#15018: LimitedStream.readline deals parameter not correct
-------------------------------------+--------------------------------------
Reporter: xjdrew | Owner: nobody
Status: reopened | Milestone:
Component: Core framework | Version: SVN
Resolution: | Keywords:
Stage: Unreviewed | Has_patch: 0
Needs_docs: 0 | Needs_tests: 0
Needs_better_patch: 0 |
-------------------------------------+--------------------------------------
Changes (by xjdrew):
* status: closed => reopened
* resolution: worksforme =>
* component: Uncategorized => Core framework
Comment:
Replying to [comment:1 russellm]:
> Your point 3 doesn't make any sense, and I can't work out what it might
be a typo for.
>
> If you want to prove there is a problem here, write a formal test case.
[http://code.djangoproject.com/browser/django/trunk/tests/regressiontests/requests/tests.py#L101
There are existing tests for LimitedStream] covering a wide range of
potential use cases; if you can encode your use case as a test that fails,
please provide that test and reopen.
[[BR]]
Thanks for your comment and the test case suggestion.[[BR]]
[[BR]]
I make a mistake about point 3, it should be:[[BR]]
'''3. and len(self.buffer) > size'''
I make a usecase as below:[[BR]]
first, I modify the function LimitedStream.readline to add some log as
below:
{{{
def readline(self, size=None):
print "buff length: %s" % len(self.buffer) ####### print buffer
size
while '\n' not in self.buffer or \
(size is not None and len(self.buffer) < size):
}}}
It will print the internal buffer status.
second, I test the class as below:
{{{
>>> stream = LimitedStream(StringIO('te\nabcdefghijkhlmnopqrstuvwst'), 10)
#### init buffer size => 10
>>> stream.readline(5)
buff length: 0
'te\n'
>>> stream.readline(1)
buff length: 2
'a'
>>> stream.readline(1)
buff length: 25 ##### buffer overflow
'b'
}}}
it's easy to explain. at the beginning, the buffer size is 0, after I call
readline(5), the buffer size becomes 2 (equals 5 - len(line) ), then I
call readline(1), now it meets the three conditions I list before. so it
execute '''chunk = self._read_limited(size - len(self.buffer))''', that is
'''chunk = self._read_limited(-1)''', for the underlay StringIO, it reads
all content. so when I call readline(1) again, you see, '''the buffer
becomes 25, it's more than the size I init.'''
If I replace StringIO with other stream, the behavior maybe different;
And if the underlay stream is a very very very big file, the program maybe
crash.
...
So I suggest, it's a potential bug.
best regards,
Drew
--
Ticket URL: <http://code.djangoproject.com/ticket/15018#comment:2>
Django <http://code.djangoproject.com/>
The Web framework for perfectionists with deadlines.
--
You received this message because you are subscribed to the Google Groups
"Django updates" group.
To post to this group, send email to django-upda...@googlegroups.com.
To unsubscribe from this group, send email to
django-updates+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/django-updates?hl=en.
