v0.1 (#5)

* better errors

* error handling when missing metadata

* remove test ticket

* original format tabular files

Author: mccallc

.github/workflows/build.yml

@@ -1,6 +1,9 @@
-name: Build
-on: push
+on: 
+  push:
+    tags:
+      - 'v*' # Push events to matching v*, i.e. v1.0, v20.15.10
 
+name: Build
 jobs:
 
   createrelease:
@@ -15,7 +18,7 @@ jobs:
       with:
         tag_name: ${{ github.ref }}
         release_name: Release ${{ github.ref }}
-        draft: false
+        draft: true
         prerelease: false
     - name: Output Release URL File
       run: echo "${{ steps.create_release.outputs.upload_url }}" > release_url.txt
@@ -42,15 +45,15 @@ jobs:
             ASSET_MIME: application/zip
           - os: windows-latest
             TARGET: windows
-            CMD_BUILD: pyinstaller -F -w -n dvcurator gui.spec
+            CMD_BUILD: pyinstaller gui.spec
             OUT_FILE_NAME: dvcurator.exe
             ASSET_MIME: application/vnd.microsoft.portable-executable
     steps:
     - uses: actions/checkout@v1
-    - name: Set up Python 3.8
+    - name: Set up Python 3.10
       uses: actions/setup-python@v2
       with:
-        python-version: 3.8
+        python-version: 3.10.4
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip

dvcurator/__main__.py

@@ -1,2 +1,2 @@
-import sys, gui
-sys.exit(gui.main())
+import sys, dvcurator.gui
+sys.exit(dvcurator.gui.main())

dvcurator/dataverse.py

@@ -5,6 +5,10 @@
 def get_citation(host, doi, token=""):
 	import requests
 
+	if not doi.startswith("doi:"):
+		print("Error: DOIs should start with \"doi:\"")
+		return None
+	
 	# Scrape data and metadata from dataverse
 	dataset_url = 'https://' + host 
 	dataset_url += '/api/datasets/:persistentId/?persistentId=' + doi
@@ -13,6 +17,11 @@ def get_citation(host, doi, token=""):
 	else:
 		key = {'X-Dataverse-Key': token}
 		dataset = requests.get(dataset_url, headers=key)
+		
+	if (dataset.json()['status']=="ERROR"):
+		print("Error: " + dataset.json()['message'])
+		return None
+		
 	citation=dataset.json()['data']['latestVersion']['metadataBlocks']['citation']['fields']
 	fields = [] # Make an index of all the metadata fields
 	values = []
@@ -36,6 +45,7 @@ def download_dataset(host, doi, token, folder_name, dropbox):
 
 	zip_url = 'https://' + host
 	zip_url += '/api/access/dataset/:persistentId/?persistentId=' + doi
+	zip_url += '&format=original'
 	if token:
 		key = {'X-Dataverse-Key': token}
 		r = requests.get(zip_url, headers=key, allow_redirects=True, stream=True)

dvcurator/github.py

@@ -2,6 +2,21 @@
 # -*- coding: utf-8 -*-
 #
 
+def check_repo(repo, key=None):
+	import requests
+	github='https://api.github.com'
+	project_url = github + "/repos/" + repo + "/issues"
+	if (not key):
+		projects = requests.get(project_url + "?per_page=100")
+	else:
+		key = {'Authorization': "token " + key}
+		projects = requests.get(project_url + "?per_page=100", headers=key)
+
+	if (projects.status_code==404):
+		return False
+	else:
+		return True
+
 def search_existing(project_name, repo, key=None):
 	import json, requests
 	github='https://api.github.com'
@@ -15,7 +30,7 @@ def search_existing(project_name, repo, key=None):
 	else:
 		key = {'Authorization': "token " + key}
 		projects = requests.get(project_url + "?per_page=100", headers=key)
-		
+				
 	# Take the first three words ("lastname - first-of-title") to search
 	project_name = ' '.join(project_name.split()[:3])
 

dvcurator/gui.py

@@ -8,7 +8,7 @@
 
 import tkinter as tk
 import sys
-import github, dataverse, pdf_metadata
+import dvcurator.github, dvcurator.dataverse, dvcurator.pdf_metadata
 
 def project_name(last_name, title):
 	import re
@@ -80,9 +80,17 @@ def load_citation(self):
 			return
 
 		#from . import dataverse
-		self.citation = dataverse.get_citation(self.host.get(), self.doi.get(), self.dv_token.get())
+		self.citation = dvcurator.dataverse.get_citation(self.host.get(), self.doi.get(), self.dv_token.get())
+		if (not self.citation):
+			return
+			
 		# citation['depositor'].split(', ')[0] is the last name of the depositor
 		self.folder_name = project_name(self.citation['depositor'].split(', ')[0], self.citation['title'])
+		
+		special_characters = ['!','#','$','%', '&','@','[',']',']','_',':',';',"'"]
+		for i in special_characters:
+			self.folder_name = self.folder_name.replace(i,'')
+			
 		print(self.folder_name)
 
 		# Enable the next step buttons
@@ -93,7 +101,7 @@ def load_citation(self):
 
 	def download_extract(self):
 		#from . import dataverse
-		extracted_path = dataverse.download_dataset(self.host.get(), self.doi.get(), self.dv_token.get(), self.folder_name, self.dropbox.get())
+		extracted_path = dvcurator.dataverse.download_dataset(self.host.get(), self.doi.get(), self.dv_token.get(), self.folder_name, self.dropbox.get())
 		if not extracted_path:
 			print("Error: folder may already exist")
 		else:
@@ -110,15 +118,19 @@ def make_github(self):
 			print("Error: no github repository specified")
 			return
 
+		if (not dvcurator.github.check_repo(self.repo.get())):
+			print("Error: github repository doesn't exist")
+			return
+			
 		#from . import github
-		existing = github.search_existing(self.folder_name, self.repo.get(), self.gh_token.get())
+		existing = dvcurator.github.search_existing(self.folder_name, self.repo.get(), self.gh_token.get())
 		if (existing):
 			print("Error: existing github issues!!")
 			return
 
 
 		# Create github project + issues
-		self.project = github.create_project(self.doi.get(), self.citation, self.folder_name, self.repo.get(), self.gh_token.get())
+		self.project = dvcurator.github.create_project(self.doi.get(), self.citation, self.folder_name, self.repo.get(), self.gh_token.get())
 		print("Created project: " + self.folder_name)
 		# Get internal issue templates from selected checkboxes
 		for issue in self.issues_selected:
@@ -128,7 +140,7 @@ def make_github(self):
 					path = os.path.join(sys._MEIPASS, "issues", path)
 				else:
 					path = resource_filename("dvcurator", "issues/" + path)
-				github.add_issue(self.folder_name, path, self.repo.get(), self.project, self.gh_token.get())
+				dvcurator.github.add_issue(self.folder_name, path, self.repo.get(), self.project, self.gh_token.get())
 				print(issue.get() + " added to project")
 
 	def set_metadata(self):
@@ -138,7 +150,7 @@ def set_metadata(self):
 			metadata_path = pdf_metadata.make_metadata_folder(self.dropbox.get(), self.folder_name)
 			if (not metadata_path): # Errors are outputted by pdf_metadata
 				return
-			pdf_metadata.standard_metadata(metadata_path, self.citation['depositor'])
+			dvcurator.pdf_metadata.standard_metadata(metadata_path, self.citation['depositor'])
 			print("PDF metadata updated in new folder")
 		else:
 			print("Error: Dropbox folder invalid")

dvcurator/issues/test.md

@@ -49,13 +49,18 @@ def standard_metadata(edit_path, author):
 	for path in pdfs:
 		pdf = pikepdf.open(path, allow_overwriting_input=True)
 		# Clean out all existing metadata
-		#del pdf.Root.Metadata
-		#def pdf.docinfo
+		try:
+			del pdf.Root.Metadata
+		except:
+			pass
+			
+		del pdf.docinfo
 
+		# Write new metadata
 		with pdf.open_metadata() as meta:
 			if meta.pdfa_status:
 				print("Warning: Edited PDF claims PDF/A")
-			#meta['dc:title'] = os.path.basename(path)
+			meta['dc:title'] = os.path.basename(path)
 			meta['dc:creator'] = author
 			meta['pdf:Author'] = author
 			meta['dc:description'] = "QDR Data Project"
@@ -64,4 +69,4 @@ def standard_metadata(edit_path, author):
 
 		pdf.save(path)
 		return True
-		#print("Metadata written to '%s'" %path)
+		print("Metadata written to '%s'" %path)

dvcurator/pdf_metadata.py

@@ -8,7 +8,7 @@ a = Analysis(['dvcurator\\gui.py'],
              pathex=['dvcurator/'],
              binaries=[],
              datas=[('dvcurator\\issues', 'issues')],
-             hiddenimports=[],
+             hiddenimports=['dvcurator'],
              hookspath=[],
              hooksconfig={},
              runtime_hooks=[],
@@ -27,6 +27,8 @@ exe = EXE(pyz,
           a.datas,  
           [],
           name='dvcurator',
+          onefile=True,
+          windowed=True, 
           debug=False,
           bootloader_ignore_signals=False,
           strip=False,