Skip to content

Commit

Permalink
Keep working.
Browse files Browse the repository at this point in the history
  • Loading branch information
@tsalo
tsalo committed Feb 19, 2024
1 parent c737248 commit 1819219
Show file tree
Hide file tree
Showing 2 changed files with 42 additions and 80 deletions.
120 changes: 41 additions & 79 deletions cubids/cubids.py
View file
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
"""Main module."""

import copy
import csv
import json
import os
Expand Down Expand Up @@ -463,64 +464,17 @@ def change_filename(self, filepath, entities):
This is the function I need to spend the most time on, since it has entities hardcoded.
"""
bids_file = self.layout.get_file(filepath)
old_ext = bids_file.entities["extension"] # assume it starts with .
old_ext = old_ext if old_ext.startswith(".") else f".{old_ext}"

suffix = entities["suffix"]
entity_file_keys = []

# Entities can be used to distinguish files.
# Think of these as non-file collection entities, or acquisition-distinguishing entities.
file_keys = ["task", "acquisition", "direction", "reconstruction", "run"]

# Find distinguishing entities from the file name.
for key in file_keys:
if key in list(entities.keys()):
entity_file_keys.append(key)

# Subject is always present, and session is often present.
# XXX: Does this enforce a longitudinal dataset design?
sub = get_key_name(filepath, "sub")
ses = get_key_name(filepath, "ses")
sub_ses = sub + "_" + ses

if "run" in list(entities.keys()) and "run-0" in filepath:
# XXX: This adds an extra leading zero to run.
entities["run"] = "0" + str(entities["run"])

# Construct a version of the filename that only includes distinguishing entities.
# XXX: This is not a good thing. The output filename should retain the original
# filename's file-collection entities (e.g., echo).
filename = "_".join([f"{key}-{entities[key]}" for key in entity_file_keys])
filename = (
filename.replace("acquisition", "acq")
.replace("direction", "dir")
.replace("reconstruction", "rec")
)
if len(filename) > 0:
filename = sub_ses + "_" + filename + "_" + suffix + old_ext
else:
raise ValueError(f"Could not construct new filename for {filepath}")

# CHECK TO SEE IF DATATYPE CHANGED
# datatype may be overridden/changed if the original file is located in the wrong folder.
# XXX: In order for this to happen, the datatype in filepath must be different from the
# datatype in entities.
dtype_orig = self.layout.get_file(filepath).get_entities()["datatype"]

if ("datatype" in entities.keys()) and (entities["datatype"] != dtype_orig):
print(f"WARNING: DATATYPE ({dtype_orig}) CHANGED TO {entities['datatype']}")

# Construct the new filename
new_path = self.layout.build_path(entities)
orig_entities = bids_file.get_entities()
updated_entities = copy.deepcopy(orig_entities)
updated_entities.update(entities)
new_path = self.layout.build_path(updated_entities)

# Add the scan path + new path to the lists of old, new filenames
self.old_filenames.append(filepath)
self.new_filenames.append(new_path)

# NOW NEED TO RENAME ASSOCIATED FILES
# XXX: What do we count as associated files?
bids_file = self.layout.get_file(filepath)
# associations = bids_file.get_associations()
associations = self.get_nifti_associations(str(bids_file))
for assoc_path in associations:
Expand Down Expand Up @@ -552,66 +506,70 @@ def change_filename(self, filepath, entities):

# Update func-specific files
# now rename _events and _physio files!
old_suffix = parse_file_entities(filepath)["suffix"]
scan_end = "_" + old_suffix + old_ext
scan_end = f"_{orig_entities['suffix']}{orig_entities['extension']}"

if "_task-" in filepath:
old_events = filepath.replace(scan_end, "_events.tsv")
if Path(old_events).exists():
self.old_filenames.append(old_events)
new_scan_end = "_" + suffix + old_ext
new_events = new_path.replace(new_scan_end, "_events.tsv")
new_events = new_path.replace(scan_end, "_events.tsv")
self.new_filenames.append(new_events)

old_ejson = filepath.replace(scan_end, "_events.json")
if Path(old_ejson).exists():
self.old_filenames.append(old_ejson)
new_scan_end = "_" + suffix + old_ext
new_ejson = new_path.replace(new_scan_end, "_events.json")
new_ejson = new_path.replace(scan_end, "_events.json")
self.new_filenames.append(new_ejson)

old_physio = filepath.replace(scan_end, "_physio.tsv.gz")
if Path(old_physio).exists():
self.old_filenames.append(old_physio)
new_scan_end = "_" + suffix + old_ext
new_physio = new_path.replace(new_scan_end, "_physio.tsv.gz")
new_physio = new_path.replace(scan_end, "_physio.tsv.gz")
self.new_filenames.append(new_physio)

# Update ASL-specific files
if "/perf/" in filepath:
old_context = filepath.replace(scan_end, "_aslcontext.tsv")
if Path(old_context).exists():
self.old_filenames.append(old_context)
new_scan_end = "_" + suffix + old_ext
new_context = new_path.replace(new_scan_end, "_aslcontext.tsv")
new_context = new_path.replace(scan_end, "_aslcontext.tsv")
self.new_filenames.append(new_context)

old_m0scan = filepath.replace(scan_end, "_m0scan.nii.gz")
if Path(old_m0scan).exists():
self.old_filenames.append(old_m0scan)
new_scan_end = "_" + suffix + old_ext
new_m0scan = new_path.replace(new_scan_end, "_m0scan.nii.gz")
new_m0scan = new_path.replace(scan_end, "_m0scan.nii.gz")
self.new_filenames.append(new_m0scan)

old_mjson = filepath.replace(scan_end, "_m0scan.json")
if Path(old_mjson).exists():
self.old_filenames.append(old_mjson)
new_scan_end = "_" + suffix + old_ext
new_mjson = new_path.replace(new_scan_end, "_m0scan.json")
new_mjson = new_path.replace(scan_end, "_m0scan.json")
self.new_filenames.append(new_mjson)

old_labeling = filepath.replace(scan_end, "_asllabeling.jpg")
if Path(old_labeling).exists():
self.old_filenames.append(old_labeling)
new_scan_end = "_" + suffix + old_ext
new_labeling = new_path.replace(new_scan_end, "_asllabeling.jpg")
new_labeling = new_path.replace(scan_end, "_asllabeling.jpg")
self.new_filenames.append(new_labeling)

# RENAME INTENDED FORS!
ses_path = self.path + "/" + sub + "/" + ses
files_with_if = []
files_with_if += Path(ses_path).rglob("fmap/*.json")
files_with_if += Path(ses_path).rglob("perf/*_m0scan.json")
files_with_if += self.layout.get(
return_type="file",
datatype="fmap",
extension="json",
subject=orig_entities["subject"],
session=orig_entities.get("session", Query.NONE),
)
files_with_if += self.layout.get(
return_type="file",
datatype="perf",
suffix="m0scan",
extension="json",
subject=orig_entities["subject"],
session=orig_entities.get("session", Query.NONE),
)
for path_with_if in files_with_if:
filename_with_if = str(path_with_if)
self.IF_rename_paths.append(filename_with_if)
Expand Down Expand Up @@ -1560,6 +1518,8 @@ def _get_param_groups(

dfs.append(selected_metadata)

print(files[0])

# Assign each file to a ParamGroup
# Round parameter groups based on precision in config
df = round_params(pd.DataFrame(dfs), grouping_config, datatype)
Expand All @@ -1571,16 +1531,15 @@ def _get_param_groups(
# Get the subset of columns by which to drop duplicates
check_cols = []
for col in list(df.columns):
if f"Cluster_{col}" not in list(df.columns) and col != "FilePath":
if (
(f"Cluster_{col}" not in df.columns)
and (f"{col}__collection" not in df.columns)
and (col != "FilePath")
):
check_cols.append(col)

# Find the unique ParamGroups and assign ID numbers in "ParamGroup"
try:
deduped = df.drop("FilePath", axis=1)
except Exception:
# XXX: When would this be reached?
return "erroneous sidecar found"

deduped = df.drop("FilePath", axis=1)
deduped = deduped.drop_duplicates(subset=check_cols, ignore_index=True)
deduped["ParamGroup"] = np.arange(deduped.shape[0]) + 1

Expand Down Expand Up @@ -1699,7 +1658,7 @@ def format_params(param_group_df, config, datatype):
The datatype-wise dictionary's keys are names of BIDS fields to derive from the
NIfTI header and include in the Parameter Groupings.
"""
to_format = config["sidecar_params"][datatype]
to_format = copy.deepcopy(config["sidecar_params"][datatype])
to_format.update(config["derived_params"][datatype])

for fc_fields in file_collection_metadata.values():
Expand Down Expand Up @@ -1803,9 +1762,12 @@ def check_for_file_collection_conflicts(layout):
"""
errors = []
files = layout.get(return_type="object", extension=[".nii", ".nii.gz"])
valid_entities = layout.get_entities(metadata=False)
for file_ in files:
entities = file_.get_entities()
entities_to_check = entities.copy()
valid_entities = {k: Query.NONE for k in valid_entities.keys()}
valid_entities.update(entities)
entities_to_check = valid_entities.copy()
for ent in file_collection_metadata.keys():
if ent in entities.keys():
entities_to_check[ent] = [entities_to_check[ent], Query.NONE]
Expand Down
2 changes: 1 addition & 1 deletion cubids/metadata_merge.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -238,7 +238,7 @@ def merge_json_into_json(from_file, to_file, raise_on_error=False):
if not merged_metadata == orig_dest_metadata:
print("OVERWRITING", to_file)
with open(to_file, "w") as tofw:
json.dump(merged_metadata, tofw, indent=4)
json.dump(merged_metadata, tofw, indent=4, sort_keys=True)

return 0

Expand Down

0 comments on commit 1819219

Please sign in to comment.