Prettify code. Add docs for macros, seed

data/snowplow_seeds.yml

@@ -0,0 +1,13 @@
+version: 2
+
+seeds:
+  - name: country_codes
+    description: >
+      English names for countries based on their two-letter ISO code, which is
+      stored in the `geo_country` column of `snowplow_page_views` and 
+      `snowplow_sessions`. Not directly used in any of the snowplow package's 
+      sessionization logic.
+    columns:
+      - name: name
+      - name: two_letter_iso_code
+      - name: three_letter_iso_code

macros/adapters/is_adapter.sql

@@ -1,7 +1,7 @@
 {% macro is_adapter(adapter='default') %}
 
     {% if adapter == 'default' %}
-        {% set adapters = ['redshift', 'snowflake', 'postgres'] %}
+        {% set adapters = ['postgres', 'redshift', 'snowflake'] %}
     {% elif adapter is string %}
         {% set adapters = [adapter] %}
     {% else %}

macros/snowplow_macros.yml

@@ -0,0 +1,87 @@
+version: 2
+
+macros:
+
+  - name: bot_any
+    description: >
+      List of text values to check if user agents are `like`. If matched,
+      the page view is probably by a bot visitor.
+  
+  - name: convert_timezone
+    description: >
+      Adapter macro that supports default and Postgres syntax for converting
+      timestamps from one timezone into another.
+    arguments:
+      - name: in_tz
+        type: string
+        description: time zone the timestamp is currently in
+      - name: out_tz
+        type: string
+        description: time zone of the output
+      - name: in_timestamp
+        type: timestamp
+        description: timestamp to convert
+
+  - name: get_max_sql
+    description: >
+      Basically, `select max(`field`) from {{relation}}`.
+      Used in `get_start_ts`.
+    arguments:
+      - name: relation
+      - name: field
+
+  - name: get_most_recent_record
+    description: >
+      Runs `get_max_sql` using `run_query` and parses the result.
+      Used in `get_start_ts`.
+    arguments:
+      - name: relation
+      - name: field
+
+  - name: get_start_ts
+    description: >
+      Determine the start timestamp/date for this incremental run based on the
+      max timestamp/date in the current version of this table.
+      
+      In the default implementation (Postgres/Redshift/Snowflake), this
+      resolves to a subquery (`select max(timestamp) from {{this}}`).
+      
+      On BigQuery, the method for executing this macro depends on the
+      incremental strategy.
+        *Merge:* Uses `run_query` to get the result of `select max(partition_field) from {{this}}`,
+          and feeds the result into the model SQL.
+        *Insert overwrite*: If user supplies `partitions` config, resolves to
+          the `least()` of all static partition values. Otherwise, resolves to 
+          `_dbt_max_partition` or `date(_dbt_max_partition)`, which dbt will generate
+          as part of the materialization script.
+    arguments:
+      - name: relation
+        type: relation
+        description: always pass `this`
+      - name: field
+        type: string
+        description: >
+          - name of the timestamp/date column to get max value
+          - default value: `collector_tstamp`
+          - not needed on Bigquery, which always uses the model's configured partition field
+
+      
+  - name: is_adapter
+    description: >
+      Determine whether a model should be enabled depending on the `target.type` 
+      of the current run. Returns `true` or `false`. All `default` models run on Postgres, Redshift,
+      and Snowflake. All `bigquery` models run on BigQuery.
+    arguments:
+      - name: adapter
+        type: string
+        description: *default* or *bigquery*
+
+  - name: timestamp_ntz
+    description: >
+      Adapter macro that supports Snowflake's good-yet-eccentric behavior
+      around timestamps and timezones. The default implementation returns the
+      input unchanged.
+    arguments:
+      - name: field
+        type: timestamp
+        description: timestamp to return as `timestampntz`

models/identification/bigquery/snowplow_id_map.sql

@@ -20,7 +20,7 @@ with all_events as (
     select * from {{ ref('snowplow_base_events') }}
 
     {% if is_incremental() %}    
-        where DATE(collector_tstamp) >= {{get_start_ts(this)}}
+        where date(collector_tstamp) >= {{get_start_ts(this)}}
     {% endif %}
 
 ),

models/page_views/bigquery/snowplow_page_views.sql

@@ -30,7 +30,7 @@ with all_events as (
 
     {% if is_incremental() %}
 
-        where DATE(collector_tstamp) >= 
+        where date(collector_tstamp) >= 
             date_sub(
                 {{get_start_ts(this)}},
                 interval {{var('snowplow:page_view_lookback_days')}} day

models/sessions/bigquery/snowplow_sessions_tmp.sql

@@ -16,7 +16,7 @@ with all_page_views as (
     select * from {{ ref('snowplow_page_views') }}
 
     {% if is_incremental() %}
-        where DATE(page_view_start) >= {{get_start_ts(this)}}
+        where date(page_view_start) >= {{get_start_ts(this)}}
     {% endif %}
 
 ),