frequency: set default --unique-text to <ALL_UNIQUE>

adding brackets, to better differentiate it as a special value separate from values in the CSV
jqnatividad · Aug 6, 2024 · feafe27 · feafe27
1 parent 80ddd7b
commit feafe27
Show file tree

Hide file tree

Showing 3 changed files with 8 additions and 8 deletions.
diff --git a/src/cmd/frequency.rs b/src/cmd/frequency.rs
@@ -19,7 +19,7 @@ to get column cardinalities. This short-circuits frequency compilation for colum
 all unique values (i.e. where rowcount == cardinality), enabling it to compute frequencies for
 larger-than-memory datasets as it doesn't need to load all the column's unique values into memory.
 
-Instead, it will use the "ALL_UNIQUE" value for columns with all unique values.
+Instead, it will use the "<ALL_UNIQUE>" value for columns with all unique values.
 
 This behavior can be adjusted with the --stats-mode option.
 
@@ -93,14 +93,14 @@ frequency options:
                             eliminates memory usage for columns with all unique values.
                             There are three modes:
                               auto: use stats cache if it already exists to get column cardinalities.
-                                    For columns with all unique values, "ALL_UNIQUE" will be used.
+                                    For columns with all unique values, "<ALL_UNIQUE>" will be used.
                               force: force stats calculation to get cardinalities.
                               none: don't use cardinality information.
                                     For columns with all unique values, the first N sorted unique
                                     values (based on the --limit and --unq-limit options) will be used.
                             [default: auto]
-   --all-unique-text <arg>  The text to use for the "ALL_UNIQUE" category.
-                            [default: ALL_UNIQUE]
+   --all-unique-text <arg>  The text to use for the "<ALL_UNIQUE>" category.
+                            [default: <ALL_UNIQUE>]
     -j, --jobs <arg>        The number of jobs to run in parallel.
                             This works much faster when the given CSV data has
                             an index already created. Note that a file handle

diff --git a/src/cmd/schema.rs b/src/cmd/schema.rs
@@ -461,7 +461,7 @@ fn get_unique_values(
         flag_ignore_case:     args.flag_ignore_case,
         // internal mode for getting frequency tables
         flag_stats_mode:      "_schema".to_string(),
-        flag_all_unique_text: "ALL UNIQUE".to_string(),
+        flag_all_unique_text: "<ALL UNIQUE>".to_string(),
         flag_jobs:            Some(util::njobs(args.flag_jobs)),
         flag_output:          None,
         flag_no_headers:      args.flag_no_headers,

diff --git a/tests/test_frequency.rs b/tests/test_frequency.rs
@@ -492,7 +492,7 @@ fn frequency_all_unique_with_stats_cache() {
     let got: Vec<Vec<String>> = wrk.read_stdout(&mut cmd);
     let expected = vec![
         svec!["field", "value", "count", "percentage"],
-        svec!["case_enquiry_id", "ALL_UNIQUE", "100", "100"],
+        svec!["case_enquiry_id", "<ALL_UNIQUE>", "100", "100"],
     ];
     assert_eq!(got, expected);
 }
@@ -512,7 +512,7 @@ fn frequency_all_unique_with_stats_cache_alt_all_unique_text() {
 
     let mut cmd = wrk.command("frequency");
     cmd.args(["--select", "1"])
-        // "ALL_UNIQUE" in German
+        // "<ALL_UNIQUE>" in German
         .args(["--all-unique-text", "<ALLE EINZIGARTIG>"])
         .arg(testdata);
 
@@ -537,7 +537,7 @@ fn frequency_all_unique_force_stats_cache() {
     let got: Vec<Vec<String>> = wrk.read_stdout(&mut cmd);
     let expected = vec![
         svec!["field", "value", "count", "percentage"],
-        svec!["case_enquiry_id", "ALL_UNIQUE", "100", "100"],
+        svec!["case_enquiry_id", "<ALL_UNIQUE>", "100", "100"],
     ];
     assert_eq!(got, expected);
 }