summary_counts returning a list object

[pihltd]

Running any query using the summmary_counts feature is (according to the documentation) supposed to return a dictionary. However, what actually gets returned is a list object that has extra bits of information. For example:

Query: summary_counts(table="subject", match_all=["file_format = bam"])

Returned object:

[   total_subject_matches
 0                  30565,
    total_related_files
 0              3055727,
   system  count
 0    GDC  24183
 1    IDC  12329
 2    CDS   8796
 3    PDC   1893,
            sex  count
 0         male  11859
 1       female  11765
 2       Female   3220
 3         Male   3100
 4                 519
 5            F     53
 6            M     48
 7  unspecified      1,
                                         race  count
 0                                      White  20167
 1                                              6410
 2                  Black or African American   2512
 3                                      Asian   1339
 4           American Indian or Alaska Native     85
 5  Native Hawaiian or Other Pacific Islander     52,
                 ethnicity  count
 0  Not Hispanic or Latino  19691
 1                          10731
 2                   White    143,
             cause_of_death  count
 0                           29712
 1           Cancer Related    672
 2       Not Cancer Related    142
 3                Infection     19
 4  Cardiovascular Disorder     12
 5    Surgical Complication      8]

That kind of looks like each entry is a line from a data frame, index included, but it's certainly not a dictionary

[ACharbonneau]

@abradyGDIT do we need to update the docs?

[ACharbonneau]

Oh. Sorry @abradyGDIT it is changed, I missed updating the site. @pihltd the text in `help(summary_counts) should have been correct, and I just pushed an update to the website to match. That at least will make the description of the results true. Do you have additional questions about the result itself? or just that the website was inaccurate?

[pihltd]

The result list itself seems a little clunky for what it's trying to represent. In other words, I can't really get to the count values without parsing each of the list elements, something that returning a dictionary would avoid. There's also what looks like pandas index numbers included in the returned results, and I'm not sure how useful those are in a list object.

[ACharbonneau]

Totally agree. The plan for the next release is to have it be more like the previous incarnation where it can display as tables, but we still wanted people to be able to access the underlying data, which was super hard before. Maybe in the next week or two you, Arthur and I can get together and talk about what a useful output would look like. From: Todd Pihl ***@***.***> Date: Friday, April 12, 2024 at 1:05 PM To: CancerDataAggregator/CDA-HelpDesk ***@***.***> Cc: Charbonneau, Amanda L ***@***.***>, Comment ***@***.***> Subject: Re: [CancerDataAggregator/CDA-HelpDesk] summary_counts returning a list object (Discussion #61) The result list itself seems a little clunky for what it's trying to represent. In other words, I can't really get to the count values without parsing each of the list elements, something that returning a dictionary would avoid. There's also ZjQcmQRYFpfptBannerStart This Message Is From an External Sender Please use caution with links, attachments, and any requests for credentials. ZjQcmQRYFpfptBannerEnd The result list itself seems a little clunky for what it's trying to represent. In other words, I can't really get to the count values without parsing each of the list elements, something that returning a dictionary would avoid. There's also what looks like pandas index numbers included in the returned results, and I'm not sure how useful those are in a list object. — Reply to this email directly, view it on GitHub<https://urldefense.us/v3/__https:/github.com/CancerDataAggregator/CDA-HelpDesk/discussions/61*discussioncomment-9098508__;Iw!!JRQnnSFuzw7wjAKq6ti6!1sVSXD8AVIyl2GVkt_72Khg2MzKPBsXBmLSVRLzJeNO8WV-xMFt2x_lGvOXmXwPL-IUmnaABziERAjfSP2wUhfu4gyBhLA$>, or unsubscribe<https://urldefense.us/v3/__https:/github.com/notifications/unsubscribe-auth/AANDYQCMMJBHJMREYZJ7KTLY5AH35AVCNFSM6AAAAABF7G54FOVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TAOJYGUYDQ__;!!JRQnnSFuzw7wjAKq6ti6!1sVSXD8AVIyl2GVkt_72Khg2MzKPBsXBmLSVRLzJeNO8WV-xMFt2x_lGvOXmXwPL-IUmnaABziERAjfSP2wUhftaKRILsQ$>. You are receiving this because you commented.Message ID: ***@***.***>

[pihltd]

Upon further review/education it turns out the function is returning a list of dataframes, not a list of strings. So updating the documentation would be sufficient to solve the issue.