terrible-ideas / butterdb Goto Github PK

I'm planning on rewriting butterdb, and I'd like to start a conversation around usage. I'd love any examples or thoughts around how you'd like to use the library.

Off the top of my head, this is what I had envisaged:

database = butterdb.GoogleSpreadsheets(oauth_key)

@butterdb.register(butterdb.model)
class FooModel(object):
    def __init__(self, bar, baz):
        self.bar = butterdb.StringField(bar)
        self.baz = butterdb.IntField(baz)

FooModel.all()
FooModel.where(bar: 'test')
FooModel.find(baz: 1)

instance = FooModel('blah', 5)
instance.save()

instance.foo = 'snaz'
instance.save()

Is there anything in there you'd change? Are there other object mappers with crazy cool features I should steal?

Please let me know if you want this feature.

Should contain technology, testing and a more detailed explanation

may your api have twisted protocol support since this affect IO calls a non blocking solution would be highly appreciated.

A separate constructor should be added, as this is super redundant.

Hi there,

Thanks for creating this library. Here is a question, i'm trying to run the example code. Reading from the spreadsheet works like a charm. However, trying to commit one cell change is throwing this error.

Error comes from the line: marianne.commit()

Any idea?

Here is the stacktrace:

Traceback (most recent call last):
File "butterdb_hello.py", line 32, in
marianne.commit()
File "/Library/Python/2.7/site-packages/butterdb-0.1.4-py2.7.egg/butterdb/fuckit.py", line 127, in commit
cell = self.database.get_cell(self.data, field.row, field.column)
File "/Library/Python/2.7/site-packages/butterdb-0.1.4-py2.7.egg/butterdb/fuckit.py", line 48, in get_cell
return data.cell(row + 1, column + 1)
TypeError: unsupported operand type(s) for +: 'builtin_function_or_method' and 'int'

Right now all I can do is fetch all rows, iterate and compare. Even if it needs to be implemented this way, this is something i'd expect an ORM to do.

No point in getting people to define models that can easily be generated.

Can be achieved by storing repr of data and using ast.literal_eval to parse.

I'd rather avoid writing the password of my google account in the middle of my poor quality server side app code full of bad practices.

(I workarounded this by creating a google account that only has access to that account, so i'm just requesting the feature now so you keep it in mind for the rewrite.)

gspread docs:

https://github.com/burnash/gspread#alternate-authorization-using-oauth2

http://burnash.github.io/gspread/oauth2.html

Got to change import too.

Hi,

i'm new to this but I have an issue I hope someone can address, sorry if this is a dumb question:
I use the Python gspread library on a Raspberry Pi to log temperature from a sensor and write to a Google Doc spreadsheet.
I can open the spreadsheet from my code fine.

However testing the instruction below:
worksheet.append_row("hello")
will not write "hello" in the first cell of the new row, but instead will write "h" in first cell, "e" in the cell next to it etc... (each letter in a new cell on the same row)

What am I doing wrong?
Thanks