2022-05-26 14:54:30 +03:00
|
|
|
- description: PG add source
|
|
|
|
url: /v1/metadata
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
message: success
|
|
|
|
query:
|
|
|
|
type: pg_add_source
|
|
|
|
args:
|
|
|
|
name: pg1
|
|
|
|
configuration:
|
|
|
|
connection_info:
|
|
|
|
database_url:
|
|
|
|
from_env: HASURA_GRAPHQL_PG_SOURCE_URL_1
|
|
|
|
pool_settings:
|
|
|
|
max_connections: 50
|
|
|
|
idle_timeout: 180
|
|
|
|
retries:
|
|
|
|
customization:
|
2022-05-27 08:55:45 +03:00
|
|
|
naming_convention: graphql-default
|
2022-05-26 14:54:30 +03:00
|
|
|
|
|
|
|
- description: create enum table
|
|
|
|
url: /v1/query
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
result_type: CommandOk
|
|
|
|
result:
|
|
|
|
query:
|
|
|
|
type: run_sql
|
|
|
|
args:
|
|
|
|
source: pg1
|
|
|
|
sql: |
|
|
|
|
CREATE TABLE days_of_the_week(
|
|
|
|
day_name text PRIMARY KEY
|
|
|
|
);
|
|
|
|
INSERT INTO days_of_the_week (day_name)
|
|
|
|
VALUES ('Monday'), ('Tuesday'), ('Wednesday'), ('Thursday'), ('Friday'), ('Saturday'), ('Sunday');
|
2022-07-19 09:55:42 +03:00
|
|
|
|
2022-05-26 14:54:30 +03:00
|
|
|
CREATE TABLE users(
|
|
|
|
id serial PRIMARY KEY,
|
|
|
|
favourite_day text NOT NULL REFERENCES days_of_the_week
|
|
|
|
);
|
|
|
|
INSERT INTO users (favourite_day)
|
|
|
|
VALUES ('Sunday');
|
|
|
|
|
|
|
|
- description: Track tables
|
|
|
|
url: /v1/query
|
|
|
|
status: 200
|
|
|
|
response:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- message: success
|
|
|
|
- message: success
|
2022-05-26 14:54:30 +03:00
|
|
|
query:
|
|
|
|
type: bulk
|
|
|
|
args:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- type: track_table
|
|
|
|
args:
|
|
|
|
table: days_of_the_week
|
|
|
|
source: pg1
|
|
|
|
is_enum: true
|
|
|
|
- type: track_table
|
|
|
|
args:
|
|
|
|
source: pg1
|
|
|
|
table: users
|
2022-05-26 14:54:30 +03:00
|
|
|
|
|
|
|
- description: Check the enum values (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
__type:
|
|
|
|
enumValues:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- name: FRIDAY
|
|
|
|
- name: MONDAY
|
|
|
|
- name: SATURDAY
|
|
|
|
- name: SUNDAY
|
|
|
|
- name: THURSDAY
|
|
|
|
- name: TUESDAY
|
|
|
|
- name: WEDNESDAY
|
2022-05-26 14:54:30 +03:00
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
{
|
|
|
|
__type(name: "DaysOfTheWeekEnum") {
|
|
|
|
enumValues {
|
|
|
|
name
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-07-19 09:55:42 +03:00
|
|
|
- description: Insert users returning enum values (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
insertUsers:
|
|
|
|
returning:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- id: 2
|
|
|
|
favouriteDay: SUNDAY
|
|
|
|
- id: 3
|
|
|
|
favouriteDay: FRIDAY
|
|
|
|
- id: 4
|
|
|
|
favouriteDay: MONDAY
|
|
|
|
- id: 5
|
|
|
|
favouriteDay: SATURDAY
|
2022-07-19 09:55:42 +03:00
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation {
|
|
|
|
insertUsers(objects: [{favouriteDay: SUNDAY},{favouriteDay: FRIDAY},{favouriteDay: MONDAY},{favouriteDay: SATURDAY}]) {
|
|
|
|
returning {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
- description: Update users returning enum values (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
updateUsers:
|
|
|
|
returning:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- id: 1
|
|
|
|
favouriteDay: MONDAY
|
|
|
|
- id: 2
|
|
|
|
favouriteDay: MONDAY
|
2022-07-19 09:55:42 +03:00
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation {
|
|
|
|
updateUsers(where: {favouriteDay: {_eq: SUNDAY}}, _set: {favouriteDay: MONDAY}) {
|
|
|
|
returning {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
- description: Delete users returning enum values (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
deleteUsers:
|
|
|
|
returning:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- id: 4
|
|
|
|
favouriteDay: MONDAY
|
|
|
|
- id: 1
|
|
|
|
favouriteDay: MONDAY
|
|
|
|
- id: 2
|
|
|
|
favouriteDay: MONDAY
|
2022-07-19 09:55:42 +03:00
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation MyMutation {
|
|
|
|
deleteUsers(where: {favouriteDay: {_eq: MONDAY}}) {
|
|
|
|
returning {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
- description: Insert user by pk returning enum value (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
insertUsersOne:
|
|
|
|
id: 6
|
|
|
|
favouriteDay: SUNDAY
|
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation {
|
|
|
|
insertUsersOne(object: {favouriteDay: SUNDAY}) {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
- description: Update user by pk returning enum value (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
updateUsersByPk:
|
|
|
|
id: 3
|
|
|
|
favouriteDay: WEDNESDAY
|
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation {
|
|
|
|
updateUsersByPk(pk_columns: {id: 3}, _set: {favouriteDay: WEDNESDAY}) {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
- description: Delete user by pk returning enum value (should be uppercased)
|
|
|
|
url: /v1/graphql
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
data:
|
|
|
|
deleteUsersByPk:
|
|
|
|
id: 5
|
|
|
|
favouriteDay: SATURDAY
|
|
|
|
|
|
|
|
query:
|
|
|
|
query: |
|
|
|
|
mutation MyMutation {
|
|
|
|
deleteUsersByPk(id: 5) {
|
|
|
|
id
|
|
|
|
favouriteDay
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-05-26 14:54:30 +03:00
|
|
|
- description: Untrack tables
|
|
|
|
url: /v1/metadata
|
|
|
|
status: 200
|
|
|
|
response:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- message: success
|
|
|
|
- message: success
|
2022-05-26 14:54:30 +03:00
|
|
|
query:
|
|
|
|
type: bulk
|
|
|
|
args:
|
server/tests-py: Split the naming convention tests.
This splits two naming convention tests into four classes (and four YAML
files), which might seem overkill, but allows us to provision sources
declaratively in the future. As each class will require a custom source
configuration, we are able to annotate them accordingly, which means the
test cases are decoupled from the source database URL, letting us
generate a new database for each test case and automatically add it as a
source to HGE.
The future changes are already prepared, but this has been extracted out
as it splits the YAML files, which is a large change best reviewed in
isolation.
The test case `test_type_and_field_names` has been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefault`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefault`
The test case `test_type_and_field_names_with_prefix_and_suffix` has
been split into:
* `TestNamingConventionsTypeAndFieldNamesGraphqlDefaultWithPrefixAndSuffix`
* `TestNamingConventionsTypeAndFieldNamesHasuraDefaultWithPrefixAndSuffix`
The YAML files have been split in the same way. This was fairly trivial
as each test case would add a source, run some tests with
the `graphql_default` naming convention, drop the source, and then
repeat for the `hasura_default` naming convention. I simply split the
file in two. There is a little bit of duplication for provisioning the
various database tables, which I think is worth it.
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/5496
GitOrigin-RevId: 94825e755c427a5414230f69985b534991b3aad6
2022-08-17 07:35:03 +03:00
|
|
|
- type: pg_untrack_table
|
|
|
|
args:
|
|
|
|
table: days_of_the_week
|
|
|
|
source: pg1
|
|
|
|
- type: pg_untrack_table
|
|
|
|
args:
|
|
|
|
source: pg1
|
|
|
|
table: users
|
2022-05-26 14:54:30 +03:00
|
|
|
|
|
|
|
- description: drop table
|
|
|
|
url: /v1/query
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
result_type: CommandOk
|
|
|
|
result:
|
|
|
|
query:
|
|
|
|
type: run_sql
|
|
|
|
args:
|
|
|
|
source: pg1
|
|
|
|
sql: |
|
|
|
|
drop table users;
|
|
|
|
drop table days_of_the_week;
|
|
|
|
|
|
|
|
- description: PG Drop Source 1
|
|
|
|
url: /v1/metadata
|
|
|
|
status: 200
|
|
|
|
response:
|
|
|
|
message: success
|
|
|
|
query:
|
|
|
|
type: pg_drop_source
|
|
|
|
args:
|
|
|
|
name: pg1
|