我正在尝试将外观视图文件转换为 csv/xls 文件，所有键都作为列标题，相应的值作为每一行。
这里的最终目标是动态识别键并将它们作为列标题。
查看文件
view: sf_block_group_facts {
sql_table_name: DATABLOCKS.ACS.FAST_FACTS ;;
dimension: logrecno_bg_map_block_group {
hidden: yes
type: string
sql: ${TABLE}.LOGRECNO_BG_MAP_BLOCK_GROUP ;;
}
measure: total_population {
description: "Total Population"
type: sum
sql: ${TABLE}.total_population ;;
}
# Household Measures
measure: housing_units {
type: sum
group_label: "Households"
sql: ${TABLE}.housing_units ;;
}
measure: avg_persons_house {
type: number
group_label: "Households"
label: "Average Persons per Household"
sql: ${total_population}/NULLIF(${housing_units}, 0) ;;
value_format_name: decimal_2
}
# Income Measures
measure: aggregate_income {
hidden: yes
type: sum
group_label: "Households"
sql: ${TABLE}.aggregate_income ;;
}
measure: avg_income_house {
type: number
group_label: "Households"
label: "Average Income per Household"
sql: ${aggregate_income}/NULLIF(${housing_units}, 0) ;;
value_format_name: usd_0
}
# Sex Measures
measure: female_population {
type: sum
sql: ${TABLE}.female ;;
group_label: "Sex"
}
measure: male_population {
type: sum
sql: ${TABLE}.male ;;
group_label: "Sex"
}
measure: pct_male{
type: number
label: "Male % of Population"
group_label: "Sex"
sql: ${male_population}/NULLIF(${total_population}, 0) ;;
value_format_name: percent_2
}
measure: pct_female{
type: number
label: "Female % of Population"
group_label: "Sex"
sql: ${female_population}/NULLIF(${total_population}, 0) ;;
value_format_name: percent_2
}
# Racial Measures
measure: white_alone_or_in_combo {
label: "White Population (Alone or in Combo with Other Races)"
group_label: "Race"
type: sum
sql: ${TABLE}.white_alone_or_in_combo ;;
}
measure: black_alone_or_in_combo {
label: "Black or African American Population (Alone or in Combo with Other Races)"
group_label: "Race"
type: sum
sql: ${TABLE}.black_alone_or_in_combo ;;
}
measure: amind_alone_or_in_combo {
label: "American Indian or Native Alaskan Population (Alone or in Combo with Other Races)"
group_label: "Race"
type: sum
sql: ${TABLE}.amind_alone_or_in_combo ;;
}
measure: asian_alone_or_in_combo {
label: "Asian Population (Alone or in Combo with Other Races)"
group_label: "Race"
type: sum
sql: ${TABLE}.asian_alone_or_in_combo ;;
}
measure: nat_haw_alone_or_in_combo {
label: "Native Hawaiian or Other Pacific Islander (Alone or in Combo with Other Races)"
group_label: "Race"
type: sum
sql: ${TABLE}.nat_haw_alone_or_in_combo ;;
}
measure: white_non_hisp {
label: "White, Non-Hispanic Population"
group_label: "Race"
type: sum
sql: ${TABLE}.white_non_hisp ;;
}
measure: pct_white {
label: "White % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${white_alone_or_in_combo}/NULLIF(${total_population}, 0) ;;
}
measure: pct_black {
label: "Black/African American % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${black_alone_or_in_combo}/NULLIF(${total_population}, 0) ;;
}
measure: pct_asian {
label: "Asian % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${asian_alone_or_in_combo}/NULLIF(${total_population}, 0) ;;
}
measure: pct_amind {
label: "American Indian or Native Alaskan % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${amind_alone_or_in_combo}/NULLIF(${total_population}, 0) ;;
}
measure: pct_nathaw {
label: "Native Hawaiian or Other Pacific Islander % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${nat_haw_alone_or_in_combo}/NULLIF(${total_population}, 0) ;;
}
measure: pct_white_nh {
label: "White, Non-Hispanic % of Population"
group_label: "Race"
type: number
value_format_name: percent_2
sql: ${white_non_hisp}/NULLIF(${total_population}, 0) ;;
}
measure: hispanic_or_latino {
label: "Hispanic or Latino Population (Any Race)"
type: sum
sql: ${TABLE}.hispanic_or_latino ;;
group_label: "Hispanic/Latino"
}
measure: pct_hispanic_or_latino {
label: "Hispanic or Latino % of Population (Any Race)"
type: number
sql: ${hispanic_or_latino}/NULLIF(${total_population}, 0) ;;
group_label: "Hispanic/Latino"
value_format_name: percent_2
}
measure: pct_non_hispanic_or_latino {
label: "Non-Hispanic, Non-Latino % of Population (Any Race)"
type: number
sql: 1-(${hispanic_or_latino}/NULLIF(${total_population}, 0)) ;;
group_label: "Hispanic/Latino"
value_format_name: percent_2
}
measure: pct_white_nh2 {
label: "White, Non-Hispanic % of Population"
group_label: "Hispanic/Latino"
type: number
value_format_name: percent_2
sql: ${white_non_hisp}/NULLIF(${total_population}, 0) ;;
}
# Age Group Measures
measure: under_18 {
label: "Population 17 years and younger"
type: sum
sql: ${TABLE}.under_18 ;;
group_label: "Age"
}
measure: eighteen_to_64 {
label: "Population 18 to 64 years"
type: sum
sql: ${TABLE}.eighteen_to_64 ;;
group_label: "Age"
}
measure: sixty_five_and_over {
label: "Population 65 years and older"
type: sum
sql: ${TABLE}.sixty_five_and_over ;;
group_label: "Age"
}
measure: pct_under_18 {
label: "% of Population 17 years and younger"
type: number
sql: ${under_18}/NULLIF(${total_population}, 0) ;;
group_label: "Age"
value_format_name: percent_2
}
measure: pct_18_64 {
label: "% of Population 18 and 64 years"
type: number
sql: ${eighteen_to_64}/NULLIF(${total_population}, 0) ;;
group_label: "Age"
value_format_name: percent_2
}
measure: pct_65_over {
label: "% of Population 65 and older"`enter code here`
type: number
sql: ${sixty_five_and_over}/NULLIF(${total_population}, 0) ;;
group_label: "Age"
value_format_name: percent_2
}
}
Python 示例代码
import lkml
import json
import csv
import pandas as pd
from pandas.io.json import json_normalize
with open('./test_looker_view.view.lkml', 'r') as file:
parsed = lkml.load(file)
flat = json_normalize(parsed['views'])
flat.head(3)
预期产出
csv 文件应如下所示：
object_type,name,hidden,type,sql,description,label,view,sql_table_name...
dimension,logrecno_bg_map_block_group,yes,string,${TABLE}.LOGRECNO_BG_MAP_BLOCK_GROUP,null,null,sf_block_group_facts,DATABLOCKS.ACS.FAST_FACTS...
measure,total_population,null,sum,${TABLE}.total_population,Total Population,null,sf_block_group_facts,DATABLOCKS.ACS.FAST_FACTS...
但是我无法展平和分组内部字典。我想动态地展平它们而不必担心 json 键。
我尝试使用记录路径，但它更多的是硬编码
dim_data = json_normalize(data= parsed['views'],record_path=['dimensions'], record_prefix="dim.")
这是我到目前为止所尝试的。可能还有另一种方法来展平我可能没有看到的 json 数据，但这就是我在反复试验中取得的进展。