... as they otherwise appear in the about:networking list as "false"
while having been resolved by TRR.
MozReview-Commit-ID: 9g9fUExvyjS
--HG--
extra : rebase_source : 3098b7c3f7d01e55f5a8c031fc6a73e53f7edb05
... to avoid a catch-22 as CP needs name resolve to work.
MozReview-Commit-ID: DC1CjlUy4cJ
--HG--
extra : rebase_source : 3bc0f146071f04757d44b76f352e48b2abb4dad0
* Deserialization now only happens via a mutator
* The CID for URI implementations actually returns the nsIURIMutator for each class
* The QueryInterface of mutators implementing nsISerializable will now act as a finalizer if passed the IID of an interface implemented by the URI it holds
MozReview-Commit-ID: H5MUJOEkpia
--HG--
extra : rebase_source : 01c8d16f7d31977eda6ca061e7889cedbf6940c2
* Deserialization now only happens via a mutator
* The CID for URI implementations actually returns the nsIURIMutator for each class
* The QueryInterface of mutators implementing nsISerializable will now act as a finalizer if passed the IID of an interface implemented by the URI it holds
MozReview-Commit-ID: H5MUJOEkpia
--HG--
extra : rebase_source : 8ebb459445cab23288a6c4c86e4e00c6ee611e34
These are the only occurrences of "browser.cache.use_new_backend" in the tree.
MozReview-Commit-ID: IOSw0uUD5FW
--HG--
extra : rebase_source : c6c7e6abfdf2c3ff23c1f18e36110781f6b5580b
The original code (from bug 1200802) declared an XPCOM object as a static bare
pointer, which for future reference is probably never the right thing to do. It
might have worked if it was cleared before shutdown but it never was.
MozReview-Commit-ID: EMe7wgzm6zv
--HG--
extra : rebase_source : 572ce6822e297692bab3311a65e1143785b913c4
Previously it would store the time even for early AAAA responses that
weren't used until the A response came in, thus getting the timing
wrong.
MozReview-Commit-ID: KctUjMjH5FR
--HG--
extra : rebase_source : fafd7ad25cd1b7a43d4b4a5653600cad830b471d
To prevent it from sitting waiting for an event that was sent before the
TRRService started.
MozReview-Commit-ID: 9F0IlWGdA9O
--HG--
extra : rebase_source : 6a70210a4e538d8a1b240684a0b3ed5fed38e6ad
There's no need to allocate a completely new nsCString when all we want
to do is parse a character string into an integer. We can allocate a
dependent string instead, which will avoid some memory churn.
Otherwise it will just load back the same (problematic) addresses from the
cache again the second time. This introduces a new resolver bit
(REFRESH_CACHE) that also invalidates the existing cache entry while doing the
new resolve.
MozReview-Commit-ID: 5Bc2KiAGYYA
--HG--
extra : rebase_source : ae368c88a5db27f0980b9928439d27588bc84815
What we're really doing in CacheEntry::CheckRequest is checking:
a) Whether the method is contained in our allowed methods; and
b) Whether all of the headers are contained in our allowed headers.
nsTArray lets us check directly for containing elements, so let's use
that facility rather than rolling our own.
nsCaseInsensitiveCStringComparator ought to be cheap to construct, but
the object actually has a vtable to install and whatnot. So it's
beneficial to pull the construction of it outside of the headers loop.
The entries in mMethods and mHeaders aren't sorted in any special way,
so we can remove expired entries using UnorderedRemoveElementAt, which
is faster than RemoveElementAt.
DNSRequestChild's constructor takes `const nsCString&` parameters. If
you have a `const nsACString&` argument to pass in to the constructor,
you're forced to construct a temporary nsCString object to satisfy the
prototype. But the temporary string will just get copied inside the
constructor and you'll have to destroy the temporary string object you
created.
Let's skip all this, and just have DNSRequestChild take `const
nsACString&` arguments instead, and avoid the temporary object.
Early AAAA responses might cause issues on hosts without working native
IPv6 connectivity, of course especially notable in TRR-only mode.
MozReview-Commit-ID: 6ZqE6AKnucH
--HG--
extra : rebase_source : ff42cb8daf941a3fa1f7e783c76d823e879024c3
mCleanedUp is a VarCache variable, which mirrors the canonical value of the
network.predictor.cleaned-up pref. When the canonical pref value is modified,
e.g. by SetBool(), then mCleanedUp is also updated.
But the updating relationship is one-way -- if mCleanedUp is modified, the
canonical value of the pref is not updated. Such an inconsistency is bad! For
example, Predictor.cpp will use mCleanedUp's value, but about:config will show
the canonical value.
(For this reason, VarCache prefs are meant to be read-only outside of libpref.
Bug 1436655 will enforce this.)
This patch changes mCleanedUp so it's not a VarCache variable, avoiding the
mirroring issue.
MozReview-Commit-ID: LIG02gMkRjF
--HG--
extra : rebase_source : 273b2372ce718b0f346695a0dc96a189cd3ba233
Older versions of clang complain about this code, with:
netwerk/protocol/http/nsCORSListenerProxy.cpp:230:5: error: default initialization of an object of const type 'const struct CheckHeaderToken' without a user-provided default constructor
Later versions of the standard were amended to make this not an error,
but clang still warns. Pacify clang by making the object not const.
WebPlatformTests expect that when calling
url.host = "host:" // port missing
url.host = "host:65536" // port too big
url.host = "host:bla" // invalid port
that the hostname will be set, but the port will be left unchanged.
Since DOM APIs are the only consumers for SetHostPort it means we can change this behaviour to match the WPT expectations.
As such, SetHostPort will return NS_OK if setting the host succeded, and will ignore if setting the port failed.
MozReview-Commit-ID: LoMw8hCWlCv
--HG--
extra : rebase_source : db28b73d98060c2f66f899afe1a4ae26f4db85db
We normally fail in nsStandardURL::SetPassword if the username is empty.
But if the password we are trying to set is also empty, we should't really fail.
MozReview-Commit-ID: FIDqkPrb1gp
--HG--
extra : rebase_source : 9080c44e91e27acd210f3ace3a234528513928c3
* Code in XMLHttpRequestMainThread is converted to set the username and password individually. This is because when the parameters are empty, it ended up calling SetUserPass(":") which always returns an error.
MozReview-Commit-ID: 3cK5HeyzjFE
--HG--
extra : rebase_source : f34400c11245d88648b0ae9c196637628afa9517
Calling NS_ENSURE_SUCCESS at the begining of the methods shows the error in the console, but what we really care about is where the error code comes from. So it is best to use if (NS_FAILED(mStatus) {} at the begining of the methods, and NS_ENSURE_SUCCESS right after calling the appropriate method on mMutator.
MozReview-Commit-ID: 5vVuHk3N4FU
--HG--
extra : rebase_source : f477f0f87b3303422a595f27d9b39ac25335d701
* Removes setHostAndPort from nsIURIMutator as it only has one use
* Instead, the consumer sets the port to -1 before calling setHostPort()
MozReview-Commit-ID: Jx9UMW440hq
--HG--
extra : rebase_source : cb60e76c905db6bb308ddfd8fa548cc13d3afe06
RFC 1035 section 4.1.1 documents this bit as:
RD - Recursion Desired - this bit may be set in a query and is copied
into the response. If RD is set, it directs the name server to pursue
the query recursively. Recursive query support is optional.
MozReview-Commit-ID: 8iHDgNtA1L1
--HG--
extra : rebase_source : a29010a2894fd00ebfbfb869f5938cf507345a2f
... and remove Cancel() from nsHostRecord::RemoveOrRefresh since we
don't need to cancel TRR resolves due to network changes.
MozReview-Commit-ID: Akuhpzgg4N5
We instead add a templated method NS_MutatorMethod that returns a std::function<nsresult(nsIURIMutator*)> which Apply then calls with mMutator as an argument.
The function returned by NS_MutatorMethod performs a QueryInterface, then calls the passed method with arguments on the result.
MozReview-Commit-ID: Jjqp7gGLG1D
--HG--
extra : rebase_source : f2a17aee7bb66a7ba8652817d43b9aa7ec7ef710
We instead add a templated method NS_MutatorMethod that returns a std::function<nsresult(nsIURIMutator*)> which Apply then calls with mMutator as an argument.
The function returned by NS_MutatorMethod performs a QueryInterface, then calls the passed method with arguments on the result.
MozReview-Commit-ID: Jjqp7gGLG1D
--HG--
extra : rebase_source : 592d13349a8c4627c7ce3146ec592f577b39f3cc
... and also store allow-rfc1918 bool locally to remove later accesses
to TRRservice.
MozReview-Commit-ID: KkO4u2N9gfE
--HG--
extra : rebase_source : 2fdfecb127987cdbfdccd0e77f7b4bb65f6f5f5d
When the native resolve needs to run again, the rec->mResolving counter
must not be decreased, as otherwise it triggers an assert when the next
resolve completes.
MozReview-Commit-ID: 3UQGkDSOmGi
--HG--
extra : rebase_source : 1caf047070a95b451acf0ddca6b5986f4e4d7c69
The test would set a timer for 200ms then expect to be called back more than 200ms later (as in 201ms or more)
This change makes sure that we don't fail if the callback comes back exactly 200ms later.
MozReview-Commit-ID: 1OCyO3juAdZ
--HG--
extra : rebase_source : 5e123337f73e6b95f5f0afa7d63604d5558dfd94
nsBase64Encoder::Finish currently works by calling PL_Base64Encode to
allocate a base64-encoded string, then Assign()'ing that to the result
string, another allocation. mozilla::Base64Encode enables us to base64
encode directly into the result string with a single allocation, saving
an allocation. (Base64Encode is also slightly more efficient, because
we don't have to do a strlen() on the string being Assign()'d.) Let's
use Base64Encode instead.
In the previous code, a race condition could cause us to call SendSetPriority() after calling SendDeleteSelf.
For example:
T1: SendDeleteSelf()
T2: if (!mIPCClosed) SendSetPriority()
T1: mIPCClosed = true
MozReview-Commit-ID: 3XOwCaphb2o
--HG--
extra : source : 4ebdab0e332892378558817e30d0138c95199ce5
Provides an optional resolver mechanism for Firefox that allows running
together with or instead of the native resolver.
TRR offers resolving of host names using a dedicated DNS-over-HTTPS server
(HTTPS is required, HTTP/2 is preferable).
DNS-over-HTTPS (DOH) allows DNS resolves with enhanced privacy, secure
transfers and improved performance.
To keep the failure rate at a minimum, the TRR system manages a dynamic
persistent blacklist for host names that can't be resolved with DOH but works
with the native resolver. Blacklisted entries will not be retried over DOH for
a couple of days. "localhost" and names in the ".local" TLD will not be
resolved via DOH.
TRR is preffed OFF by default and you need to set a URI for an available DOH
server to be able to use it. Since the URI for DOH is set with a name itself,
it may have to use the native resolver for bootstrapping. (Optionally, the
user can set the IP address of the DOH server in a pref to avoid the required
initial native resolve.)
When TRR starts up, it will first verify that it works by checking a
"confirmation" domain name. This confirmation domain is a pref by default set
to "example.com". TRR will also by default await the captive-portal detection
to raise its green flag before getting activated.
All prefs for TRR are under the "network.trr" hierarchy.
The DNS-over-HTTPS spec: https://tools.ietf.org/html/draft-ietf-doh-dns-over-https-03
MozReview-Commit-ID: GuuU6vjTjlm
--HG--
extra : rebase_source : 53fcca757334090ac05fec540ef29d109d5ceed3
The change to RootAccessible.cpp fixes an obvious bug introduced in bug 741707.
The visibility changes in gfx/thebes are because NS_DECL_ISUPPORTS has a
trailing "public:" that those classes were relying on to have public
constructors.
MozReview-Commit-ID: IeB8KIJCGhU
This macro is identical to NS_INTERFACE_MAP_END and encourages the
reader to think that there's something extra-special threadsafe about QI
implementations that use the macro, when in reality there's nothing of
the sort.
See the comment on "Address test failures caused by bumping timer precision to 2 ms"
for more details.
MozReview-Commit-ID: LrsucEPdZIo
--HG--
extra : rebase_source : 8147c034f7dc93f678eebc80b0afabf55729d804
This annotates two SQL statements that are responsible for a large amount of
warning spam due to "suboptimal indexes." Given we no longer wish to update
appcache code we can just disable the warning with a commment instead.
--HG--
extra : rebase_source : 519678e38f69cfa2aab75161a058bcdaf548fad0
In the previous code, a race condition could cause us to call SendSetPriority() after calling SendDeleteSelf.
For example:
T1: SendDeleteSelf()
T2: if (!mIPCClosed) SendSetPriority()
T1: mIPCClosed = true
MozReview-Commit-ID: 3XOwCaphb2o
--HG--
extra : source : 4ebdab0e332892378558817e30d0138c95199ce5
In the previous code, a race condition could cause us to call SendSetPriority() after calling SendDeleteSelf.
For example:
T1: SendDeleteSelf()
T2: if (!mIPCClosed) SendSetPriority()
T1: mIPCClosed = true
MozReview-Commit-ID: 3XOwCaphb2o
--HG--
extra : rebase_source : 609b9d111d97a8e4a60117dd79ecd741725ec611
