Improve API and CLI about camelot HOT 6 CLOSED

Current API:

>>> from camelot.pdf import Pdf
>>> from camelot.lattice import Lattice
>>> manager = Pdf(Lattice(), "/path/to/pdf")
>>> tables = manager.extract()

This is redundant (since two imports are needed to create a 'manager' that parses PDFs and not intuitive (since the user doesn't know what is returned in tables, which is a dict with multiple keys that make him/her/them look at the documentation, these multiple keys contain different stats and data is present in the 'data' key as a 2-D list).

from camelot.

Edit: Commented again.

from camelot.

Will list down properties of the Table object here later, the idea is to let the user get the data without having to look at the docs often, either by having a df property which contains a pandas DataFrame object or a 2-D list again perhaps? Other properties can be the various stats.

from camelot.

Make configuration parameters for parsing methods more intuitive (either through docs or making them simpler or removing them?), for example, user has no idea what threshold_constant is.

from camelot.

New API:

>>> import camelot
>>> tables = camelot.read_pdf("foo.pdf")
>>> tables
<TableList n=2>
>>> tables.to_csv(zip=True) # to_json, to_excel, to_html
>>> tables[0]
<Table shape=(3,4)>
>>> tables[0].parsing_report
{
    "accuracy": 96,
    "whitespace": 80,
    "time_taken": 0.5,
    "page": 1
}
>>> df = tables[0].df
>>> tables[0].to_csv("foo.csv") # to_json, to_excel, to_html

Each table is internally represented as a pandas DataFrame.

from camelot.

Opening another issue #95 for configuration parameter names.

from camelot.