Polar channels

CHANGELOG.md

@@ -1,3 +1,79 @@
+# Unreleased
+
+- Enabling magnetism no longer jumps to another page. Ticking a layer's
+  "Magn." box while the current engine cannot model magnetism now asks whether
+  to switch the project to refl1d and does both in one step. 
+  The calculation engine has its own group on the Sample page.
+- Selecting an engine that cannot model magnetism, while the sample already has
+  magnetic layers, is refused with a message on both pages.
+- Fixed the calculation-engine selector showing the wrong engine after loading
+  a project that used a different one.
+- Magnetic depth profiles on the SLD chart (Sample and Analysis share the
+  chart):
+  - Once a layer is magnetic, the chart adds the spin-up and spin-down
+    potentials ρ ± ρM·cos(θM − A) for each magnetic model, dashed in the
+    model's colour. ρM and the moment angle θM each have a checkbox; θM uses
+    its own right-hand axis. The y-range covers every visible curve and grows
+    when a curve is switched on, so ρ + ρM is not clipped. θM is only defined
+    where there is a moment, and that curve is drawn in pieces, not joined
+    across the gaps.
+  - The switches are in a new "Magnetic profile" group under Magnetism on the
+    Sample page, and in Analysis under "Plot control". Both use the same
+    selection.
+  - If no model is magnetic, the chart, legend and sidebar are unchanged.
+- Spin-asymmetry view, SA(q) = (R↑↑ − R↓↓)/(R↑↑ + R↓↓):
+  - A "Spin asymmetry" tab on the Experiment page (measured data) and a third
+    tab on the Analysis page's lower panel (measured data plus the model).
+    Neither is shown unless the experiment measured both non-spin-flip
+    channels. Without one, Experiment has no tab strip.
+  - Error bars come from the channel uncertainties. Points where R↑↑ + R↓↓ is
+    not significantly above zero (the background-dominated tail) are dropped,
+    with a note of how many, so they do not wreck the axis. Same for points
+    where the two channels do not share the same q. The axis shows the full
+    [−1, 1] window and expands if background-subtracted data go outside it.
+- Polarized (spin-channel) experiment import and display:
+  - New "Load polarized experiment (file per channel)" flow: multi-select one
+    file per spin channel, then review and edit the automatic assignment (from
+    the ORSO `polarization` header or the file name). Any number of channels
+    can be assigned, including a single channel. Files can be marked "not
+    used". Duplicate or unknown channels, missing files, or nothing assigned
+    are rejected with a message.
+  - The experiment chart draws one measured series (plus error bounds) per
+    visible spin channel in a fixed palette (↑↑ pp, ↑↓ pm, ↓↑ mp, ↓↓ mm), with
+    a channel selector in the sidebar and a per-channel legend. At least one
+    measured channel stays visible. With several experiments selected, each
+    polarized experiment contributes one series per visible channel, using the
+    experiment colour as the hue base.
+  - Experiment lists mark polarized experiments with a `⇅N` badge for the
+    number of measured spin channels.
+  - Report figures plot each channel's spin cross-section in its channel
+    colour. A channel that cannot be calculated (for example spin-flip on a
+    non-magnetic model) is shown as measured data only.
+  - **Limitations:** one resolution function per polarized experiment (taken
+    from the first assigned channel; the import dialog says so); Bayesian
+    sampling of polarized experiments is not supported yet.
+  - Project save/load now fully supports polarized experiments.
+- Magnetism editing and polarized fitting:
+  - New "Magnetism" group on the Sample page: one row per layer of the current
+    assembly, with a magnetic on/off checkbox, magnetic SLD (ρM) and in-plane
+    moment angle (θM). Only refl1d can model magnetic layers. The group says
+    so and offers to switch the project's calculation engine when a layer is
+    made magnetic (see above).
+  - ρM and θM appear in the Analysis parameter table like other layer
+    parameters, named after their assembly and model (`Model Fe rho_m`), with
+    default limits, fit checkboxes and constraints. The name filter accepts
+    "magnetic" as a keyword.
+  - Fitting a polarized experiment fits all its measured spin channels at once
+    against the shared model. Thickness, roughness, nuclear SLD, scale and
+    background are common to every channel; ρM and θM are constrained by all
+    of them. Polarized and ordinary experiments can be fitted together. This
+    used to report "not supported yet".
+  - The analysis and residual charts draw one measured/calculated pair per
+    visible spin channel, each with that channel's cross-section. Previously
+    only the first visible channel was shown. A channel the model cannot
+    calculate (spin-flip on a non-magnetic sample) shows measured points only
+    and contributes no residuals.
+
 # Version 1.4.0 (3 Aug 2026)
 
 - Added Bayesian analysis: run MCMC sampling alongside classical fitting, with posterior median and credibility intervals on the main chart, trace/corner-style plots per parameter, a dedicated status display with cancellation support, and plot export.

EasyReflectometryApp/Backends/Mock/Analysis.qml

@@ -11,6 +11,9 @@ QtObject {
 
     readonly property var experimentsAvailable: ['experiment_1', 'experiment_2', 'experiment_3']
     readonly property int experimentCurrentIndex: 2
+    // Polarization badge / channel count columns of the experiment lists
+    readonly property var experimentsPolarized: [false, false, false]
+    readonly property var experimentsChannelCount: [0, 0, 0]
 
     // Minimizer
     readonly property double minimizerTolerance: 1.0

EasyReflectometryApp/Backends/Mock/Experiment.qml

@@ -22,4 +22,32 @@ QtObject {
     function load(path) {
         console.debug(`Loading experiment from ${path}`)
     }
+
+    // Filename-token suggestion, so the assignment dialog can be exercised
+    // against the mock backend as well.
+    function suggestPolarizedChannels(paths) {
+        console.debug(`Suggesting polarized channels for ${paths}`)
+        const list = Array.isArray(paths) ? paths : String(paths).split(',')
+        const tokens = {uu: 'pp', pp: 'pp', dd: 'mm', mm: 'mm', ud: 'pm', pm: 'pm', du: 'mp', mp: 'mp'}
+        return list.filter(path => path !== '').map(path => {
+            const name = String(path).split(/[\\/]/).pop()
+            let channel = ''
+            for (const token in tokens) {
+                if (name.toLowerCase().indexOf('_' + token) !== -1) {
+                    channel = tokens[token]
+                    break
+                }
+            }
+            return {path: path, name: name, channel: channel}
+        })
+    }
+
+    function loadPolarized(assignments) {
+        console.debug(`Loading polarized experiment from ${assignments.length} file(s)`)
+    }
+
+    // Emitted with a user-facing message when an import is rejected.
+    signal loadFailed(string message)
+    // Emitted with the list position of a newly imported experiment.
+    signal experimentLoaded(int index)
 }

EasyReflectometryApp/Backends/Mock/Plotting.qml

@@ -28,6 +28,63 @@ QtObject {
 
     property int modelCount: 1
 
+    // Polarized experiment (spin channel) support
+    property bool currentExperimentIsPolarized: false
+    property var experimentChannelList: []
+    signal channelSelectionChanged()
+    signal experimentChannelsChanged()
+    function setChannelVisible(channel, visible) {
+        console.debug(`setChannelVisible ${channel} ${visible}`)
+    }
+    function getExperimentChannels(index) {
+        return []
+    }
+    function getExperimentChannelDataPoints(index, channel) {
+        return []
+    }
+
+    // Magnetic depth profiles (no magnetic model in the mock)
+    property bool anyModelHasMagnetism: false
+    property var visibleSldCurves: ['spin_up', 'spin_down']
+    property double sldThetaMinY: 0
+    property double sldThetaMaxY: 360
+    signal magneticProfileChanged()
+    function modelHasMagnetism(index) {
+        return false
+    }
+    function getMagneticSldDataPointsForModel(index, curve) {
+        return []
+    }
+    function getMagneticSldSegmentsForModel(index, curve) {
+        return []
+    }
+    function getMagneticSldSegment(index, curve, segment) {
+        return []
+    }
+    function sldCurveVisible(curve) {
+        return visibleSldCurves.indexOf(curve) !== -1
+    }
+    function setSldCurveVisible(curve, visible) {
+        console.debug(`setSldCurveVisible ${curve} ${visible}`)
+    }
+
+    // Spin asymmetry (no polarized experiment in the mock)
+    property bool spinAsymmetryAvailable: false
+    property bool spinAsymmetryCalculatedAvailable: false
+    property int spinAsymmetryMaskedPoints: 0
+    property int spinAsymmetryOutOfOverlapPoints: 0
+    property double spinAsymmetryMinX: 0
+    property double spinAsymmetryMaxX: 1
+    property double spinAsymmetryMinY: -1
+    property double spinAsymmetryMaxY: 1
+    signal spinAsymmetryChanged()
+    function getSpinAsymmetryPoints(index) {
+        return []
+    }
+    function getSpinAsymmetryCalculatedPoints(index) {
+        return []
+    }
+
     // Plot mode properties
     property bool plotRQ4: false
     property string yMainAxisTitle: 'R(q)'

EasyReflectometryApp/Backends/Mock/Sample.qml

@@ -3,6 +3,20 @@ pragma Singleton
 import QtQuick
 
 QtObject {
+
+    // Calculation engine (project-wide setting, also shown on the Sample page)
+    property var calculationEngines: ['refnx', 'refl1d']
+    property int calculationEngineIndex: 0
+    property var calculationEnginesSupportingMagnetism: ['refl1d']
+    signal calculationEngineChanged()
+    signal magnetismNeedsEngine(int index, string engine)
+    function setCalculationEngineIndex(value) {
+        console.debug(`setCalculationEngineIndex ${value}`)
+    }
+    function enableMagnetismWithEngineAtIndex(index, engine) {
+        console.debug(`enableMagnetismWithEngineAtIndex ${index} ${engine}`)
+    }
+
     // Signals to match the Python backend
     signal constraintsChanged
     // MATERIALS
@@ -250,6 +264,23 @@ QtObject {
         console.debug(`setCurrentLayerSolvation ${value}`)
     }
 
+    // Layer magnetism (polarized analysis)
+    readonly property bool magnetismSupported: true
+    readonly property var layersMagnetism: [
+        { 'label': 'label 1', 'magnetic': 'True', 'rho_m': '5.0', 'theta_m': '40.0' },
+        { 'label': 'label 2', 'magnetic': 'False', 'rho_m': '0.0', 'theta_m': '270.0' },
+        { 'label': 'label 3', 'magnetic': 'False', 'rho_m': '0.0', 'theta_m': '270.0' },
+    ]
+    function setLayerMagneticAtIndex(index, value) {
+        console.debug(`setLayerMagneticAtIndex ${index} ${value}`)
+    }
+    function setLayerRhoMAtIndex(index, value) {
+        console.debug(`setLayerRhoMAtIndex ${index} ${value}`)
+    }
+    function setLayerThetaMAtIndex(index, value) {
+        console.debug(`setLayerThetaMAtIndex ${index} ${value}`)
+    }
+
     // Table functions
     function removeLayer(value) {
         console.debug(`removeLayer ${value}`)