Sampling size argument support for text

README.md

@@ -84,6 +84,8 @@ Options:
           URL for image redact endpoint for MsPresidio redacter
       --gemini-model <GEMINI_MODEL>
           Gemini model name for Gemini LLM redacter. Default is 'models/gemini-1.5-flash'
+      --sampling-size <SAMPLING_SIZE>
+          Sampling size in bytes before redacting files. Disabled by default
   -h, --help
           Print help
 ```

src/args.rs

@@ -132,6 +132,12 @@ pub struct RedacterArgs {
         help = "Gemini model name for Gemini LLM redacter. Default is 'models/gemini-1.5-flash'"
     )]
     pub gemini_model: Option<GeminiLlmModelName>,
+
+    #[arg(
+        long,
+        help = "Sampling size in bytes before redacting files. Disabled by default"
+    )]
+    pub sampling_size: Option<usize>,
 }
 
 impl TryInto<RedacterOptions> for RedacterArgs {
@@ -189,6 +195,7 @@ impl TryInto<RedacterOptions> for RedacterArgs {
             allow_unsupported_copies: self.allow_unsupported_copies,
             csv_headers_disable: self.csv_headers_disable,
             csv_delimiter: self.csv_delimiter.map(|c| c as u8),
+            sampling_size: self.sampling_size,
         })
     }
 }

src/commands/copy_command.rs

@@ -43,20 +43,27 @@ pub async fn command_copy(
     redacter_options: Option<RedacterOptions>,
 ) -> AppResult<CopyCommandResult> {
     let bold_style = Style::new().bold();
-    let redacted_output = if let Some(ref options) = redacter_options {
+    let redacted_output = if let Some(ref options) = redacter_options.as_ref() {
         bold_style
             .clone()
             .green()
-            .apply_to(format!("✓ Yes ({})", options))
+            .apply_to(format!("✓ Yes ({})", &options))
     } else {
         bold_style.clone().red().apply_to("✗ No".to_string())
     };
+    let sampling_output =
+        if let Some(ref sampling_size) = redacter_options.as_ref().and_then(|o| o.sampling_size) {
+            Style::new().apply_to(format!("{} bytes.", sampling_size))
+        } else {
+            Style::new().dim().apply_to("-".to_string())
+        };
     term.write_line(
         format!(
-            "Copying from {} to {}.\nRedacting: {}.",
+            "Copying from {} to {}.\nRedacting: {}.\nSampling: {}\n",
             bold_style.clone().white().apply_to(source),
             bold_style.clone().yellow().apply_to(destination),
-            redacted_output
+            redacted_output,
+            sampling_output
         )
         .as_str(),
     )?;

src/redacters/aws_comprehend.rs

@@ -151,6 +151,7 @@ mod tests {
             allow_unsupported_copies: false,
             csv_headers_disable: false,
             csv_delimiter: None,
+            sampling_size: None,
         };
 
         let redacter = AwsComprehendRedacter::new(

src/redacters/gcp_dlp.rs

@@ -400,6 +400,7 @@ mod tests {
             allow_unsupported_copies: false,
             csv_headers_disable: false,
             csv_delimiter: None,
+            sampling_size: None,
         };
 
         let redacter = GcpDlpRedacter::new(

src/redacters/gemini_llm.rs

@@ -236,6 +236,7 @@ mod tests {
             allow_unsupported_copies: false,
             csv_headers_disable: false,
             csv_delimiter: None,
+            sampling_size: None,
         };
 
         let redacter = GeminiLlmRedacter::new(

src/redacters/mod.rs

@@ -52,6 +52,7 @@ pub struct RedacterOptions {
     pub allow_unsupported_copies: bool,
     pub csv_headers_disable: bool,
     pub csv_delimiter: Option<u8>,
+    pub sampling_size: Option<usize>,
 }
 
 #[derive(Debug, Clone)]
@@ -189,9 +190,19 @@ pub async fn redact_stream<
         {
             let all_chunks: Vec<bytes::Bytes> = input.try_collect().await?;
             let all_bytes = all_chunks.concat();
-            let content = String::from_utf8(all_bytes).map_err(|e| AppError::SystemError {
-                message: format!("Failed to convert bytes to string: {}", e),
-            })?;
+            let whole_content =
+                String::from_utf8(all_bytes).map_err(|e| AppError::SystemError {
+                    message: format!("Failed to convert bytes to string: {}", e),
+                })?;
+            let content = if let Some(sampling_size) = redacter.options().sampling_size {
+                let sampling_size = std::cmp::min(sampling_size, whole_content.len());
+                whole_content
+                    .chars()
+                    .take(sampling_size)
+                    .collect::<String>()
+            } else {
+                whole_content
+            };
             Ok(RedacterDataItem {
                 content: RedacterDataItemContent::Value(content),
                 file_ref: file_ref.clone(),

src/redacters/ms_presidio.rs

@@ -257,6 +257,7 @@ mod tests {
             allow_unsupported_copies: false,
             csv_headers_disable: false,
             csv_delimiter: None,
+            sampling_size: None,
         };
 
         let redacter = MsPresidioRedacter::new(