Bug 1056516 - submit Hyphenation Control reftests to CSSWG. r=jfkthame MozReview-Commit-ID: 636GBMjvej6

Bug 1056516 - add tests to our local reftests folder. r=jfkthame MozReview-Commit-ID: 1Az4xBe6Nts

Bug 1056516 - let auto hyphen honor manual hyphen when hyphens:auto is set. r=jfkthame MozReview-Commit-ID: BWzg3pb2lDH

Bug 1056516 - use HyphenType to store different types of hyphenations. r=jfkthame MozReview-Commit-ID: 22ICYSaY1I2

Bug 1056516 - use AutoTArray for hyphenBuffer in BreakAndMeasureText. r=jfkthame MozReview-Commit-ID: 1fQGiosLYG

Backed out 3 changesets (bug 1344346) for mochitest shutdownleaks Backed out changeset 71e3443cdc43 (bug 1344346) Backed out changeset cd8993e4d0dd (bug 1344346) Backed out changeset 8e03e57d44aa (bug 1344346)

servo: Merge #15783 - Replace manual host parsing code with parse-host crate (from ferjm:issue15756-parse-hosts); r=avadacatavra This patch is replacing the code to parse the hosts file with the [parse-hosts](https://crates.io/crates/parse-hosts) crate. This crate has a [CC0 1.0 Universal License](https://creativecommons.org/publicdomain/zero/1.0/deed.en). I could have used [HostsFile::load()](https://clarcharr.github.io/parse-hosts/parse_hosts/struct.HostsFile.html#method.load) directly, but this method loads `/etc/hosts` by default and does not allow to override the default path (for example with [env::var("HOST_FILE")](https://dxr.mozilla.org/servo/source/components/net_traits/hosts.rs#19)), so I kept the existing code to open and read the content of `env::var("HOST_FILE")` and also kept the [parse_hostsfile](https://dxr.mozilla.org/servo/source/components/net_traits/hosts.rs#42) method (used by the unit tests), but I modified it to use [HostsFile::read_buffered](https://clarcharr.github.io/parse-hosts/parse_hosts/struct.HostsFile.html#method.read_buffered), which is doing the actual hosts parsing for a given string buffer. - [X] `./mach build -d` does not report any errors - [X] `./mach test-tidy` does not report any errors - [X] These changes fix #15756 . - [X] There are tests for these changes (tests/unit/net/resource_thread.rs and tests/unit/net/http_loader.rs) Source-Repo: https://github.com/servo/servo Source-Revision: 8f7c69e15f0b9beddad2ba67e3419f88aca0d579

Bug 1329093 - Part 5: Update test expectations; r=bz MozReview-Commit-ID: 4mfC99WEQFG

Bug 1329093 - Part 4: stylo: Delay SVG mapped attr resolution till later; r=bz MozReview-Commit-ID: 2GvHPg1egjS

Bug 1329093 - Part 3: Overload FromContent() to work with const; r=bz MozReview-Commit-ID: BUFmPtcf6LM

Bug 1329093 - Part 1: stylo: Handle SVG presentation attributes; r=bz MozReview-Commit-ID: 9cymo3c9HIn

Bug 1344589 - Wait for the httpServer to shutdown in head_search.js to avoid reported leaks. r=adw The test slow_timeout in test_searchSuggest.js was keeping a request open for 10s so the server wouldn't stop immediately. MozReview-Commit-ID: EGFPRKMNa51

Bug 1329793 - Re-subscribe for a push channel periodically r=eoger,nalexander On startup and at the beginning of a sync we check how long it has been since we've subscribed to a channel for fxa service. If it's been over 21 days, request re-subscription. MozReview-Commit-ID: GzvPecZ9hTy

Bug 1291483 part 2: Use UniquePtr/MakeUnique more thoroughly in chain-of-custody for gfxFont::mVerticalMetrics. r=jfkthame This member-var has type 'UniquePtr', but (up until this patch) its value is set up using "new" and raw pointers. This patch improves that codepath by using UniquePtr & MakeUnique, for stronger ownership guarantees. MozReview-Commit-ID: KWZVpvr9bYj

Bug 1291483 part 1: Use 'MakeUnique' instead of 'new' to allocate gfxFont::mGlyphChangeObservers. r=jfkthame MozReview-Commit-ID: 7FO1HAukp5j

servo: Merge #15893 - Support SVG presentation attributes (from Manishearth:stylo-svg-pres); r=bz r=bz from https://bugzilla.mozilla.org/show_bug.cgi?id=1329093 Source-Repo: https://github.com/servo/servo Source-Revision: ee223798cc72bf81a4b2bb6849697de617c5f028

Bug 1344346 - [mozlog] Use less verbose formatting on 'process_output' messages, r=jgraham This simplifies the 'process_output' formatting in both the mach and tbpl formatters. It will also add the string 'pid' somewhere in the format, but only if the process id is actually a positive int. MozReview-Commit-ID: 6nc5q06cDfM

Bug 1344346 - [mozharness] Ensure StructuredOutputParser runs 'process_output' messages through the error lists, r=jgraham Currently the StructuredOutputParser validates all unstructured output against a series of error lists (regexes). However, before this happens, the mochitest harness is converting all unstructured output to structured 'process_output' messages. This means that Gecko output is not being checked against the error regexes. This change ensure that in addition to unstructured output, we also validate 'process_output' messages against the error lists. MozReview-Commit-ID: DG6sZqpg5aw

Bug 1344346 - [mochitest] Log unstructured output as 'process_output', r=jgraham Mochitest currently converts unstructured logs (e.g output from gecko) to 'info' messages. But this means those messages won't be validated against mozharness' error logs. This change first gets unstructured messages logged as process_output, and also ensures the StructuredOutputParser in mozharness checks process_output messages against the error list. MozReview-Commit-ID: KPTQnulwzyK

Bug 1345222 - Fix user-set ClearType params detection. r=mchang MozReview-Commit-ID: B7bCPbZ9id9