Support downloading XLSX file inside Zip file, to comply with DMP+Zenodo

README.md

@@ -187,12 +187,13 @@ FADN_FILES_LOCATION | Directory where FADN datasets are downloaded and cached |
 CACHE_FILE_LOCATION | Directory where SDMX datasets are downloaded and cached | "/srv/sdmx_datasets_cache" |
 REDIS_HOST_FILESYSTEM_DIR | If REDIS_HOST='filesystem:local_session', directory where sessions are stored | "/srv/sessions" |
 SSP_FILES_DIR | Not used | "" |
+NIS_FILES_LIST | A comma-separated list of URLs to CSV files where NIS case studies or parts of them are enumerated. Each CSV file starts with a header, with four columns: name, url, description and example (True if it is an example) | "" |  
 REDIS_HOST | "localhost" expects a REDIS server available at localhost:6379; "redis-local" creates a local REDIS instance; "filesystem:local_session" uses filesystem to store sessions (a good option for execution in PC/laptop) | "" |
 TESTING | "True"| "True" |
 SELF_SCHEMA | Name of the host where Backend RESTful service responds, preceded by the protocol (http or https) | "https://one.nis.magic-nexus.eu/" |
 FS_TYPE | "Webdav" | "Webdav" |
 FS_SERVER | Host name of the WebDAV server | "nextcloud.data.magic-nexus.eu" |
-FS_USER | User name used. Files and folders must be readable and writeable by this user | "<webdav user>" |
+FS_USER | User name used. Files and folders must be readable and writable by this user | "<webdav user>" |
 FS_PASSWORD | Password for the previous user | "<password in clear>" |
 GAPI_CREDENTIALS_FILE | Path to a file obtained from Google API management web, to directly access a NIS workbook file in Google Sheets | "/srv/credentials.json" |
 GAPI_TOKEN_FILE | Path to a file used to stored authorization token | "/srv/tocken.pickle" |

nexinfosys/common/helper.py

@@ -31,6 +31,7 @@
 from nexinfosys.ie_imports.google_drive import download_xlsx_file_id
 from nexinfosys.models import log_level
 import os
+from zipfile import ZipFile
 
 logger = logging.getLogger(__name__)
 logger.setLevel(log_level)
@@ -1196,6 +1197,48 @@ def download_file(location, wv_user=None, wv_password=None, wv_host_name=None):
         data = urllib.request.urlopen(location).read()
         data = io.BytesIO(data)
 
+    # If it is a ZIP file...
+    if pr.path.lower().endswith(".zip"):
+        zipfile = ZipFile(data)
+        # If no anchor, return the XLSX file (find it and return it)
+        file_to_extract = None
+        candidate_files_to_extract = []
+        if pr.fragment:
+            file_to_extract = pr.fragment
+            if "#" in file_to_extract:
+                pos = file_to_extract.find("#")
+                sub_fragment = file_to_extract[pos + 1:]
+                file_to_extract = file_to_extract[:pos]
+            if file_to_extract.startswith("/"):
+                file_to_extract = file_to_extract[1:]
+        found = False
+        for name in zipfile.namelist():
+            if not file_to_extract:
+                if name.lower().endswith(".xlsx"):
+                    candidate_files_to_extract.append(name)
+                    found = True
+            else:
+                if strcmp(file_to_extract, name):
+                    found = True
+                    break
+        if found:
+            if not file_to_extract:
+                # Try to find best option from the list of candidates
+                file_to_extract = candidate_files_to_extract[0]
+                for f in candidate_files_to_extract:
+                    if f.lower().startswith("msm/cs"):
+                        file_to_extract = f
+                        break  # Best option (DMP): a file named "cs_<whatever>.xlsx" in "msm" folder
+                    if f.lower().startswith("msm/"):
+                        file_to_extract = f  # Second best option: an XLSX file inside "msm" folder
+
+            if file_to_extract:
+                data = io.BytesIO(zipfile.read(file_to_extract))
+            else:
+                data = None
+        else:
+            data = None
+
     return data