The previous check was too conservative and acquired a lock on the
schema cache in situations where it was unnecessary. This change
exposes the logic run_sql uses to determine whether to use the
metadata check to make the check more precise.
* Update graphql-parser-hs and hence use `Scientific` directly
The new version of graphql-parser-hs returns Scientific and Integer
rather than Double and Int32, respectively. So we now need to do less
work in graphql-engine, and we can process larger numbers.
In practice, this means that when inserting a bigint, we no longer
need to specify the inserted integer as text. This is also
represented in the updated tests.
* Generate int overflow error on insert
* Document bigint insertion support in changelog
* allow underscore prefix and special characters in json path
* server: Rewrite/refactor JSONPath parser
The JSONPath parser is also rewritten, the previous implementation
was written in a very explicitly “recursive descent” style, but the whole
point of using attoparsec is to be able to backtrack! Taking advantage
of the combinators makes for a much simpler parser.
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
Co-authored-by: Alexis King <lexi.lambda@gmail.com>
Co-authored-by: Aleksandra Sikora <ola.zxcvbnm@gmail.com>
Co-authored-by: Shahidh K Muhammed <shahidh@hasura.io>
* Allow `_inc` to update other numeric types in addition to integers
* Add support for PostgreSQL's `money` field type
* Add support for _inc on money types
* Add note of generalized `_inc` support to changelog
* add support for action queries
* a new parameter `type` is added in the ArgumentDefinition, its value
can be either `query` or `mutation` and it defaults to the latter
* throw 400 when a query action is tried to explain
* update the actions docs to include query actions
* refactor the ToJSON and ToOrdJSON of ActionDefinition
Co-authored-by: Rishichandra Wawhal <rishi@hasura.io>
Co-authored-by: Tirumarai Selvan <tiru@hasura.io>
* add new optional field `claims_namespace_path` in JWT config
* return value when empty array is found in executeJSONPath
* update the docs related to claims_namespace_path
* improve encodeJSONPath, add property tests for parseJSONPath
* throw error if both claims_namespace_path and claims_namespace are set
* refactor the Data.Parser.JsonPath to Data.Parser.JSONPathSpec
* update the JWT docs
Co-Authored-By: Marion Schleifer <marion@hasura.io>
Co-authored-by: Marion Schleifer <marion@hasura.io>
Co-authored-by: rakeshkky <12475069+rakeshkky@users.noreply.github.com>
Co-authored-by: Tirumarai Selvan <tirumarai.selvan@gmail.com>
* allow re-using Postgres scalars in custom types, close#4125
* add pytest tests
* update CHANGELOG.md
* add a doc pointer for reusable postgres scalars
* document the code, improve the CHANGELOG entry
As suggested by @lexi-lambda
* a bit more source code documentation, use WriterT to collect reused scalars
* Apply suggestions from code review
Co-Authored-By: Marion Schleifer <marion@hasura.io>
* improve doc for Postgres scalars in custom graphql types
* Add some more references to Note; fix Haddock syntax
Also a few very minor tweaks:
* Use HashSet instead of [] more pervasively
* Export execWriterT from Hasura.Prelude
* Use pattern guards in multi-way if
* Tweak a few names/comments
* Pull buildActions out of buildAndCollectInfo, use buildInfoMap
* Tweak wording in documentation
* incorporate changes in console code
* account Postgres scalars for action input arguments
-> Avoid unnecessary 'throw500' in making action schema
* Review changes
Co-authored-by: Marion Schleifer <marion@hasura.io>
Co-authored-by: Alexis King <lexi.lambda@gmail.com>
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
Co-authored-by: Aleksandra Sikora <ola.zxcvbnm@gmail.com>
* unlock the locked-events during graceful shutdown
* Some events can still be delivered multiple times due to
ungraceful shutdown
* modify the preparevents documentation
* modify the prepareEvents doc
* update changelog
Co-authored-by: Tirumarai Selvan <tiru@hasura.io>
* type is not required for jwk_url
* remove type from JWTConfig
* Omit type field in JWTConfig serialization if jwk_url is provided
* remove type from jwk_url test suite
* add changelog
* fix docs with new format
Co-authored-by: Alexis King <lexi.lambda@gmail.com>
* add expiry time to webhook user info
This also adds an optional message to webhook errors: if we fail to
parse an expiry time, we will log a warning with the parse error.
* refactored Auth
This change had one main goal: put in common all expiry time
extraction code between the JWT and WebHook parts of the
code. Furthermore, this change also moves all WebHook specific code to
its own module, similarly to what is done for JWT.
* Remove dependency on string-conversions in favor of text-conversions
string-conversions silently uses UTF8 instead of being explicit about
it, and it uses lenientDecode when decoding ByteStrings when it’s
usually better to reject invalid UTF8 input outright. text-conversions
solves both those problems.
Co-authored-by: Alexis King <lexi.lambda@gmail.com>
* do not perform the metadata check in read-only mode
* improve the isAltrDropReplace regex
* quote the regex at compile-time to handle syntax errors statically
Co-authored-by: Alexis King <lexi.lambda@gmail.com>
* add new column "pg_version" while sending telemetry data
* make a new type for PGVersion and use serverVersion func
* define runTxIO action to run transaction(which exits on error)
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
* Improve performance of replace_metadata on tracking tables. Closes#3802
On the 1000 table track case this went from >20min to 8 seconds, the
bottleneck being all the former calls (for each table) to
buildSchemaCache.
* refactor replace metadata
Only save metadata in hdb_catalog schema and build schema cache (strictly)
* remove `withBuildSchemaCache` function
Co-authored-by: rakeshkky <12475069+rakeshkky@users.noreply.github.com>
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
- The metrics will include
- no of synchronous actions
- no of asynchronous actions
- no of type relationships with the output
- no of custom types defined
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
* add 'ID' to default scalars for custom types, fix#4061
* preserve cookie headers from sync action webhook, close#4021
* validate action webhook response to conform to output type, fix#3977
* fix tests, don't run actions' tests on PG version < 10
* update CHANGELOG.md
* no-op refactor, use types from http-network more
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
Writing to a mutable var is a particularly potent source of leaks since
it mostly defeats GHC's analysis. Here we add assertions to all mutable
writes, and fix a couple spots where we wrote some thunks to a mutable
var (compiled with -O2).
Some of these thunks were probably benign, but others looked liked they
might be retaining big args. Didn't do much analysis, just fixed.
Actually pretty happy with how easy this was to use and as a diagnostic,
once I sorted out some issues. We should consider using it elsewhere,
and maybe extending so that we can use it with tests, enable when
`-fenable-assertsions` etc.
Relates #3388
Also simplified codepaths that use `AcceptWith`, which has unnecessary
`Maybe` fields.
We’re lucky that this never bit us. For the most part, these rules
aren’t actually used; most code programs against ArrowCache and doesn’t
get specialized enough for these rules to fire.
Even if we did have code that could trigger this rule, the situations
where it would actually fire are slim. In order for the rule to
typecheck at all, both sides of the pair being passed through the arrow
must have exactly the same type. Of course, that would just make this
even more hellish to debug.
Rewrite rules are dangerous.
* Test working through a backlog of change events
* Use a slightly more performant threaded http server in eventing pytests
This helped locally but not on CI it seems...
* Rework event processing for backpressure. Closes#3839
With loo low `HASURA_GRAPHQL_EVENTS_FETCH_INTERVAL` and/or slow webhooks
and/or too small `HASURA_GRAPHQL_EVENTS_HTTP_POOL_SIZE` we might
previously check out events from the DB faster than we can service them,
leading to space leaks, weirdness, etc.
Other changes:
- avoid fetch interval sleep latency when we previously did a non-empty
fetch
- prefetch event batch while http pool is working
- warn when it appears we can't keep up with events being generated
- make some effort to process events in creation order so we don't
starve older ones.
ALSO NOTE: HASURA_GRAPHQL_EVENTS_FETCH_INTERVAL changes semantics
slightly, since it only comes into play after an empty fetch. The old
semantics weren't documented in detail, so I think this is fine.
This is the result of a general audit of how we fork threads, with a
detour into how we're using mutable state especially in websocket
codepaths, making more robust to async exceptions and exceptions
resulting from bugs.
Some highlights:
- use a wrapper around 'immortal' so threads that die due to bugs are
restarted, and log the error
- use 'withAsync' some places
- use bracket a few places where we might break invariants
- log some codepaths that represent bugs
- export UnstructuredLog for ad hoc logging (the alternative is we
continue not logging useful stuff)
I had to timebox this. There are a few TODOs I didn't want to address.
And we'll wait until this is merged to attempt #3705 for
Control.Concurrent.Extended
* basic doc for actions
* custom_types, sync and async actions
* switch to graphql-parser-hs on github
* update docs
* metadata import/export
* webhook calls are now supported
* relationships in sync actions
* initialise.sql is now in sync with the migration file
* fix metadata tests
* allow specifying arguments of actions
* fix blacklist check on check_build_worthiness job
* track custom_types and actions related tables
* handlers are now triggered on async actions
* default to pgjson unless a field is involved in relationships, for generating definition list
* use 'true' for action filter for non admin role
* fix create_action_permission sql query
* drop permissions when dropping an action
* add a hdb_role view (and relationships) to fetch all roles in the system
* rename 'webhook' key in action definition to 'handler'
* allow templating actions wehook URLs with env vars
* add 'update_action' /v1/query type
* allow forwarding client headers by setting `forward_client_headers` in action definition
* add 'headers' configuration in action definition
* handle webhook error response based on status codes
* support array relationships for custom types
* implement single row mutation, see https://github.com/hasura/graphql-engine/issues/3731
* single row mutation: rename 'pk_columns' -> 'columns' and no-op refactor
* use top level primary key inputs for delete_by_pk & account select permissions for single row mutations
* use only REST semantics to resolve the webhook response
* use 'pk_columns' instead of 'columns' for update_by_pk input
* add python basic tests for single row mutations
* add action context (name) in webhook payload
* Async action response is accessible for non admin roles only if
the request session vars equals to action's
* clean nulls, empty arrays for actions, custom types in export metadata
* async action mutation returns only the UUID of the action
* unit tests for URL template parser
* Basic sync actions python tests
* fix output in async query & add async tests
* add admin secret header in async actions python test
* document async action architecture in Resolve/Action.hs file
* support actions returning array of objects
* tests for list type response actions
* update docs with actions and custom types metadata API reference
* update actions python tests as per #f8e1330
Co-authored-by: Tirumarai Selvan <tirumarai.selvan@gmail.com>
Co-authored-by: Aravind Shankar <face11301@gmail.com>
Co-authored-by: Rakesh Emmadi <12475069+rakeshkky@users.noreply.github.com>
* Add check expresion to update permissions (close#384)
* wip on conflict behavior
* Handle upserts for views properly
* Use insert check if there is no update check
* Fix the test
* Improve error message slightly
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
* Add downgrade command
* Add docs per @lexi-lambda's suggestions
* make tests pass
* Update hdb_version once, from Haskell
* more work based on feedback
* Improve the usage message
* Small docs changes
* Test downgrades exist for each tag
* Update downgrading.rst
* Use git-log to find tags which are ancestors of the current commit
Co-authored-by: Vamshi Surabhi <0x777@users.noreply.github.com>
* fix nested insert with returning computed fields gives error, fix#3609
* revert using ordered hashmaps, sort columns based on ordinal postion
* fix 1. keys order 2. json/jsonb column value in nested insert returning
* add a note for sorted columns
* cast 'VALUES' expression as table row type
* use single CTE expression for generating returning for nested inserts
We upload a set of accumulating timers and counters to track service
time for different types of operations, across several dimensions (e.g.
did we hit the plan cache, was a remote involved, etc.)
Also...
Standardize on DiffTime as a standard duration type, and try to use it
consistently.
See discussion here:
https://github.com/hasura/graphql-engine/pull/3584#pullrequestreview-340679369
It should be possible to overwrite that module so the new threadDelay
sticks per the pattern in #3705 blocked on #3558
Rename the Control.Concurrent.Extended.threadDelay to `sleep` since a
naive use with a literal argument would be very bad!
We catch a bug in 'computeTimeDiff'.
Add convenient 'Read' instances to the time unit utility types. Make
'Second' a newtype to support this.
This fixes#3759. Also, while we’re at it, also improve the way
invalidations are synced across instances so enums and remote schemas
are appropriately reloaded by the schema syncing process.
* WIP: Remove hdb_views for inserts
* Show failing row in check constraint error
* Revert "Show failing row in check constraint error"
This reverts commit dd2cac29d0.
* Use the better query plan
* Simplify things
* fix cli test
* Update downgrading.rst
* remove 1.1 asset for cli
- Move MonadBase/MonadBaseControl instances for TxE into pg-client-hs
- Set the -qn2 RTS option by default to limit the parallel GC to 2
threads
- Remove eventlog instrumentation
- Don’t rebuild the schema cache again after running a query that needs
it to be rebuilt, since we do that explicitly now.
- Remove some redundant checks, and relocate a couple others.
This changes TableCoreCacheT to internally record dependencies at a
per-table level. In practice, this dramatically improves the performance
of building permissions: it makes it far, far less likely for
permissions to be needlessly rebuilt because some unrelated table
changed.
These aren't suitable e.g. for running in CI since some take far too
long (and an impossibly long-time when running under criterion's normal
bootstrapping sampling regime.
We might try to improve this ourselves:
https://github.com/bos/criterion/issues/218
An initial summary analysis will be in #3530.
* export metadata without nulls, empty arrays
* property tests for 'ReplaceMetadata' using QuickCheck
-> Derive Arbitrary class for 'ReplaceMetadata' dependant types
* reduce property test cases number to 30
QuickCheck generates the `ReplaceMetadata` value really large
for higher number test cases. Encoded JSON for such values is large and
consumes more memory. Thus, CI is giving up while running property
tests.
* circle-ci: Add property tests as saperate job
* add no command mode to tests
* add yaml.v2 to go mod
* remove indirect comment for yaml.v2 dependency
The connection handler in websocket transport was not using the
'UserAuthentication' interface to resolve user info. Fix resolving
user info in websocket transport to use the common
'UserAuthentication' interface
Instead of
'WITH some_alias (SELECT * from some_func()) SELECT <rows> FROM some_alias'
for SQL function queries, Use
'SELECT <rows> FROM some_func() AS some_alias'
* save permissions, relationships and collections in catalog with 'is_system_defined'
* Use common stanzas in the .cabal file
* Refactor migration code into lib instead of exe
* Add new server test suite that exercises migrations
* Make graphql-engine clean succeed even if the schema does not exist
This fix is a little ugly, but it’s the only simple solution without a
significant refactoring that restructures the relationship between
GraphQL/Validate and GraphQL/Resolve. The ugliness should go away if we
implement something like #2801.
* Separate DB and metadata migrations
* Refactor Migrate.hs to generate list of migrations at compile-time
* Replace ginger with shakespeare to improve performance
* Improve migration log messages
Although brotli itself is MIT-licensed, the Haskell brotli library that provides bindings to it is GPL-licensed, so we cannot use it unless we get a response on haskell-hvr/brotli#1.
* allow customizing GraphQL root field names, close#981
* document v2 track_table API in reference
* support customising column field names in GraphQL schema
* [docs] add custom column fields doc in API reference
* add tests
* rename 'ColField' to 'ColumnField'
* embed column's graphql field in 'PGColumnInfo'
-> Value constructor of 'PGCol' is not exposed
-> Using 'parseJSON' to construct 'PGCol' in 'FromJSON' instances
* avoid using 'Maybe TableConfig'
* refactors & 'custom_column_fields' -> 'custom_column_names'
* cli-test: add configuration field in metadata export test
* update expected keys in `FromJSON` instance of `TableMeta`
* use `buildSchemaCacheFor` to update configuration in v2 track_table
* remove 'GraphQLName' type and use 'isValidName' exposed from parser lib
* point graphql-parser-hs library git repo to hasura
* support 'set_table_custom_fields' query API & added docs and tests
This fixes an issue where queries could incorrectly be considered
reusable if a variable was used in two positions: one where it affected
SQL generation and one where it did not.
* initial raster support
* _st_intersects_geom -> _st_intersects_geom_nband
* add tests
* update docs
* improve docs
As requested by @marionschleifer
* new type for raster values
Suggested by @lexi-lambda
* replace `SEUnsafe "NULL"` with SENull
* use positional arguments in SQL functions
* only allow omitting set of last arguments in functions
* disallow omitting of a non default argument in functions
These changes also add a new type, PGColumnType, between PGColInfo and
PGScalarType, and they process PGRawColumnType values into PGColumnType
values during schema cache generation.
This mostly simplifies the RootFlds type to make it clearer what it’s
used for, but it has the convenient side-effect of preventing some
“impossible” cases using the type system.
* allow altering type of a column iff session vars are defined in permissions
* use a sum type to define dependency reason
* set jwt expiry test's expiry time to 4 seconds
* derive Data instance for necessary types to simplify 'hasStaticExp'
query templates is a little known feature that lets you template rql
queries and serve them as rest apis. This is not relevant anymore
given the GraphQL interface and getting rid of it reduces the dev
time when adding features in few subsystems.
This feature has never been used outside hasura's internal projects or
documented or exposed through console and hence can safely be removed.
* fix bug in audience check while verifying JWT
- previously the check was converting the audience type into a string
and then comparing with the conf value. all audience types (as it is a
string or URI) will convert to plain strings
- use the Audience type from the jose library for comparing
* add docs for audience
* add issuer check as well
* docs minor syntax fix
* skip audience check if not given in conf
* minor docs update
* qualify import jose library
Currently, we allow tracking of a table with the same name as an already tracked function, and vice-versa. This causes an issue when querying from GraphQL since it will only query the table and not the function. I've made changes to disallow this by throwing an error.
* Bump node-sass to version ^4.12.0
This avoids sass/node-sass#2632, which causes compilation failures on
Node v12.x.
* Do most of the work in /pg_dump in Haskell instead of shell
The shell version caused problems on non-Linux systems since it relied
on the non-POSIX -i option for sed, which works slightly differently on
BSD and macOS.
This PR builds console static assets into the server docker image at `/srv/console-assets`. When env var `HASURA_GRAPHQL_CONSOLE_ASSETS_DIR=/srv/console-assets` or flag `--console-assets-dir=/srv/console-assets` is set on the server, the files in this directory are served at `/console/assets/*`.
The console html template will have a variable called `cdnAssets: false` when this flag is set and it loads assets from server itself instead of CDN.
The assets are moved to a new bucket with a new naming scheme:
```
graphql-engine-cdn.hasura.io/console/assets/
/common/{}
/versioned/<version/{}
/channel/<channel>/<version>/{}
```
Console served by CLI will still load assets from CDN - will fix that in the next release.
Changes compared to `/v1alpha1/graphql`
* Changed all graphql responses in **/v1/graphql** endpoint to be 200. All graphql clients expect responses to be HTTP 200. Non-200 responses are considered transport layer errors.
* Errors in http and websocket layer are now consistent and have similar structure.
1. Reuses postgres connections during startup which reduces the overhead of opening and closing connections.
2. Faster schema cache building. This is done by fetching all the required data in a single sql statement.
* build schema cache function without db setup
The setup shouldn't happen for sync. The database is already setup by the instance which generated the event. This means that the sync is now faster.
* use SQL loop to drop hdb_views schema views and routines with ordering
This avoids deadlocks when schema is being changed concurrently
* schema sync now only processes the latest event
This becomes useful when a lot of schema change
events happen while we are still processing an
earlier event.
* split stm transactions when snapshotting to make it faster
* mx subs: push to both old and new sinks at the same time
* expose dev APIs through allowed APIs flag
* add types to represent unparsed http gql requests
This will help when we add caching of frequently used ASTs
* query plan caching
* move livequery to execute
* add multiplexed module
* session variable can be customised depending on the context
Previously the value was always "current_setting('hasura.user')"
* get rid of typemap requirement in reusable plan
* subscriptions are multiplexed when possible
* use lazytx for introspection to avoid acquiring a pg connection
* refactor to make execute a completely decoupled module
* don't issue a transaction for a query
* don't use current setting for explained sql
* move postgres related types to a different module
* validate variableValues on postgres before multiplexing subs
* don't user current_setting for queries over ws
* plan_cache is only visible when developer flag is enabled
* introduce 'batch size' when multiplexing subscriptions
* bump stackage to 13.16
* fix schema_stitching test case error code
* store hashes instead of actual responses for subscriptions
* internal api to dump subscriptions state
* remove PlanCache from SchemaCacheRef
* allow live query options to be configured on server startup
* capture metrics for multiplexed subscriptions
* more metrics captured for multiplexed subs
* switch to tvar based hashmap for faster snapshotting
* livequery modules do not expose internal details
* fix typo in live query env vars
* switch to hasura's pg-client-hs
* read version from env var at build time (close#1398)
* remove un-used imports, edit makefile
* edit makefile to add new targets and export variables
* only export VERSION in makefile
* read version by executing the script if env var is absent
From `alpha-40` we've been using a `WHERE` clause to fetch required rows and generate mutation response. This has a few limitations like the requirement of a primary key/unique constraint. This also returns inconsistent data on `delete` mutation as mentioned in #1794.
Now, we're using `VALUES (..)` (refer [here](https://www.postgresql.org/docs/current/sql-values.html)) expression to form virtual table rows in `SQL` to generate mutation response.
Internal changes:-
- Not to use primary key/unique constraint columns:-
- Revert back to `ConstraintName` from `TableConstraint` in `TableInfo` type
- Remove `tcCols` field in `TableConstraint` type
- Modify `table_info.sql` and `fetchTableMeta` function `SQL`
- A test case to perform `delete` mutation and returning relational objects.
1. Haskel library `pg-client-hs` has been updated to expose a function that helps listen to `postgres` notifications over a `channel` in this [PR](https://github.com/hasura/pg-client-hs/pull/5)
2. The server records an event in a table `hdb_catalog.hdb_cache_update_event` whenever any `/v1/query` (that changes metadata) is requested. A trigger notifies a `cache update` event via `hasura_cache_update` channel
3. The server runs two concurrent threads namely `listener` and `processor`. The `listener` thread listens to events on `hasura_cache_update` channel and pushed into a `Queue`. The `processor` thread fetches events from that `Queue` and processes it. Thus server rebuilds schema cache from database and updates.
If returning field contains nested selections then mutation is performed in two steps
1. Mutation is performed with returning columns of any primary key and unique constraints
2. returning fields are queried on rows returned by selecting from table by filtering with column values returned in Step 1.
Since mutation takes two courses based on selecting relations in returning field, it is hard to maintain sequence of prepared arguments (PrepArg) generated while resolving returning field. So, we're using txtConverter instead of prepare to resolve mutation fields.
* read cookie while initialising websocket connection (fix#1660)
* add tests for cookie on websocket init
* fix logic for tests
* enforce cors, and flag to force read cookie when cors disabled
- as browsers don't enforce SOP on websockets, we enforce CORS policy
on websocket handshake
- if CORS is disabled, by default cookie is not read (because XSS
risk!). Add special flag to force override this behaviour
* add log and forward origin header to webhook
- add log notice when cors is disabled, and cookie is not read on
websocket handshake
- forward origin header to webhook in POST mode. So that when CORS is
disabled, webhook can also enforce CORS independently.
* add docs, and forward all client headers to webhook
Rename the admin secret key header used to access GraphQL engine from X-Hasura-Access-Key to X-Hasura-Admin-Secret.
Server CLI and console all support the older flag but marks it as deprecated.
* support jsonb and geometry operators on RQL bool exps, close#1408
* add tests for jsonb operators in /v1/query
TODO:-
-> add tests for geometry (postgis) operators
* support parsing session variables for st_d_within and has_key ops
-> Add tests for boolExp operators and select permissions
* improve parsing $st_d_within op's json value logic
