junifer 0.0.4.dev782__py3-none-any.whl → 0.0.4.dev814__py3-none-any.whl

junifer/datagrabber/utils.py

@@ -6,7 +6,68 @@
 
 from typing import Dict, List
 
-from ..utils import raise_error
+from ..utils import logger, raise_error
+
+
+# Define schema for pattern-based datagrabber's patterns
+PATTERNS_SCHEMA = {
+    "T1w": {
+        "mandatory": ["pattern", "space"],
+        "optional": ["mask_item"],
+    },
+    "T1w_mask": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "T2w": {
+        "mandatory": ["pattern", "space"],
+        "optional": ["mask_item"],
+    },
+    "T2w_mask": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "BOLD": {
+        "mandatory": ["pattern", "space"],
+        "optional": ["mask_item"],
+    },
+    "BOLD_confounds": {
+        "mandatory": ["pattern", "format"],
+        "optional": [],
+    },
+    "BOLD_mask": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "Warp": {
+        "mandatory": ["pattern", "src", "dst"],
+        "optional": [],
+    },
+    "VBM_GM": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "VBM_WM": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "probseg_CSF": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "probseg_GM": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "probseg_WM": {
+        "mandatory": ["pattern", "space"],
+        "optional": [],
+    },
+    "DWI": {
+        "mandatory": ["pattern"],
+        "optional": [],
+    },
+}
 
 
 def validate_types(types: List[str]) -> None:
@@ -30,7 +91,7 @@ def validate_types(types: List[str]) -> None:
 
 
 def validate_replacements(
-    replacements: List[str], patterns: Dict[str, str]
+    replacements: List[str], patterns: Dict[str, Dict[str, str]]
 ) -> None:
     """Validate the replacements.
 
@@ -44,38 +105,41 @@ def validate_replacements(
     Raises
     ------
     TypeError
-        If ``replacements`` is not a list or if the values are not string or
-        if ``patterns`` is not a dictionary.
+        If ``replacements`` is not a list or if the values are not string.
     ValueError
-        If a value in ``replacements`` is not in ``pattern`` or if no value in
-        ``patterns`` contain all values in ``replacements``.
+        If a value in ``replacements`` is not part of a data type pattern or
+        if no data type patterns contain all values in ``replacements``.
 
     """
     if not isinstance(replacements, list):
         raise_error(msg="`replacements` must be a list.", klass=TypeError)
 
-    if not isinstance(patterns, dict):
-        raise_error(msg="`patterns` must be a dict.", klass=TypeError)
-
     if any(not isinstance(x, str) for x in replacements):
         raise_error(
             msg="`replacements` must be a list of strings.", klass=TypeError
         )
 
     for x in replacements:
-        if all(x not in y for y in patterns.values()):
-            raise_error(msg=f"Replacement {x} is not part of any pattern.")
+        if all(
+            x not in y
+            for y in [
+                data_type_val["pattern"] for data_type_val in patterns.values()
+            ]
+        ):
+            raise_error(msg=f"Replacement: {x} is not part of any pattern.")
 
     # Check that at least one pattern has all the replacements
     at_least_one = False
-    for _, v in patterns.items():
-        if all(x in v for x in replacements):
+    for data_type_val in patterns.values():
+        if all(x in data_type_val["pattern"] for x in replacements):
             at_least_one = True
     if at_least_one is False:
         raise_error(msg="At least one pattern must contain all replacements.")
 
 
-def validate_patterns(types: List[str], patterns: Dict[str, str]) -> None:
+def validate_patterns(
+    types: List[str], patterns: Dict[str, Dict[str, str]]
+) -> None:
     """Validate the patterns.
 
     Parameters
@@ -87,12 +151,17 @@ def validate_patterns(types: List[str], patterns: Dict[str, str]) -> None:
 
     Raises
     ------
+    KeyError
+        If any mandatory key is missing for a data type.
+    RuntimeError
+        If an unknown key is found for a data type.
     TypeError
         If ``patterns`` is not a dictionary.
     ValueError
         If length of ``types`` and ``patterns`` are different or
         if ``patterns`` is missing entries from ``types`` or
-        if ``patterns`` contain '*' as value.
+        if unknown data type is found in ``patterns`` or
+        if data type pattern key contains '*' as value.
 
     """
     # Validate the types
@@ -110,9 +179,60 @@ def validate_patterns(types: List[str], patterns: Dict[str, str]) -> None:
         raise_error(
             msg="`patterns` must contain all `types`", klass=ValueError
         )
-    # Wildcard check in patterns
-    if any("}*" in pattern for pattern in patterns.values()):
-        raise_error(
-            msg="`patterns` must not contain `*` following a replacement",
-            klass=ValueError,
-        )
+    # Check against schema
+    for data_type_key, data_type_val in patterns.items():
+        # Check if valid data type is provided
+        if data_type_key not in PATTERNS_SCHEMA:
+            raise_error(
+                f"Unknown data type: {data_type_key}, "
+                f"should be one of: {list(PATTERNS_SCHEMA.keys())}"
+            )
+        # Check mandatory keys for data type
+        for mandatory_key in PATTERNS_SCHEMA[data_type_key]["mandatory"]:
+            if mandatory_key not in data_type_val:
+                raise_error(
+                    msg=(
+                        f"Mandatory key: `{mandatory_key}` missing for "
+                        f"{data_type_key}"
+                    ),
+                    klass=KeyError,
+                )
+            else:
+                logger.debug(
+                    f"Mandatory key: `{mandatory_key}` found for "
+                    f"{data_type_key}"
+                )
+        # Check optional keys for data type
+        for optional_key in PATTERNS_SCHEMA[data_type_key]["optional"]:
+            if optional_key not in data_type_val:
+                logger.debug(
+                    f"Optional key: `{optional_key}` missing for "
+                    f"{data_type_key}"
+                )
+            else:
+                logger.debug(
+                    f"Optional key: `{optional_key}` found for "
+                    f"{data_type_key}"
+                )
+        # Check stray key for data type
+        for key in data_type_val.keys():
+            if key not in (
+                PATTERNS_SCHEMA[data_type_key]["mandatory"]
+                + PATTERNS_SCHEMA[data_type_key]["optional"]
+            ):
+                raise_error(
+                    msg=(
+                        f"Key: {key} not accepted for {data_type_key} "
+                        "pattern, remove it to proceed"
+                    ),
+                    klass=RuntimeError,
+                )
+        # Wildcard check in patterns
+        if "}*" in data_type_val["pattern"]:
+            raise_error(
+                msg=(
+                    f"`{data_type_key}.pattern` must not contain `*` "
+                    "following a replacement"
+                ),
+                klass=ValueError,
+            )

junifer/datareader/default.py

@@ -53,23 +53,22 @@ class DefaultDataReader(PipelineStepMixin, UpdateMetaMixin):
         # Nothing to validate, any input is fine
         return input
 
-    def get_output_type(self, input: List[str]) -> List[str]:
+    def get_output_type(self, input_type: str) -> str:
         """Get output type.
 
         Parameters
         ----------
-        input : list of str
-            The input to the reader. The list must contain the
-            available Junifer Data dictionary keys.
+        input_type : str
+            The data type input to the reader.
 
         Returns
         -------
-        list of str
-            The updated list of output types, as reading possibilities.
+        str
+            The data type output by the reader.
 
         """
         # It will output the same type of data as the input
-        return input
+        return input_type
 
     def _fit_transform(
         self,

{junifer-0.0.4.dev782.dist-info → junifer-0.0.4.dev814.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: junifer
-Version: 0.0.4.dev782
+Version: 0.0.4.dev814
 Summary: JUelich NeuroImaging FEature extractoR
 Author-email: Fede Raimondo <f.raimondo@fz-juelich.de>, Synchon Mandal <s.mandal@fz-juelich.de>
 Maintainer-email: Fede Raimondo <f.raimondo@fz-juelich.de>, Synchon Mandal <s.mandal@fz-juelich.de>

{junifer-0.0.4.dev782.dist-info → junifer-0.0.4.dev814.dist-info}/RECORD

@@ -1,5 +1,5 @@
 junifer/__init__.py,sha256=x1UR2jUcrUdm2HNl-3Qvyi4UUrU6ms5qm2qcmNY7zZk,391
-junifer/_version.py,sha256=J-m5LNYnI6A-KaVita_ud_ULV_D6TEhdcZxT2nse54U,428
+junifer/_version.py,sha256=EyqFRoxnT-66sDsNvzVoTVLeESpR_C5kEImKt8y_MD8,428
 junifer/stats.py,sha256=sU5IZ2qFZWbzgSutQS_z42miIVItpSGmQYBn6KkD5fA,6162
 junifer/api/__init__.py,sha256=YILu9M7SC0Ri4CVd90fELH2OnK_gvCYAXCoqBNCFE8E,257
 junifer/api/cli.py,sha256=auw38tIH7ipTnaADM7on0or7zauY-BFII8nd-eRUEJs,13664
@@ -39,11 +39,11 @@ junifer/api/tests/data/partly_cloudy_agg_mean_tian.yml,sha256=nS8K_R1hEuV71Vv-i9
 junifer/configs/__init__.py,sha256=r6BU6vW7FVapSD81j24QeQiZe1oKspsJJRRPjXnCk00,120
 junifer/configs/juseless/__init__.py,sha256=Ws98DvlLEMHfwW6BjmvHQmqTlFRDps9r4pLAfNjfEiM,149
 junifer/configs/juseless/datagrabbers/__init__.py,sha256=tqCLmelWqB1xfElvknnaJ5oVRPp9XVXtZLzIpxYIghg,452
-junifer/configs/juseless/datagrabbers/aomic_id1000_vbm.py,sha256=A4N8tpG66vWmqdWEBAh8Y8MHWsa_afcr3WTMNVhVIEA,1350
-junifer/configs/juseless/datagrabbers/camcan_vbm.py,sha256=GGbWDTvf0aHxFi_iS6jmlttTiD3k5gN5FYPIxemEnG4,1398
-junifer/configs/juseless/datagrabbers/ixi_vbm.py,sha256=UpOLpKFkfL_LeaYasUvpvpQSiX_RlZE99Nbfdexz8zk,2184
-junifer/configs/juseless/datagrabbers/ucla.py,sha256=Aea1ixDqh8X4BefgAQg2lT-ELEnxmsM8BnSGeMml0Ag,4302
-junifer/configs/juseless/datagrabbers/ukb_vbm.py,sha256=YWPL-VV84wyWApnakwsiNo6o2-mPtdAAWbU8hwEgBT0,1382
+junifer/configs/juseless/datagrabbers/aomic_id1000_vbm.py,sha256=HNyfxjU_D7hkwnaeSW1r_PllOlc3phBq2LP25zZGziE,1474
+junifer/configs/juseless/datagrabbers/camcan_vbm.py,sha256=PtzjrfL2b6nxXLQ4TU8cKftHnD4g7nk4bf2ib6WLqGg,1505
+junifer/configs/juseless/datagrabbers/ixi_vbm.py,sha256=OE2D311Sfey0DrfoKqWB_hFIBApzP3-r1M9sliKZPC4,2309
+junifer/configs/juseless/datagrabbers/ucla.py,sha256=44T9JMJ4usowRib9V6EqdTRQy_2xkAWZpEmmoFWF4Mg,4915
+junifer/configs/juseless/datagrabbers/ukb_vbm.py,sha256=lb9pLXFnlMOeQUX-BBFD9dKECdaU2bEIHA-iD9KkOAQ,1489
 junifer/configs/juseless/datagrabbers/tests/test_aomic_id1000_vbm.py,sha256=Y_npFmmj0MN0TYv8jaxVMgFHLMQLZ8vXl8lWQKvOYbc,1001
 junifer/configs/juseless/datagrabbers/tests/test_camcan_vbm.py,sha256=o0dzptS97pxrWaY7I1m0dpJtsnAwmIXNqdU9ABTWCqI,975
 junifer/configs/juseless/datagrabbers/tests/test_ixi_vbm.py,sha256=8jxpNZelXwpJGvA5LOfpso2X8yt1chvERAYmv76hS_g,1252
@@ -83,33 +83,33 @@ junifer/data/tests/test_masks.py,sha256=cZmj6UZmuyqXI-8zMqAkuim6DWYecjaMCPo00cr_
 junifer/data/tests/test_parcellations.py,sha256=ZEU1VHIK0AyxpclcJhG_0rQU0phaBU_dHP7Erfi3mN8,38222
 junifer/data/tests/test_template_spaces.py,sha256=PJulN7xHpAcSOTY-UzTG_WPywZEBSlAZGiNG4gzk1_8,3144
 junifer/datagrabber/__init__.py,sha256=pZHJIY8nAlbVngsyRScE6a6GKbytiwjJB7SdJNqIbl4,680
-junifer/datagrabber/base.py,sha256=clErdDX5jz8LTjwPTQPCv3jfELCJ2vObg7ks8imAx8Y,6224
-junifer/datagrabber/datalad_base.py,sha256=dDaBiIePPP6-G4ycgBMxTcXxs4vkg-yDS3OBURK4VGs,10731
-junifer/datagrabber/dmcc13_benchmark.py,sha256=C1ZzeQV7NsdIL5HCkjcZjE6uvecqt6ZY244OfvKD-Zo,11521
+junifer/datagrabber/base.py,sha256=KgMSKfkwd4yLW4PhoJDoWMgcDkGmDoIs6jkgKyOJd9A,6303
+junifer/datagrabber/datalad_base.py,sha256=SsGUJdefdDgAJARBG5kHcbLK2CvvnoEto0TpGZUgnWE,10659
+junifer/datagrabber/dmcc13_benchmark.py,sha256=ybgjJbl_4MD70FzhWUr_LvUZyjEft2_FYvLXI2kW64I,12896
 junifer/datagrabber/multiple.py,sha256=eXQIsvSNvD8GuEITjMaMoi1GwoeyWXXbQMRi-f2qgc4,4923
-junifer/datagrabber/pattern.py,sha256=uGobDTrRhq-XPNIz0z3eIeIzmNOYFajEXwa7OWqno5g,10720
-junifer/datagrabber/pattern_datalad.py,sha256=SyXhM0XQEF45poNaWyVAG054BSCYyqJ-JUwbRo05nyk,1653
-junifer/datagrabber/utils.py,sha256=-i5Pw7w72RO4SRapWYglIC0Lve8e8Mo3Gmu0z6A_PbY,3516
+junifer/datagrabber/pattern.py,sha256=5XQe0k3iiTID5fbUsShjFKaqkueuPU34ZpzOi1JzFEw,12700
+junifer/datagrabber/pattern_datalad.py,sha256=y9ptDZEhSj41pf05WNySrj8j4kbi5uIXveFXv4dkKpk,4092
+junifer/datagrabber/utils.py,sha256=aiaWMBAGsZ5VPgOXeUHralLnAoI3fs0inyHbv8K2Km8,7157
 junifer/datagrabber/aomic/__init__.py,sha256=R7yrRVBWsBW25CH0fw-KHpFwb_EC-MlPKDzssGfj5A0,281
-junifer/datagrabber/aomic/id1000.py,sha256=Cfp55jGzillWJ_SSeSKF33Cacl5FvVEb80B2Ieooy3I,5698
-junifer/datagrabber/aomic/piop1.py,sha256=wSqQUyHAoSf-1BPDQmBrowoRviFZznkox-iBWJ3NxOo,7718
-junifer/datagrabber/aomic/piop2.py,sha256=SIegiR2ybC8Uok0KDhRauSwhVi21z5cyOJiI698-k90,7337
+junifer/datagrabber/aomic/id1000.py,sha256=qO2WRq8CcK-cwohigqRewKHjAB75QZIhlqwK1hXWAs0,6125
+junifer/datagrabber/aomic/piop1.py,sha256=kYdDbDqSj-pQ12JY7RfD0OtUEkQS1YNivnzAjOaGaF8,8544
+junifer/datagrabber/aomic/piop2.py,sha256=ZHd3smD0pibBGbU170pDufLa1-TM7JlBVzWPjX1Uaq4,8163
 junifer/datagrabber/aomic/tests/test_id1000.py,sha256=eejuGiNQdX_s_8klGROd8OD8O5OprQywWN6O-JU3jD4,4824
 junifer/datagrabber/aomic/tests/test_piop1.py,sha256=Trp905MCT4jhnCg1leP0PzpEVc_icoKJZrI0TwVCwSE,5578
 junifer/datagrabber/aomic/tests/test_piop2.py,sha256=KAMU8eoQbKV3n9qRWbpZ24g-qN9uzZ55ZlofcSB9BIA,5377
 junifer/datagrabber/hcp1200/__init__.py,sha256=zy4Qq1_m3vECEhioG-UDteco2b5cni_8xuElICaRtt4,189
 junifer/datagrabber/hcp1200/datalad_hcp1200.py,sha256=p5Bbg09qoM46km9eFSlspwLwOe6LMJFSIZ9NC9E2lmc,2432
-junifer/datagrabber/hcp1200/hcp1200.py,sha256=6lNk3wjd6elVFeiQfmU140fsNoZwKFQjipbhlf1KXMk,6134
+junifer/datagrabber/hcp1200/hcp1200.py,sha256=ohXHmrQHWjqRZSEXI6E5zla1d_t6LZTbv-cAj79w3PU,6091
 junifer/datagrabber/hcp1200/tests/test_hcp1200.py,sha256=KJ-Jq01l0a6TaboG898qjBdPTHG1E3PZtHCjJ7n-1X0,10765
-junifer/datagrabber/tests/test_base.py,sha256=TBv14m7JZCvF6OPKG2MesMI7MPm-BkCAzOHb1GmMrG8,3738
-junifer/datagrabber/tests/test_datagrabber_utils.py,sha256=08fdnXRehAhrWoNdMP_2Ecgam7a1uMbt1i0wIpderEg,2795
-junifer/datagrabber/tests/test_datalad_base.py,sha256=GORMEzpIIK1StqJLv7jE8dZeLjXMZTPL_d_KDafDATc,16496
+junifer/datagrabber/tests/test_base.py,sha256=fZdVhNhvfht9lpTHrAUf5E6mAfNNUP7OTQ5KLaBQ1gI,3506
+junifer/datagrabber/tests/test_datagrabber_utils.py,sha256=SR2Zc9DJaCtuULhqCz10JCK7zc4VGnAcKT2pxloM1ys,6311
+junifer/datagrabber/tests/test_datalad_base.py,sha256=SYxUB9_4YPMfrb7iJM-aJCWbGa3EJfYz31wAUCNa03s,16285
 junifer/datagrabber/tests/test_dmcc13_benchmark.py,sha256=24T2ioumTTWC2xMrt7hbuasOY_KfFrJ61ztDxa23TFU,8123
-junifer/datagrabber/tests/test_multiple.py,sha256=Buu-ajH6AC8jVZz7RdEXm8kBIHwvFZvoSPH1DETxoFA,4955
-junifer/datagrabber/tests/test_pattern.py,sha256=coHa3ugiVs34Lhy4Md-_sAWBX878LHIbqGniQpsYxCE,9823
-junifer/datagrabber/tests/test_pattern_datalad.py,sha256=QoeM6F1T-W0XAU_PDh0T2gU4Cqrnu-q7vZ8_O9gT5sU,6077
+junifer/datagrabber/tests/test_multiple.py,sha256=Mx3xfDrQiWG2W5MW24P5L2XiSeALpJ2-jFlzWkKtu9w,5659
+junifer/datagrabber/tests/test_pattern.py,sha256=Zmwg79f-qs6AEPVoFpooOquK7rm1hsmgkzuo11BG5PE,8019
+junifer/datagrabber/tests/test_pattern_datalad.py,sha256=hxw_aXBwHjUo-aUrHescBA2dn1bSJxh-0oV8495iIEA,6483
 junifer/datareader/__init__.py,sha256=WWoiIz6y0EdExWO3WCGiERNw3Y3WUsXyHaWnHilP2kE,263
-junifer/datareader/default.py,sha256=9KjCbtrWgQQzoe4Z-uwD5B-flXI2f6yGFK5n9tXXaDk,4953
+junifer/datareader/default.py,sha256=BFtk_HJueia7EfhaIpgebe-RWHSKaLEO5btpHA--wo0,4848
 junifer/datareader/tests/test_default_reader.py,sha256=9dPZSkba1YQjFsA0XwdUbx5sq8DVIEZoy_WfMAcvRus,5220
 junifer/external/__init__.py,sha256=58gL7gNN_PI1GyN6r9CO035TzvJdi8S-zFg5reDt4Xc,116
 junifer/external/h5io/h5io/__init__.py,sha256=LG7ru_Rt3EOE2H4PGYfBhC12Iax3yeTquZkd8TICiKk,469
@@ -248,10 +248,10 @@ junifer/utils/logging.py,sha256=furcU3XIUpUvnpe4PEwzWWIWgmH4j2ZA4MQdvSGWjj0,9216
 junifer/utils/tests/test_fs.py,sha256=WQS7cKlKEZ742CIuiOYYpueeAhY9PqlastfDVpVVtvE,923
 junifer/utils/tests/test_helpers.py,sha256=k5qqfxK8dFyuewTJyR1Qn6-nFaYNuVr0ysc18bfPjyU,929
 junifer/utils/tests/test_logging.py,sha256=l8oo-AiBV7H6_IzlsNcj__cLeZBUvgIGoaMszD9VaJg,7754
-junifer-0.0.4.dev782.dist-info/AUTHORS.rst,sha256=rmULKpchpSol4ExWFdm-qu4fkpSZPYqIESVJBZtGb6E,163
-junifer-0.0.4.dev782.dist-info/LICENSE.md,sha256=MqCnOBu8uXsEOzRZWh9EBVfVz-kE9NkXcLCrtGXo2yU,34354
-junifer-0.0.4.dev782.dist-info/METADATA,sha256=RfPsY5N8UGCCkrn1mJRnyBTQcb854nsEuOca6tgPxS0,8235
-junifer-0.0.4.dev782.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-junifer-0.0.4.dev782.dist-info/entry_points.txt,sha256=DxFvKq0pOqRunAK0FxwJcoDfV1-dZvsFDpD5HRqSDhw,48
-junifer-0.0.4.dev782.dist-info/top_level.txt,sha256=4bAq1R2QFQ4b3hohjys2JBvxrl0GKk5LNFzYvz9VGcA,8
-junifer-0.0.4.dev782.dist-info/RECORD,,
+junifer-0.0.4.dev814.dist-info/AUTHORS.rst,sha256=rmULKpchpSol4ExWFdm-qu4fkpSZPYqIESVJBZtGb6E,163
+junifer-0.0.4.dev814.dist-info/LICENSE.md,sha256=MqCnOBu8uXsEOzRZWh9EBVfVz-kE9NkXcLCrtGXo2yU,34354
+junifer-0.0.4.dev814.dist-info/METADATA,sha256=mkbDgm5UJ8ihamdZE3fNkAvV6UxfW5tm-0o0Z9Jke_E,8235
+junifer-0.0.4.dev814.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+junifer-0.0.4.dev814.dist-info/entry_points.txt,sha256=DxFvKq0pOqRunAK0FxwJcoDfV1-dZvsFDpD5HRqSDhw,48
+junifer-0.0.4.dev814.dist-info/top_level.txt,sha256=4bAq1R2QFQ4b3hohjys2JBvxrl0GKk5LNFzYvz9VGcA,8
+junifer-0.0.4.dev814.dist-info/RECORD,,