Skip to content

A
Amazon-Selection-Data
Commit 3797377f by fangxingjun
Options

no message

parent 7293af95
Showing with 3 additions and 2 deletions
Pyspark_job/my_kafka/kafka_asin_detail.py
...@@ -376,7 +376,7 @@ class DimStAsinInfo(Templates): ...@@ -376,7 +376,7 @@ class DimStAsinInfo(Templates):
df = df.select("asin", "parentAsin", "title", "img_url", "variat_num", "best_sellers_rank", "best_sellers_herf", "price", "rating", df = df.select("asin", "parentAsin", "title", "img_url", "variat_num", "best_sellers_rank", "best_sellers_herf", "price", "rating",
"brand", "account_id", "account_name", "account_url", "buy_box_seller_type", "brand", "account_id", "account_name", "account_url", "buy_box_seller_type",
"volume", "weight", "weight_str", "launch_time", "total_comments", "page_inventory", "asinUpdateTime", "site_name", "node_id", "volume", "weight", "weight_str", "launch_time", "total_comments", "page_inventory", "asinUpdateTime", "site_name", "node_id",
"buy_sales", 'asin_amazon_orders', 'asin_ao_val', 'matrix_ao_val', "asin_zr_flow_proportion", 'matrix_flow_proportion') "buy_sales", 'asin_amazon_orders', 'asin_ao_val', 'matrix_ao_val', "asin_zr_flow_proportion", 'matrix_flow_proportion', 'describe')
return df return df
def rename_cols(self, df): def rename_cols(self, df):
...@@ -411,11 +411,12 @@ class DimStAsinInfo(Templates): ...@@ -411,11 +411,12 @@ class DimStAsinInfo(Templates):
df = df.withColumnRenamed("matrix_ao_val", "matrixAoVal") df = df.withColumnRenamed("matrix_ao_val", "matrixAoVal")
df = df.withColumnRenamed("asin_zr_flow_proportion", "asinZrFlowProportion") df = df.withColumnRenamed("asin_zr_flow_proportion", "asinZrFlowProportion")
df = df.withColumnRenamed("matrix_flow_proportion", "asinZrFlowProportionMatrix") df = df.withColumnRenamed("matrix_flow_proportion", "asinZrFlowProportionMatrix")
df = df.withColumnRenamed("describe", "asinDescribe")
# df = df.withColumnRenamed("fd_country_name", "fdCountryName") # df = df.withColumnRenamed("fd_country_name", "fdCountryName")
df = df.select('asin', 'parentAsin', 'title', "imgUrl", 'asinVarNum', 'oneCategoryRank', 'bestSellersRank', 'lastHerf', 'aoVal', 'matrixAoVal', 'price', 'rating', df = df.select('asin', 'parentAsin', 'title', "imgUrl", 'asinVarNum', 'oneCategoryRank', 'bestSellersRank', 'lastHerf', 'aoVal', 'matrixAoVal', 'price', 'rating',
'bsrOrders', 'bsrOrdersSale', 'brandName', 'accountId', 'accountName', 'accountUrl', 'buyBoxSellerType', 'bsrOrders', 'bsrOrdersSale', 'brandName', 'accountId', 'accountName', 'accountUrl', 'buyBoxSellerType',
'volume', 'weight', 'launchTime', 'totalComments', 'pageInventory', 'asinUpdateTime', 'asinBoughtMonth', "asinAmazonOrders", 'volume', 'weight', 'launchTime', 'totalComments', 'pageInventory', 'asinUpdateTime', 'asinBoughtMonth', "asinAmazonOrders",
"fdCountryName", "key_outer", "key_inner", "volumeFormat", "weightFormat", "isSelfAsin", "auctionsNum", "skusNumCreat", "asinZrFlowProportion", "asinZrFlowProportionMatrix") "fdCountryName", "key_outer", "key_inner", "volumeFormat", "weightFormat", "isSelfAsin", "auctionsNum", "skusNumCreat", "asinZrFlowProportion", "asinZrFlowProportionMatrix", "asinDescribe")
return df return df
def get_topic_name(self): def get_topic_name(self):
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment