no copyfile

Author: antgonza

qiita_pet/handlers/download.py

@@ -38,7 +38,6 @@
 from tempfile import mkdtemp
 from zipfile import ZipFile
 from io import BytesIO
-from shutil import copyfile
 
 
 class BaseHandlerDownload(BaseHandler):
@@ -383,6 +382,10 @@ class DownloadDataReleaseFromPrep(BaseHandlerDownload):
     @coroutine
     @execute_as_transaction
     def get(self, prep_template_id):
+        """ This method constructs an on the fly ZIP with all the files
+            required for a data-prep release/data-delivery. Mainly sample, prep
+            info, bioms and coverage
+        """
         user = self.current_user
         if user.level not in ('admin', 'web-lab admin'):
             raise HTTPError(403, reason="%s doesn't have access to download "
@@ -411,6 +414,7 @@ def get(self, prep_template_id):
             '',
         ]
 
+        # helper dict to add "user/human" friendly names to the bioms
         human_names = {
             'ec.biom': 'KEGG Enzyme (EC)',
             'per-gene.biom': 'Per gene Predictions',
@@ -421,18 +425,21 @@ def get(self, prep_template_id):
             'rna_copy_counts.biom': 'RNA copy counts'
         }
 
+        # sample-info creation
         fn = join(td, f'sample_information_from_prep_{pid}.tsv')
         readme.append(f'Sample information: {basename(fn)}')
-        files.append(fn)
+        files.append([fn, basename(fn)])
         st.to_dataframe(samples=list(pt)).to_csv(fn, sep='\t')
 
+        # prep-info creation
         fn = join(td, f'prep_information_{pid}.tsv')
         readme.append(f'Prep information: {basename(fn)}')
-        files.append(fn)
+        files.append([fn, basename(fn)])
         pt.to_dataframe().to_csv(fn, sep='\t')
 
         readme.append('')
 
+        # finding the bioms to be added
         bioms = dict()
         coverages = None
         for a in Study(sid).artifacts(artifact_type='BIOM'):
@@ -447,25 +454,27 @@ def get(self, prep_template_id):
             if biom is None:
                 continue
             biom_fn = basename(biom['fp'])
+            # there is a small but real chance that the same prep has the same
+            # artifacts so using the latests
             if biom_fn not in bioms:
                 bioms[biom_fn] = [a, biom]
             else:
                 if getctime(biom['fp']) > getctime(bioms[biom_fn][1]['fp']):
                     bioms[biom_fn] = [a, biom]
 
+        # once we have all the bioms, we can add them to the list of zips
+        # and to the readme the biom details and all the processing
         for fn, (a, fp) in bioms.items():
             aname = basename(fp["fp"])
             nname = f'{a.id}_{aname}'
-            nfile = join(td, nname)
-            copyfile(fp['fp'], nfile)
-            files.append(nfile)
+            files.append(fp['fp'], nname)
 
             hname = ''
             if aname in human_names:
                 hname = human_names[aname]
             readme.append(f'{nname}\t{hname}')
 
-            for an in a.ancestors.nodes():
+            for an in set(a.ancestors.nodes()):
                 p = an.processing_parameters
                 if p is not None:
                     c = p.command
@@ -476,22 +485,22 @@ def get(self, prep_template_id):
                     pd = p.dump()
                     readme.append(f'\t{cn}\t{sn}\t{sv}\t{pd}')
 
+        # if a coverage was found, add it to the list of files
         if coverages is not None:
-            aname = basename(coverages)
-            nfile = join(td, aname)
-            copyfile(coverages, nfile)
-            files.append(nfile)
+            fn = basename(coverages)
+            readme.append(f'{fn}\tcoverage files')
+            files.append([coverages, fn])
 
         fn = join(td, 'README.txt')
         with open(fn, 'w') as fp:
             fp.write('\n'.join(readme))
-        files.append(fn)
+        files.append([fn, basename(fn)])
 
         zp_fn = f'data_release_{pid}_{date}.zip'
         zp = BytesIO()
         with ZipFile(zp, 'w') as zipf:
-            for fp in files:
-                zipf.write(fp, basename(fp))
+            for fp, fn in files:
+                zipf.write(fp, fn)
 
         self.set_header('Content-Type', 'application/zip')
         self.set_header("Content-Disposition", f"attachment; filename={zp_fn}")