Skip to content

fix: add casts for timestamp and boolean to resolve Spark type incompatibility #497

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
orentiman wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: add casts for timestamp and boolean to resolve Spark type incompatibility #497

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
50 changes: 50 additions & 0 deletions macros/upload_individual_datasets/upload_exposures.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -189,3 +189,53 @@
{{ return("") }}
{% endif %}
{%- endmacro %}


{% macro spark__get_exposures_dml_sql(exposures) -%}

{% if exposures != [] %}
{% set exposure_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(6)) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(7) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(10) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(12)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(13)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(14)) }}
from values
{% for exposure in exposures -%}
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ exposure.unique_id | replace("'","\\'") }}', {# node_id #}
timestamp('{{ run_started_at }}'), {# run_started_at #}
'{{ exposure.name | replace("'","\\'") }}', {# name #}
'{{ exposure.type }}', {# type #}
'{{ tojson(exposure.owner) }}', {# owner #}
'{{ exposure.maturity }}', {# maturity #}
'{{ exposure.original_file_path | replace('\\', '\\\\') }}', {# path #}
'{{ exposure.description | replace("'","\\'") }}', {# description #}
'{{ exposure.url }}', {# url #}
'{{ exposure.package_name }}', {# package_name #}
'{{ tojson(exposure.depends_on.nodes) }}', {# depends_on_nodes #}
'{{ tojson(exposure.tags) }}', {# tags #}
{% if var('dbt_artifacts_exclude_all_results', false) %}
null
{% else %}
'{{ tojson(exposure) | replace("\\", "\\\\") | replace("'", "\\'") | replace('"', '\\"') }}' {# all_results #}
{% endif %}
)
{%- if not loop.last %},{%- endif %}
{%- endfor %}
{% endset %}
{{ exposure_values }}
{% else %} {{ return("") }}
{% endif %}
{% endmacro -%}
75 changes: 75 additions & 0 deletions macros/upload_individual_datasets/upload_invocations.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -373,3 +373,78 @@

{% endmacro -%}


{% macro spark__get_invocations_dml_sql() -%}
{% set invocation_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(6) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(7) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(10) }},
nullif({{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }}, ''),
nullif({{ adapter.dispatch('column_identifier', 'dbt_artifacts')(12) }}, ''),
nullif({{ adapter.dispatch('column_identifier', 'dbt_artifacts')(13) }}, ''),
nullif({{ adapter.dispatch('column_identifier', 'dbt_artifacts')(14) }}, ''),
nullif({{ adapter.dispatch('column_identifier', 'dbt_artifacts')(15) }}, ''),
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(16)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(17)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(18)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(19)) }}
from values
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ dbt_version }}', {# dbt_version #}
'{{ project_name }}', {# project_name #}
timestamp('{{ run_started_at }}'), {# run_started_at #}
'{{ flags.WHICH }}', {# dbt_command #}
boolean('{{ flags.FULL_REFRESH }}'), {# full_refresh_flag #}
'{{ target.profile_name }}', {# target_profile_name #}
'{{ target.name }}', {# target_name #}
'{{ target.schema }}', {# target_schema #}
{{ target.threads }}, {# target_threads #}

'{{ env_var('DBT_CLOUD_PROJECT_ID', '') }}', {# dbt_cloud_project_id #}
'{{ env_var('DBT_CLOUD_JOB_ID', '') }}', {# dbt_cloud_job_id #}
'{{ env_var('DBT_CLOUD_RUN_ID', '') }}', {# dbt_cloud_run_id #}
'{{ env_var('DBT_CLOUD_RUN_REASON_CATEGORY', '') }}', {# dbt_cloud_run_reason_category #}
'{{ env_var('DBT_CLOUD_RUN_REASON', '') | replace("'","\\'") }}', {# dbt_cloud_run_reason #}

{% if var('env_vars', none) %}
{% set env_vars_dict = {} %}
{% for env_variable in var('env_vars') %}
{% do env_vars_dict.update({env_variable: (env_var(env_variable, '') | replace("'", "''"))}) %}
{% endfor %}
'{{ tojson(env_vars_dict) }}', {# env_vars #}
{% else %}
null, {# env_vars #}
{% endif %}

{% if var('dbt_vars', none) %}
{% set dbt_vars_dict = {} %}
{% for dbt_var in var('dbt_vars') %}
{% do dbt_vars_dict.update({dbt_var: (var(dbt_var, '') | replace("'", "''"))}) %}
{% endfor %}
'{{ tojson(dbt_vars_dict) }}', {# dbt_vars #}
{% else %}
null, {# dbt_vars #}
{% endif %}

'{{ tojson(invocation_args_dict) | replace('\\', '\\\\') | replace("'", "\\'") }}', {# invocation_args #}

{% set metadata_env = {} %}
{% for key, value in dbt_metadata_envs.items() %}
{% do metadata_env.update({key: (value | replace("'", "''"))}) %}
{% endfor %}
'{{ tojson(metadata_env) | replace('\\', '\\\\') }}' {# dbt_custom_envs #}

)
{% endset %}
{{ invocation_values }}

{% endmacro -%}
60 changes: 60 additions & 0 deletions macros/upload_individual_datasets/upload_model_executions.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -295,3 +295,63 @@
{{ return("") }}
{% endif %}
{%- endmacro %}


{% macro spark__get_model_executions_dml_sql(models) -%}
{% if models != [] %}
{% set model_execution_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(6) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(7) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(10) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(12) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(13) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(14) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(15) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(16)) }}

from values
{% for model in models -%}
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ model.node.unique_id }}', {# node_id #}
timestamp('{{ run_started_at }}'), {# run_started_at #}

{% set config_full_refresh = model.node.config.full_refresh %}
{% if config_full_refresh is none %}
{% set config_full_refresh = flags.FULL_REFRESH %}
{% endif %}
boolean('{{ config_full_refresh }}'), {# was_full_refresh #}

'{{ model.thread_id }}', {# thread_id #}
'{{ model.status }}', {# status #}

{% set compile_started_at = (model.timing | selectattr("name", "eq", "compile") | first | default({}))["started_at"] %}
{% if compile_started_at %}timestamp('{{ compile_started_at }}'){% else %}null{% endif %}, {# compile_started_at #}
{% set query_completed_at = (model.timing | selectattr("name", "eq", "execute") | first | default({}))["completed_at"] %}
{% if query_completed_at %}timestamp('{{ query_completed_at }}'){% else %}null{% endif %}, {# query_completed_at #}

{{ model.execution_time }}, {# total_node_runtime #}
null, -- rows_affected not available {# Only available in Snowflake & BigQuery #}
'{{ model.node.config.materialized }}', {# materialization #}
'{{ model.node.schema }}', {# schema #}
'{{ model.node.name }}', {# name #}
'{{ model.node.alias }}', {# alias #}
'{{ model.message | replace("\\", "\\\\") | replace("'", "\\'") | replace('"', '\\"') }}', {# message #}
'{{ tojson(model.adapter_response) | replace("\\", "\\\\") | replace("'", "\\'") | replace('"', '\\"') }}' {# adapter_response #}
)
{%- if not loop.last %},{%- endif %}
{%- endfor %}
{% endset %}
{{ model_execution_values }}
{% else %} {{ return("") }}
{% endif %}
{%- endmacro %}
53 changes: 53 additions & 0 deletions macros/upload_individual_datasets/upload_models.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -199,3 +199,56 @@
{{ return("") }}
{% endif %}
{%- endmacro %}


{% macro spark__get_models_dml_sql(models) -%}

{% if models != [] %}
{% set model_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(6) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(7)) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(10) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(12)) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(13)) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(14) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(15)) }}
from values
{% for model in models -%}
{% set model_copy = dbt_artifacts.copy_model(model) -%}
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ model_copy.unique_id }}', {# node_id #}
timestamp('{{ run_started_at }}'), {# run_started_at #}
'{{ model_copy.database }}', {# database #}
'{{ model_copy.schema }}', {# schema #}
'{{ model_copy.name }}', {# name #}
'{{ tojson(model_copy.depends_on.nodes) | replace('\\', '\\\\') }}', {# depends_on_nodes #}
'{{ model_copy.package_name }}', {# package_name #}
'{{ model_copy.original_file_path | replace('\\', '\\\\') }}', {# path #}
'{{ model_copy.checksum.checksum | replace('\\', '\\\\') }}', {# checksum #}
'{{ model_copy.config.materialized }}', {# materialization #}
'{{ tojson(model_copy.tags) }}', {# tags #}
'{{ tojson(model_copy.config.meta) | replace("\\", "\\\\") | replace("'","\\'") | replace('"', '\\"') }}', {# meta #}
'{{ model_copy.alias }}', {# alias #}
{% if var('dbt_artifacts_exclude_all_results', false) %}
null
{% else %}
'{{ tojson(model_copy) | replace("\\", "\\\\") | replace("'","\\'") | replace('"', '\\"') }}' {# all_results #}
{% endif %}
)
{%- if not loop.last %},{%- endif %}
{%- endfor %}
{% endset %}
{{ model_values }}
{% else %} {{ return("") }}
{% endif %}
{% endmacro -%}
58 changes: 58 additions & 0 deletions macros/upload_individual_datasets/upload_seed_executions.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -308,3 +308,61 @@
{{ return("") }}
{% endif %}
{% endmacro -%}

{% macro spark__get_seed_executions_dml_sql(seeds) -%}
{% if seeds != [] %}
{% set seed_execution_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(6) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(7) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(10) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(12) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(13) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(14) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(15) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(16)) }}
from values
{% for model in seeds -%}
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ model.node.unique_id }}', {# node_id #}
timestamp('{{ run_started_at }}'), {# run_started_at #}

{% set config_full_refresh = model.node.config.full_refresh %}
{% if config_full_refresh is none %}
{% set config_full_refresh = flags.FULL_REFRESH %}
{% endif %}
boolean('{{ config_full_refresh }}'), {# was_full_refresh #}

'{{ model.thread_id }}', {# thread_id #}
'{{ model.status }}', {# status #}

{% set compile_started_at = (model.timing | selectattr("name", "eq", "compile") | first | default({}))["started_at"] %}
{% if compile_started_at %}timestamp('{{ compile_started_at }}'){% else %}null{% endif %}, {# compile_started_at #}
{% set query_completed_at = (model.timing | selectattr("name", "eq", "execute") | first | default({}))["completed_at"] %}
{% if query_completed_at %}timestamp('{{ query_completed_at }}'){% else %}null{% endif %}, {# query_completed_at #}

{{ model.execution_time }}, {# total_node_runtime #}
null, -- rows_affected not available {# Only available in Snowflake #}
'{{ model.node.config.materialized }}', {# materialization #}
'{{ model.node.schema }}', {# schema #}
'{{ model.node.name }}', {# name #}
'{{ model.node.alias }}', {# alias #}
'{{ model.message | replace("\\", "\\\\") | replace("'", "\\'") | replace('"', '\\"') }}', {# message #}
'{{ tojson(model.adapter_response) | replace("\\", "\\\\") | replace("'", "\\'") | replace('"', '\\"') }}' {# adapter_response #}
)
{%- if not loop.last %},{%- endif %}
{%- endfor %}
{% endset %}
{{ seed_execution_values }}
{% else %} {{ return("") }}
{% endif %}
{% endmacro -%}
46 changes: 46 additions & 0 deletions macros/upload_individual_datasets/upload_seeds.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -174,3 +174,49 @@
{{ return("") }}
{% endif %}
{%- endmacro %}

{% macro spark__get_seeds_dml_sql(seeds) -%}

{% if seeds != [] %}
{% set seed_values %}
select
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(1) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(2) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(3) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(4) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(5) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(6) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(7) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(8) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(9) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(10)) }},
{{ adapter.dispatch('column_identifier', 'dbt_artifacts')(11) }},
{{ adapter.dispatch('parse_json', 'dbt_artifacts')(adapter.dispatch('column_identifier', 'dbt_artifacts')(12)) }}
from values
{% for seed in seeds -%}
(
'{{ invocation_id }}', {# command_invocation_id #}
'{{ seed.unique_id }}', {# node_id #}
timestamp('{{ run_started_at }}'), {# run_started_at #}
'{{ seed.database }}', {# database #}
'{{ seed.schema }}', {# schema #}
'{{ seed.name }}', {# name #}
'{{ seed.package_name }}', {# package_name #}
'{{ seed.original_file_path | replace('\\', '\\\\') }}', {# path #}
'{{ seed.checksum.checksum | replace('\\', '\\\\') }}', {# checksum #}
'{{ tojson(seed.config.meta) | replace("\\", "\\\\") | replace("'","\\'") | replace('"', '\\"') }}', {# meta #}
'{{ seed.alias }}', {# alias #}
{% if var('dbt_artifacts_exclude_all_results', false) %}
null
{% else %}
'{{ tojson(seed) | replace("\\", "\\\\") | replace("'","\\'") | replace('"', '\\"') }}' {# all_results #}
{% endif %}
)
{%- if not loop.last %},{%- endif %}
{%- endfor %}
{% endset %}
{{ seed_values }}
{% else %} {{ return("") }}
{% endif %}
{% endmacro -%}

Loading
Loading