Updating your code from cdapython beta

We've added a few features while also simplifying the cdapython interface to make search faster and easier. This page provides a one-to-one mapping for all of the previous functions of cda-python to their new forms. If you need more help updating your code base, please contact us cancerdataaggregator @ gmail.com

Global changes

Install

• old: install from github
• new: pip install git+https://github.com/CancerDataAggregator/cdapython.git

*** If you previously installed cdapython in a VM, discard that VM and install in a fresh one. The new cdapython has dependency conflicts with the old version and will not install properly over the older version ***

Import

• old: from cdapython import Q, columns, unique_terms
• new: from cdapython import tables, columns, column_values, fetch_rows, summary_counts

See help text

• old: N/A

• new: help(<functionname>)

• old: N/A

• new: in any function, add a debug=True parameter to get detailed error information

Output Type

• old: some functions could have to_dataframe(), to_list, or to_csv appended, others had a parameter
• new: all functions that return matrix-like data (columns, column_values, and fetch_rows) take the two parameters return_data_as and output_file. return_data_as can be set to dataframe, list or tsv. When tsv is specified, use output_file to supply a path/to/filename.tsv

List all tables

• old: N/A
• new: tables() simple function that returns a list of currently queryable table names

Specific function updates

List all columns

function is still columns()

columns flags

• old: page_size, limit, and description parameters have been removed

• new: columns always returns all unique values and their counts by default, however there are several new parameters

• new: sort_by=<column:asc/desc> sort results by any column

• new: filters=<variable or list of variables> You can now filter out any column(s) entirely, or apply filters by row to any column(s). Full filter list.

See columns for more details

See all unique values for a given column

• old: unique_terms()
• new: column_values()

parameters

• old: page_size, limit, and count parameters have been removed

• new: column_values always returns all unique values and their counts by default, however there are several new parameters

• old: system=<data source>

• new: data_source=<data source> can now take a list, as in data_source=["GDC", "PDC"]

• new: sort_by=<column:asc/desc> sort results by any column

• new: force=<True/False> For columns with an extremely large number of unique values, such as filename, the query will fail with a large data warning. You can override the warning with Force=True

See column_values for more details

Summarize results

• old: <queryobject>.<table>.count.run()
• new: summary_counts(table=<table>, <optional parameters>) running this command with no parameters will return counts for the entire table.

parameters

• new: match_all=<filter or list of filters>. This is effectively AND for all of the listed filters, any of which can take a * wildcard e.g. match_all=["sex = male", "data_type = *sv"]

• new: match_any=<filter or list of filters>. This is effectively OR for all of the listed filters, all of which can take a * wildcard e.g. match_all=["sex = male", "data_type = *sv"]

• new: data_source restrict the results to a data_source(s), e.g. data_source=['GDC', 'IDC']

See summary_counts for more details

Returning a matrix of results

• old: all of the functions previously used with, or chained onto Q()...run() have been replaced with the single function fetch_rows()
• new: `fetch_rows(table=, )

  parameters

  • new: match_all=<filter or list of filters>. This is effectively AND for all of the listed filters, any of which can take a * wildcard e.g. match_all=["sex = male", "data_type = *sv"]

  • new: match_any=<filter or list of filters>. This is effectively OR for all of the listed filters, all of which can take a * wildcard e.g. match_all=["primary_diagnosis = neuro*", "days_to_birth > 600"]

  • new: data_source restrict the results to a data_source(s), e.g. data_source=['GDC', 'IDC']

  • new: link_to_table=<table> will return your results joined to the specifed table

  • new: provenance=<True/False> will return your results expanded to show which data_source each row came from, and the accompanying identifiers

  • new: count_only=<True/False> will return a simple row count for your results rather than the results themselves

  See fetch_rows for more details