have a permanent patch ownership issue. LGTM modulo comments.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I recall this patch is needed to make fuzzing test happy. I was going to make another pass to see if tests can be fixed to get rid of the patch.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, the last bit of this patch is related to fuzzing test. I agree we should take a deeper look on tests and modify it.
This reverts commit be0f0b8. Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
/retest |
Retrying Azure Pipelines: |
/retest |
Retrying Azure Pipelines: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you add a release note please?
/wait
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
Signed-off-by: Dhi Aurrahman <[email protected]>
/retest |
Retrying Azure Pipelines: |
/retest |
Retrying Azure Pipelines: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM; I'd like @alyssawilk to provide a final go/no-go before we merge, as another input on the runtime default direction, but good from my end. Thanks so much @dio for working through an area of large ambiguity and integration complexity, eliminating our fork here is a huge win.
Generally we encourage new code paths on by default unless we think it's very high risk. If this is roughly the same code, just a different source, I think on by default totally make sense :-) |
@htuch @yanavlasov Sorry, fixing the conflicted current.rst dismissed the reviews. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks!
/azp run |
Azure Pipelines successfully started running 1 pipeline(s). |
@htuch or @moderation we need lgtm deps? :) |
/lgtm deps |
Commit Message: This patch adds runtime feature flag to switch from chromium_url to url lib from googleurl. The new one is enabled by default.
Additional Description: This is provided as an alternative of #13808.
Risk Level: Medium
Testing: Existing, plus added unit tests for testing different runtime flags.
Docs Changes: N/A
Release Notes: N/A
Platform Specific Features: N/A