Early date format checking.

ukdataservice · Aug 3, 2018 · 2d9feb86ce6ce3a6e2a77605cb1cb4f403e105b2 · 2d9feb8
1 parent c6129af
commit 2d9feb86ce6ce3a6e2a77605cb1cb4f403e105b2
Unified Split

Showing with 69 additions and 17 deletions.

+1 −0 .gitignore

+0 −6 src/bin/main.rs

+1 −1 src/check/mod.rs

+55 −4 src/check/post.rs

+11 −6 src/readstat/handler.rs

+1 −0 src/report/mod.rs
diff --git a/.gitignore b/.gitignore
@@ -3,6 +3,7 @@
 **/*.rs.bk
 output.json
 index.html
+  
+*scratch*
 # Created by https://www.gitignore.io/api/vim,emacs

diff --git a/src/bin/main.rs b/src/bin/main.rs
@@ -8,8 +8,6 @@ extern crate toml;
 use qamd::readstat::read::read;
 use qamd::config::{ Config, Valid };
 use qamd::html::to_html;
-  
-// use qamd::config::{ VariableConfig, ValueConfig, Setting, Level };
-  
-// use qamd::report::Report;
 use std::io;
 use std::io::prelude::*;
@@ -88,12 +86,8 @@ fn main() {
        _ => false,
    };
-  
-    // println!("file_path: {}\nconfig_path: {}\nlocators: {}", file_path, config_path, include_locators);
-  
-
    match parse_config(&config_path) {
        Ok(ref mut config) => {
-  
-            //println!("Config: {:#?}", config);
-  
-
            config.include_locators = override_config(config.include_locators,
                                                  include_locators);
            config.progress = override_config(config.progress,

diff --git a/src/check/mod.rs b/src/check/mod.rs
@@ -48,6 +48,6 @@ impl fmt::Debug for Check {
 pub fn contains(string: &str, patterns: &Vec<String>) -> bool {
    patterns.iter()
        .map(|p| string.contains(p))
-  
-        .fold(false, |a, b| a || b)
+  
+        .fold(false, |a, b| { a || b })
 }
diff --git a/src/check/post.rs b/src/check/post.rs
@@ -3,15 +3,61 @@ use readstat::context::Context;
 use config::Config;
 use report::{ Report, Status, Locator };
 use report::missing::Missing;
-  
-use check::PostCheckFn;
+  
+use check::{ PostCheckFn, contains };
 /// Returns a vec of the functions provided by this module
 pub fn register() -> Vec<PostCheckFn> {
    vec!(primary_variable,
+  
+         date_variables,
         system_missing_over_threshold,
         variables_with_unique_values)
 }
+  
+fn date_variables(context: &Context,
+  
+                  _config: &Config,
+  
+                  _report: &mut Report) {
+  
+    // refer here for the docs on the date format. ReadStat internally
+  
+    // attempts to treat data as-if it were just Stata.
+  
+    // https://www.stata.com/help.cgi?datetime_display_formats
+  
+
+  
+    for variable in context.variables.iter() {
+  
+        // println!("{:#?}", variable.value_format);
+  
+        let date_time_specifiers = vec!(
+  
+            // date
+  
+            "JJJ",     // Day within year (001 - 366)
+  
+            "jjj",     // Day within year (1 - 366)
+  
+            "DD",      // Day within month (01 - 31)
+  
+            "dd",      // Day within month (1 - 31)
+  
+            "DAYNAME", // day of week        Sunday, Monday, ... (aligned)
+  
+            "Dayname", // day of week        Sunday, Monday, ... (unaligned)
+  
+            "Day",     // day of week        Sun, Mon, ...
+  
+            "Da",      // day of week        Su, Mo, ...
+  
+            "day",     // day of week        sun, mon, ...
+  
+            "da",      //day of week        su, mo, ...
+  
+            // time
+  
+            "HH",   // hour               00 - 23
+  
+            "Hh",   // hour               00 - 12
+  
+            "hH",   // hour               0 - 23
+  
+            "hh",   // hour               0 - 12
+  
+            "MM",   // minute             00 - 59
+  
+            "mm",   // minute             0 - 59
+  
+            "SS",   // second             00 - 60 (sic, due to leap seconds)
+  
+            "ss",   // second             0 - 60 (sic, due to leap seconds)
+  
+            ".s",   // tenths             .0 - .9
+  
+            ".ss",  // hundredths         .00 - .99
+  
+            ".sss", // thousandths        .000 - .999
+  
+            "Y",
+  
+            "y",
+  
+        ).iter().map(|s| s.to_string()).collect();
+  
+
+  
+        if contains(&variable.value_format, &date_time_specifiers) {
+  
+            println!("variable {} is a date! : {}",
+  
+                     variable.name,
+  
+                     variable.value_format);
+  
+        }
+  
+    }
+  
+}
+  
+
 /// Count the number of cases using the provided primary variable_count
 fn primary_variable(context: &Context,
                    config: &Config,
@@ -21,7 +67,8 @@ fn primary_variable(context: &Context,
            report.metadata.case_count = Some(0);
        }
-  
-        if let Some((_variable, map)) = context.frequency_table.iter().find(|(variable, _)| {
+  
+        if let Some((_variable, map)) = context.frequency_table
+  
+            .iter().find(|(variable, _)| {
            variable.name == primary_variable.setting
        }) {
            // report count of distinct cases for this variable
@@ -66,7 +113,10 @@ fn system_missing_over_threshold(context: &Context,
                // compare with config threhold
                // and increment pass/fail
-  
-                if let Some((_, count)) = map.iter().find(|(value, _)| value.missing == Missing::SYSTEM_MISSING) {
+  
+                if let Some((_, count)) = map
+  
+                    .iter().find(|(value, _)| {
+  
+                        value.missing == Missing::SYSTEM_MISSING
+  
+                    }) {
                    let sys_miss = (*count as f32 / sum as f32) * 100.0;
                    if sys_miss > setting.setting as f32 {
                        status.fail += 1;
@@ -90,7 +140,8 @@ fn variables_with_unique_values(context: &Context,
        include_check!(report.summary.variables_with_unique_values,
                       "Detects values as outliers if they unique.");
-  
-        if let Some(ref mut status) = report.summary.variables_with_unique_values {
+  
+        if let Some(ref mut status) = report.summary
+  
+            .variables_with_unique_values {
            for (variable, map) in context.frequency_table.iter() {
                if let Some(_) = map.iter().find(|(_value, occ)| {
                    *occ <= &setting.setting

diff --git a/src/readstat/handler.rs b/src/readstat/handler.rs
@@ -1,9 +1,7 @@
-  
-//use report::Report;
 use report::{ Value, Variable };
 use report::missing::Missing;
 use report::anyvalue::AnyValue;
-  
-//use check::Check;
 use readstat::bindings::*;
 use readstat::context::Context;
@@ -50,17 +48,23 @@ pub unsafe extern "C" fn metadata_handler(metadata: *mut readstat_metadata_t,
 /// Variable callback
 pub unsafe extern "C" fn variable_handler(index: c_int,
-  
-                                      variable: *mut readstat_variable_t,
-  
-                                      val_labels: *const c_char,
-  
-                                      ctx: *mut c_void) -> c_int {
+  
+                                          variable: *mut readstat_variable_t,
+  
+                                          val_labels: *const c_char,
+  
+                                          ctx: *mut c_void) -> c_int {
    let context = ctx as *mut Context;
    let variable_name = ptr_to_str!(readstat_variable_get_name(variable));
    let label = if readstat_variable_get_label(variable) != ptr::null() {
        ptr_to_str!(readstat_variable_get_label(variable))
    } else {
-  
-        "".to_string()
+  
+        String::new()
+  
+    };
+  
+
+  
+    let value_format = if readstat_variable_get_format(variable) != ptr::null() {
+  
+        ptr_to_str!(readstat_variable_get_format(variable))
+  
+    } else {
+  
+        String::new()
    };
    let value_labels = if val_labels != ptr::null() {
@@ -74,6 +78,7 @@ pub unsafe extern "C" fn variable_handler(index: c_int,
        index: index as i32 + 1,
        name: variable_name,
        label: label,
+  
+        value_format: value_format,
        value_labels: value_labels,
    };

diff --git a/src/report/mod.rs b/src/report/mod.rs
@@ -173,6 +173,7 @@ pub struct Variable {
    pub index: i32,
    pub name: String,
    pub label: String,
+  
+    pub value_format: String,
    pub value_labels: String,
 }