add function to concatenate HTML files, such as reports

Randomidous · Randomidous · commit 5d9fb3fb7855 · 2025-04-23T14:07:29.000+02:00
diff --git a/doc/changes/devel/13223.newfeature.rst b/doc/changes/devel/13223.newfeature.rst
@@ -0,0 +1 @@
+Add functionality to concatenate HTML files, such as mne.Reports, by `Roy Eric Wieske`_.
diff --git a/mne/report/report.py b/mne/report/report.py
@@ -23,6 +23,7 @@
 from shutil import copyfile
 
 import numpy as np
+from bs4 import BeautifulSoup, Comment, Tag
 
 from .. import __version__ as MNE_VERSION
 from .._fiff.meas_info import Info, read_info
@@ -675,6 +676,93 @@ def open_report(fname, **params):
     return report
 
 
+def concatenate_reports(html_files, output_file):
+    """Concatenate multiple HTML files into one.
+
+    This function reads multiple HTML files, extracts their <head> and <body> sections,
+    and combines them into a single HTML file. It also handles duplicate IDs in the body
+    by removing them.
+
+    Parameters
+    ----------
+    html_files : list of str or Path
+        List of paths to the HTML files to be concatenated.
+    output_file : str or Path
+        Path to the output HTML file.
+
+    Returns
+    -------
+    final_html : BeautifulSoup
+        A BeautifulSoup object representing the combined HTML content.
+    """
+    combined_head = BeautifulSoup("<head></head>", "lxml").head
+    combined_body = BeautifulSoup("<body></body>", "lxml").body
+    used_ids = set()
+
+    for file in html_files:
+        file = Path(file)
+        with open(file, encoding="utf-8") as f:
+            soup = BeautifulSoup(f, "lxml")
+
+        toc = soup.find(id="toc") or soup.find(class_="toc")
+        if toc:
+            toc.decompose()
+
+        # handle head
+        if soup.head:
+            seen_styles = set()
+            seen_links = set()
+            seen_scripts = set()
+
+            for tag in soup.head.find_all(["script", "link", "style"], recursive=True):
+                tag_str = str(tag)
+                if tag.name == "style":
+                    if tag_str in seen_styles:
+                        continue
+                    seen_styles.add(tag_str)
+                elif tag.name == "link":
+                    if tag_str in seen_links:
+                        continue
+                    seen_links.add(tag_str)
+                elif tag.name == "script":
+                    if tag_str in seen_scripts:
+                        continue
+                    seen_scripts.add(tag_str)
+
+                combined_head.append(tag)
+
+        # handle body
+        if soup.body:
+            section = soup.new_tag("section")
+            section.append(soup.new_tag("hr"))
+            section.append(Comment(f"START {file.name}"))
+
+            for tag in soup.body.contents:
+                if isinstance(tag, Tag):
+                    for t in tag.find_all(True):
+                        id_ = t.get("id")
+                        if id_:
+                            if id_ in used_ids:
+                                del t["id"]
+                            else:
+                                used_ids.add(id_)
+
+                section.append(tag)
+
+            combined_body.append(section)
+
+    # create final HTML
+    final_html = BeautifulSoup("<html></html>", "lxml")
+    final_html.html.append(combined_head)
+    final_html.html.append(combined_body)
+
+    output_file = Path(output_file)
+    with output_file.open("w", encoding="utf-8") as f:
+        f.write(final_html.prettify())
+
+    return final_html
+
+
 ###############################################################################
 # HTML scan renderer
 
diff --git a/mne/report/tests/test_report.py b/mne/report/tests/test_report.py
@@ -13,6 +13,7 @@
 
 import numpy as np
 import pytest
+from bs4 import BeautifulSoup
 from matplotlib import pyplot as plt
 
 from mne import (
@@ -33,6 +34,7 @@
 from mne.report.report import (
     _ALLOWED_IMAGE_FORMATS,
     CONTENT_ORDER,
+    concatenate_reports,
 )
 from mne.utils import Bunch, _record_warnings
 from mne.utils._testing import assert_object_equal
@@ -634,6 +636,54 @@ def test_open_report(tmp_path):
     assert h5io.read_hdf5(hdf5, title="companion") == "test"
 
 
+def test_concatenate_reports(tmp_path, sample_meg_dir):
+    """Test the concatenate_reports function."""
+    raw_path = sample_meg_dir / "sample_audvis_raw.fif"
+    raw = read_raw_fif(raw_path, preload=True)
+    raw.set_annotations(None)
+    raw.crop(0, 20)
+
+    with tmp_path as tmp_dir:
+        tmp_path = Path(tmp_dir)
+
+        # Report 1 with custom content
+        report1 = Report(title="Report eeg_preprocessing #1")
+        report1.add_html(
+            "<div class='custom-note'>This is report one</div>", title="Note 1"
+        )
+        report1.add_raw(raw, title="Raw data", psd=False)
+        report1.save(tmp_path / "report1.html", overwrite=True, open_browser=False)
+
+        # Report 2 with different custom content
+        report2 = Report(title="Report eeg_preprocessing #2")
+        report2.add_html(
+            "<div class='custom-note'>This is report two</div>", title="Note 2"
+        )
+        report2.add_raw(raw, title="Raw data", psd=False)
+        report2.save(tmp_path / "report2.html", overwrite=True, open_browser=False)
+
+        file1 = tmp_path / "report1.html"
+        file2 = tmp_path / "report2.html"
+        output_file = tmp_path / "combined.html"
+
+        _ = concatenate_reports([file1, file2], output_file)
+
+        assert output_file.exists()
+
+        with open(output_file, encoding="utf-8") as f:
+            out_html = BeautifulSoup(f, "lxml")
+
+        assert out_html.head is not None
+        assert out_html.body is not None
+
+        styles = out_html.find_all("style")
+        scripts = out_html.find_all("script")
+        divs = out_html.find_all("div")
+
+        assert out_html.find(text=lambda t: "This is report one" in t) is not None
+        assert out_html.find(text=lambda t: "This is report two" in t) is not None
+
+
 def test_remove():
     """Test removing figures from a report."""
     r = Report()
diff --git a/pyproject.toml b/pyproject.toml
@@ -88,6 +88,7 @@ full = ["mne[full-no-qt]", "PyQt6 != 6.6.0", "PyQt6-Qt6 != 6.6.0, != 6.7.0"]
 # and mne[full-pyside6], which will install PySide6 instead of PyQt6.
 full-no-qt = [
   "antio >= 0.5.0",
+  "beautifulsoup4",
   "darkdetect",
   "defusedxml",
   "dipy",

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1 @@`
	`1`	+Add functionality to concatenate HTML files, such as mne.Reports, by `Roy Eric Wieske`_.