Retiring OutputAction for arbitrary additional CSV report in favor of…

… CSVFileType

Fixing mc medias command
Fixing ct summary command

minet/cli/argparse.py

@@ -345,6 +345,25 @@ def __call__(self, string):
         return d
 
 
+class CSVFileType:
+    def __init__(self, mode="r"):
+        self.mode = mode
+
+    def __call__(self, string):
+        if string == "-":
+            return acquire_cross_platform_stdout()
+
+        try:
+            # As per #254: newline='' is necessary for CSV output on windows to avoid
+            # outputting extra lines because of a '\r\r\n' end of line...
+            return open(string, self.mode, encoding="utf-8", newline="")
+        except OSError as e:
+            raise ArgumentTypeError(
+                "can't open '%(filename)s': %(error)s"
+                % {"filename": string, "error": e}
+            )
+
+
 class SplitterType:
     def __init__(self, splitchar=","):
         self.splitchar = splitchar

minet/cli/crowdtangle/__init__.py

@@ -6,7 +6,7 @@
 #
 from casanova import RowCountResumer, LastCellResumer
 
-from minet.cli.argparse import BooleanAction, SplitterType, OutputAction
+from minet.cli.argparse import BooleanAction, SplitterType, CSVFileType
 
 # TODO: lazyloading issue
 from minet.crowdtangle.constants import (
@@ -282,7 +282,7 @@
         {
             "name": "--posts",
             "help": "Path to a file containing the retrieved posts.",
-            "action": OutputAction,
+            "type": CSVFileType("w"),
         },
         {
             "flag": "--sort-by",

minet/cli/mediacloud/__init__.py

@@ -7,7 +7,7 @@
 from minet.cli.argparse import (
     command,
     ConfigAction,
-    OutputAction,
+    CSVFileType,
     SplitterType,
 )
 
@@ -23,7 +23,7 @@
         {
             "flag": "--feeds",
             "help": "If given, path of the CSV file listing media RSS feeds.",
-            "action": OutputAction,
+            "type": CSVFileType("w"),
         }
     ],
 )

minet/cli/mediacloud/medias.py

@@ -15,35 +15,42 @@
 )
 
 
+def get_headers(cli_args):
+    headers = MEDIACLOUD_MEDIA_CSV_HEADER[1:]
+
+    if cli_args.feeds is not None:
+        headers.append("feeds")
+
+    return headers
+
+
 @with_mediacloud_fatal_errors
 @with_enricher_and_loading_bar(
-    headers=MEDIACLOUD_MEDIA_CSV_HEADER[1:], title="Fetching medias", unit="medias"
+    headers=get_headers, title="Fetching medias", unit="medias"
 )
 def action(cli_args, enricher, loading_bar):
-    added_headers = MEDIACLOUD_MEDIA_CSV_HEADER[1:]
-
     feeds_writer = None
 
     if cli_args.feeds:
-        added_headers.append("feeds")
         feeds_writer = casanova.writer(
             cli_args.feeds, fieldnames=MEDIACLOUD_FEED_CSV_HEADER
         )
 
     client = MediacloudAPIClient(cli_args.token)
 
     for row, media_id in enricher.cells(cli_args.column, with_rows=True):
-        result = client.media(media_id)
-        result = result.as_csv_row()[1:]
+        with loading_bar.step():
+            result = client.media(media_id)
+            result = result.as_csv_row()[1:]
 
-        if cli_args.feeds:
-            feeds = client.feeds(media_id)
+            if cli_args.feeds:
+                assert feeds_writer is not None
 
-            enricher.writerow(row, result + [len(feeds)])
+                feeds = client.feeds(media_id)
 
-            for feed in feeds:
-                feeds_writer.writerow(feed.as_csv_row())
-        else:
-            enricher.writerow(row, result)
+                enricher.writerow(row, result + [len(feeds)])
 
-        loading_bar.advance()
+                for feed in feeds:
+                    feeds_writer.writerow(feed.as_csv_row())
+            else:
+                enricher.writerow(row, result)