[ENH] improve error messages for invalid input data (#150)

* improve doc * add extra check * improve error msg * fix
cpp-lln-lab · Aug 1, 2024 · de73e5b · de73e5b
1 parent 7435e94
commit de73e5b
Show file tree

Hide file tree

Showing 14 changed files with 180 additions and 29 deletions.
diff --git a/.gitignore b/.gitignore
@@ -1,3 +1,7 @@
+node_modules
+package-lock.json
+package.json
+
 # folders
 outputs
 download

diff --git a/bidsmreye/bids_utils.py b/bidsmreye/bids_utils.py
@@ -41,7 +41,11 @@ def check_layout(cfg: Config, layout: BIDSLayout, for_file: str = "bold") -> Non
         or "DatasetType" in desc
         and desc["DatasetType"] != "derivative"
     ):
-        raise RuntimeError("DatasetType must be 'derivative' in dataset_description.json")
+        raise RuntimeError(
+            "DatasetType must be 'derivative' in dataset_description.json\n."
+            "Check the FAQ for more information: "
+            "https://bidsmreye.readthedocs.io/en/latest/FAQ.html"
+        )
 
     this_filter = get_bids_filter_config()[for_file]
 
@@ -71,7 +75,10 @@ def check_layout(cfg: Config, layout: BIDSLayout, for_file: str = "bold") -> Non
     if bf == []:
         raise RuntimeError(
             f"Input dataset {layout.root} does not have "
-            "any data to process for filter\n{this_filter}"
+            f"any data to process for filter\n{this_filter}.\n"
+            "Is your dataset a BIDS derivative dataset?\n"
+            "Check the FAQ for more information: "
+            "https://bidsmreye.readthedocs.io/en/latest/FAQ.html"
         )
 
 

diff --git a/bidsmreye/configuration.py b/bidsmreye/configuration.py
@@ -87,8 +87,17 @@ def __attrs_post_init__(self) -> None:
             database_path=database_path,
             reset_database=self.reset_database,
         )
-
-        log.debug(f"Layout in:\n{layout_in}")
+        log.debug(f"Layout in:\n{layout_in.root}")
+
+        value = layout_in.get(return_type="id", target="subject", datatype="func")
+        if not value:
+            raise RuntimeError(
+                f"Input dataset {layout_in.root} does not have "
+                f"any data to process.\n"
+                "Is your dataset a BIDS derivative dataset?\n"
+                "Check the FAQ for more information: "
+                "https://bidsmreye.readthedocs.io/en/latest/FAQ.html"
+            )
 
         if not database_path.is_dir():
             layout_in.save(database_path)

diff --git a/bidsmreye/quality_control.py b/bidsmreye/quality_control.py
@@ -232,17 +232,11 @@ def compute_robust_outliers(
     points (assuming we don't want to 'remove' too many).
 
     References:
-    - `Rousseeuw, P. J., and Croux, C. (1993). Alternatives to the the median
-      absolute deviation. J. Am. Stat. Assoc. 88, 1273-1263.
-      <https://www.tandfonline.com/doi/abs/10.1080/01621459.1993.10476408>`_
+        - :cite:t:`rousseeuw_alternatives_1993`
+        - :cite:t:`carling_resistant_2000`
+        - :cite:t:`hoaglin_performance_1986`
 
-    - `Carling, K. (2000). Resistant outlier rules and the non-Gaussian case.
-      Stat. Data Anal. 33, 249:258.
-      <http://www.sciencedirect.com/science/article/pii/S0167947399000572>`_
 
-    - `Hoaglin, D.C., Iglewicz, B. (1987) Fine-tuning some resistant rules for
-      outlier labelling. J. Amer. Statist. Assoc., 82 , 1147:1149
-      <http://www.tandfonline.com/doi/abs/10.1080/01621459.1986.10478363>`_
     """
     if outlier_type is None:
         outlier_type = "S-outliers"

diff --git a/bidsmreye/templates/CITATION.bib b/bidsmreye/templates/CITATION.bib
@@ -17,6 +17,7 @@ @article{deepmreye
 }
 
 
+
 @article{alexander_open_2017,
 	title = {An open resource for transdiagnostic research in pediatric mental health and learning disorders},
 	volume = {4},
@@ -119,3 +120,36 @@ @article{carling_resistant_2000
 	pages = {249--258},
 	file = {Submitted Version:/home/remi/Zotero/storage/VHQYXF6K/Carling - 2000 - Resistant outlier rules and the non-Gaussian case.pdf:application/pdf},
 }
+
+@article{hoaglin_performance_1986,
+	title = {Performance of {Some} {Resistant} {Rules} for {Outlier} {Labeling}},
+	volume = {81},
+	issn = {0162-1459, 1537-274X},
+	url = {http://www.tandfonline.com/doi/abs/10.1080/01621459.1986.10478363},
+	doi = {10.1080/01621459.1986.10478363},
+	language = {en},
+	number = {396},
+	urldate = {2023-05-24},
+	journal = {Journal of the American Statistical Association},
+	author = {Hoaglin, David C. and Iglewicz, Boris and Tukey, John W.},
+	month = dec,
+	year = {1986},
+	pages = {991--999},
+}
+
+
+@article{rousseeuw_alternatives_1993,
+	title = {Alternatives to the {Median} {Absolute} {Deviation}},
+	volume = {88},
+	issn = {0162-1459, 1537-274X},
+	url = {http://www.tandfonline.com/doi/abs/10.1080/01621459.1993.10476408},
+	doi = {10.1080/01621459.1993.10476408},
+	language = {en},
+	number = {424},
+	urldate = {2023-05-24},
+	journal = {Journal of the American Statistical Association},
+	author = {Rousseeuw, Peter J. and Croux, Christophe},
+	month = dec,
+	year = {1993},
+	pages = {1273--1283},
+}
diff --git a/docs/questions/validation.question.md b/docs/questions/validation.question.md
@@ -5,12 +5,44 @@ alt_titles:
   - "What input format does bidsmreye expect?"
 ---
 
-bidsmreye requires your input fmri data:
-
- - to be minimally preprocessed
- - with filenames and structure that conforms to a BIDS derivative dataset.
+bidsmreye requires a BIDS **preprocessed** dataset as input.
 
 Two bids apps are available to generate those types of preprocessed data:
 
 - [fmriprep](https://fmriprep.org/en/stable/)
 - [bidspm](https://bidspm.readthedocs.io/en/latest/general_information.html)
+
+bidsmreye requires your input fmri data:
+
+ - to be minimally preprocessed
+ - with [filenames and structure that conforms to a BIDS derivative dataset](https://bids-specification.readthedocs.io/en/latest/derivatives/imaging.html#preprocessed-coregistered-andor-resampled-volumes).
+
+More specifically the dataset should look like this:
+
+```
+dataset_description.json
+sub-{sub}
+    [ses-{session}]
+        func (func_dir)
+            sub-{sub}[_ses-{session}]_task-{task}[_acq-{acq}][_ce-{ce}][_dir-{dir}][_rec-{rec}][_run-{run_index}]_space-{space}[_res-{res}]_desc-preproc_bold.nii[.gz]
+[participants.tsv]
+[README]
+[CHANGES]
+[LICENSE]
+```
+
+- Filename entities, files or directories between square brackets
+  (for example, `[_ses-<label>]`) are OPTIONAL.
+  Note that for bidsmreye to work, the `space` entity is required.
+- `[.gz]` means that both the unzipped and gzipped versions of the extension are valid.
+
+Moreover the dataset_description.json file specify
+that the input dataset is a derivative dataset:
+
+```json
+{
+    "Name": "my_dataset",
+    "BIDSVersion": "1.8.0",
+    "DatasetType": "derivative",
+}
+```
diff --git a/docs/source/FAQ.md b/docs/source/FAQ.md
@@ -7,16 +7,48 @@
 <a name="how-i-should-structure-my-input-data"></a>
 ## How I should structure my input data?
 
-bidsmreye requires your input fmri data:
-
- - to be minimally preprocessed
- - with filenames and structure that conforms to a BIDS derivative dataset.
+bidsmreye requires a BIDS **preprocessed** dataset as input.
 
 Two bids apps are available to generate those types of preprocessed data:
 
 - [fmriprep](https://fmriprep.org/en/stable/)
 - [bidspm](https://bidspm.readthedocs.io/en/latest/general_information.html)
 
+bidsmreye requires your input fmri data:
+
+ - to be minimally preprocessed
+ - with [filenames and structure that conforms to a BIDS derivative dataset](https://bids-specification.readthedocs.io/en/latest/derivatives/imaging.html#preprocessed-coregistered-andor-resampled-volumes).
+
+More specifically the dataset should look like this:
+
+```
+dataset_description.json
+sub-{sub}
+    [ses-{session}]
+        func (func_dir)
+            sub-{sub}[_ses-{session}]_task-{task}[_acq-{acq}][_ce-{ce}][_dir-{dir}][_rec-{rec}][_run-{run_index}]_space-{space}[_res-{res}]_desc-preproc_bold.nii[.gz]
+[participants.tsv]
+[README]
+[CHANGES]
+[LICENSE]
+```
+
+- Filename entities, files or directories between square brackets
+  (for example, `[_ses-<label>]`) are OPTIONAL.
+  Note that for bidsmreye to work, the `space` entity is required.
+- `[.gz]` means that both the unzipped and gzipped versions of the extension are valid.
+
+Moreover the dataset_description.json file specify
+that the input dataset is a derivative dataset:
+
+```json
+{
+    "Name": "my_dataset",
+    "BIDSVersion": "1.8.0",
+    "DatasetType": "derivative",
+}
+```
+
 <a name="is-the-"prepare"-only-suitable-for-the-datasets-that-have-eye-tracking-info"></a>
 ## Is the "prepare" only suitable for the datasets that have eye-tracking info?
 

diff --git a/docs/source/conf.py b/docs/source/conf.py
@@ -42,6 +42,7 @@
     "sphinx_copybutton",
     "myst_parser",
     "sphinxarg.ext",
+    "sphinxcontrib.bibtex",
 ]
 
 # Add any paths that contain templates here, relative to this directory.
@@ -85,6 +86,10 @@
 # Configuration of sphinx.ext.coverage
 coverage_show_missing_items = True
 
+bibtex_bibfiles = ["references.bib"]
+bibtex_style = "unsrt"
+bibtex_reference_style = "author_year"
+
 # -- Options for HTML output -------------------------------------------
 
 # The theme to use for HTML and HTML Help pages.  See the documentation for

diff --git a/docs/source/index.rst b/docs/source/index.rst
@@ -10,6 +10,7 @@ Welcome to bidsMReye's documentation!
    demo
    FAQ
    modules
+   references
 
 Indices and tables
 ==================

diff --git a/docs/source/references.bib b/docs/source/references.bib
@@ -0,0 +1 @@
+../../bidsmreye/templates/CITATION.bib
diff --git a/docs/source/references.rst b/docs/source/references.rst
@@ -0,0 +1,8 @@
+References
+**********
+
+Bilbiography
+============
+
+.. bibliography::
+    :all:
diff --git a/pyproject.toml b/pyproject.toml
@@ -57,7 +57,8 @@ doc = [
     "sphinx",
     "sphinx-argparse",
     "sphinx-copybutton",
-    "sphinx-rtd-theme"
+    "sphinx-rtd-theme",
+    "sphinxcontrib-bibtex"
 ]
 docs = ["bidsmreye[doc]"]
 style = [

diff --git a/tests/test_bids_utils.py b/tests/test_bids_utils.py
@@ -4,6 +4,8 @@
 import shutil
 from pathlib import Path
 
+import pytest
+
 from bidsmreye.bids_utils import (
     check_layout,
     create_bidsname,
@@ -102,3 +104,28 @@ def test_check_layout_prepare_data():
         reset_database=True,
     )
     check_layout(cfg, layout_in)
+
+
+def test_check_layout_error_no_space_entity(tmp_path):
+    shutil.copytree(pybids_test_dataset(), tmp_path, dirs_exist_ok=True)
+    for file in tmp_path.rglob("*_space-*"):
+        file.unlink()
+
+    cfg = Config(
+        tmp_path,
+        tmp_path / "foo",
+    )
+
+    print(cfg.input_dir)
+
+    layout_in = get_dataset_layout(
+        cfg.input_dir,
+        use_database=False,
+        config=["bids", "derivatives"],
+        reset_database=True,
+    )
+
+    with pytest.raises(
+        RuntimeError, match="does not have any data to process for filter"
+    ):
+        check_layout(cfg, layout_in)
diff --git a/tests/test_quality_control.py b/tests/test_quality_control.py
@@ -135,11 +135,9 @@ def test_quality_control_output():
     quality_control_output(cfg)
 
 
-def test_quality_control_input():
-    input_dir = Path().resolve().joinpath("tests", "data", "ds000201-der")
-    output_dir = input_dir.joinpath("derivatives")
-
-    rm_dir(output_dir)
+def test_quality_control_input(tmp_path):
+    input_dir = Path("tests") / "data" / "ds000201-der"
+    output_dir = tmp_path / "derivatives"
 
     cfg = Config(
         input_dir,
@@ -148,8 +146,6 @@ def test_quality_control_input():
 
     quality_control_input(cfg)
 
-    rm_dir(output_dir)
-
 
 def test_perform_quality_control():
     create_basic_json()