src/localize/datetime.rs

@@ -9,6 +9,11 @@ use std::path::Path;
 static FILE_NAME_REGEX: Lazy<Regex> =
     Lazy::new(|| Regex::new(r"^(?P<date>\d{8})(?:-(?P<time>\d{4,6}))?.+\.md$").unwrap());
 
+/// Regex for extracting a file-type prefix from file names.
+/// Matches filenames like `20260412-123456_daily.md` or `20260412_daily Some Title.md`.
+static FILE_TYPE_REGEX: Lazy<Regex> =
+    Lazy::new(|| Regex::new(r"^\d{8}(?:-\d{4,6})?_([a-zA-Z0-9]+)").unwrap());
+
 /// Regex for validating datetime marker format (14 digits).
 static DATETIME_MARKER_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new(r"^\d{14}$").unwrap());
 
@@ -62,6 +67,28 @@ pub fn extract_datetime_from_file_name(file_name: &str, tz: Tz) -> Option<DateTi
         .and_then(|dt| naive_to_utc(dt, tz))
 }
 
+/// Extract the file-type prefix from a filename.
+///
+/// Filenames with a `_prefix` segment after the timestamp (and optional time component)
+/// are recognised. The prefix must consist of alphanumeric characters only.
+///
+/// # Examples
+/// - `"20260412-123456_daily.md"` → `Some("daily")`
+/// - `"20260412_daily Some Title.md"` → `Some("daily")`
+/// - `"20260412-123456 Some Title.md"` → `None`
+/// - `"/path/to/20260412-123456_daily.md"` → `Some("daily")`
+pub fn extract_file_type_from_file_name(file_name: &str) -> Option<String> {
+    let base_name = Path::new(file_name)
+        .file_name()
+        .and_then(|s| s.to_str())
+        .unwrap_or(file_name);
+
+    FILE_TYPE_REGEX
+        .captures(base_name)
+        .and_then(|c| c.get(1))
+        .map(|m| m.as_str().to_string())
+}
+
 /// Parse a 14-digit marker string as a NaiveDateTime without timezone conversion.
 fn parse_naive_datetime_from_marker(marker: &str) -> Option<NaiveDateTime> {
     if !DATETIME_MARKER_REGEX.is_match(marker) {
@@ -155,6 +182,51 @@ mod tests {
     use chrono::TimeZone;
     use chrono_tz::UTC;
 
+    #[test]
+    fn test_extract_file_type_with_time() {
+        assert_eq!(
+            extract_file_type_from_file_name("20260412-123456_daily.md"),
+            Some("daily".to_string())
+        );
+    }
+
+    #[test]
+    fn test_extract_file_type_with_time_and_title() {
+        assert_eq!(
+            extract_file_type_from_file_name("20260412-123456_daily Some Title.md"),
+            Some("daily".to_string())
+        );
+    }
+
+    #[test]
+    fn test_extract_file_type_without_time() {
+        assert_eq!(
+            extract_file_type_from_file_name("20260412_daily.md"),
+            Some("daily".to_string())
+        );
+    }
+
+    #[test]
+    fn test_extract_file_type_without_prefix() {
+        assert_eq!(
+            extract_file_type_from_file_name("20260412-123456 Some Title.md"),
+            None
+        );
+    }
+
+    #[test]
+    fn test_extract_file_type_with_full_path() {
+        assert_eq!(
+            extract_file_type_from_file_name("/path/to/20260412-123456_daily.md"),
+            Some("daily".to_string())
+        );
+    }
+
+    #[test]
+    fn test_extract_file_type_no_timestamp() {
+        assert_eq!(extract_file_type_from_file_name("notes.md"), None);
+    }
+
     #[test]
     fn test_extract_date_from_file_name_valid() {
         let file_name = "20230101-123456 Some Text.md";

src/localize/mod.rs

@@ -9,7 +9,7 @@ pub use configuration::{
 };
 pub use datetime::{
     extract_date_from_marker, extract_datetime_from_file_name, extract_datetime_from_marker,
-    extract_datetime_from_marker_list, extract_time_from_marker,
+    extract_datetime_from_marker_list, extract_file_type_from_file_name, extract_time_from_marker,
 };
 pub use preconfigured::TaskConfiguration;
 pub use shard::{localize_shard, localize_stream_file};

src/localize/preconfigured.rs

@@ -20,6 +20,12 @@ pub static TaskConfiguration: Lazy<RepositoryConfiguration> = Lazy::new(|| {
                 .with_comment("Project the task is attached to")
                 .with_propagate(true),
         )
+        .with_dimension(
+            "file_type",
+            Dimension::new("File Type")
+                .with_comment("Type of file derived from filename prefix (e.g. 'daily')")
+                .with_propagate(true),
+        )
         .with_marker(
             "Task",
             Marker::new("Task").with_placements(vec![

src/localize/shard.rs

@@ -5,7 +5,10 @@ use indexmap::{IndexMap, IndexSet};
 use crate::error::StreamdError;
 use crate::models::{LocalizedShard, RepositoryConfiguration, Shard, StreamFile};
 
-use super::datetime::{extract_datetime_from_file_name, extract_datetime_from_marker_list};
+use super::datetime::{
+    extract_datetime_from_file_name, extract_datetime_from_marker_list,
+    extract_file_type_from_file_name,
+};
 
 /// Localize a shard within the repository's coordinate system.
 ///
@@ -102,6 +105,9 @@ pub fn localize_stream_file(
 
     let mut initial_location = IndexMap::new();
     initial_location.insert("file".to_string(), stream_file.file_name.clone());
+    if let Some(file_type) = extract_file_type_from_file_name(&stream_file.file_name) {
+        initial_location.insert("file_type".to_string(), file_type);
+    }
 
     Ok(localize_shard(
         shard,