Possible Duplicate:
Favorite 3rd-party Python Libraries?
there is a question Favorite 3rd-party Python Libraries?
i don't want know about favorite libraries, i want know a list of essential libraries.
what libraries that every python programmer should know?
Depends on what kind of programming the given Python programmer does! If it involves computation on numerical arrays, numpy; for more general scientific programming, other parts of scipy; for scraping often-badly-written HTML, BeautifulSoup (perhaps but not necessarily as included in lxml; for SSH connections, tunneling, etc, paramiko; and so on, and so forth.
Few programmers find a need to develop all these kinds of software (and many, many more besides). Besides, in many important and popular areas, there isn't a single package that dominates the scene so utterly that "every" Python programmer "has" to know it -- such areas as GUIs and web server/application frameworks, for example, offer several worthwhile alternatives. Nowhere it is written that (for any given area of software development) "there must be only one";-).
SQLAlchemy, for all your relational DB needs.
lxml and numpy as Alex Martelli mentioned.
Other than that, Stdlib is pretty consistent with providing for basic needs, cannot think of any other 'essentials'.
I second what Alex Martelli wrote but I'd like to add my own opinion too. I think that the standard library is a good place to start. After all this years I can still find new interesting module or APIs inside Python's huge standard library. Sometimes is a method or a function that I didn't know existed, sometimes is some welcomed addition to the library. Nowadays, to conclude, there's a lot of new stuff on the 2.x line due to the numerous backports from the 3.x development line. They've been added to help people migrate to 3.0 but they're still useful per se.
I can name you Twisted Matrix as a great place to start developing networking applications but I'd start from one of the very foundations of Python's usefulness: its standard library.
Doug Hellman did a terrific job over the times in summarizing and explaining most of the available modules: http://www.doughellmann.com/PyMOTW/