Running Lighthouse on DevTools lead to these changes:
* Label our <select> toolbar items
* Don't use tabIndex=1 for our canvas elements
* Don't use li items in the network panel
* add lang=en to our html tags
Change-Id: I9fef3d731dbbcbb56c3f73d81fd5669d65fa9827
Reviewed-on: https://chromium-review.googlesource.com/1093399
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Joel Einbinder <einbinder@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#566195}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 237b755cd007b58b585ad34137353e11e2bc3ccd
We used to listen to resize events from the root DevTools window or a
SplitWidget, but not when the viewport grows/shrinks due to toggling
the filter bar.
This CL updates the viewport in that case, too.
Bug: 807917
Change-Id: I7c5cfbe20f7e293f808e8138b2c7da0ab76df845
Reviewed-on: https://chromium-review.googlesource.com/1077757
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Erik Luo <luoe@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#563331}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 4dcc9eb7b3652120ef1b3bd07bdd9b918c9a7d60
It is useful when debugging the certificate especially when failed to
load the signed exchange.
(When succeeded, we can view the certificate in Security panel)
This CL also changes "certUrl" to optional in PDL file, because this
could be null according to the spec.
Bug: 830505
Change-Id: I710d91bc43bd414d03aca7d4e3dfdb9342b10500
Reviewed-on: https://chromium-review.googlesource.com/1055407
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Kunihiko Sakamoto <ksakamoto@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#561758}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 4defbb6af0fda52f8a5ccc32ad5eb03cb31ce645
This incorporats with dgozman's comments in crrev.com/c/1004890
- Change "from SignedExchange" to "from signed-exchange" in NetworkRequestNode.
- Show the signed exchagne url as the initiator of certificate request.
- Show "from signed-exchange" in the headers tab of the inner request.
- Add "View request" link in the preview tab of signed exchange to show the inner request.
- Add "Learn more" link in the preview tab of signed exchange.
Bug: 830505
Change-Id: Ica7ae52e563efb283c600c812be196951a12a336
Reviewed-on: https://chromium-review.googlesource.com/1065558
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Kunihiko Sakamoto <ksakamoto@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#560287}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 4ac04d63b06db0e81a986691346f67ba8ad7f865
SignedExchangeDevToolsProxy sends the SignedExchangeHeader and the certificate
information and the error messages to DevTools.
When the user clicks a signed exchange response in Network panel, these
information will be displayed in the Preview panel.
And the network grid view will show “from SignedExchange” in the size column of
the content of the signed exchange like “from disk cache”.
See the doc for more details: https://bit.ly/2HXVbCU
This CL depends on https://crrev.com/c/1029755
Bug: 830505
Change-Id: Ice3743f2e1675b6d604eeaba731ebd8494c2b5a1
Reviewed-on: https://chromium-review.googlesource.com/1004890
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Reviewed-by: Kunihiko Sakamoto <ksakamoto@chromium.org>
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#559363}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 705b2aa35e1523b1a00b6c8aa73b3a5f0c8e50b1
This change will make it easier to finish
https://chromium-review.googlesource.com/c/chromium/src/+/1035091:
- The "other" blocked reason, that is, an unspecified blocked reason,
is now the default which we'll assume for the 0 value.
- There's a 1:1 correspondence with the enum in
browser_protocol.pdl, so we no longer need to conflate kNone and
kOther.
- The base::nullopt value for base::Optional<...> is more appropriate
because in the inspector's protocol it will correspond to the absence
of a string value (e.g. in a JSON message), as opposed to the string
"none" or "other" (both misleading).
Would like to land this first since it's easier to review / test it
this way, as opposed to mixing it into a larger PR.
Bug: 838291
Change-Id: I66da523495b38a297d22289ae7c32367a66bd6bb
Reviewed-on: https://chromium-review.googlesource.com/1056030
Commit-Queue: Johannes Henkel <johannes@chromium.org>
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Cr-Original-Commit-Position: refs/heads/master@{#558109}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: ee045a5ae4784d9223b1152db737f40c93a385da
Joel Einbinder