Hi Marek,
it would be helpful if you'd post the details of the traceback for the
"int key" megatest error, and the location / procedure for running the
megatests to reproduce the problem. If the traceback message sent by
Namshin below matches what you're seeing, something strange is
happening: the code actually guarantees that the key is a string, so
shelve reporting that the key is an int is... confusing.
def get_prefix_id(self, k):
'subdivide key into prefix, id using separator'
try:
t = k.split(self.separator)
except AttributeError:
raise KeyError('key should be string! ' + repr(k))
l = len(t)
if l == 2:
return t
... more lines of code here...
def __getitem__(self,k):
"for ID 'foo.bar', return item 'bar' in dict f associated
with prefix 'foo'"
prefix,seqID = self.get_prefix_id(k)
try:
return self.get_subitem(self.prefixDict[prefix], seqID)
As you can see, __getitem__() first calls get_prefix_id(), which
immediately verifies that k is a string (or at least has a split()
method). So seqID is guaranteed to be a string. So how can return
d[seqID] in get_subitem() raise an error claiming that seqID is an int?
-- Chris
> On Tue, Mar 24, 2009 at 1:03 PM, Namshin Kim <n...@rna.kr> wrote:
> File "/usr/local/lib/python2.5/shelve.py", line 112, in __getitem__
> f = StringIO(self.dict[key])
> TypeError: 'int' object is unsubscriptable
>
> At the end of error message, you can see shelve error saying integer
> key is not allowed.
>
> On Tue, Mar 24, 2009 at 12:40 PM, Namshin Kim <n...@rna.kr> wrote:
> Hi Marek,
>
> I think the error is coming from seqdb. It should change integer ID
> into string automatically, but it looks like it is returning integer
> ID as itself. As you know, python shelve does not take integer ID,
> only string ID.
>
> For further checking, we need to build custom annotation database
> with integer ID and check whether it handles integer<->string
> conversion correctly. As I remember, integer ID is somehow changed
> into like 'INT:1'.
>
> File "/result/pygr_mt_metabase/pygr/build/lib.linux-x86_64-2.5/
> pygr/seqdb.py", line 565, in __getitem__
> return self.get_subitem(self.prefixDict[prefix], seqID)
> File "/result/pygr_mt_metabase/pygr/build/lib.linux-x86_64-2.5/
> pygr/seqdb.py", line 558, in get_subitem
> return d[seqID]
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"pygr-dev" group.
To post to this group, send email to pygr-dev@googlegroups.com
To unsubscribe from this group, send email to
pygr-dev+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/pygr-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
