use nipype's copyfile, create hardlinks if possible

Author: oesteban

fmriprep/viz/reports.py

@@ -15,19 +15,25 @@
 import re
 
 import html
-import shutil
 
 import jinja2
-from niworkflows.nipype.utils.filemanip import loadcrash
+from niworkflows.nipype.utils.filemanip import loadcrash, copyfile
 from pkg_resources import resource_filename as pkgrf
 
+
 class Element(object):
+    """
+    Just a basic component of a report
+    """
     def __init__(self, name, title=None):
         self.name = name
         self.title = title
 
 
 class Reportlet(Element):
+    """
+    A reportlet has title, description and a list of graphical components
+    """
     def __init__(self, name, file_pattern=None, title=None, description=None, raw=False):
         self.name = name
         self.file_pattern = re.compile(file_pattern)
@@ -39,6 +45,9 @@ def __init__(self, name, file_pattern=None, title=None, description=None, raw=Fa
 
 
 class SubReport(Element):
+    """
+    SubReports are sections within a Report
+    """
     def __init__(self, name, reportlets=None, title=''):
         self.name = name
         self.title = title
@@ -48,74 +57,13 @@ def __init__(self, name, reportlets=None, title=''):
         self.isnested = False
 
 
-def order_by_run(subreport):
-    ordered = []
-    run_reps = {}
-
-    for element in subreport.reportlets:
-        if len(element.source_files) == 1 and element.source_files[0]:
-            ordered.append(element)
-            continue
-
-        for filename, file_contents in zip(element.source_files, element.contents):
-            name, title = generate_name_title(filename)
-            if not filename or not name:
-                continue
-
-            new_element = Reportlet(
-                name=element.name, title=element.title, file_pattern=element.file_pattern,
-                description=element.description, raw=element.raw)
-            new_element.contents.append(file_contents)
-            new_element.source_files.append(filename)
-
-            if name not in run_reps:
-                run_reps[name] = SubReport(name, title=title)
-
-            run_reps[name].reportlets.append(new_element)
-
-    if run_reps:
-        keys = list(sorted(run_reps.keys()))
-        for key in keys:
-            ordered.append(run_reps[key])
-        subreport.isnested = True
-
-    subreport.reportlets = ordered
-    return subreport
-
-
-def generate_name_title(filename):
-    fname = os.path.basename(filename)
-    expr = re.compile('^sub-(?P<subject_id>[a-zA-Z0-9]+)(_ses-(?P<session_id>[a-zA-Z0-9]+))?'
-                      '(_task-(?P<task_id>[a-zA-Z0-9]+))?(_acq-(?P<acq_id>[a-zA-Z0-9]+))?'
-                      '(_rec-(?P<rec_id>[a-zA-Z0-9]+))?(_run-(?P<run_id>[a-zA-Z0-9]+))?')
-    outputs = expr.search(fname)
-    if outputs:
-        outputs = outputs.groupdict()
-    else:
-        return None, None
-
-    name = '{session}{task}{acq}{rec}{run}'.format(
-        session="_ses-" + outputs['session_id'] if outputs['session_id'] else '',
-        task="_task-" + outputs['task_id'] if outputs['task_id'] else '',
-        acq="_acq-" + outputs['acq_id'] if outputs['acq_id'] else '',
-        rec="_rec-" + outputs['rec_id'] if outputs['rec_id'] else '',
-        run="_run-" + outputs['run_id'] if outputs['run_id'] else ''
-    )
-    title = '{session}{task}{acq}{rec}{run}'.format(
-        session=" Session: " + outputs['session_id'] if outputs['session_id'] else '',
-        task=" Task: " + outputs['task_id'] if outputs['task_id'] else '',
-        acq=" Acquisition: " + outputs['acq_id'] if outputs['acq_id'] else '',
-        rec=" Reconstruction: " + outputs['rec_id'] if outputs['rec_id'] else '',
-        run=" Run: " + outputs['run_id'] if outputs['run_id'] else ''
-    )
-    return name.strip('_'), title
-
-
 class Report(object):
-
+    """
+    The full report object
+    """
     def __init__(self, path, config, out_dir, run_uuid, out_filename='report.html'):
         self.root = path
-        self.sub_reports = []
+        self.sections = []
         self.errors = []
         self.out_dir = out_dir
         self.out_filename = out_filename
@@ -127,7 +75,7 @@ def _load_config(self, config):
         with open(config, 'r') as configfh:
             config = json.load(configfh)
 
-        self.index(config['sub_reports'])
+        self.index(config['sections'])
 
     def index(self, config):
         fig_dir = 'figures'
@@ -142,21 +90,21 @@ def index(self, config):
             reportlets = []
             for reportlet_cfg in subrep_cfg['reportlets']:
                 rlet = Reportlet(**reportlet_cfg)
-                for f in reportlet_list:
-                    ext = f.split('.')[-1]
-                    if rlet.file_pattern.search(f):
+                for src in reportlet_list:
+                    ext = src.split('.')[-1]
+                    if rlet.file_pattern.search(src):
                         contents = None
                         if ext == 'html':
-                            with open(f) as fp:
+                            with open(src) as fp:
                                 contents = fp.read().strip()
                         elif ext == 'svg':
-                            fbase = os.path.basename(f)
-                            newf = os.path.join(svg_dir, fbase)
-                            shutil.copy(f, newf)
+                            fbase = os.path.basename(src)
+                            copyfile(src, os.path.join(svg_dir, fbase),
+                                     copy=True, use_hardlink=True)
                             contents = os.path.join(subject, fig_dir, fbase)
 
                         if contents:
-                            rlet.source_files.append(f)
+                            rlet.source_files.append(src)
                             rlet.contents.append(contents)
 
                 if rlet.source_files:
@@ -166,7 +114,7 @@ def index(self, config):
                 sub_report = SubReport(
                     subrep_cfg['name'], reportlets=reportlets,
                     title=subrep_cfg.get('title'))
-                self.sub_reports.append(order_by_run(sub_report))
+                self.sections.append(order_by_run(sub_report))
 
         error_dir = os.path.join(self.out_dir, "fmriprep", subject, 'log', self.run_uuid)
         if os.path.isdir(error_dir):
@@ -226,12 +174,75 @@ def generate_report(self):
             trim_blocks=True, lstrip_blocks=True
         )
         report_tpl = env.get_template('viz/report.tpl')
-        report_render = report_tpl.render(sub_reports=self.sub_reports, errors=self.errors)
+        report_render = report_tpl.render(sub_reports=self.sections, errors=self.errors)
         with open(os.path.join(self.out_dir, "fmriprep", self.out_filename), 'w') as fp:
             fp.write(report_render)
         return len(self.errors)
 
 
+def order_by_run(subreport):
+    ordered = []
+    run_reps = {}
+
+    for element in subreport.reportlets:
+        if len(element.source_files) == 1 and element.source_files[0]:
+            ordered.append(element)
+            continue
+
+        for filename, file_contents in zip(element.source_files, element.contents):
+            name, title = generate_name_title(filename)
+            if not filename or not name:
+                continue
+
+            new_element = Reportlet(
+                name=element.name, title=element.title, file_pattern=element.file_pattern,
+                description=element.description, raw=element.raw)
+            new_element.contents.append(file_contents)
+            new_element.source_files.append(filename)
+
+            if name not in run_reps:
+                run_reps[name] = SubReport(name, title=title)
+
+            run_reps[name].reportlets.append(new_element)
+
+    if run_reps:
+        keys = list(sorted(run_reps.keys()))
+        for key in keys:
+            ordered.append(run_reps[key])
+        subreport.isnested = True
+
+    subreport.reportlets = ordered
+    return subreport
+
+
+def generate_name_title(filename):
+    fname = os.path.basename(filename)
+    expr = re.compile('^sub-(?P<subject_id>[a-zA-Z0-9]+)(_ses-(?P<session_id>[a-zA-Z0-9]+))?'
+                      '(_task-(?P<task_id>[a-zA-Z0-9]+))?(_acq-(?P<acq_id>[a-zA-Z0-9]+))?'
+                      '(_rec-(?P<rec_id>[a-zA-Z0-9]+))?(_run-(?P<run_id>[a-zA-Z0-9]+))?')
+    outputs = expr.search(fname)
+    if outputs:
+        outputs = outputs.groupdict()
+    else:
+        return None, None
+
+    name = '{session}{task}{acq}{rec}{run}'.format(
+        session="_ses-" + outputs['session_id'] if outputs['session_id'] else '',
+        task="_task-" + outputs['task_id'] if outputs['task_id'] else '',
+        acq="_acq-" + outputs['acq_id'] if outputs['acq_id'] else '',
+        rec="_rec-" + outputs['rec_id'] if outputs['rec_id'] else '',
+        run="_run-" + outputs['run_id'] if outputs['run_id'] else ''
+    )
+    title = '{session}{task}{acq}{rec}{run}'.format(
+        session=" Session: " + outputs['session_id'] if outputs['session_id'] else '',
+        task=" Task: " + outputs['task_id'] if outputs['task_id'] else '',
+        acq=" Acquisition: " + outputs['acq_id'] if outputs['acq_id'] else '',
+        rec=" Reconstruction: " + outputs['rec_id'] if outputs['rec_id'] else '',
+        run=" Run: " + outputs['run_id'] if outputs['run_id'] else ''
+    )
+    return name.strip('_'), title
+
+
 def run_reports(reportlets_dir, out_dir, subject_label, run_uuid):
     """
     Runs the reports