*** Reason for rollback ***
This code is fine, and I will be resurrecting it unaltered next week. It's only being rolled back for operational reasons related to the timing of our internal pushes.
*** Original change description ***
Switch over to non-ReferenceUnion fields on DomainBase
This is the second phase of a three phase migration to remove
ReferenceUnions. As of the end of this phase, ReferenceUnions are no longer read
from in any active code paths, but are still written to in case a rollback to
the previous version is necessary. The third and final phase will remove the
ReferenceUnions entirely.
***
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=136642759
This is the second phase of a three phase migration to remove
ReferenceUnions. As of the end of this phase, ReferenceUnions are no longer read
from in any active code paths, but are still written to in case a rollback to
the previous version is necessary. The third and final phase will remove the
ReferenceUnions entirely.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=136388057
It is important to get at least this one commit in before the public Nomulus
release so that none of our public users will have to go through this data
migration (although we will have to).
The migration strategy is as follows:
1. Dual-write to non-ReferenceUnion fields in addition to the current
ReferenceUnion fields in use, and add new indexes (this commit). Deploy.
2. Run the ResaveAllEppResourcesAction backfill [].
3. Switch all code over to using the new fields. Dual-write is still in effect,
except it is now copying over the values of the new fields to the old
fields. Switch over all BigQuery reporting scripts to use the new
fields. Deploy.
4. Remove all of the old code and indexes. Deploy.
5. (Optional, at our leisure) Re-run the ResaveAllEppResourcesAction backfill
[] to delete the old obsolete fields.
Note that this migration strategy is rollback-safe at every step -- new data is
not read until it has already been written out in the previous step, and old
data is not removed immediately following a step in which it was still being
read, so the previous step is safe to roll back to.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=136196988
This is to better distinguish between an LRP "token" (the string passed along in EPP) and the datastore entity that contains the token and all metadata.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=135943480
It is no longer needed since we don't do log-replay and it's
annoying to support in the flat flows.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=135805088
There was a circular reference when hydrating a domain with a
subordinate host, since the host references the domain. To fix
this, I redid @DoNotHydrate to be the way it should have been,
rather than the hack I had originally submitted. I also beefed
up the unit tests of the epp resource types to check for cycles.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=135792416
Very few flows actually check the phase. Push the checks down to the leaf
flows so that we can remove the inherited code from ResourceFlow and replace
it with utility methods. In the process, document and test two places that
throw the exception but did not previously test it.
This introduces a temporary hack in BaseDomainCreateFlow that does something
specific for DomainApplicationCreateFlow. It will go away literally tomorrow
when I flatten that flow.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=135480538
The reason field is 1:1 mapped to skus in billing reports. Need to add a new reason for EAP for this type of billing event for reporting to work correctly. Also map that reason to the correct SKU.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=134005364
When custom effective date is passed in the check command, the response should
contain that date as an acknowledgemant that the check is performed at a time
different from now.
Also when the fee(s) returned contains a validDateRange (i. e. EAP fees that
are only valid during a certain period), the response will contain a notAfter
field which is the date after which the quoted fee(s) are no longer valid. (i.
e. the earliest of the end dates of all fees that would expire.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133989775
It is replaced by loadByForeignKey(), which does the same thing that
loadByUniqueId() did for contacts, hosts, and domains, and also
loadDomainApplication(), which loads domain application by ROID. This eliminates
the ugly mode-switching of attemping to load by other foreign key or ROID.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133980156
It's best to be consistent and use the same thing everywhere. "clientId" was
already used in more places and is shorter and no more ambiguous, so it's the
logical one to win out.
Note that this CL is almost solely a big Eclipse-assisted refactoring. There are
two places that I did not change clientIdentifier -- the actual entity field on
Registrar (though I did change all getters and setters), and the name of a
column on the exported registrar spreadsheet. Both would require data
migrations.
Also fixes a few minor nits discovered in touched files, including an incorrect
test in OfyFilterTest.java and some superfluous uses of String.format() when
calling checkArgument().
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133956465
Currently EapFee is a separate class that has no inheritance from either
BaseFee and Fee. With this CL its functionality is merged into the Fee class
and the type of the fee can be identified by the FeeType enum in the Fee class.
Future custom fees can follow the same pattern.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133627570
We almost certainly had intended to record a String representation of the
numeric Code all along, but a bad toString() method on the enum resulted
in the wrong thing happening. This is more evidence of why overriding
toString() on enums is bad.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133583683
Added support to specify custom effective date to run the fee check command.
Such functionality is useful for TLDs with creation price as a function of
time, such as those with EAP. However the implementation is not limited EAP or
create price check. Any fee check can specify a date, as long as its XML schema
supports a date.
Currently conforming to fee extension v12, subject to schema changes.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133420255
Although the delta implies that this is actually adding code, it's
better than it looks, because some of the stuff in ContactFlowUtils
is duplicating more generic methods in ResourceFlowUtils, which
can be deleted when the domain and host flows are cut over.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=133149104
This CL enhances various domain flows (check, create, delete, renew, restore, transfer, update) so that they invoke the appropriate methods on the object implementing the TLD's RegistryExtraFlowLogic (if any). TldSpecificLogicProxy is also updated to invoke RegistryExtraFlowLogic proxy (if any) to fetch the appropriate price. The tests use a made-up extra flow logic object which can be attached to a test TLD to make sure that the proper routines are being invoked.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=132486734
The default production value of 10,000 was unnecessarily large for testing
purposes.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=132441792
When EAP is involed we current have one billing event for domain create that
has the create fee and EAP fee lumped together. Change it to record two
separate billing events for each.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=132335349
It's better that it always takes a non-null ImmutableSet, which may either be
empty or contain elements. That way the ugliness of nullness is contained
just to the entity class itself, and all other code that interacts with it
can always be assured of having a real set to deal with.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=132066238
The backfill for these is no longer necessary as all of the Registry entities in all environments have been backfilled.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=131984138
This change replaces all Ref objects in the code with Key objects. These are
stored in datastore as the same object (raw datastore keys), so this is not
a model change.
Our best practices doc says to use Keys not Refs because:
* The .get() method obscures what's actually going on
- Much harder to visually audit the code for datastore loads
- Hard to distinguish Ref<T> get()'s from Optional get()'s and Supplier get()'s
* Implicit ofy().load() offers much less control
- Antipattern for ultimate goal of making Ofy injectable
- Can't control cache use or batch loading without making ofy() explicit anyway
* Serialization behavior is surprising and could be quite dangerous/incorrect
- Can lead to serialization errors. If it actually worked "as intended",
it would lead to a Ref<> on a serialized object being replaced upon
deserialization with a stale copy of the old value, which could potentially
break all kinds of transactional expectations
* Having both Ref<T> and Key<T> introduces extra boilerplate everywhere
- E.g. helper methods all need to have Ref and Key overloads, or you need to
call .key() to get the Key<T> for every Ref<T> you want to pass in
- Creating a Ref<T> is more cumbersome, since it doesn't have all the create()
overloads that Key<T> has, only create(Key<T>) and create(Entity) - no way to
create directly from kind+ID/name, raw Key, websafe key string, etc.
(Note that Refs are treated specially by Objectify's @Load method and Keys are not;
we don't use that feature, but it is the one advantage Refs have over Keys.)
The direct impetus for this change is that I am trying to audit our use of memcache,
and the implicit .get() calls to datastore were making that very hard.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=131965491
Checks are added to ensure that fees are always non-negative, while credits are always negative, as required by the EPP fee extension specification. Also, BaseFee is made specifically abstract.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=131620110
This is needed for a soon-to-be-submitted CL that changes
all Refs to Keys and therefore removes the logic in
toHydratedString that doesn't expand Keys. We need to be
able to tag types as unexpandable to avoid cycles. It
would be better to tag fields, not types, but that is a
much harder change and not currently needed by the use
case of the following CL, so for now this suffices.
While I am in here, add tests for all of the features
of toHydratedString.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=131618634
According to the ICAAN operation profile:
2.1.7. Registries MUST support lookup for entities with the registrar role within other objects using the handle (as described in 3.1.5 of RFC7482). The handle of the entity with the registrar role MUST be equal to IANA Registrar ID. The entity with the registrar role in the RDAP response MUST contain a publicIDs member to identify the IANA Registrar ID from the IANA’s Registrar ID registry. The type value of the publicID object MUST be equal to IANA Registrar ID.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=130452501
See b/30806813 for more context. Copied from there: This appears to be happening if we get an EPP domain create command that is missing any contacts (but has a registrant; with no registrant we exercise a different codepath). In that case, JAXB leaves the contacts field on the Create null, and we try to pass it into Sets.union() as a result of Corey's refactoring in [] that changed contact loading to load the contacts and registrant all at once. The fix is just to apply nullToEmpty() first.
Note that it's always an error to try to create a domain without any non-registrant contacts, but that's supposed to happen later on in BaseDomainCreateFlow.verifyCreateIsAllowed() via validateRequiredContactsPresent(), which will produce a nice error message.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=130309019
Adds the ability to search for entities (contacts and registrars) by name.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=130305930
