Commit Graph

51 Commits

Author SHA1 Message Date
Alexey Volkov 9a1d47a185 SDK - Capturing function dependencies when creating lightweight components (#1372)
* Transitively capturing code dependencies
Using cloudpickle.

* Got rid of func_type_declarations_code variable

* Extracted the function code extraction functions

* Improved support for capturing module-level dependencies

* Added test for capturing module-level dependencies

* Removed the _capture_function_code_using_source_copy function
As requested by Ning
2019-05-28 18:18:18 -07:00
Alexey Volkov a41bd106a1 SDK - Removing unneeded uses of dsl.Pipeline (#1229)
* SDK - Removing unneeded usages of dsl.Pipeline

* Fixed the naming-related issue
2019-05-14 18:48:18 -07:00
Alexey Volkov b61bef04a3 SDK - Renamed ModelBase.from_struct/to_struct to from_dict/to_dict (#1290) 2019-05-07 14:06:35 -07:00
Alexey Volkov 819d91d2f1 Retaining the component url, digest or tag when loading (#1090) 2019-05-03 16:55:38 -07:00
Alexey Volkov f40a22a3f4 SDK - Made ComponentSpec.implementation field optional (#1188)
* SDK - Made ComponentSpec.implementation field optional
Improved the error message when trying to convert tasks to ContainerOp.

* Switched from attribute checking to type checking
2019-04-24 12:54:46 -07:00
Alexey Volkov 6920aceeba SDK - Removed SourceSpec structure (#1119)
It has never been used and ComponentSpec.metadata.annotations['source'] is a better place for such metadata.
2019-04-24 12:06:26 -07:00
Alexey Volkov 848d4fb99c SDK - Replaced insecure yaml.load with yaml.safe_load (#1170)
This improves security and gets rid of security warnings.
See https://github.com/yaml/pyyaml/wiki/PyYAML-yaml.load(input)-Deprecation
2019-04-23 15:26:00 -07:00
Alexey Volkov 929ff52fd2 Passing the annotations and labels to the ContainerOp (#1077)
Currently the annotations and labels are not passed from component to the ContainerOp. This PR fixes that.

Fixes https://github.com/kubeflow/pipelines/issues/1013
2019-04-08 22:03:05 -07:00
Alexey Volkov 291691a9f9 SDK/Components - Handling public GCS URIs in load_component (#1057) 2019-03-28 15:55:56 -07:00
Eterna2 825f64d672 Feature: sidecar for ContainerOp (#879)
* Feature: sidecar for ContainerOp

* replace f-string with string format for compatibility with py3.5

* ContainerOp now can be updated with any k8s V1Container attributes as well as sidecars with Sidecar class. ContainerOp accepts PipelineParam in any valid k8 properties.

* WIP: fix conflicts and bugs with recent master. TODO: more complex template with pipeline params

* fix proxy args

* Fixed to work with latest master head

* Added container_kwargs to ContainerOp to pass in k8s container kwargs

* Fix comment bug, updated with example in ContainerOp docstring

* fix copyright year

* expose match_serialized_pipelineparam as public for compiler to process serialized pipeline params

* fixed pydoc example and removed unnecessary ContainerOp.container.parent

* Fix conflicts in compiler tests
2019-03-28 11:11:30 -07:00
Alexey Volkov e452385a55 Fixed handling parameters with default values in task factory construction (#1047)
* Fixed handling default inputs in task factory construction

* Added tests.
2019-03-26 19:14:47 -07:00
Ning 1c4f9eb431
exposing type checking (#1022)
* exposing types under dsl.types
2019-03-26 09:33:16 -07:00
Alexey Volkov 9b804688d3 Added the metadata property to ComponentSpec (#1023)
The `metadata` section contains the `annotations` and `labels` dictionaries.
2019-03-23 16:27:05 -07:00
Alexey Volkov 07aa5db70f Fixed bug in docstring construction (#1012) 2019-03-21 14:57:36 -07:00
Alexey Volkov 665d088030 Added the component name to the docstring (#976) 2019-03-19 21:50:24 -07:00
Ning c829115574 Add type check (#938)
* add core types and type checking function

* fix unit test bug

* avoid defining dynamic classes

* typo fix

* add component metadata format

* add a construct for the component decorator

* add default values for the meta classes

* add input/output types to the metadata

* add from_dict in TypeMeta

* small fix

* add unit tests

* use python struct for the openapi schema

* add default in parameter

* add default value

* remove the str restriction for the param default

* bug fix

* add pipelinemeta

* add pipeline metadata

* ignore annotation if it is not str/BaseType/dict

* update param name in the check_type functions
remove schema validators for GCRPath, and adjust for GCRPath, GCSPath
change _check_valid_dict to _check_valid_type_dict to avoid confusion
fix typo in the comments
adjust function order for readability

* remove default values for non-primitive types in the function signature
update the _check_valid_type_dict name

* pass metadata from component decorator and task factory to containerOp

* pass pipeline metadata to Pipeline

* fix unit test

* typo in the comments

* move the metadata classes to a separate module

* fix unit test

* small change

* add __eq__ to meta classes
not export _metadata classes

* nothing

* fix unit test

* unit test python component

* unit test python pipeline

* fix bug: duplicate variable of args

* fix unit tests

* move python_component and _component decorator in _component file

* remove the print

* change parameter default value to None

* add functools wraps around _component decorator

* TypeMeta accept both str and dict

* fix indent, add unit test for type as strings

* do not set default value for the name field in ParameterMeta, ComponentMeta, and PipelineMeta

* add type check in task factory

* output error message

* add type check in component decorator; move the metadata assignment out of the containerop __init__ function

* fix bug; add unit test

* add more unit tests

* more unit tests; fix bugs

* more unit tests; fix bugs

* add unit tests

* more unit tests

* add type check switch; add unit tests

* add compiler option for type check

* resolving pr comments

* add unit test for pipeline param check with component types; fix the bug; also fix the bug when there are not a single return annotations
2019-03-11 11:22:12 -07:00
Alexey Volkov 6d080c70f9
Added support for loading zip-packed components (#931)
The zip-packed components are supported in all load_component APIs:
`kfp.components.load_component`
`kfp.components.load_component_from_file`
`kfp.components.load_component_from_url`
`kfp.components.ComponentStore.load_component`
2019-03-06 23:00:03 -08:00
Alexey Volkov fa02e750da SDK/Components - Added naming.generate_unique_name_conversion_table (#716)
generate_unique_name_conversion_table replaces _make_name_unique_by_adding_index and simplifies code in several places.
2019-03-06 15:12:58 -08:00
Ning 974d602b74
Pass meta to containerop and pipeline (#905)
pass metadata from python conf to containerop and the pipeline
2019-03-06 13:42:23 -08:00
Alexey Volkov 5ab368ac10 Added support for default values to Lightweight python components (#890) 2019-03-01 14:51:18 -08:00
Alexey Volkov f5bdf2474e Added support for default values to load_component (#889) 2019-03-01 14:12:32 -08:00
Alexey Volkov 85738cbaaf Passing the environment variables to ContainerOp (#877)
When the DSL bridge code was written, ContainerOp did not support env, so we did not pass it. Now we're adding the passing code.
Added test that chacks that the env variables get to the ContainerOp.
2019-02-28 19:29:54 -08:00
Alexey Volkov d15c72470f SDK/Components - Improved error when type checking fails in constructor (#732) 2019-01-25 14:44:15 -08:00
Alexey Volkov edf9b5471a SDK/Components - convert_object_to_struct now uses __init__ to get field list (#733)
This stops serialization of any additional attributes set on an object
2019-01-24 20:01:23 -08:00
Alexey Volkov 8c4f5de1f7 SDK/Components - Command line args can only be strings or placeholders (#711)
Ultimately, command line is an array of strings. Component yaml files should have the arguments as strings instead of Python SDK doing conversion sometimes.
2019-01-24 19:13:50 -08:00
Alexey Volkov 4457e7e940 SDK/Components - More meaningful error when trying to convert graph component to ContainerOp (#710) 2019-01-24 18:15:07 -08:00
Alexey Volkov a53cb586fc SDK/Components - Added _naming._convert_to_human_name function (#715)
* SDK/Components - Moved naming-related functions to _naming.py

* SDK/Components - Added _naming._convert_to_human_name function
2019-01-24 16:07:46 -08:00
Alexey Volkov 32475bfafb SDK/Components/Python - Improved Python2 compatibility (#718)
Improved Python2 compatibility in Lightweight python components
2019-01-24 14:42:03 -08:00
Alexey Volkov 9b4088626c SDK/Components/Python - Made the typing.NamedTuple import optional (#717)
Now it's only imported if the return type is NamedTuple.
2019-01-23 16:31:13 -08:00
Alexey Volkov b12d5d8f8e SDK/Components - Added /data to the generated file paths (#663)
This is needed for the future storage system based on volume mounts:
If outputs were written to files in the same dir (e.g. /outputs/out1.txt and /outputs/out2.txt), then we cannot separate them and mount to the downstream task containers independently.
2019-01-15 18:01:40 -08:00
Alexey Volkov fd282d67cd SDK/Components - Simplified _create_task_factory_from_component_spec function (#662) 2019-01-14 18:02:16 -08:00
Alexey Volkov b8f5348fb3 SDK/Components - Renamed serialized command-line argument tokens (#659) 2019-01-10 00:49:37 -08:00
Alexey Volkov 83e9ffe5bc SDK/Components - Reworked the component model structures. (#642)
* Reworked the Component structures.
Rewrote parsing, type checking and serialization code.
Improved the graph component structures.
Added most of the needed k8s structures.
Added model validation (input/output existence etc).
Added task cycle detection and topological sorting to GraphSpec.
All container component tests now work.
Added some graph component tests.

* Fixed incompatibilities with python <3.7

* Added __init__.py to make the Travis tests work.

* Adding kubernetes structures to setup.py

* Addressed PR feedback: Renamed _original_names to _serialized_names

* Addressed PR feedback: Reduced indentation.

* Added descriptions for all component structures.

* Fixed a bug in ComponentSpec._post_init()

* Added documentation for ModelBase class and functions.

* Added __eq__/__ne__ and improved __repr__

* Added ModelBase tests
2019-01-09 15:51:34 -08:00
qimingj 875efea1f9 Support replacable arguments in command as well (besides arguments) in container op. (#623)
* Support replacable arguments in command as well (besides arguments) in container op.

* Fix components builder.

* Fix tests.

* Follow up CR comments.

* Fix test.
2019-01-07 07:57:36 -08:00
Ning 85c6413a2e Refactor Python SDK (#568)
* add some comments

* remove unused import; add license to dsl_bridge

* move_convert_k8s_obj_to_dic from compiler to k8s_helper

* move unit test
2018-12-20 09:51:09 -08:00
Alexey Volkov 17582a8fab SDK/Components - Made component search locations configurable (#519)
* SDK/Components - Made component search locations configurable

* Raise proper error on existing, but malformed components instead of saying the component was not found.
Component name must be non-empty.
Addressed PR feedback.
2018-12-12 17:08:51 -08:00
Alexey Volkov e64a76656b SDK/Components - Do not crash on non-hashable objectsApparently Python's `dict.get` throws exception when it thinks that the object is not suitable for key. (#511) 2018-12-11 00:05:51 -08:00
Alexey Volkov fa1886c39d SDK/Components - Allow dict type spec (#490) 2018-12-08 02:29:37 -08:00
Alexey Volkov 96ec194260 SDK/Components - Removed outputs from task factory function signature (#388)
This realizes the outputs handling roadmap and solves problems with input and output name clashes.
2018-12-03 14:52:32 -08:00
Alexey Volkov e06dc88316 SDK/Components - Renamed container.arguments to container.args (#437)
This aligns us with Kubernetes spec
2018-12-03 11:02:15 -08:00
Alexey Volkov 9110296e57 SDK/Components - Support for optional inputs (#214)
* Renamed "required" to "optional"

* Added support for optional inputs

* Added tests for optional inputs. "If then *" tests now also work.
2018-11-30 13:30:09 -08:00
Alexey Volkov dd0bd45aa3 SDK/Components - Renamed DockerContainer spec to to Container (#323) 2018-11-20 12:47:49 -08:00
Alexey Volkov 6a37388a79 SDK/Components/Python - Removed python_op in favor of python_component (#85)
* SDK/Components/Python - Removed python_op in favor of python_component

* Added test for python_component + func_to_container_op.
* dsl.PythonComponent class is removed, because the metadata is now properly stored in the function object itself.
* build_python_component now optionally accepts base_image and no longer requires the decorator (but can use decorator's metadata).
* Made staging_gcs_path optional since it's only needed when build_image == True
* Added more validation
* Added description and parameter help to python_component

* Fixed the pipeline path string after merge.

* Addressed the PR comments.
Added more detailed explanation of base_image selection to the docstring.

* Removed extra empty lines
2018-11-16 19:32:08 -08:00
Alexey Volkov 0c6fef8870 SDK/Components - Fixes and more tests (#213)
* Fixed string boolean handling in if condition

* Fixed bug in isPresent

* Fixed list expansion when an item expands to a list

* Renamed two tests

* Fixed resolving primitive types (yaml supports and decodes them)

* Added test that checks handling arguments of all yaml types

* Added tests for handling true and false booleand and string literals in conditional expressions
2018-11-15 14:26:15 -08:00
Alexey Volkov 09e4146e8c SDK/Components - Switched the generated inputs/outputs structure from dict to list (#173) 2018-11-10 15:17:26 -08:00
Alexey Volkov 2a7aeee184 SDK/Components - Removed the old argument syntax (#168) 2018-11-10 14:42:56 -08:00
Alexey Volkov 199a962e42 SDK - Relative imports (#156)
Made all SDK import relative so that they files always refer to the sibling files instead of the installed package. This makes debugging and development easier since you can be sure the correct files are used.
2018-11-10 13:56:12 -08:00
Alexey Volkov 06f8821abf Components - Python - Stopped using the Fire library (#40) 2018-11-05 17:58:11 -08:00
Alexey Volkov 5409792183 SDK/Components - Switching to map-based syntax for the arguments. (#29)
* Switching to map-based syntax for the arguments.

The list-based syntax is going to be deprecated.

* Switched to map-style arguments in _func_to_component_spec

* Updated testdata
2018-11-05 14:59:23 -08:00
Alexey Volkov 6f4386884c Components - Removed debug print 2018-11-04 23:40:50 -08:00