Obtaining the document causing DuplicateKeyError exception to be thrown

[K E]

I'm using PyMongo 2.8 in my application, and I'm trying to get the document existing in the database that caused the DuplicateKeyError

When I add a document to the database and there is a conflict, then a DuplicateKeyError exception is thrown. Under what circumstances is details on the exception not None?

It seemed like I should be able to do something along these lines:

    except DuplicateKeyError as e:

        doc = e.details

        # Handle exception

[Bernie Hackett]

e.details includes the error document returned by the server:

>>> c.test.test.insert({'_id': 1})

1

>>> from pymongo.errors import DuplicateKeyError

>>> try:

...     c.test.test.insert({'_id': 1})

... except DuplicateKeyError as exc:

...     print exc.details

... 

{u'index': 0, u'code': 11000, u'errmsg': u'E11000 duplicate key error index: test.test.$_id_ dup key: { : 1 }'}

Unfortunately you would have to parse the value of the "errmsg" field to find the field and value that caused the error. If you are getting this when calling Collection.insert, you already now the values of all the keys (insert() modifies the document you passed in to add _id if the document doesn't already include it). If you're getting DuplicateKeyError from update() you will have to parse the error document to find this information.

[K E]

I'm not getting anything from details. I'm getting None.

Using Ming and PyMongo:

        try:

            model = Model(**data)

            model.__class__.__mongometa__.session.flush(model)

        except DuplicateKeyError as e:

            import pdb; pdb.set_trace()

(pdb) p e.details

None

-----------------------------

Since I'm using Ming though, I was able to use __mongometa__ to work around it, but it seems overly complicated.

            if hasattr(metadata, 'custom_indexes') and metadata.custom_indexes:

                indexes = metadata.custom_indexes

                expressions = []

                for index in indexes:

                    if index.get('unique', False):

                        fields = index['fields']

                        query = {}

                        for field in fields:

                            query[field] = data.get(field, None)

                        expressions.append(query)

            if hasattr(metadata, 'unique_indexes') and metadata.unique_indexes:

                indexes = metadata.unique_indexes

                expressions = []

                for index in indexes:

                    query = {}

                    for field in index:

                        query[field] = data.get(field, None)

                    expressions.append(query)

            model = Model.query.find({'$or': expressions}).first()

[K E]

I forgot to include in my work around that metadata = Model.__mongometa__

[Bernie Hackett]

Are you using mim? It seems to raise DuplicateKeyError itself, not added any details about what happened:

http://sourceforge.net/p/merciless/code/ci/master/tree/ming/mim.py#l395

[K E]

I was using MIM. I was using MIM with a test case to ensure the functionality would work before deciding if I would want to use it on a live site.

[K E]

Posted prematurely. I'll try modifying that line in Ming. Thanks for your help.