support not parsing some columns as int or float when saving to xlsx files.

Author: ribbondz

README.md

@@ -109,7 +109,7 @@ rsv clean --help                                 # help info on all flags
 
 ```shell
 rsv unique data.csv               # default to drop duplicates on all columns,
-                                    # default keep first record of duplicates
+                                  # default keep first record of duplicates
 rsv unique -c 0 data.csv          # drop on first column
 rsv unique -c 0,1 data.csv        # drop on first and second columns
 rsv unique --keep-last data.csv   # keep the last record when dropping
@@ -286,6 +286,14 @@ rsv slice -s 10 -e 15 data.csv | rsv table      # convert result to an aligned t
 rsv table --help                                # help info on all flags
 ```
 
+- **rsv to**
+
+```shell
+rsv head data.csv | rsv to data.xlsx                      # save to the data.xlsx file
+rsv head data.csv | rsv to data.xlsx --text-columns 0,1   # regard first two columns as texts
+rsv to --help                                             # help info on all flags
+```
+
 ## Command pipeline
 
 - **two commands pipelined**

rsv-lib/src/csv_lib/stats.rs

@@ -17,6 +17,7 @@ pub fn csv_stats(
     quote: char,
     no_header: bool,
     cols: String,
+    text_columns: &Vec<usize>,
 ) -> CliResultData {
     let mut result_data = ResultData::new();
     result_data.insert_header(CStat::get_fields.iter().map(|f| f.to_string()).collect());
@@ -25,7 +26,9 @@ pub fn csv_stats(
     let cols = Columns::new(cols.as_str())
         .total_col_of(file, sep, quote)
         .parse();
-    let Some(col_type) = ColumnTypes::guess_from_csv(file, sep, quote, no_header, &cols)? else {
+    let Some(col_type) =
+        ColumnTypes::guess_from_csv(file, sep, quote, no_header, &cols, text_columns)?
+    else {
         return Ok(Some(result_data));
     };
 

rsv-lib/src/lib.rs

@@ -58,10 +58,11 @@ pub fn file_stats(
     no_header: bool,
     cols: String,
     sheet: usize,
+    text_columns: &Vec<usize>,
 ) -> CliResultData {
     let path = full_path(file);
     match is_excel(&path) {
         true => excel_stats(&path, no_header, cols, sheet),
-        false => csv_stats(&path, sep, quote, no_header, cols),
+        false => csv_stats(&path, sep, quote, no_header, cols, text_columns),
     }
 }

rsv-lib/src/utils/column.rs

@@ -202,7 +202,7 @@ impl<'a> Columns<'a> {
             .collect::<Vec<_>>()
     }
 
-    pub fn col_vec_or_length_of(&self, n: usize) -> Vec<usize> {
+    pub fn col_index_vec(&self, n: usize) -> Vec<usize> {
         match self.select_all {
             true => (0..n).collect::<Vec<_>>(),
             false => self.cols.clone(),

rsv-lib/src/utils/column_type.rs

@@ -52,6 +52,7 @@ impl ColumnTypes {
         quote: char,
         no_header: bool,
         cols: &column::Columns,
+        text_columns: &Vec<usize>,
     ) -> Result<Option<Self>, Box<dyn Error>> {
         // reader
         let rdr = BufReader::new(File::open(path)?).lines();
@@ -72,9 +73,15 @@ impl ColumnTypes {
             .collect::<Vec<_>>();
 
         let guess = cols
-            .col_vec_or_length_of(lines[0].len())
+            .col_index_vec(lines[0].len())
             .into_par_iter()
-            .map(|n| (n, parse_col_type_at(n, &lines), max_length_at(n, &lines)))
+            .map(|n| {
+                if text_columns.contains(&n) {
+                    (n, ColumnType::String, max_length_at(n, &lines))
+                } else {
+                    (n, parse_col_type_at(n, &lines), max_length_at(n, &lines))
+                }
+            })
             .collect::<Vec<_>>()
             .iter()
             .fold(ColumnTypes(vec![]), |mut a, b| {
@@ -98,7 +105,7 @@ impl ColumnTypes {
         }
 
         let mut guess = ColumnTypes(vec![]);
-        for c in cols.col_vec_or_length_of(lines[0].len()) {
+        for c in cols.col_index_vec(lines[0].len()) {
             // max_length is meaningless for excel, so set default to 0
             guess.push(c, parse_excel_col_type_at(c, &lines), 0)
         }
@@ -107,12 +114,16 @@ impl ColumnTypes {
     }
 
     // sequential guess given that io is usually small
-    pub fn guess_from_io(v: &[Vec<&str>], cols: &Columns) -> Self {
+    pub fn guess_from_io(v: &[Vec<&str>], cols: &Columns, text_columns: &Vec<usize>) -> Self {
         let v = if v.len() < 5000 { v } else { &v[..5000] };
 
         let mut guess = ColumnTypes(vec![]);
-        for c in cols.col_vec_or_length_of(v[0].len()) {
-            guess.push(c, parse_col_type_at(c, v), max_length_at(c, v))
+        for c in cols.col_index_vec(v[0].len()) {
+            if text_columns.contains(&c) {
+                guess.push(c, ColumnType::String, max_length_at(c, v))
+            } else {
+                guess.push(c, parse_col_type_at(c, v), max_length_at(c, v))
+            }
         }
 
         guess

rsv-lib/src/utils/to.rs

@@ -2,7 +2,7 @@ use super::{cli_result::CliResult, filename::new_file, reader::ExcelReader, writ
 use crate::utils::{column::Columns, column_type::ColumnTypes, row_split::CsvRowSplitter};
 use std::{
     fs::File,
-    io::{stdin, BufRead, BufReader, BufWriter, Write},
+    io::{BufRead, BufReader, BufWriter, Write, stdin},
     path::{Path, PathBuf},
 };
 use xlsxwriter::{Workbook, Worksheet};
@@ -64,7 +64,14 @@ pub fn excel_to_csv(path: &Path, sheet: usize, sep: &str, out: &str) -> CliResul
     Ok(())
 }
 
-pub fn csv_to_excel(path: &Path, sep: char, quote: char, out: &str, no_header: bool) -> CliResult {
+pub fn csv_to_excel(
+    path: &Path,
+    sep: char,
+    quote: char,
+    out: &str,
+    no_header: bool,
+    text_columns: &Vec<usize>,
+) -> CliResult {
     // out path
     let out = out_filename(out);
 
@@ -75,10 +82,11 @@ pub fn csv_to_excel(path: &Path, sep: char, quote: char, out: &str, no_header: b
 
     // column type
     let cols = Columns::new("").total_col_of(path, sep, quote).parse();
-    let ctypes = match ColumnTypes::guess_from_csv(path, sep, quote, no_header, &cols)? {
-        Some(v) => v,
-        None => return Ok(()),
-    };
+    let ctypes =
+        match ColumnTypes::guess_from_csv(path, sep, quote, no_header, &cols, text_columns)? {
+            Some(v) => v,
+            None => return Ok(()),
+        };
     ctypes.update_excel_column_width(&mut sheet)?;
     let ctypes = Some(ctypes);
 
@@ -94,7 +102,13 @@ pub fn csv_to_excel(path: &Path, sep: char, quote: char, out: &str, no_header: b
     Ok(())
 }
 
-pub fn io_to_excel(sep: char, quote: char, no_header: bool, out: &str) -> CliResult {
+pub fn io_to_excel(
+    sep: char,
+    quote: char,
+    no_header: bool,
+    out: &str,
+    text_columns: &Vec<usize>,
+) -> CliResult {
     // out path
     let out = out_filename(out);
 
@@ -119,7 +133,8 @@ pub fn io_to_excel(sep: char, quote: char, no_header: bool, out: &str) -> CliRes
     let ctypes = if equal_width(&lines) {
         // column type
         let cols = Columns::new("").total_col(lines[0].len()).parse();
-        let ctypes = ColumnTypes::guess_from_io(&lines[(1 - no_header as usize)..], &cols);
+        let ctypes =
+            ColumnTypes::guess_from_io(&lines[(1 - no_header as usize)..], &cols, text_columns);
         ctypes.update_excel_column_width(&mut sheet)?;
         Some(ctypes)
     } else {

rsv/src/args.rs

@@ -251,6 +251,9 @@ pub struct Stats {
     /// Get the nth worksheet of EXCEL file
     #[arg(short = 'S', long, default_value_t = 0)]
     pub sheet: usize,
+    /// Comma-separated column indexes to format as text (e.g., "0,2")
+    #[arg(short, long, value_delimiter = ',', allow_hyphen_values = true)]
+    pub text_columns: Vec<usize>,
 }
 
 #[derive(Debug, Args)]
@@ -383,6 +386,9 @@ pub struct To {
     /// Get the nth worksheet of EXCEL file
     #[arg(short = 'S', long, default_value_t = 0)]
     pub sheet: usize,
+    /// Comma-separated column indexes to format as text (e.g., "0,2")
+    #[arg(short, long, value_delimiter = ',', allow_hyphen_values = true)]
+    pub text_columns: Vec<usize>,
 }
 
 #[derive(Debug, Args)]

rsv/src/cmd_desc.rs

@@ -323,15 +323,16 @@ Usage:
   rsv stats -c 0,1 --export data.xlsx      # EXCEL file
 
 Arguments:
-  <FILENAME>             File to open, including CSV, TXT, and EXCEL
+  <FILENAME>                         File to open, including CSV, TXT, and EXCEL
 
 Options:
-  -s, --sep <SEP>        Separator [default: ,]
-      --no-header        Whether the file has a header
-  -c, --cols <COLS>      Columns to generate statistics, Default to select all
-  -E, --export           Export results to a file named current-file-selected.csv
-  -S, --sheet <SHEET>    Get the nth worksheet of EXCEL file [default: 0]
-  -h, --help             Print help information
+  -s, --sep <SEP>                    Separator [default: ,]
+      --no-header                    Whether the file has a header
+  -c, --cols <COLS>                  Columns to generate statistics, Default to select all
+  -E, --export                       Export results to a file named current-file-selected.csv
+  -S, --sheet <SHEET>                Get the nth worksheet of EXCEL file [default: 0]
+  -t, --text-columns <TEXT_COLUMNS>  Comma-separated column indexes to format as text [default:]
+  -h, --help                         Print help information
 
 Column selection syntax:
   -c 0,1,2,5       -->   cols [0,1,2,5]
@@ -455,16 +456,17 @@ Usage:
   rsv head data | rsv to out.xlsx
 
 Arguments:
-  <OUT>                  Output file, a file name or a file format
-  [FILENAME]             File to open
+  <OUT>                              Output file, a file name or a file format
+  [FILENAME]                         File to open
 
 Options:
-      --no-header        Whether the file has a header
-  -s, --sep <SEP>        Input file Separator [default: ,]
-  -q, --quote <QUOTE>    Quote char [default: "]
-  -o, --outsep <OUTSEP>  Output file Separator [default: ,]
-  -S, --sheet <SHEET>    Get the nth worksheet of EXCEL file [default: 0]
-  -h, --help             Print help
+      --no-header                    Whether the file has a header
+  -s, --sep <SEP>                    Input file Separator [default: ,]
+  -q, --quote <QUOTE>                Quote char [default: "]
+  -o, --outsep <OUTSEP>              Output file Separator [default: ,]
+  -S, --sheet <SHEET>                Get the nth worksheet of EXCEL file [default: 0]
+  -t, --text-columns <TEXT_COLUMNS>  Comma-separated column indexes to format as text [default:]
+  -h, --help                         Print help
 "#;
 
 pub const SAMPLE_DESC: &str = r#"

rsv/src/csv/stats.rs

@@ -20,8 +20,14 @@ impl Stats {
         let cols = Columns::new(&self.cols)
             .total_col_of(path, self.sep, self.quote)
             .parse();
-        let Some(col_type) =
-            ColumnTypes::guess_from_csv(path, self.sep, self.quote, self.no_header, &cols)?
+        let Some(col_type) = ColumnTypes::guess_from_csv(
+            path,
+            self.sep,
+            self.quote,
+            self.no_header,
+            &cols,
+            &self.text_columns,
+        )?
         else {
             return Ok(());
         };

rsv/src/csv/to.rs

@@ -9,9 +9,14 @@ impl To {
 
         match out.as_str() {
             v if is_valid_plain_text(v) => csv_or_io_to_csv(Some(path), &out)?,
-            v if is_valid_excel(v) => {
-                csv_to_excel(path, self.sep, self.quote, &out, self.no_header)?
-            }
+            v if is_valid_excel(v) => csv_to_excel(
+                path,
+                self.sep,
+                self.quote,
+                &out,
+                self.no_header,
+                &self.text_columns,
+            )?,
             _ => return Err(format!("output file format <{out}> is un-recognized.").into()),
         };