CharlyWargnier

df[['httpCode','header']] = df["header"].str.split(" ", 1, expand=True)

CharlyWargnier

df[['httpProto','header']] = df["header"].str.split("\" ", 1, expand=True)

CharlyWargnier

df[['url','header']] = df["header"].str.split(" ", 1, expand=True)
# Remove slash (/) at the start of each URL
df['url'] = df['url'].str.replace('/','',n=1)

CharlyWargnier

df[['Method','header']] = df["header"].str.split(" ", 1, expand=True)

CharlyWargnier

df[['Domain','header']] = df["header"].str.split(" \"", 1, expand=True)

CharlyWargnier

# Split Date & Time
df[['Date_Time', 'header']] = df['header'].str.split("\] ", 1,
        expand=True)
df['Date_Time'] = df['Date_Time'].str.replace(':.*', '')

# Convert the Date_Time column from string to datetime format
df['Date'] = pd.to_datetime(df['Date_Time'], errors='coerce',
                            infer_datetime_format=True)

# Remove Time column, for clarity's sake

CharlyWargnier

# Split IP addresses
df[['IP','header']] = df["header"].str.split(" - - \[", 1, expand=True)
# Remove slash
df['IP'] = df['IP'].str.replace("Row\(\('","",n=1)

CharlyWargnier

df.rename(columns={0:'header'}, inplace=True)

CharlyWargnier

# Note that we convert a list of lists, not just a list.
df = pd.DataFrame.from_records(log_sample)
df.info()

CharlyWargnier

## Form elements ##

table_name = 'log_sample' #@param {type:"string"}
SQL_1st_Filter ='ch.loggy' #@param {type:"string"}
SQL_Useragent_Filter = "Googlebot/2.1" #@param ["Googlebot/2.1", "YandexBot", "BingBot", "DuckDuckBot", "Baiduspider"] {allow-input: true}

# Temporary Bigquery table name
table_id = table_name

## Full SQL query ##