layout: Make sure anonymous table flows are statically positioned.
The failing `float-applies-to-*` CSS 2.1 tests never really should have
been passing in the first place; they depend on floats inside
fixed-layout tables working properly, which they don't.
Closes#6078.
Closes#6709.
Closes#6858.
r? @mbrubeck
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6977)
<!-- Reviewable:end -->
The failing `float-applies-to-*` CSS 2.1 tests never really should have
been passing in the first place; they depend on floats inside
fixed-layout tables working properly, which they don't.
Closes#6078.
Closes#6709.
Closes#6858.
layout: Take inline margins into account when determining the intrinsic sizes of fragments.
Avoids a needless wrapped line in the repository name on GitHub.
r? @mbrubeck
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7011)
<!-- Reviewable:end -->
Use Box<str> instead of Box<String> for UnscannedTextFragmentInfo
This removes an extraneous layer of indirection, without increasing the size of the `SpecificFragmentInfo` enum (because `Box<str>` is a fat pointer, which is the same size as the `FlowRef` trait object that is already stored in `InlineAbsoluteHypotheticalFragmentInfo`)
r? @pcwalton
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7014)
<!-- Reviewable:end -->
Fixes#6879 (Improper connections to the devtools server)
Modifies how we behave in the case that something attempts to communicate with the devtools server improperly. This includes...
- Invalid encoding (Non `UTF8`) of the packet length / error parsing / none specified
- JSON encoding error (such as a `Parser::SyntaxError` or a `Parser::IoError`)
Happy to make changes if anyone has an issue with this or feels another way is more idiomatic!
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7010)
<!-- Reviewable:end -->
Lock stderr while printing the alert() message.
2a7f262b7d was unsufficient for the case where
the interleaved output was actually on stderr rather than stdout, such as
output from the error macro.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/7027)
<!-- Reviewable:end -->
Prevent flake8 from crashing tidy
If someone were to write:
def hello() :
print "hello world"
flake8 would warn:
stdin:54:45: E203 whitespace before ':'
Normally there are only three colons in a flake8 error message, but this
one has four, which causes issue with this line:
_, line_num, _, message = error.split(":")
...causing this error:
ValueError: too many values to unpack
This commit updates the `str.split` call to utilize the `maxsplit`
parameter to prevent this error from occurring.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6996)
<!-- Reviewable:end -->
This removes an extraneous layer of indirection, without increasing the size
of the `SpecificFragmentInfo` enum (because `Box<str>` is a fat pointer, which
is the same size as the `FlowRef` trait object that is already stored in
`InlineAbsoluteHypotheticalFragmentInfo`)
If someone were to write:
def hello() :
print "hello world"
flake8 would warn:
stdin:54:45: E203 whitespace before ':'
Normally there are only three colons in a flake8 error message, but this
one has four, which causes issue with this line:
_, line_num, _, message = error.split(":")
...causing this error:
ValueError: too many values to unpack
This commit updates the `str.split` call to utilize the `maxsplit`
parameter to prevent this error from occurring.
