Before this CL, each contact and host was independently
loaded via the ReferenceUnion adapter. Since fields are
processed serially by JAXB, this means worst-case there
were 17 loads, best case 3 (the 3 required contacts) and
usual case 5-6 (some hosts). This CL reduces that to 1
datastore roundtrip in all cases.
A side effect of this CL is the further hollowing-out of
ReferenceUnion, since it no longer is involved in
marshalling at all.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=123712842
This completes the command extensions for the regType 0.2 extension.
Up next will be the response extensions.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=123322887
Note that it doesn't do anything yet beyond basic XML validation
because the default registry system doesn't use registration types,
but this serves as a template for the other domain commands using
registration types and provides a method that TLDs implementing custom
logic can use. This also explicitly doesn't yet handle the response
extensions.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=123245388
This is so we can associate history records with all mutations when doing
database maintenance.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=123209304
This will make the utility a lot more versatile (allow both open and closed intervals) and hand the range validation and comparison off to the Range class.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122867110
Second-level domain name isn't accurate because we support multi-part
TLDs, so standardize on the "fullyQualifiedDomainName" name that is
used throughout the code base.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122693009
This also renames the existing FlowRegistry to FlowPicker to avoid
overloaded uses of the word "registry". Absent this renaming, the new
package would've been google.registry.flows.registry, which gives
entirely the wrong impression as it makes it sound like the home for
flows that affect TLDs.
This is a preparatory CL for adding flow picker engines that will
allow customized flows to run on a per-TLD basis.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122671260
RFC5731 leaves the registrant field as optional. We
currently don't show it on un-authed info, but we do
show it in WHOIS, so there's no strong reason not to
show it here. By doing so, we maintain the invariant
that registrant must always be present on a
DomainResource object.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122645356
Currently the build breaks due to a surprise migration to dnsjava 1.6.4,
which is not available in Maven. dnsjava 2.x is available internally,
but it jarjar's the package name. So we'll need to deal with that the
same way we deal with the Objectify package rename.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122447929
Old ForeignKeyIndex entities pointing to hosts are soft-deleted when a
host is renamed, and a new ForeignKeyIndex is created that correctly
points at the new host. This is an expected state of the system, so
don't throw an error when it is observed.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122447608
ReferenceUnion is a hack to work around the mismatch between how
we store references (by roid) and how they are represented in EPP
(by foreign key). If it ever needed to exist (not entirely clear...)
it should have remained tightly scoped within the domain commands
and resources. Instead it has leaked everywhere in the project,
causing lots of boilerplate. This CL hides all of that behind
standard Refs, and should be followed by work to remove ReferenceUnion
completely.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122424416
Synthetic BillingEvent.OneTimes need to have a pointer to their source Recurring event, in order to match up properly against Cancellations.
Also improving the exception message for mismatched Flag.SYNTHETIC and syntheticCreationTime.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=122155636
In order to clean up potentially bad BillingEvent.Recurring expansions, we'll need to be able to trace synthetic billing events back to particular runs of the []. This field will be set to the cursor time at the start of the MR (all expansions in one MR job will have the same timestamp).
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121999938
The tool needs to:
* Replace all hosts on a domain with a provided set of hosts
* Add 3 server locks to the domain
* Print an undo command that restores the domain to its original state
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121944934
1) By our convention, entity fields are package-private, not private.
2) Static analysis can't tell that checkArgumentNotNull prevents
continuing with a null on the next line. Change the code to make
a warning in eclipse go away.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121944468
This refactors the existing premium list functionality into the new
class StaticPremiumListPricingEngine, which implements PricingEngine.
A backfill @OnLoad is provided to default existing Registry entities
into the static implementation. For now there is just this one
implementation. Dagger map multibinding is used to generate the total
set of allowed pricing engines, and allows other parties to plug in
their own implementations.
The pricing engine is a required field on the Registry object. If you
don't want a particular Registry to actually have a premium list, then
use the static pricing engine but don't actually set a premium list.
A subsequent CL will refactor the Key<PremiumList> field on the
Registry entity class to be handled solely by the
StaticPremiumListPricingEngine implementation. Going forward, all
configuration and implementation details that are specific to a given
pricing engine should be handled by that pricing engine, and not as
fields on the Registry object.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121850176
Expanding recurring billing events will require a global cursor as opposed to a Registry-scoped cursor, so this CL creates a more generic Cursor type and adds a dual-write for the old RegistryCursor (for both old and new styles) on save. We can then touch any stragglers using the UpdateCursorsCommand and simply drop the old RegistryCursor once all have been migrated.
See [] for migration tracking.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121706885
Utilizing this function in the recurring billing event [], in order to abstract a lot of the expansion logic out of the MR itself.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121579246
We moved a few internal tools to third_party/ as part of large java code move but there's not much use for
these files to be in opensource drop so I pull it back to our project directory.
Basically this CL does:
- mv j{,t}/c/[] d/r/tools/
- rm third_party/java_src/gtld/j/[]
- corresponding tweaks
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121406277
Java's stock regex implementation doesn't guarantee linear time
complexity which makes it a security liability.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=121159875
The dark lord Gosling designed the Java package naming system so that
ownership flows from the DNS system. Since we own the domain name
registry.google, it seems only appropriate that we should use
google.registry as our package name.
This change renames directories in preparation for the great package
rename. The repository is now in a broken state because the code
itself hasn't been updated. However this should ensure that git
correctly preserves history for each file.
Plus refactoring of the KillAllXXXActions and tests to share
common code.
This essentially completes the KillAll []s. We can now reliably
clean out everything except for:
* Lock - harmless to leave alone or delete from the gae console
* GaeUserIdConverter - same
* RdeRevision - filed [] to track, but harmless if not cleaned up
* ForeignKeyHostIndex of renamed hosts - tracked in []
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=120781975
This refactoring allows the schemas to live in the same package they
are used, rather than all having to be inside the bigquery package.
This is a follow-up to []
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=120551011
This is needed to eliminate the last non-injected use of BigqueryFactory. It gets rid of the deprecated HttpServletUtils now that we don't have lots of separate servlets, and changes the check snapshot servlet's usage of such. It also fixes up some related minor style issues in the update snapshot action (and its injectable parameters).
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=120547998
In [] I made Registry a BGR in prep for each Registry being in
its own entity group. But we decided not to do that, and in []
mcilwain@ moved Registry under the cross-tld root. So there's no reason
for Registry to implement BackupGroupRoot anymore.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=120402417
Our whois format was flagged as wrong in the .meet PDT. Although
we had followed the AWIP samples from ICANN, the definitive list
of field names is from Specification 4 of our contract, available at
https://newgtlds.icann.org/sites/default/files/agreements/agreement-approved-09jan14-en.htm
and indeed our fields are incorrect. (The remaining formatting issues
are ambiguous but the PDT testers' interpretation is probably correct.)
Since the footer format is now somewhat more complicated, I also denormalized
the disclaimer field into all of the testdata files. (I spent some time
debugging an extra newline between the content and the disclaimer, and
it would have been far clearer to solve if the files had been this way.)
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=120338930
