Warn when clobbering non-normalized data in the cache. #5833
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Here's an example of one of the warnings: |
benjamn
force-pushed
the
warn-about-replacing-non-normalized-data-in-cache
branch
from
86a2bba to
2fe7b28
Compare
One consequence of #5603 is that replacing non-normalized data in the cache can result in loss of useful data. In almost every case, the right solution is to make sure the data can be normalized, or (if that isn't possible) to define a custom merge function for the replaced field, within the parent TypePolicy. It turns out we can give a very detailed warning in all such situations, so that's what this commit does. Looking at the output for our test suite, every warning is legitimate and worth fixing. I will resolve the warnings and test failures that our test suite generates in subsequent commits.
Replacing a Reference in the cache is always safe, even with non-normalized data, since the Reference refers to data safely stored elsewhere in the cache.
It's much easier for the Policies code to check whether there's a merge function defined for a given parent __typename and field name.
benjamn
force-pushed
the
warn-about-replacing-non-normalized-data-in-cache
branch
from
cbeeafc to
bc2712d
Compare
Warning about data loss is rightfully a concern of the StoreWriter, not Policies. Also, StoreWriter has easy access to policies.hasMergeFunction, which was the main selling point for putting warnAboutDataLoss in the policies.ts module.
benjamn
added a commit
that referenced
this pull request
Jan 25, 2020
Follow-up to #5680. Assuming keyArgs:false just because a merge function (or a read function) was defined means that the following pattern can actually change cache behavior in surprising ways: const cache = new InMemoryCache({ typePolicies: { Person: { fields: { friends: { // This is supposed to be identical to what the cache does by // default, except without warnings about data loss, but it // subtly alters the default argument-based field identity // behavior when arguments are provided to the friends field. merge(_, incoming) { return incoming; }, }, }, }, }, }); Although this merge function mirrors how the cache behaves without a merge function, it has the benefit of making that behavior explicit, thereby silencing the warnings introduced by #5833. In order to recommend this strategy for silencing warnings when a last-write-wins merge strategy is desired, it seems unacceptable for this pattern to have any unexpected consequences beyond silencing the warnings. Thinking about this default keyArgs:false behavior more, disabling field identity with keyArgs:false doesn't really make sense if you have only a merge function, or only a read function, since any fancy argument handling done by one of those functions cannot be balanced in the other direction.
benjamn
added a commit
that referenced
this pull request
Jan 25, 2020
Follow-up to #5680. Assuming keyArgs:false just because a merge function (or a read function) was defined means that the following pattern can actually change cache behavior in surprising ways: const cache = new InMemoryCache({ typePolicies: { Person: { fields: { friends: { // This is supposed to be identical to what the cache does by // default, except without warnings about data loss, but it // subtly alters the default argument-based field identity // behavior when arguments are provided to the friends field. merge(_, incoming) { return incoming; }, }, }, }, }, }); Although this merge function mirrors how the cache behaves without a merge function, it has the benefit of making that behavior explicit, thereby silencing the warnings introduced by #5833. In order to recommend this strategy for silencing warnings when a last-write-wins merge strategy is desired, it seems unacceptable for this pattern to have any unexpected consequences beyond silencing the warnings. Thinking about this default keyArgs:false behavior more, disabling field identity with keyArgs:false doesn't really make sense if you have only a merge function, or only a read function, since any fancy argument handling done by one of those functions cannot be balanced in the other direction.
benjamn
added a commit
that referenced
this pull request
Jan 28, 2020
Follow-up to #5680. Assuming keyArgs:false just because a merge function (or a read function) was defined means that the following pattern can actually change cache behavior in surprising ways: const cache = new InMemoryCache({ typePolicies: { Person: { fields: { friends: { // This is supposed to be identical to what the cache does by // default, except without warnings about data loss, but it // subtly alters the default argument-based field identity // behavior when arguments are provided to the friends field. merge(_, incoming) { return incoming; }, }, }, }, }, }); Although this merge function mirrors how the cache behaves without a merge function, it has the benefit of making that behavior explicit, thereby silencing the warnings introduced by #5833. In order to recommend this strategy for silencing warnings when a last-write-wins merge strategy is desired, it seems unacceptable for this pattern to have any unexpected consequences beyond silencing the warnings. Thinking about this default keyArgs:false behavior more, disabling field identity with keyArgs:false doesn't really make sense if you have only a merge function, or only a read function, since any fancy argument handling done by one of those functions cannot be balanced in the other direction.
Codecov Report
@@ Coverage Diff @@
## master #5833 +/- ##
=======================================
Coverage 94.53% 94.53%
=======================================
Files 91 91
Lines 3679 3698 +19
Branches 897 869 -28
=======================================
+ Hits 3478 3496 +18
Misses 179 179
- Partials 22 23 +1 Continue to review full report at Codecov.
|
hwillson
reviewed
Feb 9, 2020
|
|
||
| For more information about these options, please refer to the documentation: | ||
|
|
||
| * Ensuring entity objects have IDs: https://go.apollo.dev/c/generating-unique-identifiers |
There was a problem hiding this comment.
@benjamn These new links are temporarily redirecting to the longer deploy
preview links (we can update them to point to the proper prod docs location when this PR is merged). I took a guess at the slug names in these links; let me know if you want them changed/shortened further.
benjamn
added a commit
that referenced
this pull request
Jun 1, 2020
> This is a revival of PR #5833, which has accumulated too many merge conflicts over the last few months to be worth rebasing. One consequence of #5603 is that replacing non-normalized data in the cache can result in loss of useful data, which is preferable to mistakenly merging unrelated objects, but not ideal. In almost every case, the right solution is to make sure the data can be normalized, or (if that isn't possible) to define a custom `merge` function for the field in question, within the parent type policy. It turns out we can give a very detailed warning about such situations in development, and that's what this commit does. For example: Cache data may be lost when replacing the d field of a Query object. To address this problem (which is not a bug in Apollo Client), either ensure all objects of type D have IDs, or define a custom merge function for the Query.d field, so InMemoryCache can safely merge these objects: existing: {"__typename":"D","e":4} incoming: {"__typename":"D","h":{"__typename":"H","i":7}} For more information about these options, please refer to the documentation: * Ensuring entity objects have IDs: https://go.apollo.dev/c/generating-unique-identifiers * Defining custom merge functions: https://go.apollo.dev/c/merging-non-normalized-objects Looking at the output for our test suite, every warning seems legitimate and worth fixing. I will resolve the warnings in subsequent commits.
|
Superseded by #6372. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
One consequence of #5603 is that replacing non-normalized data in the cache can result in loss of useful data, which is preferable to mistakenly merging unrelated objects.
In almost every case, the right solution is to make sure the data can be normalized, or (if that isn't possible) to define a custom
mergefunction for the replaced field, within the parentTypePolicy.It turns out we can give a very detailed warning in all such situations, so that's what this commit does. Looking at the output for our test suite, every warning is legitimate and worth fixing. I will resolve the warnings and test failures that our test suite generates in subsequent commits.
TypeName.fieldNamemore than once.