Parse date (parsedate)

Description

Parses a string representing a date using a custom format and converts it into a timestamp, optionally specifying a time zone to see the date in one different than yours and a language locale to adapt it. This is the inverse of the Format date (formatdate) operation.

How does it work in the search window?

Select Create column in the search window toolbar, then select the Parse date operation. You need to specify at least two arguments:

Argument	Data type	More information
String to be parsed ^mandatory	string
Date format ^mandatory	string	Enter the format of the string in the first argument exactly as it appears. You must use the notation for valid format patterns described below. It must coincide with the string to be parsed, otherwise you will get the value "null".
Time zone	string	You need a valid string format the app can recognize so it returns meaningful results. If you leave the field empty or introduce a value the app cannot recognize, the default Time Zone is UTC. You can use one of the following methods: A time zone code→ UTC, PST, CST, etc. Check the full list here. The time zone database name → America/Los_Angeles, Europe/Amsterdam, Asia/Tokyo, etc. For the full list, see here. The full name of the time zone → Universal Time Coordinated, Pacific Standard Time, Central Standard Time, etc. Check the full list here. The name of the region as it appears in Devo → Asia/Shanghai, Europe/Athens, America/New York, etc. You can check the time zone codes and region names in the Configure Timezone window in Devo. Be aware that some of the codes coincide, as CST could mean Central Standard Time or China Standard Time. In that case, it would be advisable to avoid using codes and introduce any of the other formats mentioned.
Locale	string	Applies the correct language to the date based on the locale you specify. If this argument is not added, the date will be in your own locale, which is determined by your user preferences. The accepted formats for the locale argument correspond to the ISO 639-1 codes, which consist of two-character codes ("en" for English, "es" for Spanish, "fr" for Frech or "de" for German), or a combination of this with the ISO 3166-2, which consists of two-character codes for specific countries ("gb" for United Kingdom, "es" for Spain...) and together they result in a four-character code specifying the subdivision ("en-us" for American English, etc.). It must not be confused with the ISO codes that consist of three-character codes. As not all codes are recognized as valid locales, entering an invalid code will return the locale set in your user preferences. Check the full list of locales here.

The data type of the values in the new column is timestamp.

In the Date format argument, enter the exact format of the strings to be converted using the following patterns.

Type	Valid format patterns	Example
Year	YY, YYYY	19, 2019
Month	M, MM, MMM, MMMM	7, 07, Jul, July
Day in month	D, DD	9, 09
Day in week	ddd	thu, fri
Hour (0...23, 00...23)	H, HH	7, 07
Hour (1...12, 01...12)	h, hh	7, 07
AM/PM	A
Minute	m, mm	3, 03
Second	s, ss	8, 08
Millisecond (000...999)	SSS	056
Weekyear (15, 2015)	GG, GGGG
Week of week year (1...53, 01...53)	W, WW	5, 05
Parse TimeZone offset (-0700, ...+0700)	ZZ

If the string to be parsed contains literal text, enclose it inside square brackets [] so it can be correctly isolated and omitted when transformed. In the following example you can see how the literal text and date format must be treated:

String to be parsed → Event occurred on January 23 of 2020 at 10:38
Date format → [Event occurred on] MMMM DD [of] YYYY [at] HH:mm

Example 1

Let's transform the Timestamp field in the demo.ecommerce.data table into timestamp format.

Click Create column and select the Parse date operation. Assign a name to the new column - let's call it simply parsedate. Now, add the following arguments:

String to be parsed: select the timestamp column.
Date format: enter the following pattern, which corresponds to the format the string to be parsed presents: DD/MMM/YYYY:HH:mm:ss +SSSS

Click Create column and you will see the following result:

Example 2

Let's say we want to transform the Str_Timestamp field, which has the months in Spanish, into Timestamp format.

Click Create column and select the Parse date operation. Assign a name to the new column - let's call it simply parsedate. Now, add the following arguments and click Create column:

String to be parsed: select the Str_Timestamp column.
Date format: enter the following pattern, which corresponds to the format the string to be parsed presents: DD-MMMM-YYYY
Time zone: you can select any time zone, however, we chose CET because it is the one in which Spain is located.
Locale: es, which corresponds to the Spanish locale.

How does it work in LINQ?

Use the operator select... as... and add the operation syntax to create the new column. These are the valid formats of the operation:

parsedate(string_to_be_parsed, format_string)
parsedate(string_to_be_parsed, format_string, time_zone_string)
parsedate(string_to_be_parsed, format_string, time_zone_string, locale_string)

Example 1

Copy the following LINQ script and use it to replicate the first example on the demo.ecommerce.data table.

from demo.ecommerce.data
  select parsedate(timestamp, "DD/MMM/YYYY:HH:mm:ss +SSSS") as parsedate

Example 2

Copy the following LINQ script and use it to replicate the second example on the my.upload.sample.data table.

from my.upload.sample.data
select split(message, ";", 20) as Str_Timestamp
  select parsedate(Str_Timestamp, "DD-MMMM-YYYY", "CET", "es") as parsedate