Add --delimiter option to set CSV field separator (#21)

First suggested in #14.

docs/changelog.md

@@ -3,6 +3,7 @@
 ## Version 0.3.0 (unreleased)
 
 - Switch to using [uv](https://github.com/astral-sh/uv) for project management
+- Add `--delimiter` command-line option. Use it to choose the character that separates CSV fields in the output
 
 ## Version 0.2.0 (3 Jan 2022)
 

docs/usage.md

@@ -195,6 +195,10 @@ By default, `NA`, `N/A`, `.`, and `-` are considered null values. These are not
 
 The short form of this option is `-n`.
 
+### `--delimiter`
+
+Let's you choose the character used to separate fields in the CSV output. By default a comma (`,`) is used as the delimiter. If you want to use a tab as a delimiter, pass `--delimiter $'\t'` (Bash and Zsh) or `--delimiter (printf '\t')` (Fish).
+
 ### `--version`
 
 Show the version of HTMLTab you have installed, and exit.

src/htmltab/cli.py

@@ -69,6 +69,14 @@
     "strings. Use multiple times if you have more than one "
     "currency symbol  [default: '{}']".format("', '".join(DEFAULT_CURRENCY_SYMBOLS)),
 )
+@click.option(
+    "--delimiter",
+    "-e",
+    type=str,
+    default=",",
+    show_default=True,
+    help="Character used to separate fields in the CSV output",
+)
 @click.option(
     "--output",
     "-o",
@@ -85,6 +93,7 @@ def main(
     group_symbol: str,
     decimal_symbol: str,
     currency_symbol: list[str],
+    delimiter: str,
     output: IO[Any],
     html_file: Callable[[], str],
 ):
@@ -127,6 +136,11 @@ def main(
         # Do nothing on platforms without signals or ``SIGPIPE``.
         pass
 
+    # Check the CSV delimiter is a single character before requesting and
+    # parsing the HTML.
+    if len(delimiter) != 1:
+        raise click.UsageError("delimiter must be a single character")
+
     # Parse file contents as HTML.
     try:
         doc = parse_html(html_file())
@@ -219,7 +233,7 @@ def main(
             rows.append(row)
 
     # Output the CSV to stdout.
-    out = csv.writer(output)
+    out = csv.writer(output, delimiter=delimiter)
     for row in rows:
         # Extra empty cells are added to the row as required, to ensure that
         # all rows have the same number of fields (as required by the closest

tests/conftest.py

@@ -14,6 +14,13 @@ def basic_csv():
     return file_contents
 
 
+@pytest.fixture(scope="session")
+def basic_tsv():
+    with open("tests/fixtures/basic.tsv") as fh:
+        file_contents = fh.read()
+    return file_contents
+
+
 @pytest.fixture(scope="session")
 def basic_european_csv():
     with open("tests/fixtures/basic_european.csv") as fh:

tests/fixtures/basic.tsv

@@ -0,0 +1,6 @@
+Col A	Col B
+1	2
+3	4
+5	6
+3.142	3142
+3142	3.142

tests/test_cli.py

@@ -160,6 +160,18 @@ def test_keep_numbers(runner, three_csv_table_two_keep):
     assert result2.output == three_csv_table_two_keep
 
 
+def test_delimiter(runner, basic_tsv):
+    result = runner.invoke(main, ["-e", "	", "tests/fixtures/basic.html"])
+    assert result.exit_code == 0
+    assert result.output == basic_tsv
+
+
+def test_delimiter_must_be_single_character(runner):
+    result = runner.invoke(main, ["-e", "foo", "tests/fixtures/basic.html"])
+    assert result.exit_code != 0
+    assert "Error: delimiter must be a single character" in result.output
+
+
 def test_bad_input(runner):
     result = runner.invoke(main, input="<")
     assert result.exit_code != 0