Documentation for tokenize contains typos, functions detect_encoding() and tokenize() at least. The lines are:

The tokenize() generator requires one argment, readline,...

It requires one argment, readline,

...where argment is supposed to be 'argument' i'm sure. Things like that bug me in my own code/docs, and since it's been there since 2.7 (Also in 3.3 and 3.4), I figured no one had noticed.

Oops, forgot to mention the documentation I am speaking of is the actual doc strings. The actual test is to run:
import tokenize
help(tokenize)

or just:

`python -c "import tokenize;help(tokenize)" | grep "argment"`

Thanks for catching this, would you like to make a patch?
http://docs.python.org/devguide gives help to do that.

New changeset 9b170d74a0a2 by Ezio Melotti in branch '2.7':
#19620: Fix typo in docstring (noticed by Christopher Welborn).
http://hg.python.org/cpython/rev/9b170d74a0a2

New changeset 3f99564b712e by Ezio Melotti in branch '3.3':
#19620: Fix typo in docstring (noticed by Christopher Welborn).
http://hg.python.org/cpython/rev/3f99564b712e

New changeset 0d5da548b80a by Ezio Melotti in branch 'default':
#19620: merge with 3.3.
http://hg.python.org/cpython/rev/0d5da548b80a

Fixed, thanks for the report!

I should've made the patch, sorry about that. I got really busy and by the time I came back it had been fixed already. (fast turnaround :)
I'm just glad it got fixed. I'm removing myself from the nosy-list.