fix: incorrect filenames in output introduced in 06be420

lazear · Sep 19, 2023 · b04bfbe · b04bfbe
1 parent 51001ad
commit b04bfbe
Show file tree

Hide file tree

Showing 7 changed files with 30 additions and 15 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -4,8 +4,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [v0.14.1-alpha.1]
+### Fixed
+- Fix bug introduced by 06be420 where `file_id` was not being propagated into scans, resulting in only the first filename being written to output files
+
 ## [v0.14.1-alpha]
-## Added
+### Added
 - Added additional output showing search progress if `SAGE_LOG=trace` environment variable is set
 - Added additional warnings about precursor tolerances
 - Added configuration option `precursor_charge` to make it explicit what charge states are being searched in the case where the mzML does not contain charge state information, or where `wide_window` is turned on.

diff --git a/Cargo.lock b/Cargo.lock
diff --git a/crates/sage-cli/Cargo.toml b/crates/sage-cli/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "sage-cli"
-version = "0.14.1-alpha"
+version = "0.14.1-alpha.1"
 authors = ["Michael Lazear <michaellazear92@gmail.com"]
 edition = "2021"
 rust-version = "1.62"

diff --git a/crates/sage-cloudpath/Cargo.toml b/crates/sage-cloudpath/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "sage-cloudpath"
-version = "0.14.1-alpha"
+version = "0.14.1-alpha.1"
 authors = ["Michael Lazear <michaellazear92@gmail.com"]
 edition = "2021"
 rust-version = "1.62"

diff --git a/crates/sage-cloudpath/src/mzml.rs b/crates/sage-cloudpath/src/mzml.rs
@@ -57,7 +57,6 @@ const SELECTED_ION_CHARGE: &[u8] = b"MS:1000041";
 const ISO_WINDOW_LOWER: &[u8] = b"MS:1000828";
 const ISO_WINDOW_UPPER: &[u8] = b"MS:1000829";
 
-#[derive(Default)]
 pub struct MzMLReader {
     ms_level: Option<u8>,
     // If set to Some(level) and noise intensities are present in the MzML file,
@@ -73,10 +72,10 @@ impl MzMLReader {
     /// # Example
     ///
     /// A minimum level of 2 will not parse or return MS1 scans
-    pub fn with_level_filter(ms_level: u8) -> Self {
+    pub fn with_file_id_and_level_filter(file_id: usize, ms_level: u8) -> Self {
         Self {
             ms_level: Some(ms_level),
-            file_id: 0,
+            file_id,
             signal_to_noise: None,
         }
     }
@@ -115,7 +114,7 @@ impl MzMLReader {
         let mut binary_dtype = Dtype::F64;
         let mut binary_array = None;
 
-        let mut spectrum = RawSpectrum::default();
+        let mut spectrum = RawSpectrum::default_with_file_id(self.file_id);
         let mut precursor = Precursor::default();
         let mut iso_window_lo: Option<f32> = None;
         let mut iso_window_hi: Option<f32> = None;
@@ -194,7 +193,7 @@ impl MzMLReader {
                                 let level = extract_value!(ev);
                                 if let Some(filter) = self.ms_level {
                                     if level != filter {
-                                        spectrum = RawSpectrum::default();
+                                        spectrum = RawSpectrum::default_with_file_id(self.file_id);
                                         state = None;
                                     }
                                 }
@@ -206,7 +205,7 @@ impl MzMLReader {
                                 let value = extract_value!(ev);
                                 if value == 0.0 {
                                     // No ion current, break out of current state
-                                    spectrum = RawSpectrum::default();
+                                    spectrum = RawSpectrum::default_with_file_id(self.file_id);
                                     state = None;
                                 } else {
                                     spectrum.total_ion_current = value;
@@ -357,7 +356,7 @@ impl MzMLReader {
                                 }
                                 (false, _) => {}
                             }
-                            spectrum = RawSpectrum::default();
+                            spectrum = RawSpectrum::default_with_file_id(self.file_id);
                             None
                         }
                         _ => state,
@@ -464,7 +463,7 @@ mod test {
             </binaryDataArrayList>
         </spectrum>
         "#;
-        let mut spectra = MzMLReader::default().parse(s.as_bytes()).await?;
+        let mut spectra = MzMLReader::with_file_id(0).parse(s.as_bytes()).await?;
 
         assert_eq!(spectra.len(), 1);
         let s = spectra.pop().unwrap();

diff --git a/crates/sage/Cargo.toml b/crates/sage/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "sage-core"
-version = "0.14.1-alpha"
+version = "0.14.1-alpha.1"
 authors = ["Michael Lazear <michaellazear92@gmail.com"]
 edition = "2021"
 rust-version = "1.62"

diff --git a/crates/sage/src/spectrum.rs b/crates/sage/src/spectrum.rs
@@ -75,6 +75,7 @@ pub struct ProcessedSpectrum {
 
 #[derive(Default, Debug, Clone)]
 /// An unprocessed mass spectrum, as returned by a parser
+/// *CRITICAL*: Users must set all fields manually, including `file_id`
 pub struct RawSpectrum {
     pub file_id: usize,
     /// MSn level
@@ -97,6 +98,17 @@ pub struct RawSpectrum {
     pub intensity: Vec<f32>,
 }
 
+impl RawSpectrum {
+    /// Return a [`RawSpectrum`] with default values, but with the `file_id` field
+    /// properly set
+    pub fn default_with_file_id(file_id: usize) -> Self {
+        Self {
+            file_id,
+            ..Default::default()
+        }
+    }
+}
+
 #[derive(Copy, Clone, Debug, PartialEq, Eq, PartialOrd, Ord)]
 pub enum Representation {
     Profile,