Re-add deprecated module 'arrayconnection' for backwards compatibility #45
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
|
One thing to note here is that all of these are available from the top level of graphql_relay. There's actually no good reason to import from deeper in the hierarchy, and I would say its dangerous to do so. I was actually bitten by this as well, but for things that I could not import from the graphql_relay top level, namely types and named tuples and stuff that I need to subclass. I think it makes sense to provide anything one might need from the top level, so nobody needs to import from deeper and won't be dinged by path renaming. |
|
@markedwards Which names are still missing on the top level that would be useful? I want to provide only as much as necessary. |
|
In my case, I hit: I'm sure that's not a comprehensive list though. To be honest, quite a few of those are my fault :-P. But not all of them. |
winged
pushed a commit
to projectcaluma/caluma
that referenced
this pull request
Feb 1, 2022
* feat(form): enable hint messages on questions (#1655) Add a hint text field on all question types except for form, static and action button questions. This functionality requires a db migration. * feat(workflow): implement WorkItem redo pattern (#1656) This commit implements a possibility to redo an already finished WorkItem and all the ones that were finished in between. Closes #1510 * chore: update dependencies (#1660) * chore: cleanup `WorkItem.get_redoable()` * chore: update dependencies * chore: release 7.15 (#1661) * chore: add a code of conduct (#1651) * chore(deps): update graphene-django Update to graphene-django 2.12.1 and fix connection field, schema. * chore(deps): add pytest-xfaillist plugin * chore(deps): switch to graphql 3.0 * chore: fix ResolveInfo fixture * chore(test): ensure name is valid identifier * chore(deps): remove XPASSing tests from xfails * chore(deps): handle enum in JSONValueFilter * chore(deps): remove passing tests from xfails.list * fix(tests): hand over UUID as string * chore(deps): get ordering value from Enum if any * chore(deps): use graphene's updated data types issue #814 * chore(deps): update and sort xfaillist * chore(deps): remove passing test from xfaillist #814 * chore(deps): handle enum in filter #814 * check if these tests are correct if so fixup commit with previous * chore(deps): upgrade pytest-xfaillist * chore(deps): pass kwargs to data source resolver (#1643) * fix(graphene): make more tests work (#1644) * Introduce our own choices field (for now): Graphene seems to already instantiate Enum instances from the request data, which causes the regular DRF choice field to fail, as the enum object isn't handled * Fix a few resolvers which now receive additional (unused by use) parameters * Graphene django v3 (#1645) * fix(datasource): accept new additional parameters from graphene * fix(filters): overload resolve_queryset() to support multi order_by Graphene does not really support order_by with multiple values like we need it. There's a bug in graphene that we need to work around: graphql-python/graphene-django#1280 While we're at it, also clean up some more data. * fix(historical): fix uuid and resolver params (#1646) * fix(historical): fix uuid and resolver params * fix(form): accept new additional resolver args * Graphene django v3 (#1647) * fix(tests): adjust test to new error message style graphene 3.0 returns a text fragment of the offending query * fix(middleware): use new visitor and ast conventions The node types are not CamelCase named, but snake_cased. Also, the operation is an AST object, not a bare string. In queries without variables, the variables MUST NOT be set to an empty string. (None works however...) * fix(cases): various fixes in csae filter and tests * fix(schema): make schema tests work again (sort of) The schema test relies on stringification of the root schema object. This currently fails due to a function within the graphql core package being unable to handle django lazy translation objects. The test setup is rigged such that once upstream fixes the problem, it will trigger a failure here and we can remove the hacky workaround. * fix(tests): more tests fixed with correct (non-uuid) input objects Also remove passing some info objects to schema_executor(), it doesn't accept them anymore (just pass context_value as that's the interesting bit) * fix(tests): stringify localized fields when passing to gql * chore(coverage): improve coverage Most of the lines that were "not covered" will never happen or are error cases such as raising validation errors, or other non-problematic parts. * fix(tests): sort the schema type map (#1648) A sorted type map (as it used to be in older graphene versions) helps better comparing / diffing the schema * Graphene django v3 (#1649) * chore: coverage back to 100% * chore: fix conflict in setup.py * fix(schema): use explicit types for case status and workitem status Graphene would otherwise use an automatic naming scheme that breaks our clients * fix: coverage in new status fields (#1650) Those status fields are only used for the type system, but not as actual input types. Thus, they're never actually instantiated * fix: pin graphql-relay to <3.1.1 the minor version includes deprecation of backwards compatibility measures. See graphql-python/graphql-relay-py#45 Might include other changes that are unrelated to the above that break a number of our tests. * update snapshots for test_question * fix(test): fix UUID * fix(tests): update test_workflow snapshot * fix: remove schema workaround and update snapshot Co-authored-by: luytena <[email protected]> Co-authored-by: Fabio Ambauen <[email protected]> Co-authored-by: David Vogt <[email protected]> Co-authored-by: Christian Zosel <[email protected]> Co-authored-by: Stefan Borer <[email protected]> Co-authored-by: Jean-Louis Fuchs <[email protected]>
winged
pushed a commit
to projectcaluma/caluma
that referenced
this pull request
Feb 17, 2022
* chore(deps): update graphene-django Update to graphene-django 2.12.1 and fix connection field, schema. * chore(deps): add pytest-xfaillist plugin * chore(deps): switch to graphql 3.0 * chore: fix ResolveInfo fixture * chore(test): ensure name is valid identifier * chore(deps): remove XPASSing tests from xfails * chore(deps): handle enum in JSONValueFilter * chore(deps): remove passing tests from xfails.list * fix(tests): hand over UUID as string * chore(deps): get ordering value from Enum if any * chore(deps): use graphene's updated data types issue #814 * chore(deps): update and sort xfaillist * chore(deps): remove passing test from xfaillist #814 * chore(deps): handle enum in filter #814 * check if these tests are correct if so fixup commit with previous * chore(deps): upgrade pytest-xfaillist * chore(deps): pass kwargs to data source resolver (#1643) * fix(graphene): make more tests work (#1644) * Introduce our own choices field (for now): Graphene seems to already instantiate Enum instances from the request data, which causes the regular DRF choice field to fail, as the enum object isn't handled * Fix a few resolvers which now receive additional (unused by use) parameters * Graphene django v3 (#1645) * fix(datasource): accept new additional parameters from graphene * fix(filters): overload resolve_queryset() to support multi order_by Graphene does not really support order_by with multiple values like we need it. There's a bug in graphene that we need to work around: graphql-python/graphene-django#1280 While we're at it, also clean up some more data. * fix(historical): fix uuid and resolver params (#1646) * fix(historical): fix uuid and resolver params * fix(form): accept new additional resolver args * Graphene django v3 (#1647) * fix(tests): adjust test to new error message style graphene 3.0 returns a text fragment of the offending query * fix(middleware): use new visitor and ast conventions The node types are not CamelCase named, but snake_cased. Also, the operation is an AST object, not a bare string. In queries without variables, the variables MUST NOT be set to an empty string. (None works however...) * fix(cases): various fixes in csae filter and tests * fix(schema): make schema tests work again (sort of) The schema test relies on stringification of the root schema object. This currently fails due to a function within the graphql core package being unable to handle django lazy translation objects. The test setup is rigged such that once upstream fixes the problem, it will trigger a failure here and we can remove the hacky workaround. * fix(tests): more tests fixed with correct (non-uuid) input objects Also remove passing some info objects to schema_executor(), it doesn't accept them anymore (just pass context_value as that's the interesting bit) * fix(tests): stringify localized fields when passing to gql * chore(coverage): improve coverage Most of the lines that were "not covered" will never happen or are error cases such as raising validation errors, or other non-problematic parts. * fix(tests): sort the schema type map (#1648) A sorted type map (as it used to be in older graphene versions) helps better comparing / diffing the schema * Graphene django v3 (#1649) * chore: coverage back to 100% * chore: fix conflict in setup.py * fix(schema): use explicit types for case status and workitem status Graphene would otherwise use an automatic naming scheme that breaks our clients * fix: coverage in new status fields (#1650) Those status fields are only used for the type system, but not as actual input types. Thus, they're never actually instantiated * rebase graphene update branch on main (#1664) * feat(form): enable hint messages on questions (#1655) Add a hint text field on all question types except for form, static and action button questions. This functionality requires a db migration. * feat(workflow): implement WorkItem redo pattern (#1656) This commit implements a possibility to redo an already finished WorkItem and all the ones that were finished in between. Closes #1510 * chore: update dependencies (#1660) * chore: cleanup `WorkItem.get_redoable()` * chore: update dependencies * chore: release 7.15 (#1661) * chore: add a code of conduct (#1651) * chore(deps): update graphene-django Update to graphene-django 2.12.1 and fix connection field, schema. * chore(deps): add pytest-xfaillist plugin * chore(deps): switch to graphql 3.0 * chore: fix ResolveInfo fixture * chore(test): ensure name is valid identifier * chore(deps): remove XPASSing tests from xfails * chore(deps): handle enum in JSONValueFilter * chore(deps): remove passing tests from xfails.list * fix(tests): hand over UUID as string * chore(deps): get ordering value from Enum if any * chore(deps): use graphene's updated data types issue #814 * chore(deps): update and sort xfaillist * chore(deps): remove passing test from xfaillist #814 * chore(deps): handle enum in filter #814 * check if these tests are correct if so fixup commit with previous * chore(deps): upgrade pytest-xfaillist * chore(deps): pass kwargs to data source resolver (#1643) * fix(graphene): make more tests work (#1644) * Introduce our own choices field (for now): Graphene seems to already instantiate Enum instances from the request data, which causes the regular DRF choice field to fail, as the enum object isn't handled * Fix a few resolvers which now receive additional (unused by use) parameters * Graphene django v3 (#1645) * fix(datasource): accept new additional parameters from graphene * fix(filters): overload resolve_queryset() to support multi order_by Graphene does not really support order_by with multiple values like we need it. There's a bug in graphene that we need to work around: graphql-python/graphene-django#1280 While we're at it, also clean up some more data. * fix(historical): fix uuid and resolver params (#1646) * fix(historical): fix uuid and resolver params * fix(form): accept new additional resolver args * Graphene django v3 (#1647) * fix(tests): adjust test to new error message style graphene 3.0 returns a text fragment of the offending query * fix(middleware): use new visitor and ast conventions The node types are not CamelCase named, but snake_cased. Also, the operation is an AST object, not a bare string. In queries without variables, the variables MUST NOT be set to an empty string. (None works however...) * fix(cases): various fixes in csae filter and tests * fix(schema): make schema tests work again (sort of) The schema test relies on stringification of the root schema object. This currently fails due to a function within the graphql core package being unable to handle django lazy translation objects. The test setup is rigged such that once upstream fixes the problem, it will trigger a failure here and we can remove the hacky workaround. * fix(tests): more tests fixed with correct (non-uuid) input objects Also remove passing some info objects to schema_executor(), it doesn't accept them anymore (just pass context_value as that's the interesting bit) * fix(tests): stringify localized fields when passing to gql * chore(coverage): improve coverage Most of the lines that were "not covered" will never happen or are error cases such as raising validation errors, or other non-problematic parts. * fix(tests): sort the schema type map (#1648) A sorted type map (as it used to be in older graphene versions) helps better comparing / diffing the schema * Graphene django v3 (#1649) * chore: coverage back to 100% * chore: fix conflict in setup.py * fix(schema): use explicit types for case status and workitem status Graphene would otherwise use an automatic naming scheme that breaks our clients * fix: coverage in new status fields (#1650) Those status fields are only used for the type system, but not as actual input types. Thus, they're never actually instantiated * fix: pin graphql-relay to <3.1.1 the minor version includes deprecation of backwards compatibility measures. See graphql-python/graphql-relay-py#45 Might include other changes that are unrelated to the above that break a number of our tests. * update snapshots for test_question * fix(test): fix UUID * fix(tests): update test_workflow snapshot * fix: remove schema workaround and update snapshot Co-authored-by: luytena <[email protected]> Co-authored-by: Fabio Ambauen <[email protected]> Co-authored-by: David Vogt <[email protected]> Co-authored-by: Christian Zosel <[email protected]> Co-authored-by: Stefan Borer <[email protected]> Co-authored-by: Jean-Louis Fuchs <[email protected]> * chore: cleanup schema test workaround * fix(coverage): raise cov and fix a typo * chore(deps): update graphene-django * chore(deps): Update to graphene-django 2.12.1 and fix connection field, schema. * chore(deps): add pytest-xfaillist plugin * chore(deps): switch to graphql 3.0 * chore: fix ResolveInfo fixture * chore(test): ensure name is valid identifier * chore(deps): handle enum in JSONValueFilter * fix(tests): hand over UUID as string * chore(deps): get ordering value from Enum if any * chore(deps): use graphene's updated data types issue #814 * chore(deps): handle enum in filter #814 * chore(deps): pass kwargs to data source resolver (#1643) * fix(graphene): make more tests work (#1644) * Introduce our own choices field (for now): Graphene seems to already instantiate Enum instances from the request data, which causes the regular DRF choice field to fail, as the enum object isn't handled * Fix a few resolvers which now receive additional (unused by use) parameters * Graphene django v3 (#1645) * fix(datasource): accept new additional parameters from graphene * fix(filters): overload resolve_queryset() to support multi order_by Graphene does not really support order_by with multiple values like we need it. There's a bug in graphene that we need to work around: graphql-python/graphene-django#1280 While we're at it, also clean up some more data. * fix(historical): fix uuid and resolver params (#1646) * fix(historical): fix uuid and resolver params * fix(form): accept new additional resolver args * Graphene django v3 (#1647) * fix(tests): adjust test to new error message style graphene 3.0 returns a text fragment of the offending query * fix(middleware): use new visitor and ast conventions The node types are not CamelCase named, but snake_cased. Also, the operation is an AST object, not a bare string. In queries without variables, the variables MUST NOT be set to an empty string. (None works however...) * fix(cases): various fixes in csae filter and tests * fix(schema): make schema tests work again (sort of) The schema test relies on stringification of the root schema object. This currently fails due to a function within the graphql core package being unable to handle django lazy translation objects. The test setup is rigged such that once upstream fixes the problem, it will trigger a failure here and we can remove the hacky workaround. * fix(tests): more tests fixed with correct (non-uuid) input objects Also remove passing some info objects to schema_executor(), it doesn't accept them anymore (just pass context_value as that's the interesting bit) * fix(tests): stringify localized fields when passing to gql * chore(coverage): improve coverage Most of the lines that were "not covered" will never happen or are error cases such as raising validation errors, or other non-problematic parts. * fix(tests): sort the schema type map (#1648) A sorted type map (as it used to be in older graphene versions) helps better comparing / diffing the schema * Graphene django v3 (#1649) * chore: coverage back to 100% * chore: fix conflict in setup.py * fix(schema): use explicit types for case status and workitem status Graphene would otherwise use an automatic naming scheme that breaks our clients * fix: coverage in new status fields (#1650) Those status fields are only used for the type system, but not as actual input types. Thus, they're never actually instantiated * rebase graphene update branch on main (#1664) * feat(form): enable hint messages on questions (#1655) Add a hint text field on all question types except for form, static and action button questions. This functionality requires a db migration. * feat(workflow): implement WorkItem redo pattern (#1656) This commit implements a possibility to redo an already finished WorkItem and all the ones that were finished in between. Closes #1510 * chore: update dependencies (#1660) * chore: cleanup `WorkItem.get_redoable()` * chore: update dependencies * chore: release 7.15 (#1661) * chore: add a code of conduct (#1651) * chore(deps): update graphene-django Update to graphene-django 2.12.1 and fix connection field, schema. * chore(deps): add pytest-xfaillist plugin * chore(deps): switch to graphql 3.0 * chore: fix ResolveInfo fixture * chore(test): ensure name is valid identifier * chore(deps): remove XPASSing tests from xfails * chore(deps): handle enum in JSONValueFilter * chore(deps): remove passing tests from xfails.list * fix(tests): hand over UUID as string * chore(deps): get ordering value from Enum if any * chore(deps): use graphene's updated data types issue #814 * chore(deps): update and sort xfaillist * chore(deps): remove passing test from xfaillist #814 * chore(deps): handle enum in filter #814 * check if these tests are correct if so fixup commit with previous * chore(deps): upgrade pytest-xfaillist * chore(deps): pass kwargs to data source resolver (#1643) * fix(graphene): make more tests work (#1644) * Introduce our own choices field (for now): Graphene seems to already instantiate Enum instances from the request data, which causes the regular DRF choice field to fail, as the enum object isn't handled * Fix a few resolvers which now receive additional (unused by use) parameters * Graphene django v3 (#1645) * fix(datasource): accept new additional parameters from graphene * fix(filters): overload resolve_queryset() to support multi order_by Graphene does not really support order_by with multiple values like we need it. There's a bug in graphene that we need to work around: graphql-python/graphene-django#1280 While we're at it, also clean up some more data. * fix(historical): fix uuid and resolver params (#1646) * fix(historical): fix uuid and resolver params * fix(form): accept new additional resolver args * Graphene django v3 (#1647) * fix(tests): adjust test to new error message style graphene 3.0 returns a text fragment of the offending query * fix(middleware): use new visitor and ast conventions The node types are not CamelCase named, but snake_cased. Also, the operation is an AST object, not a bare string. In queries without variables, the variables MUST NOT be set to an empty string. (None works however...) * fix(cases): various fixes in csae filter and tests * fix(schema): make schema tests work again (sort of) The schema test relies on stringification of the root schema object. This currently fails due to a function within the graphql core package being unable to handle django lazy translation objects. The test setup is rigged such that once upstream fixes the problem, it will trigger a failure here and we can remove the hacky workaround. * fix(tests): more tests fixed with correct (non-uuid) input objects Also remove passing some info objects to schema_executor(), it doesn't accept them anymore (just pass context_value as that's the interesting bit) * fix(tests): stringify localized fields when passing to gql * chore(coverage): improve coverage Most of the lines that were "not covered" will never happen or are error cases such as raising validation errors, or other non-problematic parts. * fix(tests): sort the schema type map (#1648) A sorted type map (as it used to be in older graphene versions) helps better comparing / diffing the schema * Graphene django v3 (#1649) * chore: coverage back to 100% * chore: fix conflict in setup.py * fix(schema): use explicit types for case status and workitem status Graphene would otherwise use an automatic naming scheme that breaks our clients * fix: coverage in new status fields (#1650) Those status fields are only used for the type system, but not as actual input types. Thus, they're never actually instantiated * fix: pin graphql-relay to <3.1.1 the minor version includes deprecation of backwards compatibility measures. See graphql-python/graphql-relay-py#45 Might include other changes that are unrelated to the above that break a number of our tests. * update snapshots for test_question * fix(test): fix UUID * fix(tests): update test_workflow snapshot * fix: remove schema workaround and update snapshot Co-authored-by: luytena <[email protected]> Co-authored-by: Fabio Ambauen <[email protected]> Co-authored-by: David Vogt <[email protected]> Co-authored-by: Christian Zosel <[email protected]> Co-authored-by: Stefan Borer <[email protected]> Co-authored-by: Jean-Louis Fuchs <[email protected]> * chore: cleanup schema test workaround * feat(form): enable hint messages on questions (#1655) Add a hint text field on all question types except for form, static and action button questions. This functionality requires a db migration. * feat(workflow): implement WorkItem redo pattern (#1656) This commit implements a possibility to redo an already finished WorkItem and all the ones that were finished in between. Closes #1510 * fix(coverage): raise cov and fix a typo * fixme: question snapshot differs * fixme: caluma analytics snapshot differs and tests fail * fix: incomplete merge * chore(analytics): compatibility cleanups Co-authored-by: Stefan Borer <[email protected]> Co-authored-by: Jean-Louis Fuchs <[email protected]> Co-authored-by: David Vogt <[email protected]> Co-authored-by: luytena <[email protected]> Co-authored-by: Fabio Ambauen <[email protected]> Co-authored-by: Christian Zosel <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Adds back the
arrayconnectionmodule which simply exports the names fromarray_connection. This was necessary to maintain backwards compatibility with the previous versions.Note: this should be removed after a major version increment.