Member Avatar for tomoconnell

I am getting very strange results from the following:

def list(self)
import shelve
db = shelve.open(testfile)
list = []
cnt = 0
for id in db.keys():
print db[id].info()

the testfile was written with an class defined object;
What i am finding, is if I write 50 objects, the above code prints
only 38 objects, and there are odd gaps,. like,with the name of the
objects: test1, test2, test3...test50, the print outputs
test1, test3, test5, test6, ... etc with gaps in the file names output.

But using another function, that reads the files one by one, I can read all 50 object entries, so I know they are there on the shelf. Note when I read them one by one, I supply the id,
not from the "in db.keys()" method.
I have a feeling when you list a shelved set of objects, that scanning it
via "keys" is not correct, and that is why I am getting weird results.
Actually, when you think about it, what IS the "key", for a saved object?
There is a __dict__.keys for each record, but that implies multiple keys
within one record.
Man, I am very confused....
Well, then how DO you walk through a variable length list, that you stored
via shelve?

  • 3 Contributors
  • 2 Replies
  • 515 Views
  • 1 Day Discussion Span
  • Latest Post Latest Post by vegaseat

Dani AI

Generated

Clarification first: a shelve is just a persistent dictionary — the "key" is the string you used when you assigned the object (for example db['test1'] = myobj), not the object's __dict__ keys. So if you stored 50 objects under 50 distinct string keys you should be able to iterate every stored key like you would a normal dict. was right about that behavior.

A few concrete fixes and checks (also echoing : don’t shadow built-ins). Avoid using names like list or id for your variables. Take a stable snapshot of keys before walking the shelf and inspect the raw key values so hidden characters don’t hide entries:

keys = list(db.keys())        # make a snapshot
for k in sorted(keys):
    print(repr(k), db[k].info())

Troubleshooting checklist (do these in order):

  • Compare counts: len(keys) vs len(set(keys)) to find duplicates.
  • Print repr(key) and type(key) to reveal leading/trailing whitespace, strange encodings, or non-string keys.
  • Re-open the shelf in a fresh interpreter after writing to confirm the file actually contains all items (this rules out caching or process concurrency issues).
  • Ensure writes were flushed: call db.sync() or db.close() after writing. If you mutated objects in-place when saving, either reassign them back to db[key] or use writeback=True.
  • If the shelf is shared by multiple processes, different DBM backends can behave differently; note the Python version and backend if deeper help is needed.

Those steps will expose whether keys were never written, were overwritten, or are simply hidden by whitespace/encoding issues. If a detailed reproducible snippet is available (Python version + exact store code + platform), it becomes straightforward to pinpoint the root cause.

Recommended Answers

All 2 Replies

Member Avatar for BearofNH

When providing sample code, please use [code=python] (or is it [code=Python]?) and [/code] tags so as to produce accurate spacing and syntax highlighting.

That said, I don't know the answer off the top of my head. But

  1. What is that naked def doing at the front of the code?
  2. I strongly recommend you avoid using datatypes such as "list" as variable and function names. You are overriding Python's internal namespace.

Fix those and the problem may just disappear. Or not.

Member Avatar for vegaseat

I am not sure if you are working correctly with the module shelve. For an example see Starting Python post:
http://www.daniweb.com/forums/post490861-110.html

Remember shelve behaves like a 'persistent to file' dictionary.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.