From dc407db86bd014e75c6bff3fa240fe0cb6a81237 Mon Sep 17 00:00:00 2001
From: JulianKimmig <julian.kimmig@gmx.net>
Date: Thu, 4 Sep 2025 14:46:12 +0200
Subject: [PATCH] feat(peak_finder): add optional re-centering of peak indices
 based on selected source

- Implemented logic to adjust peak indices using either the original signal or centroid method.
- Enhanced peak detection accuracy by allowing for dynamic peak repositioning within specified bounds.
---
 pyproject.toml                      |  2 +-
 src/funcnodes_span/__init__.py      |  2 +-
 src/funcnodes_span/peak_analysis.py | 21 +++++++++++++++++++++
 3 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index caaa005..3b40658 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "funcnodes-span"
-version = "1.0.0"
+version = "1.0.1"
 description = "SPectral ANalysis (SPAN) for funcnodes"
 readme = "README.md"
 classifiers = [ "Development Status :: 4 - Beta", "Intended Audience :: Developers", "Intended Audience :: Education", "Intended Audience :: Science/Research", "License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)",]
diff --git a/src/funcnodes_span/__init__.py b/src/funcnodes_span/__init__.py
index fc052fc..e864b32 100644
--- a/src/funcnodes_span/__init__.py
+++ b/src/funcnodes_span/__init__.py
@@ -7,7 +7,7 @@
 from funcnodes_lmfit import NODE_SHELF as LMFIT_NODE_SHELF
 from .curves import CURVES_NODE_SHELF
 
-__version__ = "1.0.0"
+__version__ = "1.0.1"
 
 NODE_SHELF = fn.Shelf(
     name="Spectral Analysis",
diff --git a/src/funcnodes_span/peak_analysis.py b/src/funcnodes_span/peak_analysis.py
index 8f88b0d..4b38304 100644
--- a/src/funcnodes_span/peak_analysis.py
+++ b/src/funcnodes_span/peak_analysis.py
@@ -270,6 +270,27 @@ def peak_finder(
                     if y[peak] > height:
                         peak_lst.append([left_min, peak, right_min])
 
+    # Optionally re-center peak indices based on selected source
+    if len(peak_lst) > 0:
+        for k, (i_idx, idx, f_idx) in enumerate(peak_lst):
+            i = max(0, int(i_idx))
+            f = min(len(_y) - 1, int(f_idx))
+            if f <= i:
+                continue
+            if index_source == "original" and on is not None:
+                # Use argmax on the original (or differently processed) signal within bounds
+                local_argmax = int(np.argmax(_y[i : f + 1])) + i
+                peak_lst[k] = [i_idx, local_argmax, f_idx]
+            elif index_source == "centroid":
+                # Center-of-mass within bounds on original signal
+                yy = _y[i : f + 1]
+                w = yy - np.min(yy)
+                w = np.maximum(w, 0)
+                if np.sum(w) > 0:
+                    cm = int(np.round(np.average(np.arange(i, f + 1), weights=w)))
+                    cm = int(np.clip(cm, i, f))
+                    peak_lst[k] = [i_idx, cm, f_idx]
+
     peak_properties_list = []
 
     for peak_nr, peak in enumerate(peak_lst):