From 4a738a56ca5c7d4798d52f1b37e41d7bc92c7af6 Mon Sep 17 00:00:00 2001
From: Jessy Barrette <30420025+JessyBarrette@users.noreply.github.com>
Date: Tue, 20 Aug 2024 12:05:01 -0400
Subject: [PATCH] add input-path-list (#108)

* add input-path-list
* add os path separator compatibility to both cli and configuration
---
 CHANGELOG.md                                  |  2 ++
 ocean_data_parser/batch/convert.py            | 12 +++++++--
 .../batch/default-batch-config.yaml           |  4 ++-
 tests/test_batch.py                           | 27 +++++++++++++++++++
 tests/test_cli.py                             | 27 +++++++++++++++++++
 5 files changed, 69 insertions(+), 3 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index afdd12dc..9805e655 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -13,6 +13,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ability to list multiple file glob expression and associated metadata.
 - Add Onset.csv timestamp format: "\d+\/\d+\/\d\d\d\d\s+\d+\:\d+\:\d+" = "%m/%d/%Y %H:%M:%S"
 - Rely on ruff for format and linter testing
+- Add option to pass a list of input_path paths via the configuration file or a 
+os path seperator list via the command line interface or the configuration
 
 ### Fixed
 
diff --git a/ocean_data_parser/batch/convert.py b/ocean_data_parser/batch/convert.py
index 10da5752..d6d97d35 100644
--- a/ocean_data_parser/batch/convert.py
+++ b/ocean_data_parser/batch/convert.py
@@ -86,7 +86,11 @@ def validate_parser_kwargs(ctx, _, value):
     "-i",
     "--input-path",
     type=str,
-    help="Input path to file list. It can be a glob expression (ex: *.cnv)",
+    help=(
+        "Input path to file list. It can be a glob expression (ex: *.cnv)"
+        " or a list of paths separated by a colons [:] (linux,mac) "
+        "and semi-colons [;] (windows)."
+    ),
 )
 @click.option(
     "--exclude",
@@ -248,9 +252,13 @@ def get_excluded_files(self) -> list:
 
     def get_source_files(self) -> list:
         excluded_files = self.get_excluded_files()
+        paths = self.config["input_path"]
+        paths = paths.split(os.pathsep) if isinstance(paths, str) else paths
+
         return [
             Path(file)
-            for file in glob(self.config["input_path"], recursive=True)
+            for path in paths
+            for file in glob(path, recursive=True)
             if file not in excluded_files
         ]
 
diff --git a/ocean_data_parser/batch/default-batch-config.yaml b/ocean_data_parser/batch/default-batch-config.yaml
index 3919e034..07ecf695 100644
--- a/ocean_data_parser/batch/default-batch-config.yaml
+++ b/ocean_data_parser/batch/default-batch-config.yaml
@@ -1,7 +1,9 @@
 ---
 
 name: Batch Conversion Name
-input_path: ""  # file or glob expression
+input_path: ""  # glob expression of files to convert.
+  # It can be a str, a list of os path separator (: for unix, ; for windows)
+  # separated list of paths or a list of paths
 input_table:  # retrieve files to convert from tables (ignore input_path if set)
   # input_table is used to retrieve files to convert from tables
   # that are listing different glob expressions to retrieve files and associated metadata
diff --git a/tests/test_batch.py b/tests/test_batch.py
index 030726a3..5d88212b 100644
--- a/tests/test_batch.py
+++ b/tests/test_batch.py
@@ -1,3 +1,4 @@
+import os
 from pathlib import Path
 
 import pandas as pd
@@ -389,6 +390,32 @@ def test_batch_input_path(self, input_path):
         assert len(source_files) == len(list(glob(input_path)))
         assert set(source_files) == set(Path(file) for file in glob(input_path))
 
+    def test_batch_input_path_with_os_path_seperator(self):
+        input_path = (
+            "tests/parsers_test_files/dfo/odf/bio/CTD/*.ODF"
+            + os.pathsep
+            + "tests/parsers_test_files/seabird/**/*.btl"
+        )
+        batch = BatchConversion(input_path=input_path)
+        source_files = batch.get_source_files()
+        expected_files = [
+            file for path in input_path.split(os.pathsep) for file in glob(path)
+        ]
+        assert source_files
+        assert len(source_files) == len(expected_files)
+
+    def test_batch_input_path_with_list(self):
+        input_path = [
+            "tests/parsers_test_files/dfo/odf/bio/CTD/*.ODF",
+            "tests/parsers_test_files/seabird/**/*.btl",
+        ]
+        batch = BatchConversion(input_path=input_path)
+        source_files = batch.get_source_files()
+        expected_files = [file for path in input_path for file in glob(path)]
+        assert source_files
+        assert len(source_files) == len(expected_files)
+        assert set(source_files) == set(expected_files)
+
     @pytest.mark.parametrize(
         "exclude",
         (
diff --git a/tests/test_cli.py b/tests/test_cli.py
index a23ceca5..7a2b925f 100644
--- a/tests/test_cli.py
+++ b/tests/test_cli.py
@@ -151,3 +151,30 @@ def test_odpy_convert_parser_kwargs(tmp_path):
     assert (
         "Load weather data from metqa file" in results.output
     ), "Parser kwargs not passed to parser"
+
+
+def test_multiple_input_paths(tmp_path):
+    args = (
+        "--log-level",
+        "DEBUG",
+        "convert",
+        "--input-path",
+        "../tests/parsers_test_files/dfo/nafc/pcnv/ctd/cab041_2023_011.pcnv"
+        + os.pathsep
+        + "../tests/parsers_test_files/dfo/nafc/pcnv/ctd/cab041_2023_011.pcnv",
+        "--parser",
+        "dfo.nafc.pcnv",
+        "--output-path",
+        str(tmp_path),
+        "--multiprocessing",
+        "1",
+    )
+    results = run_command(cli.main, args)
+    assert results.exit_code == 0, results.output
+    assert "ERROR" not in results.output, results.output
+    assert "Run conversion" in results.output, results.output
+    assert "cab041_2023_011" in results.output, results.output
+    assert "Conversion completed" in results.output
+    assert (
+        "2/2 files needs to be converted" in results.output
+    ), "Failed to process two files input paths"