Preparing for release

CHANGELOG.md

@@ -8,32 +8,36 @@ and this project adheres to [Semantic Versioning].
 ## [Unreleased]
 
 ### Added
-- Limited async MzML reader support
+
+- Limited async mzML reader support
 - mzMLb read and write support
+- Reading mzML and MGF from `STDIN`. HDF5, and ergo mzMLb is not supported on non-seek-able I/O devices. See the `from_stdin` example.
+- Parsing of mzML `run` and `spectrumList` metadata, although they are still not a part of the common data model
+- Spectrum averaging now has eager `averaging` and `averaging_deferred` adapter implementations as iterator adapters on `SpectrumGroupIterator`.
+  The deferred adapter is preferred for distributing the process with `rayon`. See the `averaging_writer` example.
+- Added ordered parallel iteration collation with `mzdata::spectrum::Collator` to make consuming `rayon` iterators easier while preserving the
+  original order. See `averaging_writer` example.
+- The mzML and mzMLb writers now write the total ion chromatogram and base peak chromatogram
 
 ### Changed
+
 - `RandomAccessScanIterator` methods now return mutable references to self, making them actually useful in a chain.
 - Make some window size attributes smaller as they do not require double precision.
 - Clean up the internal implementation of the various internal `SpectrumBuilder` types.
-- Factor up `mzdata::spectrum::signal` to be less monolithic.
-
-### Deprecated
-
-### Removed
-
-### Fixed
-
-### Security
-
+- Factor up `mzdata::spectrum::signal` to be less monolithic and a complete redesign of the traits used to convert `mzpeaks` to and from binary arrays.
+- Massive refactoring of `mzdata::io::traits` to make more traits depend upon `ScanSource` instead of `SpectrumIterator` and to make things slightly less verbose.
+- Switched the default `mzsignal` backend to `nalgebra` instead of `intel-mkl` for simplicity.
 
 ## [0.5.0] - 2021-09-22
 
 ### Added
+
 - MzML writing via `mzdata::io::mzml::MzMLWriter`
-- Added feature flags to allow the user to choose amongst more `flate2` backends (zlib *default*, zlib-ng-compat, miniz_oxide)
+- Added feature flags to allow the user to choose amongst more `flate2` backends (zlib _default_, zlib-ng-compat, miniz_oxide)
 - Grouped iteration mode for connecting precursor and product spectra over an iterator stream using the `groups` method of `ScanSource`.
 
 ### Changed
+
 - Re-structuring and renaming of the various iterator mechanisms for more
   consistency. `ScanIterator` -> `SpectrumIterator`, et cetera. Minor refactoring
   of this sort expected to come for `ScanSource` as responsibilities are worked out.
@@ -43,16 +47,18 @@ and this project adheres to [Semantic Versioning].
 ### Removed
 
 ### Fixed
+
 - Fixed documentation in several places, particularly where it was substantially out of date.
 
 ### Security
 
-
 <!-- Links -->
+
 [keep a changelog]: https://keepachangelog.com/en/1.0.0/
 [semantic versioning]: https://semver.org/spec/v2.0.0.html
 
 <!-- Versions -->
+
 [unreleased]: https://github.com/mobiusklein/mzdata/compare/v0.5.0...HEAD
 [0.5.0]: https://github.com/mobiusklein/mzdata/compare/v0.1.0...v0.5.0
-[0.1.0]: https://github.com/mobiusklein/mzdata/releases/tag/v0.1.0
+[0.1.0]: https://github.com/mobiusklein/mzdata/releases/tag/v0.1.0

Cargo.toml

@@ -64,8 +64,8 @@ async = ["tokio", "quick-xml/async-tokio"]
 [dependencies]
 regex = "1"
 lazy_static = "1.4.0"
-serde = { version = "1.0.126", features = ["derive"] }
-serde_json = "1.0.64"
+serde = { version = "1.0.193", features = ["derive"] }
+serde_json = "1.0.108"
 quick-xml = { version = "0.30", features = [ "serialize" ] }
 base64 = "0.21.3"
 flate2 = {version = "1.0.20"}

README.md

@@ -9,19 +9,31 @@ use std::fs;
 use mzdata::prelude::*;
 use mzpeaks::{Tolerance, prelude::*};
 use mzdata::io::MzMLReader;
-use mzdata::spectrum::{SignalContinuity};
-
-let reader = MzMLReader::new(fs::File::open("./test/data/small.mzML").unwrap());
-for spectrum in reader {
-    println!("Scan {} => BP {}", spectrum.id(), spectrum.peaks().base_peak().mz);
-    if spectrum.signal_continuity() < SignalContinuity::Profile {
-        let peak_picked = spectrum.into_centroid().unwrap();
-        println!("Matches for 579.155: {:?}", peak_picked.peaks.all_peaks_for(579.155, Tolerance::Da(0.02)));
+use mzdata::spectrum::SignalContinuity;
+
+
+fn main() {
+    let mut ms1_count = 0;
+    let mut msn_count = 0;
+    let reader = MzMLReader::new(fs::File::open("./test/data/small.mzML").unwrap());
+    for spectrum in reader {
+        if spectrum.ms_level() == 1 {
+            ms1_count += 1;
+        } else {
+            msn_count += 1;
+        }
+        println!("Scan {} => BP {}", spectrum.id(), spectrum.peaks().base_peak().mz);
+        if spectrum.signal_continuity() < SignalContinuity::Profile {
+            let peak_picked = spectrum.into_centroid().unwrap();
+            println!("Matches for 579.155: {:?}", peak_picked.peaks.all_peaks_for(579.155, Tolerance::Da(0.02)));
+        }
     }
+    println!("MS1 Count: {}\nMSn Count: {}", ms1_count, msn_count);
+    assert_eq!(ms1_count, 14);
+    assert_eq!(msn_count, 34);
 }
-println!("MS1 Count: {}\nMSn Count: {}", ms1_count, msn_count);
-assert_eq!(ms1_count, 14);
-assert_eq!(msn_count, 34);
+
+
 ```
 
 ### Supported Formats

examples/readme.rs

@@ -0,0 +1,28 @@
+use std::fs;
+use mzdata::prelude::*;
+use mzpeaks::{Tolerance, prelude::*};
+use mzdata::io::MzMLReader;
+use mzdata::spectrum::SignalContinuity;
+
+
+fn main() {
+    let mut ms1_count = 0;
+    let mut msn_count = 0;
+    let reader = MzMLReader::new(fs::File::open("./test/data/small.mzML").unwrap());
+    for spectrum in reader {
+        if spectrum.ms_level() == 1 {
+            ms1_count += 1;
+        } else {
+            msn_count += 1;
+        }
+        println!("Scan {} => BP {}", spectrum.id(), spectrum.peaks().base_peak().mz);
+        if spectrum.signal_continuity() < SignalContinuity::Profile {
+            let peak_picked = spectrum.into_centroid().unwrap();
+            println!("Matches for 579.155: {:?}", peak_picked.peaks.all_peaks_for(579.155, Tolerance::Da(0.02)));
+        }
+    }
+    println!("MS1 Count: {}\nMSn Count: {}", ms1_count, msn_count);
+    assert_eq!(ms1_count, 14);
+    assert_eq!(msn_count, 34);
+}
+

src/io.rs

@@ -1,21 +1,25 @@
+mod infer_format;
 pub mod mgf;
 pub mod mzml;
 #[cfg(feature = "mzmlb")]
 pub mod mzmlb;
 mod offset_index;
 pub mod traits;
 mod utils;
-mod infer_format;
 
 pub(crate) mod compression;
 
-pub use crate::io::utils::{DetailLevel, PreBufferedStream};
-pub use crate::io::mgf::{MGFReader, MGFError, MGFWriter};
-pub use crate::io::mzml::{MzMLReader, MzMLParserError, MzMLWriter};
+pub use crate::io::infer_format::{
+    infer_format, infer_from_path, infer_from_stream, open_file, MassSpectrometryFormat,
+};
+pub use crate::io::mgf::{MGFError, MGFReader, MGFWriter};
 #[cfg(feature = "async")]
 pub use crate::io::mzml::AsyncMzMLReaderType;
+pub use crate::io::mzml::{MzMLParserError, MzMLReader, MzMLWriter};
 #[cfg(feature = "mzmlb")]
-pub use crate::io::mzmlb::{MzMLbReader, MzMLbError};
+pub use crate::io::mzmlb::{MzMLbError, MzMLbReader};
 pub use crate::io::offset_index::OffsetIndex;
-pub use crate::io::traits::{RandomAccessSpectrumIterator, ScanAccessError, SpectrumIterator, ScanSource};
-pub use crate::io::infer_format::{open_file, infer_format, infer_from_path, MassSpectrometryFormat};
+pub use crate::io::traits::{
+    RandomAccessSpectrumIterator, ScanAccessError, ScanSource, SpectrumIterator,
+};
+pub use crate::io::utils::{DetailLevel, PreBufferedStream};

src/io/infer_format.rs

@@ -62,6 +62,7 @@ pub fn infer_from_path<P: Into<path::PathBuf>,>(path: P) -> (MassSpectrometryFor
 /// stream is GZIP compressed. This assumes the stream is seekable.
 pub fn infer_from_stream<R: Read + Seek>(stream: &mut R) -> io::Result<(MassSpectrometryFormat, bool)> {
     let mut buf = Vec::with_capacity(100);
+    buf.resize(100, b'\0');
     let current_pos = stream.stream_position()?;
     stream.read_exact(buf.as_mut_slice())?;
     let is_stream_gzipped = is_gzipped(buf.as_slice());

src/io/mgf.rs

@@ -19,7 +19,8 @@ use thiserror::Error;
 
 use lazy_static::lazy_static;
 use mzpeaks::{
-    CentroidPeak, DeconvolutedPeak, IntensityMeasurement, MZLocated, PeakCollection, peak::KnownCharge
+    peak::KnownCharge, CentroidPeak, DeconvolutedPeak, IntensityMeasurement, MZLocated,
+    PeakCollection,
 };
 use regex::Regex;
 
@@ -42,7 +43,7 @@ use crate::spectrum::spectrum::{
 use crate::spectrum::PeakDataLevel;
 use crate::spectrum::SignalContinuity;
 use crate::spectrum::{
-    Precursor, PrecursorSelection, SelectedIon, SpectrumLike, SpectrumDescription,
+    Precursor, PrecursorSelection, SelectedIon, SpectrumDescription, SpectrumLike,
 };
 use crate::utils::neutral_mass;
 
@@ -273,7 +274,7 @@ impl<
             let (key, value) = line.split_once('=').unwrap();
 
             match key {
-                "TITLE" => builder.description.id = String::from(value),
+                "TITLE" => builder.description.id = value.to_string(),
                 "RTINSECONDS" => {
                     let scan_ev = builder
                         .description
@@ -285,7 +286,9 @@ impl<
                 "PEPMASS" => {
                     let mut parts = value.split_ascii_whitespace();
                     let mz: f64 = parts.next().unwrap().parse().unwrap();
-                    let intensity: f32 = parts.next().unwrap().parse().unwrap();
+                    let intensity: f32 = parts.next().and_then(|v| {
+                        Some(v.parse().unwrap())
+                    }).unwrap_or_default();
                     let charge: Option<i32> = parts.next().map(|c| c.parse().unwrap());
                     builder.description.precursor = Some(Precursor {
                         ion: SelectedIon {
@@ -672,6 +675,14 @@ impl<
     > MSDataFileMetadata for MGFReaderType<R, C, D>
 {
     crate::impl_metadata_trait!();
+
+    fn spectrum_count_hint(&self) -> Option<u64> {
+        if self.index.init {
+            Some(self.index.len() as u64)
+        } else {
+            None
+        }
+    }
 }
 
 pub type MGFReader<R> = MGFReaderType<R, CentroidPeak, DeconvolutedPeak>;
@@ -834,10 +845,7 @@ TITLE="#,
         Ok(())
     }
 
-    pub fn write<S: SpectrumLike<C, D> + 'static>(
-        &mut self,
-        spectrum: &S,
-    ) -> io::Result<usize> {
+    pub fn write<S: SpectrumLike<C, D> + 'static>(&mut self, spectrum: &S) -> io::Result<usize> {
         let description = spectrum.description();
         if description.ms_level == 1 {
             log::warn!(