Replace df with csv format to avoid pandas dependency

Pandas was used for demo and output formatting only. With instructions
on pandas usage in the README the dataframe formatting is removed.
Instead an option to output CSV format is added.

The command-line argument is renamed from `--display` to `--format`.

Signed-off-by: cwasicki <[email protected]>

Author: cwasicki

src/frequenz/client/reporting/__main__.py

@@ -9,12 +9,9 @@
 from pprint import pprint
 from typing import AsyncIterator
 
-import pandas as pd
 from frequenz.client.common.metric import Metric
 
 from frequenz.client.reporting import ReportingApiClient
-
-# Experimental import
 from frequenz.client.reporting._client import MetricSample
 
 
@@ -52,7 +49,7 @@ def main() -> None:
     parser.add_argument("--resolution", type=int, help="Resolution", default=None)
     parser.add_argument("--psize", type=int, help="Page size", default=1000)
     parser.add_argument(
-        "--display", choices=["iter", "df", "dict"], help="Display format", default="df"
+        "--format", choices=["iter", "csv", "dict"], help="Output format", default="csv"
     )
     parser.add_argument(
         "--key",
@@ -72,7 +69,7 @@ def main() -> None:
             page_size=args.psize,
             service_address=args.url,
             key=args.key,
-            display=args.display,
+            fmt=args.format,
         )
     )
 
@@ -88,7 +85,7 @@ async def run(
     page_size: int,
     service_address: str,
     key: str,
-    display: str,
+    fmt: str,
 ) -> None:
     """Test the ReportingApiClient.
 
@@ -102,10 +99,10 @@ async def run(
         page_size: page size
         service_address: service address
         key: API key
-        display: display format
+        fmt: output format
 
     Raises:
-        ValueError: if display format is invalid
+        ValueError: if output format is invalid
     """
     client = ReportingApiClient(service_address, key)
 
@@ -129,26 +126,25 @@ def data_iter() -> AsyncIterator[MetricSample]:
             page_size=page_size,
         )
 
-    if display == "iter":
+    if fmt == "iter":
         # Iterate over single metric generator
         async for sample in data_iter():
             print(sample)
 
-    elif display == "dict":
+    elif fmt == "dict":
         # Dumping all data as a single dict
         dct = await iter_to_dict(data_iter())
         pprint(dct)
 
-    elif display == "df":
-        # Turn data into a pandas DataFrame
-        data = [cd async for cd in data_iter()]
-        df = pd.DataFrame(data).set_index("timestamp")
-        # Set option to display all rows
-        pd.set_option("display.max_rows", None)
-        pprint(df)
+    elif fmt == "csv":
+        # Print header
+        print(",".join(MetricSample._fields))
+        # Iterate over single metric generator and format as CSV
+        async for sample in data_iter():
+            print(",".join(str(e) for e in sample))
 
     else:
-        raise ValueError(f"Invalid display format: {display}")
+        raise ValueError(f"Invalid output format: {fmt}")
 
     return