Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support quartiles #273

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Support quartiles #273

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0c13b84
cmd/stats: add tests for --quartiles flag
May 28, 2021
040e198
cmd/stats: add --quartiles flag
May 28, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 31 additions & 4 deletions src/cmd/stats.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,8 @@ stats options:
This requires storing all CSV data in memory.
--median Show the median.
This requires storing all CSV data in memory.
--quartiles Show the quartiles.
This requires storing all CSV data in memory.
--nulls Include NULLs in the population size for computing
mean and standard deviation.
-j, --jobs <arg> The number of jobs to run in parallel.
Expand Down Expand Up @@ -75,6 +77,7 @@ struct Args {
flag_mode: bool,
flag_cardinality: bool,
flag_median: bool,
flag_quartiles: bool,
flag_nulls: bool,
flag_jobs: usize,
flag_output: Option<String>,
Expand Down Expand Up @@ -209,7 +212,8 @@ impl Args {
range: true,
dist: true,
cardinality: self.flag_cardinality || self.flag_everything,
median: self.flag_median || self.flag_everything,
median: self.flag_median && !self.flag_quartiles && !self.flag_everything,
quartiles: self.flag_quartiles || self.flag_everything,
mode: self.flag_mode || self.flag_everything,
})).take(record_len).collect()
}
Expand All @@ -220,7 +224,10 @@ impl Args {
"mean", "stddev",
];
let all = self.flag_everything;
if self.flag_median || all { fields.push("median"); }
if self.flag_median && !self.flag_quartiles && !all { fields.push("median"); }
if self.flag_quartiles || all {
fields.push("q1"); fields.push("q2"); fields.push("q3");
}
if self.flag_mode || all { fields.push("mode"); }
if self.flag_cardinality || all { fields.push("cardinality"); }
csv::StringRecord::from(fields)
Expand All @@ -235,6 +242,7 @@ struct WhichStats {
dist: bool,
cardinality: bool,
median: bool,
quartiles: bool,
mode: bool,
}

Expand All @@ -252,25 +260,28 @@ struct Stats {
online: Option<OnlineStats>,
mode: Option<Unsorted<Vec<u8>>>,
median: Option<Unsorted<f64>>,
quartiles: Option<Unsorted<f64>>,
which: WhichStats,
}

impl Stats {
fn new(which: WhichStats) -> Stats {
let (mut sum, mut minmax, mut online, mut mode, mut median) =
(None, None, None, None, None);
let (mut sum, mut minmax, mut online, mut mode, mut median, mut quartiles) =
(None, None, None, None, None, None);
if which.sum { sum = Some(Default::default()); }
if which.range { minmax = Some(Default::default()); }
if which.dist { online = Some(Default::default()); }
if which.mode || which.cardinality { mode = Some(Default::default()); }
if which.median { median = Some(Default::default()); }
if which.quartiles { quartiles = Some(Default::default()); }
Stats {
typ: Default::default(),
sum: sum,
minmax: minmax,
online: online,
mode: mode,
median: median,
quartiles: quartiles,
which: which,
}
}
Expand Down Expand Up @@ -299,6 +310,7 @@ impl Stats {
} else {
let n = from_bytes::<f64>(sample).unwrap();
self.median.as_mut().map(|v| { v.add(n); });
self.quartiles.as_mut().map(|v| { v.add(n); });
self.online.as_mut().map(|v| { v.add(n); });
}
}
Expand Down Expand Up @@ -343,6 +355,20 @@ impl Stats {
}
Some(v) => { pieces.push(v.to_string()); }
}
match self.quartiles.as_mut().and_then(|v| v.quartiles()) {
None => {
if self.which.quartiles {
pieces.push(empty());
pieces.push(empty());
pieces.push(empty());
}
}
Some((q1, q2, q3)) => {
pieces.push(q1.to_string());
pieces.push(q2.to_string());
pieces.push(q3.to_string());
}
}
match self.mode.as_mut() {
None => {
if self.which.mode {
Expand Down Expand Up @@ -377,6 +403,7 @@ impl Commute for Stats {
self.online.merge(other.online);
self.mode.merge(other.mode);
self.median.merge(other.median);
self.quartiles.merge(other.quartiles);
self.which.merge(other.which);
}
}
Expand Down
33 changes: 31 additions & 2 deletions tests/test_stats.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,15 +91,31 @@ fn setup<S>(name: S, rows: &[&str], headers: bool,
fn get_field_value(wrk: &Workdir, cmd: &mut process::Command, field: &str)
-> String {
if field == "median" { cmd.arg("--median"); }
if field == "quartiles" { cmd.arg("--quartiles"); }
if field == "cardinality" { cmd.arg("--cardinality"); }
if field == "mode" { cmd.arg("--mode"); }

let mut rows: Vec<Vec<String>> = wrk.read_stdout(cmd);
let headers = rows.remove(0);
let mut sequence: Vec<&str> = vec![];
for row in rows.iter() {
for (h, val) in headers.iter().zip(row.iter()) {
if &**h == field {
return val.clone();
match field {
"quartiles" => {
match &**h {
"q1" | "q2" => {
sequence.push(val);
},
"q3" => {
sequence.push(val);
return sequence.join(",").clone();
},
_ => {},
}
},
_ => {
if &**h == field { return val.clone(); }
},
}
}
}
Expand Down Expand Up @@ -127,16 +143,19 @@ stats_tests!(stats_infer_null_int_float_unicode, "type",
stats_tests!(stats_no_mean, "mean", &["a"], "");
stats_tests!(stats_no_stddev, "stddev", &["a"], "");
stats_tests!(stats_no_median, "median", &["a"], "");
stats_tests!(stats_no_quartiles, "quartiles", &["a"], ",,");
stats_tests!(stats_no_mode, "mode", &["a", "b"], "N/A");

stats_tests!(stats_null_mean, "mean", &[""], "");
stats_tests!(stats_null_stddev, "stddev", &[""], "");
stats_tests!(stats_null_median, "median", &[""], "");
stats_tests!(stats_null_quartiles, "quartiles", &[""], ",,");
stats_tests!(stats_null_mode, "mode", &[""], "N/A");

stats_tests!(stats_includenulls_null_mean, "mean", &[""], "", true);
stats_tests!(stats_includenulls_null_stddev, "stddev", &[""], "", true);
stats_tests!(stats_includenulls_null_median, "median", &[""], "", true);
stats_tests!(stats_includenulls_null_quartiles, "quartiles", &[""], ",,", true);
stats_tests!(stats_includenulls_null_mode, "mode", &[""], "N/A", true);

stats_tests!(stats_includenulls_mean,
Expand Down Expand Up @@ -180,6 +199,11 @@ stats_tests!(stats_median_even, "median", &["1", "2", "3", "4"], "2.5");
stats_tests!(stats_median_even_null, "median",
&["", "1", "2", "3", "4"], "2.5");
stats_tests!(stats_median_mix, "median", &["1", "2.5", "3"], "2.5");
stats_tests!(stats_quartiles, "quartiles", &["1", "2", "3"], "1,2,3");
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you also provide test when the input is &[""]?

stats_tests!(stats_quartiles_null, "quartiles", &["", "1", "2", "3"], "1,2,3");
stats_tests!(stats_quartiles_even, "quartiles", &["1", "2", "3", "4"], "1.5,2.5,3.5");
stats_tests!(stats_quartiles_even_null, "quartiles", &["", "1", "2", "3", "4"], "1.5,2.5,3.5");
stats_tests!(stats_quartiles_mix, "quartiles", &["1", "2.0", "3", "4"], "1.5,2.5,3.5");

mod stats_infer_nothing {
// Only test CSV data with headers.
Expand Down Expand Up @@ -208,6 +232,11 @@ mod stats_zero_median {
stats_test_headers!(stats_zero_median, "median", &[], "");
}

mod stats_zero_quartiles {
use super::test_stats;
stats_test_headers!(stats_zero_quartiles, "quartiles", &[], ",,");
}

mod stats_header_fields {
use super::test_stats;
stats_test_headers!(stats_header_field_name, "field", &["a"], "header");
Expand Down