Python parser for human readable dates

Key Features • How To Use • Installation • Common use cases • You may also like... • License

Key Features

• Support for almost every existing date format: absolute dates, relative dates ("two weeks ago" or "tomorrow"), timestamps, etc.
• Support for more than 200 language locales.
• Language autodetection
• Customizable behavior through settings.
• Support for non-Gregorian calendar systems.
• Support for dates with timezones abbreviations or UTC offsets ("August 14, 2015 EST", "21 July 2013 10:15 pm +0500"...)
• Search dates in longer texts.

Online demo

Do you want to try it out without installing any dependency? Now you can test it quickly by visiting this online demo!

How To Use

The most straightforward way to parse dates with dateparser is to use the dateparser.parse() function, that wraps around most of the functionality of the module.

>>> import dateparser

>>> dateparser.parse('Fri, 12 Dec 2014 10:55:50')
datetime.datetime(2014, 12, 12, 10, 55, 50)

>>> dateparser.parse('1991-05-17')
datetime.datetime(1991, 5, 17, 0, 0)

>>> dateparser.parse('In two months')  # today is 1st Aug 2020
datetime.datetime(2020, 10, 1, 11, 12, 27, 764201)

>>> dateparser.parse('1484823450')  # timestamp
datetime.datetime(2017, 1, 19, 10, 57, 30)

>>> dateparser.parse('January 12, 2012 10:00 PM EST')
datetime.datetime(2012, 1, 12, 22, 0, tzinfo=<StaticTzInfo 'EST'>)

As you can see, dateparser works with different date formats, but it can also be used directly with strings in different languages:

>>> dateparser.parse('Martes 21 de Octubre de 2014')  # Spanish (Tuesday 21 October 2014)
datetime.datetime(2014, 10, 21, 0, 0)

>>> dateparser.parse('Le 11 Décembre 2014 à 09:00')  # French (11 December 2014 at 09:00)
datetime.datetime(2014, 12, 11, 9, 0)

>>> dateparser.parse('13 января 2015 г. в 13:34')  # Russian (13 January 2015 at 13:34)
datetime.datetime(2015, 1, 13, 13, 34)

>>> dateparser.parse('1 เดือนตุลาคม 2005, 1:00 AM')  # Thai (1 October 2005, 1:00 AM)
datetime.datetime(2005, 10, 1, 1, 0)

>>> dateparser.parse('yaklaşık 23 saat önce')  # Turkish (23 hours ago), current time: 12:46
datetime.datetime(2019, 9, 7, 13, 46)

>>> dateparser.parse('2小时前')  # Chinese (2 hours ago), current time: 22:30
datetime.datetime(2018, 5, 31, 20, 30)

You can control multiple behaviors by using the settings parameter:

>>> dateparser.parse('2014-10-12', settings={'DATE_ORDER': 'YMD'})
datetime.datetime(2014, 10, 12, 0, 0)

>>> dateparser.parse('2014-10-12', settings={'DATE_ORDER': 'YDM'})
datetime.datetime(2014, 12, 10, 0, 0)

>>> dateparser.parse('1 year', settings={'PREFER_DATES_FROM': 'future'})  # Today is 2020-09-23
datetime.datetime(2021, 9, 23, 0, 0)

>>> dateparser.parse('tomorrow', settings={'RELATIVE_BASE': datetime.datetime(1992, 1, 1)})
datetime.datetime(1992, 1, 2, 0, 0)

To see more examples on how to use the settings, check the settings section in the docs.

False positives

dateparser will do its best to return a date, dealing with multiple formats and different locales. For that reason it is important that the input is a valid date, otherwise it could return false positives.

To reduce the possibility of receiving false positives, make sure that:

• The input string it's a valid date and it doesn't contain any other words or numbers.
• If you know the language or languages beforehand you add them through the languages or locales properties.

On the other hand, if you want to exclude any of the default parsers (timestamp, relative-time...) or change the order in which they are executed, you can do so through the settings PARSERS.

Installation

Dateparser supports Python >= 3.5. You can install it by doing:

$ pip install dateparser

If you want to use the jalali or hijri calendar, you need to install the calendars extra:

$ pip install dateparser[calendars]

Common use cases

dateparser can be used with a really different number of purposes, but it stands out when it comes to:

Consuming data from different sources:

• Scraping: extract dates from different places with several different formats and languages
• IoT: consuming data coming from different sources with different date formats
• Tooling: consuming dates from different logs / sources
• Format transformations: when transforming dates coming from different files (PDF, CSV, etc.) to other formats (database, etc).

Offering natural interaction with users:

• Tooling and CLI: allow users to write “3 days ago” to retrieve information.
• Search engine: allow people to search by date in an easiest / natural format.
• Bots: allow users to interact with a bot easily

You may also like...

• price-parser - A small library for extracting price and currency from raw text strings.
• number-parser - Library to convert numbers written in the natural language to it's equivalent numeric forms.
• Scrapy - Web crawling and web scraping framework

License

BSD3-Clause