Substantial coverage, consolidation, and dependency upgrades (#18)

feature: add `ParamDescribedRead` read-only variant `ParamDescribed`

chore: needed field visibility

* chore: substantially clean up mzML parsing and improve code coverage

* chore: exercise `ThermoRawReader`

feature: add `Any` bound to `MassSpectrometryReadWriteProcess` for better customization, add `MZReaderType::Unknown` for boxing

* chore: address rewiring

fix: fix MGF indexing

fix: fix `is_thermo_raw_prefix`, which used the wrong number of trailing bytes

* chore: keep IonMobilityFrameLike consistent with SpectrumLike

change: update `mzpeaks` v1.0.0 and `mzsignal` v1.0.0

change: require `ParamDescribed` for `ChromatogramLike`

fix: Fix `DataArray::store_as`

* chore: replace recurring `get_param_by_curie` wrapper methods with `find_param_method` macro

* chore: update CV and documentation

* chore: documentation

change: update `thermorawfilereader` to v0.3.1 and fix tracking of trailer values

feature: add `mzsignal` to the exported top-level API

* chore: minor cleaning of some examples

feature: add `MZReaderBuilder` to the top-level API

* chore: feature-gate new tests using `mzsignal`

* chore: fix dependency/feature warning after denoting `tokio` as a dependency only

* chore: fix documentation warnings

Cargo.toml

@@ -101,13 +101,13 @@ chrono = "0.4.37"
 bitflags = "2.5.0"
 identity-hash = "0.1.0"
 thiserror = "2.0.2"
-mzpeaks = { version = ">=0.23.0,<1.0.0" }
+mzpeaks = { version = ">=1.0.0,<1.1.0" }
 
 # Internal parallelism
 rayon = { version = ">=1.8.0,<2.0", optional = true }
 
 # Internal signal processing
-mzsignal = { version = ">=0.27.0,<1.0.0", default-features = false, optional = true, features = [
+mzsignal = { version = ">=1.0.0,<1.1.0", default-features = false, optional = true, features = [
     'avx',
 ] }
 
@@ -123,7 +123,7 @@ bytemuck = { version = "1.18.0", features = ["extern_crate_alloc"] }
 base64-simd = "0.8.0"
 
 # Async reader features
-tokio = { version = "1.32.0", optional = true, features = [
+tokio = { version = "1.32", optional = true, features = [
     "macros",
     "rt",
     "fs",
@@ -144,7 +144,7 @@ reqwest = { version = "0.12", features = ["json", "blocking"], optional = true }
 futures = { version = "0.3", optional = true }
 
 # Thermo RAW-related features
-thermorawfilereader = { version = "0.3.0", default-features = false, optional = true }
+thermorawfilereader = { version = "0.3.1", default-features = false, optional = true }
 
 # Bruker TDF-related features
 rusqlite = { version = "0.31.0", optional = true }

Justfile

@@ -4,7 +4,7 @@ test-units:
     cargo nextest run --lib --features nalgebra,parallelism,mzsignal,zlib-ng-compat,thermo
 
 test-coverage:
-    cargo llvm-cov --lib --tests --features nalgebra,parallelism,mzsignal,zlib-ng-compat,thermo --html
+    cargo llvm-cov --lib --tests nextest --features nalgebra,parallelism,mzsignal,zlib-ng-compat,thermo,mzmlb --html
 
 alias t := test-units
 
@@ -15,7 +15,7 @@ quick-docs:
     cargo doc --no-deps -p mzdata
 
 docs:
-    cargo doc --no-deps --features nalgebra,parallelism,mzsignal,mzmlb,zlib-ng-compat,thermo,async -p mzdata
+    cargo doc --no-deps --features nalgebra,parallelism,mzsignal,mzmlb,zlib-ng-compat,thermo,async,proxi -p mzdata -p mzsignal -p mzpeaks
 
 install-mzdata:
     cargo install --path . --features nalgebra,parallelism,mzsignal,mzmlb,zlib-ng-compat,hdf5_static

cv/extract_component.py

@@ -9,7 +9,7 @@
 from typing import Tuple, Dict, Set, List
 
 import fastobo
-from fastobo.term import TermFrame, IsAClause, NameClause, RelationshipClause
+from fastobo.term import TermFrame, IsAClause, NameClause, RelationshipClause, DefClause
 
 from fastobo.doc import OboDoc
 
@@ -113,6 +113,7 @@ def find_name(term: TermFrame):
 def make_entry_for(term: TermFrame):
     name = None
     flags = ValueType.NoType
+    descr = ""
     parents = []
     for clause in term:
         if isinstance(clause, NameClause):
@@ -122,6 +123,12 @@ def make_entry_for(term: TermFrame):
         if isinstance(clause, RelationshipClause):
             if str(clause.typedef) == 'has_value_type':
                 flags |= xsd_to_type[str(clause.term)]
+        if isinstance(clause, DefClause):
+            descr = re.sub(
+                r"(\[|\])",
+                lambda m: "\\\\" + m.group(1),
+                str(clause.definition).replace('"', "'"),
+            )
 
     vname = name
     if "-" in vname:
@@ -145,6 +152,7 @@ def make_entry_for(term: TermFrame):
 
     return f"""
     #[term(cv=MS, accession={term.id.local}, name="{name}", flags={{{int(flags)}}}, parents={{{json.dumps(parents)}}})]
+    #[doc="{name} - {descr}"]
     {vname},"""
 
 

cv/extract_software.py

@@ -12,6 +12,7 @@
     TermFrame,
     IsAClause,
     NameClause,
+    DefClause,
 )
 
 from fastobo.doc import OboDoc
@@ -52,6 +53,7 @@ def make_entry_for(term: TermFrame):
     name = None
     flags = SoftwareType.NoType
     parents = []
+    descr = ''
     for clause in term:
         if isinstance(clause, NameClause):
             name = str(clause.name)
@@ -63,6 +65,12 @@ def make_entry_for(term: TermFrame):
                 flags |= SoftwareType.Analysis
             elif clause.term == ACQUISITION_SW:
                 flags |= SoftwareType.Acquisition
+        if isinstance(clause, DefClause):
+            descr = re.sub(
+                r"(\[|\])",
+                lambda m: "\\\\" + m.group(1),
+                str(clause.definition).replace('"', "'"),
+            )
 
     vname: str = name
     if "-" in vname:
@@ -84,6 +92,7 @@ def make_entry_for(term: TermFrame):
 
     return f"""
     #[term(cv=MS, accession={term.id.local}, name="{name}", flags={{{int(flags)}}}, parents={{{json.dumps(parents)}}})]
+    #[doc="{name} - {descr}"]
     {vname},"""
 
 

docs/reader_tutorial.md

@@ -31,7 +31,7 @@ whether it is random access, or what format it is stored in.
 
 ### From a file path
 
-`mzdata` can read files on disk, accessed by path, easily. In this example we'll use the [`MZReader`](crate::io::MZReader) type
+`mzdata` can read files on disk, accessed by path, easily. In this example we'll use the [`MZReader`] type
 to figure out which reader to use for us automatically.
 
 ```rust
@@ -50,7 +50,7 @@ fn from_path() -> io::Result<()> {
 
 When reading a file from disk, `mzdata` can make certain assumptions like that the file supports
 the [`io::Seek`](std::io::Seek) trait and can read or build indices over the file quickly and guarantee
-that the file supports full random access, like [`RandomAccessSpectrumIterator`](crate::io::traits::RandomAccessSpectrumIterator).
+that the file supports full random access, like [`RandomAccessSpectrumIterator`].
 
 Additionally, some binary formats like [`ThermoRawReader`] or [`MzMLbReader`] _require_ that there be a file on disk that
 exists outside of the Rust model of the file system in order to read it.
@@ -157,7 +157,7 @@ you want to use.
 
 All of the added complexity introduced by the type system can make anything that is flexible over how you come
 to open a mass spectrometry data source cumbersome. When you don't _need_ to keep the reader around beyond the
-current scope, the [`mz_read!`](crate::mz_read) macro can substantially simplify matters. It is like [`MZReader`](crate::MZReader),
+current scope, the [`mz_read!`](crate::mz_read) macro can substantially simplify matters. It is like [`MZReader`],
 but it is even more flexible, provided that the reader instance only lives as long as the enclosing scope:
 
 ```rust

docs/spectrum_tutorial.md

@@ -102,7 +102,7 @@ appropriate.
 #    let mut spectrum = reader.get_spectrum_by_index(15).unwrap();
     // An error might occur during reprofiling, or there may not be
     // any peaks to transform.
-    spectrum.reprofile_with_shape(0.025, 0.015).unwrap();
+    spectrum.reprofile_with_shape(0.0025, 0.015).unwrap();
     spectrum.description_mut().signal_continuity = SignalContinuity::Profile;
 #    Ok(())
 # }

examples/async_mzcat.rs

@@ -1,6 +1,7 @@
 use std::time;
 use std::{env, io, path};
 
+use futures::StreamExt;
 use tokio;
 use tokio::fs;
 
@@ -21,9 +22,11 @@ async fn load_file<P: Into<path::PathBuf> + Clone>(
 
 async fn scan_file(reader: &mut mzml::AsyncMzMLReader<fs::File>) {
     let start = time::Instant::now();
-    let n = reader.len();
     let mut i = 0;
-    while let Some(scan) = reader.get_spectrum_by_index(i).await {
+
+    let mut stream = reader.as_stream();
+
+    while let Some(scan) = stream.next().await {
         if i % 10000 == 0 {
             println!(
                 "\tScan {}: {}|{} ({} seconds)",
@@ -34,9 +37,6 @@ async fn scan_file(reader: &mut mzml::AsyncMzMLReader<fs::File>) {
             );
         }
         i += 1;
-        if i == n {
-            break;
-        }
     }
     let end = time::Instant::now();
     println!(

examples/mzconvert.rs

@@ -1,3 +1,4 @@
+use std::any::Any;
 use std::env;
 use std::io;
 use std::path::PathBuf;
@@ -6,6 +7,7 @@ use std::sync::mpsc::sync_channel;
 use std::thread;
 use std::time;
 
+use log::info;
 use mzdata::io::MassSpectrometryFormat;
 use mzdata::io::{checksum_file, MassSpectrometryReadWriteProcess, Sink, Source};
 use mzdata::meta::custom_software_name;
@@ -80,8 +82,9 @@ impl MassSpectrometryReadWriteProcess<CentroidPeak, DeconvolutedPeak> for MZConv
         R: RandomAccessSpectrumIterator<CentroidPeak, DeconvolutedPeak>
             + SpectrumSource<CentroidPeak, DeconvolutedPeak>
             + Send
+            + Any
             + 'static,
-        W: SpectrumWriter<CentroidPeak, DeconvolutedPeak> + Send + 'static,
+        W: SpectrumWriter<CentroidPeak, DeconvolutedPeak> + Send + Any + 'static,
     >(
         &self,
         reader: R,
@@ -95,8 +98,9 @@ impl MassSpectrometryReadWriteProcess<CentroidPeak, DeconvolutedPeak> for MZConv
             + MSDataFileMetadata
             + SpectrumSource<CentroidPeak, DeconvolutedPeak>
             + Send
+            + Any
             + 'static,
-        W: SpectrumWriter<CentroidPeak, DeconvolutedPeak> + MSDataFileMetadata + Send + 'static,
+        W: SpectrumWriter<CentroidPeak, DeconvolutedPeak> + MSDataFileMetadata + Send + Any + 'static,
     >(
         &self,
         reader: R,
@@ -106,6 +110,7 @@ impl MassSpectrometryReadWriteProcess<CentroidPeak, DeconvolutedPeak> for MZConv
     ) -> Result<(R, W), Self::ErrorType> {
         if self.inpath != "-" {
             let pb: PathBuf = self.inpath.clone().into();
+            info!("Computing checksum for {}", pb.display());
             let checksum = checksum_file(&pb)?;
             let has_already = reader
                 .file_description()