Fixes the names and tags issue for atlas

MANIFEST.in

@@ -0,0 +1 @@
+include requirements.txt

README.md

@@ -47,6 +47,16 @@ $ docker run -p 5000:5000 amundsendev/amundsen-metadata
 $ curl -v http://localhost:5000/healthcheck
 ```
 
+## Instructions to start the service from Docker image with gunicorn (production use case)
+Note that there below command uses default config of gunicorn. Please visit [Gunicorn homepage](https://gunicorn.org/ "Gunicorn") for more information.
+```bash
+$ docker pull amundsendev/amundsen-metadata:latest
+$ docker run -p 5000:5000 amundsendev/amundsen-metadata gunicorn --bind 0.0.0.0:5000 metadata_service.metadata_wsgi
+
+-- In different terminal, verify getting HTTP/1.0 200 OK
+$ curl -v http://localhost:5000/healthcheck
+```
+
 ## Production environment
 By default, Flask comes with Werkzeug webserver, which is for development. For production environment use production grade web server such as [Gunicorn](https://gunicorn.org/ "Gunicorn").
 

metadata_service/config.py

@@ -34,11 +34,6 @@ class Config:
     # The relationalAttribute name of Atlas Entity that identifies the database entity.
     ATLAS_DB_ATTRIBUTE = 'db'
 
-    # FixMe: Once GUIDs are in place, then change it to 'name' (display name)
-    # Display name of Atlas Entities that we use for amundsen project.
-    # Atlas uses qualifiedName as indexed attribute. but also supports 'name' attribute.
-    ATLAS_NAME_ATTRIBUTE = 'qualifiedName'
-
 
 class LocalConfig(Config):
     DEBUG = False

metadata_service/entity/popular_table.py

@@ -1,21 +1,19 @@
 from typing import Optional
 
+import attr
+from marshmallow_annotations.ext.attrs import AttrsSchema
 
+
+@attr.s(auto_attribs=True, kw_only=True)
 class PopularTable:
+    database: str = attr.ib()
+    cluster: str = attr.ib()
+    schema: str = attr.ib()
+    name: str = attr.ib()
+    description: Optional[str] = None
 
-    def __init__(self, *,
-                 database: str,
-                 cluster: str,
-                 schema: str,
-                 name: str,
-                 description: Optional[str] = None) -> None:
-        self.database = database
-        self.cluster = cluster
-        self.schema = schema
-        self.name = name
-        self.description = description
 
-    def __repr__(self) -> str:
-        return """Table(database={!r}, cluster={!r}, schema={!r}, name={!r}, description={!r})"""\
-            .format(self.database, self.cluster,
-                    self.schema, self.name, self.description)
+class PopularTableSchema(AttrsSchema):
+    class Meta:
+        target = PopularTable
+        register_as_scheme = True

metadata_service/entity/table_detail.py

@@ -1,175 +1,176 @@
-from typing import Iterable, Optional
+from typing import List, Optional
 
+import attr
+from marshmallow_annotations.ext.attrs import AttrsSchema
 
+
+@attr.s(auto_attribs=True, kw_only=True)
 class User:
-    def __init__(self, *,
-                 email: str,
-                 first_name: str =None,
-                 last_name: str =None) -> None:
-        self.email = email
-        self.first_name = first_name
-        self.last_name = last_name
+    email: str = attr.ib()
+    first_name: Optional[str] = None
+    last_name: Optional[str] = None
+
 
-    def __repr__(self) -> str:
-        return 'User(email={!r}, first_name={!r}, last_name={!r})'.format(self.email, self.first_name, self.last_name)
+class UserSchema(AttrsSchema):
+    class Meta:
+        target = User
+        register_as_scheme = True
 
 
+@attr.s(auto_attribs=True, kw_only=True)
 class Reader:
-    def __init__(self, *,
-                 user: User,
-                 read_count: int) -> None:
-        self.user = user
-        self.read_count = read_count
+    user: User = attr.ib()
+    read_count: int = attr.ib()
 
-    def __repr__(self) -> str:
-        return 'Reader(user={!r}, read_count={!r})'.format(self.user, self.read_count)
 
+class ReaderSchema(AttrsSchema):
+    class Meta:
+        target = Reader
+        register_as_scheme = True
 
+
+@attr.s(auto_attribs=True, kw_only=True)
 class Tag:
-    def __init__(self, *,
-                 tag_type: str,
-                 tag_name: str) -> None:
-        self.tag_name = tag_name
-        self.tag_type = tag_type
+    tag_type: str = attr.ib()
+    tag_name: str = attr.ib()
+
 
-    def __repr__(self) -> str:
-        return 'Tag(tag_name={!r}, tag_type={!r})'.format(self.tag_name,
-                                                          self.tag_type)
+class TagSchema(AttrsSchema):
+    class Meta:
+        target = Tag
+        register_as_scheme = True
 
 
+@attr.s(auto_attribs=True, kw_only=True)
 class Watermark:
-    def __init__(self, *,
-                 watermark_type: str =None,
-                 partition_key: str =None,
-                 partition_value: str =None,
-                 create_time: str =None) -> None:
-        self.watermark_type = watermark_type
-        self.partition_key = partition_key
-        self.partition_value = partition_value
-        self.create_time = create_time
-
-    def __repr__(self) -> str:
-        return 'Watermark(watermark_type={!r}, ' \
-               'partition_key={!r}, ' \
-               'partition_value={!r}, ' \
-               'create_time={!r}))'.format(self.watermark_type,
-                                           self.partition_key,
-                                           self.partition_value,
-                                           self.create_time)
+    watermark_type: Optional[str] = None
+    partition_key: Optional[str] = None
+    partition_value: Optional[str] = None
+    create_time: Optional[str] = None
 
 
+class WatermarkSchema(AttrsSchema):
+    class Meta:
+        target = Watermark
+        register_as_scheme = True
+
+
+@attr.s(auto_attribs=True, kw_only=True)
 class Statistics:
-    def __init__(self, *,
-                 stat_type: str,
-                 stat_val: str =None,
-                 start_epoch: int =None,
-                 end_epoch: int =None) -> None:
-        self.stat_type = stat_type
-        self.stat_val = stat_val
-        self.start_epoch = start_epoch
-        self.end_epoch = end_epoch
-
-    def __repr__(self) -> str:
-        return 'Statistics(stat_type={!r}, ' \
-               'stat_val={!r},' \
-               'start_epoch={!r},' \
-               'end_epoch={!r})'.format(self.stat_type,
-                                        self.stat_val,
-                                        self.start_epoch,
-                                        self.end_epoch)
+    stat_type: str = attr.ib()
+    stat_val: Optional[str] = None
+    start_epoch: Optional[int] = None
+    end_epoch: Optional[int] = None
+
 
+class StatisticsSchema(AttrsSchema):
+    class Meta:
+        target = Statistics
+        register_as_scheme = True
 
+
+@attr.s(auto_attribs=True, kw_only=True)
 class Column:
-    def __init__(self, *,
-                 name: str,
-                 description: Optional[str],
-                 col_type: str,
-                 sort_order: int,
-                 stats: Iterable[Statistics] =()) -> None:
-        self.name = name
-        self.description = description
-        self.col_type = col_type
-        self.sort_order = sort_order
-        self.stats = stats
-
-    def __repr__(self) -> str:
-        return 'Column(name={!r}, description={!r}, col_type={!r}, sort_order={!r}, stats={!r})'\
-            .format(self.name,
-                    self.description,
-                    self.col_type,
-                    self.sort_order,
-                    self.stats)
+    name: str = attr.ib()
+    description: Optional[str] = None
+    col_type: str = attr.ib()
+    sort_order: int = attr.ib()
+    stats: List[Statistics] = []
+
+
+class ColumnSchema(AttrsSchema):
+    class Meta:
+        target = Column
+        register_as_scheme = True
+
+
+@attr.s(auto_attribs=True, kw_only=True)
+class PiiDetail:
+    pii_semantic_type: Optional[str] = None
+    pii_forget_type: Optional[str] = None
+
 
+class PiiDetailSchema(AttrsSchema):
+    class Meta:
+        target = PiiDetail
+        register_as_scheme = True
 
+
+@attr.s(auto_attribs=True, kw_only=True)
+class ForeignKey:
+    table_uri: Optional[str] = None
+    column_name: Optional[str] = None
+
+
+class ForeignKeySchema(AttrsSchema):
+    class Meta:
+        target = ForeignKey
+        register_as_scheme = True
+
+
+@attr.s(auto_attribs=True, kw_only=True)
+class DataDetail:
+    data_semantic_type: Optional[str] = None
+    data_semantic_reference: Optional[str] = None
+
+
+class DataDetailchema(AttrsSchema):
+    class Meta:
+        target = DataDetail
+        register_as_scheme = True
+
+
+@attr.s(auto_attribs=True, kw_only=True)
 class Application:
-    def __init__(self, *,
-                 application_url: str,
-                 description: str,
-                 id: str,
-                 name: str) -> None:
-        self.application_url = application_url
-        self.description = description
-        self.name = name
-        self.id = id
+    application_url: str = attr.ib()
+    description: str = attr.ib()
+    id: str = attr.ib()
+    name: str = attr.ib()
+
 
-    def __repr__(self) -> str:
-        return 'Application(application_url={!r}, description={!r}, name={!r}, id={!r})'\
-            .format(self.application_url, self.description, self.name, self.id)
+class ApplicationSchema(AttrsSchema):
+    class Meta:
+        target = Application
+        register_as_scheme = True
 
 
+@attr.s(auto_attribs=True, kw_only=True)
 class Source:
-    def __init__(self, *,
-                 source_type: str,
-                 source: str) -> None:
-        self.source_type = source_type
-        self.source = source
+    source_type: str = attr.ib()
+    source: str = attr.ib()
+
+
+class SourceSchema(AttrsSchema):
+    class Meta:
+        target = Source
+        register_as_scheme = True
+
 
-    def __repr__(self) -> str:
-        return 'Source(source_type={!r}, ' \
-               'source={!r})'.format(self.source_type,
-                                     self.source)
+# this is a temporary hack to satisfy mypy. Once https://github.com/python/mypy/issues/6136 is resolved, use
+# `attr.converters.default_if_none(default=False)`
+def default_if_none(arg: Optional[bool]) -> bool:
+    return arg or False
 
 
+@attr.s(auto_attribs=True, kw_only=True)
 class Table:
-    def __init__(self, *,
-                 database: str,
-                 cluster: str,
-                 schema: str,
-                 name: str,
-                 tags: Iterable[Tag] =(),
-                 table_readers: Iterable[Reader] = (),
-                 description: Optional[str] = None,
-                 columns: Iterable[Column],
-                 owners: Iterable[User] = (),
-                 watermarks: Iterable[Watermark] = (),
-                 table_writer: Optional[Application] = None,
-                 last_updated_timestamp: Optional[int],
-                 source: Optional[Source] = None,
-                 is_view: Optional[bool] = None,
-                 ) -> None:
-
-        self.database = database
-        self.cluster = cluster
-        self.schema = schema
-        self.name = name
-        self.tags = tags
-        self.table_readers = table_readers
-        self.description = description
-        self.columns = columns
-        self.owners = owners
-        self.watermarks = watermarks
-        self.table_writer = table_writer
-        self.last_updated_timestamp = last_updated_timestamp
-        self.source = source
-        self.is_view = is_view or False
-
-    def __repr__(self) -> str:
-        return """Table(database={!r}, cluster={!r}, schema={!r}, name={!r}, tags={!r}, table_readers={!r},
-                        description={!r}, columns={!r}, owners={!r}, watermarks={!r}, table_writer={!r},
-                        last_updated_timestamp={!r}, source={!r}, is_view={!r})"""\
-            .format(self.database, self.cluster,
-                    self.schema, self.name, self.tags,
-                    self.table_readers, self.description,
-                    self.columns, self.owners, self.watermarks,
-                    self.table_writer, self.last_updated_timestamp,
-                    self.source, self.is_view)
+    database: str = attr.ib()
+    cluster: str = attr.ib()
+    schema: str = attr.ib()
+    name: str = attr.ib()
+    columns: List[Column] = attr.ib()
+    tags: List[Tag] = []
+    table_readers: List[Reader] = []
+    description: Optional[str] = None
+    owners: List[User] = []
+    watermarks: List[Watermark] = []
+    table_writer: Optional[Application] = None
+    last_updated_timestamp: Optional[int] = None
+    source: Optional[Source] = None
+    is_view: Optional[bool] = attr.ib(default=None, converter=default_if_none)
+
+
+class TableSchema(AttrsSchema):
+    class Meta:
+        target = Table
+        register_as_scheme = True