Merge pull request #187 from KnowledgeCaptureAndDiscovery/dev

Fix bugs with codemeta export

.gitignore

@@ -10,3 +10,4 @@ Lib/*
 Scripts/*
 .idea/*
 test.json
+out.json

src/somef/cli.py

@@ -347,7 +347,8 @@ def get_path(obj, path):
         print(dockerfiles)
 
     if len(notebooks) > 0:
-        filtered_resp["hasExecutableNotebook"] = [convert_to_raw_usercontent(x, owner, repo_name, repo_ref) for x in notebooks]
+        filtered_resp["hasExecutableNotebook"] = [convert_to_raw_usercontent(x, owner, repo_name, repo_ref) for x in
+                                                  notebooks]
     if len(dockerfiles) > 0:
         filtered_resp["hasBuildFile"] = [convert_to_raw_usercontent(x, owner, repo_name, repo_ref) for x in dockerfiles]
     if len(docs) > 0:
@@ -369,6 +370,7 @@ def get_path(obj, path):
 def convert_to_raw_usercontent(partial, owner, repo_name, repo_ref):
     return f"https://raw.githubusercontent.com/{owner}/{repo_name}/{repo_ref}/{urllib.parse.quote(partial)}"
 
+
 ## Function takes readme text as input and divides it into excerpts
 ## Returns the extracted excerpts
 def create_excerpts(string_list):
@@ -595,16 +597,19 @@ def merge(header_predictions, predictions, citations, dois, binder_links, long_t
 def format_output(git_data, repo_data):
     print("formatting output")
     for i in git_data.keys():
+        # print(i)
+        # print(git_data[i])
         if i == 'description':
             if 'description' not in repo_data.keys():
                 repo_data['description'] = []
-            repo_data['description'].append({'excerpt': git_data[i], 'confidence': [1.0], 'technique': 'GitHub API'})
+            if git_data[i] != "":
+                repo_data['description'].append(
+                    {'excerpt': git_data[i], 'confidence': [1.0], 'technique': 'GitHub API'})
         else:
             if i == 'hasExecutableNotebook' or i == 'hasBuildFile' or i == 'hasDocumentation':
                 repo_data[i] = {'excerpt': git_data[i], 'confidence': [1.0], 'technique': 'File Exploration'}
-            else:
+            elif git_data[i] != "" and git_data[i] != []:
                 repo_data[i] = {'excerpt': git_data[i], 'confidence': [1.0], 'technique': 'GitHub API'}
-
     return repo_data
 
 
@@ -670,13 +675,18 @@ def average_confidence(x):
                       reverse=True)
     descriptions_text = [x["excerpt"] for x in descriptions]
 
+    published_date = ""
+    try:
+        published_date = format_date(release_path(["datePublished"]))
+    except:
+        print("Published date is not available")
+
     codemeta_output = {
         "@context": "https://doi.org/10.5063/schema/codemeta-2.0",
         "@type": "SoftwareSourceCode",
         "license": data_path(["license", "excerpt", "url"]),
         "codeRepository": "git+" + code_repository + ".git",
         "dateCreated": format_date(data_path(["dateCreated", "excerpt"])),
-        "datePublished": format_date(release_path(["datePublished"])),
         "dateModified": format_date(data_path(["dateModified", "excerpt"])),
         "downloadUrl": data_path(["downloadUrl", "excerpt"]),
         "issueTracker": code_repository + "/issues",
@@ -694,6 +704,8 @@ def average_confidence(x):
             }
         ]
     }
+    if published_date != "":
+        codemeta_output["datePublished"] = published_date
 
     pruned_output = {}