added structure for supplying extra json records with (optional) file…

… metadata for the uploaded sample files. (#5)
IQSS · Oct 7, 2019 · cb0f21f · cb0f21f
1 parent 7d00b37
commit cb0f21f
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 2 deletions.
diff --git a/create_sample_data.py b/create_sample_data.py
@@ -60,6 +60,7 @@ def check_dataset_lock(dataset_dbid):
         dataset_pid = resp.json()['data']['persistentId']
         dataset_dbid = resp.json()['data']['id']
         files_dir = path.replace(json_file, '') + 'files'
+        filemetadata_dir = path.replace(json_file, '') + '.filemetadata'
         print(files_dir)
         for path,subdir,files in os.walk(files_dir):
            for name in files:
@@ -73,9 +74,19 @@ def check_dataset_lock(dataset_dbid):
                 ## This lock check and sleep is here to prevent the dataset from being permanently
                 ## locked because a tabular file was uploaded first.
                 check_dataset_lock(dataset_dbid)
-                file_metadata = {}
                 # TODO: Think more about where the description comes from. A "sidecar" file as proposed at https://github.com/IQSS/dataverse/issues/5924#issuecomment-499605672 ?
-                #file_metadata['description'] = 'Sidecar?'
+                # L.A.: I implemented something along these lines - an (optional) directory called ".filemetadata" 
+                # in the dataset directory, where files containing extra json filemetadata records may be 
+                # placed for each of the files in the "files" directory. 
+                # (since the file names must be unique per dataset, even with folders, the .filemetadata 
+                # directory structure is flat)
+                # check for optional filemetadata file:
+                filemetadatapath = os.path.join(filemetadata_dir, filename);
+                if (os.path.exists(filemetadatapath)):
+                    with open(filemetadatapath) as m:
+                        file_metadata = json.load(m)
+                else:
+                    file_metadata = {}
                 file_metadata['directoryLabel'] = directoryLabel
                 jsonData = json.dumps(file_metadata)
                 data = { 'jsonData' : jsonData }

diff --git a/...erses/open-source-at-harvard/datasets/open-source-at-harvard/.filemetadata/2019-02-25.tsv b/...erses/open-source-at-harvard/datasets/open-source-at-harvard/.filemetadata/2019-02-25.tsv
@@ -0,0 +1,3 @@
+{
+   "description": "The tabular file contains information on known Harvard repositories on GitHub, such as the number of stars, programming language, day last updated, number of open issues, size, number of forks, repository URL, create date, and description."	
+}