fixed spaces in filenames issue

annotator.py

@@ -5,7 +5,7 @@
 import re, os
 import ast
 import json
-from urllib.parse import urlparse, unquote
+from urllib.parse import urlparse, unquote, quote
 from dateutil.parser import parse as date_parse
 from csv import Sniffer
 import requests
@@ -337,6 +337,9 @@ def __init__(
             self.url, self.encoding, self.authorization
         )
         self.file_domain = self.url.rsplit(self.file_name, 1)[0]
+        # use escaped name, no spaces allowed
+        self.file_name = quote(self.file_name)
+
         self.meta_file_name = self.file_name.rsplit(".", 1)[0] + "-metadata.json"
         self.csv_namespace = self.file_domain + self.file_name + "/"
         self.context = [
@@ -376,9 +379,16 @@ def annotate(self) -> dict:
         self.result_dict = self.process_data()
         return self.result_dict
 
-    def convert(self, format: str) -> str:
+    def graph(self) -> Graph:
         g = Graph()
         g.parse(data=json.dumps(self.result_dict), format="json-ld")
+        # with open("test.json", "w") as f:
+        #     json.dump(self.result_dict, f, indent=4)
+        return g
+
+    def convert(self, format: str) -> str:
+        g = self.graph()
+        # g.serialize("test.ttl", format="json-ld")
         self.meta_file_name = self.meta_file_name.rsplit(".", 1)[0]
         if format in ["turtle", "longturtle"]:
             self.meta_file_name += ".ttl"

app.py

@@ -1,7 +1,7 @@
 # app.py
 import os
 import base64
-from urllib.parse import urlparse
+from urllib.parse import urlparse, quote
 
 import uvicorn
 from starlette_wtf import StarletteForm, CSRFProtectMiddleware, csrf_protect
@@ -355,12 +355,13 @@ async def annotate_upload(
     # add prov o documentation
     result = {**result, **annotate_prov(request.url._url)}
     data = annotator.convert(format=return_type.value)
+
     data_bytes = BytesIO(data.encode())
     filename = annotator.meta_file_name
     # delete the temp csv file
-    if os.path.isfile(file.filename):
-        os.remove(file.filename)
-    return RDFStreamingResponse(content=data_bytes, filename=filename)
+    # if os.path.isfile(file.filename):
+    #     os.remove(file.filename)
+    return RDFStreamingResponse(content=data_bytes, filename=quote(filename))
 
 
 @app.post("/api/rdf", response_class=RDFStreamingResponse)

csvw_parser.py

@@ -7,7 +7,7 @@
 from rdflib.namespace import CSVW, RDF, XSD, PROV, RDFS, DC
 from datetime import datetime
 from urllib.request import urlopen
-from urllib.parse import urlparse, unquote
+from urllib.parse import urlparse, unquote, quote
 import io, os
 import logging
 import requests
@@ -186,7 +186,7 @@ def __init__(
         # self.metagraph.serialize('metagraph.ttl')
         print("meta_root: " + self.meta_root)
         # print('csv_url: '+url)
-        self.base_url = "{}/".format(str(self.meta_root).rsplit("/", 1)[0])
+        self.base_url = "{}/".format(quote(str(self.meta_root).rsplit("/", 1)[0]))
         parsed_url = urlparse(url)
         if parsed_url.scheme in ["https", "http", "file"]:
             self.csv_url = url