patchwork_csv_utils 0.1.22-arm64-darwin → 0.1.24-arm64-darwin

data/ext/csv_utils/src/utils/mod.rs

@@ -1,20 +1,28 @@
+use ::csv::{ByteRecord, StringRecord};
+use chrono::{NaiveDate, NaiveDateTime};
+use magnus::Ruby;
 use std::collections::{HashMap, HashSet};
 use std::error::Error;
 use std::ffi::OsStr;
 use std::path::Path;
-use ::csv::{ByteRecord, StringRecord};
-use magnus::Ruby;
 
 pub mod csv;
 pub mod dedup;
+pub mod shared;
 pub mod xls;
 
 fn missing_header(ruby: &Ruby, header: &str) -> magnus::Error {
-    magnus::Error::new(ruby.exception_standard_error(), format!("Missing '{}' header", header))
+    magnus::Error::new(
+        ruby.exception_standard_error(),
+        format!("Missing '{}' header", header),
+    )
 }
 
 fn missing_value(ruby: &Ruby, header: &str) -> magnus::Error {
-    magnus::Error::new(ruby.exception_standard_error(), format!("Missing value for '{}' header", header))
+    magnus::Error::new(
+        ruby.exception_standard_error(),
+        format!("Missing value for '{}' header", header),
+    )
 }
 
 fn headers_as_byte_record(headers: Vec<String>) -> ByteRecord {
@@ -23,18 +31,32 @@ fn headers_as_byte_record(headers: Vec<String>) -> ByteRecord {
 }
 
 fn magnus_err<E: Error>(ruby: &Ruby, e: E, msg: &str) -> magnus::Error {
-    magnus::Error::new(ruby.exception_standard_error(), format!("{}: {}", msg, e.to_string()))
+    magnus::Error::new(ruby.exception_standard_error(), format!("{}: {}", msg, e))
 }
 
 fn to_datetime_error(ruby: &Ruby, value: &str, row: usize, col: &str) -> magnus::Error {
-    magnus::Error::new(ruby.exception_standard_error(), format!("Could not parse datetime '{}', row: {}, col: {}", value, row, col))
+    magnus::Error::new(
+        ruby.exception_standard_error(),
+        format!(
+            "Could not parse datetime '{}', row: {}, col: {}",
+            value, row, col
+        ),
+    )
 }
 
-fn check_mandatory_headers(ruby: &Ruby, headers: &Vec<String>, mandatory_headers: &Vec<String>, message: &str) -> Option<magnus::error::Result<()>> {
+fn check_mandatory_headers(
+    ruby: &Ruby,
+    headers: &[String],
+    mandatory_headers: &[String],
+    message: &str,
+) -> Option<magnus::error::Result<()>> {
     let csv_mandatory_headers = filter_headers(headers, mandatory_headers);
 
     if csv_mandatory_headers.is_empty() {
-        return Some(Err(magnus::Error::new(ruby.exception_standard_error(), format!("{} has no mandatory headers", message))));
+        return Some(Err(magnus::Error::new(
+            ruby.exception_standard_error(),
+            format!("{} has no mandatory headers", message),
+        )));
     }
 
     let csv_mandatory_headers = csv_mandatory_headers.to_owned().clone();
@@ -45,22 +67,45 @@ fn check_mandatory_headers(ruby: &Ruby, headers: &Vec<String>, mandatory_headers
     let difference = set2.difference(&set1).collect::<Vec<_>>();
 
     if !difference.is_empty() {
-        let missing_headers = difference.iter().map(|h| h.to_string()).collect::<Vec<String>>();
-        return Some(Err(magnus::Error::new(ruby.exception_standard_error(), format!("{} is missing mandatory headers: {}", message, missing_headers.join(", ")))));
+        let missing_headers = difference
+            .iter()
+            .map(|h| h.to_string())
+            .collect::<Vec<String>>();
+        return Some(Err(magnus::Error::new(
+            ruby.exception_standard_error(),
+            format!(
+                "{} is missing mandatory headers: {}",
+                message,
+                missing_headers.join(", ")
+            ),
+        )));
     }
     None
 }
 
-fn index_of_header_in_mandatory_list(mandatory_headers_list: Vec<String>, column_name: String) -> Option<usize> {
-    mandatory_headers_list.iter().position(|h| h.to_string() == column_name)
+fn index_of_header_in_mandatory_list(
+    mandatory_headers_list: Vec<String>,
+    column_name: String,
+) -> Option<usize> {
+    mandatory_headers_list
+        .iter()
+        .position(|h| h == &column_name)
 }
 
-fn filter_headers(csv_headers: &Vec<String>, expected_headers: &Vec<String>) -> Vec<String> {
-    csv_headers.iter().map(|v| v.to_string()).filter(|h| expected_headers.contains(h)).collect::<Vec<String>>()
+fn filter_headers(csv_headers: &[String], expected_headers: &[String]) -> Vec<String> {
+    csv_headers
+        .iter()
+        .filter(|h| expected_headers.contains(h))
+        .cloned()
+        .collect()
 }
 
-fn create_header_map(headers: &Vec<String>) -> HashMap<String, usize> {
-    headers.iter().enumerate().map(|(i, h)| (h.to_string(), i)).collect()
+fn create_header_map(headers: &[String]) -> HashMap<String, usize> {
+    headers
+        .iter()
+        .enumerate()
+        .map(|(i, h)| (h.to_string(), i))
+        .collect()
 }
 
 pub trait FileExtension {
@@ -70,7 +115,7 @@ pub trait FileExtension {
 
 impl<P: AsRef<Path>> FileExtension for P {
     fn has_extension<S: AsRef<str>>(&self, extensions: &[S]) -> bool {
-        if let Some(ref extension) = self.as_ref().extension().and_then(OsStr::to_str) {
+        if let Some(extension) = self.as_ref().extension().and_then(OsStr::to_str) {
             return extensions
                 .iter()
                 .any(|x| x.as_ref().eq_ignore_ascii_case(extension));
@@ -83,3 +128,22 @@ impl<P: AsRef<Path>> FileExtension for P {
     }
 }
 
+pub fn string_to_datetime(s: &str) -> Option<NaiveDateTime> {
+    let maybe_correct = correct_datetime(s);
+    if maybe_correct.is_some() {
+        return maybe_correct;
+    }
+
+    // Try YYYY-MM-DD format
+    if let Ok(date) = NaiveDate::parse_from_str(s, "%Y-%m-%d") {
+        return date.and_hms_opt(0, 0, 0);
+    }
+
+    NaiveDate::parse_from_str(s, "%d-%b-%y")
+        .ok()
+        .and_then(|d| d.and_hms_opt(0, 0, 0))
+}
+
+pub fn correct_datetime(s: &str) -> Option<NaiveDateTime> {
+    NaiveDateTime::parse_from_str(s, "%Y-%m-%d %H:%M:%S").ok()
+}

data/ext/csv_utils/src/utils/shared/datetime.rs

@@ -0,0 +1,79 @@
+use crate::utils::{correct_datetime, to_datetime_error};
+use chrono::{NaiveDateTime, NaiveTime};
+use magnus::Ruby;
+
+pub struct DateTimeProcessor {
+    date_value: NaiveDateTime,
+}
+
+impl DateTimeProcessor {
+    pub fn new(date_value: NaiveDateTime) -> Self {
+        DateTimeProcessor { date_value }
+    }
+
+    pub fn process_time_column(
+        &self,
+        ruby: &Ruby,
+        value: &str,
+        row_index: usize,
+        column_name: &str,
+    ) -> magnus::error::Result<String> {
+        if let Some(correct) = correct_datetime(value) {
+            return Ok(correct.to_string());
+        }
+
+        let time = string_to_time(value)
+            .ok_or_else(|| to_datetime_error(ruby, value, row_index, column_name))?;
+
+        Ok(self.combine_date_time(time).to_string())
+    }
+
+    pub fn combine_date_time(&self, time: NaiveTime) -> NaiveDateTime {
+        NaiveDateTime::new(self.date_value.date(), time)
+    }
+
+    pub fn combine_datetime_parts(&self, time_source: NaiveDateTime) -> NaiveDateTime {
+        NaiveDateTime::new(self.date_value.date(), time_source.time())
+    }
+}
+
+fn string_to_time(s: &str) -> Option<NaiveTime> {
+    NaiveTime::parse_from_str(s, "%H:%M").ok()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use chrono::NaiveDate;
+
+    #[test]
+    fn test_combine_date_time() {
+        let date = NaiveDate::from_ymd_opt(2024, 1, 15)
+            .unwrap()
+            .and_hms_opt(0, 0, 0)
+            .unwrap();
+        let processor = DateTimeProcessor::new(date);
+
+        let time = NaiveTime::from_hms_opt(14, 30, 0).unwrap();
+        let result = processor.combine_date_time(time);
+
+        assert_eq!(result.to_string(), "2024-01-15 14:30:00");
+    }
+
+    #[test]
+    fn test_combine_datetime_parts() {
+        let date = NaiveDate::from_ymd_opt(2024, 1, 15)
+            .unwrap()
+            .and_hms_opt(0, 0, 0)
+            .unwrap();
+        let processor = DateTimeProcessor::new(date);
+
+        let time_source = NaiveDate::from_ymd_opt(2020, 1, 1)
+            .unwrap()
+            .and_hms_opt(14, 30, 0)
+            .unwrap();
+        let result = processor.combine_datetime_parts(time_source);
+
+        assert_eq!(result.to_string(), "2024-01-15 14:30:00");
+    }
+}

data/ext/csv_utils/src/utils/shared/filters.rs

@@ -0,0 +1,130 @@
+use crate::utils::string_to_datetime;
+use calamine::{Data, DataType};
+use chrono::NaiveDateTime;
+
+pub trait FilterableRecord {
+    fn is_empty(&self) -> bool;
+    fn has_empty_first_column(&self) -> bool;
+    fn get_request_id(&self, index: usize) -> Option<String>;
+    fn get_status(&self, index: Option<usize>) -> Option<String>;
+    fn get_date(&self, index: usize) -> Option<NaiveDateTime>;
+}
+
+pub struct RowFilters {
+    exclusions: Vec<String>,
+    status_exclusions: Vec<String>,
+    earliest_start_date: Option<NaiveDateTime>,
+}
+
+impl RowFilters {
+    pub fn new(
+        exclusions: Vec<String>,
+        status_exclusions: Vec<String>,
+        earliest_start_date: Option<NaiveDateTime>,
+    ) -> Self {
+        RowFilters {
+            exclusions,
+            status_exclusions,
+            earliest_start_date,
+        }
+    }
+
+    pub fn should_skip<R: FilterableRecord>(
+        &self,
+        record: &R,
+        request_id_index: usize,
+        status_index: Option<usize>,
+        date_index: usize,
+    ) -> bool {
+        record.is_empty()
+            || record.has_empty_first_column()
+            || self.should_skip_by_exclusion(record, request_id_index, status_index)
+            || self.should_skip_by_status(record, status_index)
+            || self.should_skip_by_date(record, date_index)
+    }
+
+    fn should_skip_by_exclusion<R: FilterableRecord>(
+        &self,
+        record: &R,
+        request_id_index: usize,
+        status_index: Option<usize>,
+    ) -> bool {
+        if let Some(status) = record.get_status(status_index) {
+            if status == "Recalled" {
+                return false;
+            }
+        }
+
+        record
+            .get_request_id(request_id_index)
+            .map(|id| self.exclusions.contains(&id))
+            .unwrap_or(false)
+    }
+
+    fn should_skip_by_status<R: FilterableRecord>(
+        &self,
+        record: &R,
+        status_index: Option<usize>,
+    ) -> bool {
+        record
+            .get_status(status_index)
+            .map(|status| self.status_exclusions.contains(&status))
+            .unwrap_or(false)
+    }
+
+    fn should_skip_by_date<R: FilterableRecord>(&self, record: &R, date_index: usize) -> bool {
+        self.earliest_start_date
+            .and_then(|earliest| record.get_date(date_index).map(|date| date <= earliest))
+            .unwrap_or(false)
+    }
+}
+
+use csv::StringRecord;
+
+impl FilterableRecord for StringRecord {
+    fn is_empty(&self) -> bool {
+        self.iter().all(|r| r.is_empty())
+    }
+
+    fn has_empty_first_column(&self) -> bool {
+        self.get(0).map(|s| s.is_empty()).unwrap_or(true)
+    }
+
+    fn get_request_id(&self, index: usize) -> Option<String> {
+        self.get(index).map(|s| s.to_string())
+    }
+
+    fn get_status(&self, index: Option<usize>) -> Option<String> {
+        index.and_then(|idx| self.get(idx).map(|s| s.to_string()))
+    }
+
+    fn get_date(&self, index: usize) -> Option<NaiveDateTime> {
+        self.get(index).and_then(string_to_datetime)
+    }
+}
+
+impl FilterableRecord for Vec<&Data> {
+    fn is_empty(&self) -> bool {
+        self.iter().all(|c| *c == &Data::Empty)
+    }
+
+    fn has_empty_first_column(&self) -> bool {
+        self.first().map(|d| *d == &Data::Empty).unwrap_or(true)
+    }
+
+    fn get_request_id(&self, index: usize) -> Option<String> {
+        self.get(index).map(|d| d.to_string())
+    }
+
+    fn get_status(&self, index: Option<usize>) -> Option<String> {
+        index.and_then(|idx| self.get(idx).and_then(|d| d.as_string()))
+    }
+
+    fn get_date(&self, index: usize) -> Option<NaiveDateTime> {
+        self.get(index).and_then(|data| match data {
+            Data::DateTime(d) => d.as_datetime(),
+            Data::DateTimeIso(s) => string_to_datetime(s),
+            _ => None,
+        })
+    }
+}

data/ext/csv_utils/src/utils/shared/mod.rs

@@ -0,0 +1,4 @@
+pub mod datetime;
+pub mod filters;
+pub mod types;
+pub mod validation;

data/ext/csv_utils/src/utils/shared/types.rs

@@ -0,0 +1,97 @@
+use crate::utils::{missing_header, string_to_datetime};
+use chrono::NaiveDateTime;
+use magnus::{RArray, Ruby};
+use std::collections::HashMap;
+
+#[derive(Debug, PartialOrd, PartialEq, Eq, Ord)]
+pub struct MandatoryColumn<T> {
+    pub value: T,
+    pub index: usize,
+}
+
+impl<T> MandatoryColumn<T> {
+    pub fn new(value: T, index: usize) -> Self {
+        MandatoryColumn { value, index }
+    }
+}
+
+pub struct ProcessingConfig {
+    pub exclusions: Vec<String>,
+    pub status_exclusions: Vec<String>,
+    pub mandatory_headers: Vec<String>,
+    pub expected_trust_name: String,
+    pub is_streamed_file: bool,
+    pub earliest_start_date: Option<NaiveDateTime>,
+}
+
+impl ProcessingConfig {
+    pub fn from_ruby(
+        exclusions: RArray,
+        mandatory_headers: RArray,
+        status_exclusions: RArray,
+        expected_trust_name: String,
+        is_streamed_file: bool,
+        earliest_start_date: Option<String>,
+    ) -> magnus::error::Result<Self> {
+        Ok(ProcessingConfig {
+            exclusions: RArray::to_vec(exclusions)?,
+            status_exclusions: RArray::to_vec(status_exclusions)?,
+            mandatory_headers: RArray::to_vec(mandatory_headers)?,
+            expected_trust_name,
+            is_streamed_file,
+            earliest_start_date: earliest_start_date.and_then(|s| string_to_datetime(&s)),
+        })
+    }
+}
+
+pub struct HeaderConfig {
+    pub request_id: usize,
+    pub date: usize,
+    pub start: usize,
+    pub end: usize,
+    pub actual_start: usize,
+    pub actual_end: usize,
+    pub status: Option<usize>,
+    pub trust_name: usize,
+}
+
+impl HeaderConfig {
+    pub fn from_header_map(
+        map: &HashMap<String, usize>,
+        ruby: &Ruby,
+    ) -> magnus::error::Result<Self> {
+        Ok(HeaderConfig {
+            request_id: *map
+                .get("Request Id")
+                .ok_or_else(|| missing_header(ruby, "Request Id"))?,
+            date: *map
+                .get("Date")
+                .ok_or_else(|| missing_header(ruby, "Date"))?,
+            start: *map
+                .get("Start")
+                .ok_or_else(|| missing_header(ruby, "Start"))?,
+            end: *map.get("End").ok_or_else(|| missing_header(ruby, "End"))?,
+            actual_start: *map
+                .get("Actual Start")
+                .ok_or_else(|| missing_header(ruby, "Actual Start"))?,
+            actual_end: *map
+                .get("Actual End")
+                .ok_or_else(|| missing_header(ruby, "Actual End"))?,
+            status: map.get("Status").copied(),
+            trust_name: *map
+                .get("Trust")
+                .ok_or_else(|| missing_header(ruby, "Trust"))?,
+        })
+    }
+
+    pub fn is_time_column(&self, index: usize) -> bool {
+        index == self.start
+            || index == self.end
+            || index == self.actual_start
+            || index == self.actual_end
+    }
+
+    pub fn is_date_column(&self, index: usize) -> bool {
+        index == self.date
+    }
+}

data/ext/csv_utils/src/utils/shared/validation.rs

@@ -0,0 +1,34 @@
+use magnus::{Error, Ruby};
+
+pub struct TrustValidator {
+    expected_name: String,
+    is_streamed: bool,
+}
+
+impl TrustValidator {
+    pub fn new(expected_name: String, is_streamed: bool) -> Self {
+        TrustValidator {
+            expected_name,
+            is_streamed,
+        }
+    }
+
+    pub fn validate(&self, ruby: &Ruby, actual_name: &str) -> magnus::error::Result<()> {
+        if self.is_streamed {
+            return Ok(());
+        }
+
+        let trimmed = actual_name.trim();
+        if trimmed != self.expected_name {
+            return Err(Error::new(
+                ruby.exception_standard_error(),
+                format!(
+                    "Trust actual name: '{}' is not as expected: '{}'",
+                    trimmed, self.expected_name
+                ),
+            ));
+        }
+
+        Ok(())
+    }
+}