Change count to default to stdout and rename delim argument

Author: ConstantineLignos

README.md

@@ -419,7 +419,7 @@ and document count is not printed.
 The `count` subcommand can produce the counts of chunks in the input
 file. Unlike `summarize`, it counts chunk-type pairs, not just types.
 For example, if we run
-`seqscore count --labels BIO samples/reference.bio counts.csv`,
+`seqscore count --labels BIO samples/reference.bio --output-file counts.csv`,
 tab-delimited counts would be written to `counts.csv` as follows:
 
 ```
@@ -428,6 +428,14 @@ tab-delimited counts would be written to `counts.csv` as follows:
 1	LOC	Pennsylvania
 ```
 
+You can also call `count` without the `--output-file` argument to print counts to
+standard output. However, you may encounter Unicode issues if your terminal is not
+configured properly.
+
+You can use the `--output-delim` argument to change the delimiter used in the counts.
+The default delimiter of tab is strongly recommended, as there is no escaping or
+quoting of the names in the output.
+
 ## Process
 
 The `process` subcommand can remove entity types from a file or map them to

seqscore/scripts/seqscore.py

@@ -1,6 +1,7 @@
 import json
 import sys
 from collections import Counter
+from contextlib import nullcontext
 from typing import Callable, Optional
 
 import click
@@ -273,32 +274,36 @@ def process(
 
 @cli.command(help="show counts for all the mentions contained in a file")
 @_multi_input_file_arguments
-@click.argument("output_file")
+@click.option(
+    "--output-file",
+    default=None,
+    help="path to write output to [default: stdout]",
+)
 @_repair_option()
 @_labels_option_default_bio()
 @click.option(
-    "--delim",
+    "--output-delim",
     default="\t",
     help="the delimiter to be used for output (has no effect on input) [default: tab]",
 )
 @_quiet_option()
 def count(
     file: list[str],  # Name is "file" to make sense on the command line, but it's a list
     file_encoding: str,
-    output_file: str,
+    output_file: Optional[str],
     labels: str,
     *,
     ignore_document_boundaries: bool,
     parse_comment_lines: bool,
-    delim: str,
+    output_delim: str,
     repair_method: str,
     quiet: bool,
 ) -> None:
     if repair_method == REPAIR_NONE:
         repair_method = None
 
-    delim = _normalize_tab(delim)
-    if delim != "\t":
+    output_delim = _normalize_tab(output_delim)
+    if output_delim != "\t":
         print(
             "Warning: Using a delimiter other than tab is not recommended as fields are not quoted",
             file=sys.stderr,
@@ -322,9 +327,16 @@ def count(
                     key = (mention.type, sequence.mention_tokens(mention))
                     counts[key] += 1
 
-    with open(output_file, "w", encoding=file_encoding) as output:
+    with (
+        open(output_file, "w", encoding=file_encoding)
+        if output_file
+        else nullcontext(sys.stdout) as output
+    ):
         for item, item_count in counts.most_common():
-            print(delim.join((str(item_count), item[0], " ".join(item[1]))), file=output)
+            print(
+                output_delim.join((str(item_count), item[0], " ".join(item[1]))),
+                file=output,
+            )
 
 
 @cli.command(help="show counts of the documents, sentences, and entity types")

tests/test_count_click.py

@@ -29,6 +29,7 @@ def test_count_BIO() -> None:
             "--labels",
             "BIO",
             os.path.join("tests", "conll_annotation", "minimal.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -39,6 +40,23 @@ def test_count_BIO() -> None:
     )
 
 
+def test_count_BIO_stdout() -> None:
+    runner = CliRunner()
+    result = runner.invoke(
+        count,
+        [
+            "--labels",
+            "BIO",
+            os.path.join("tests", "conll_annotation", "minimal.bio"),
+        ],
+    )
+    assert result.exit_code == 0
+    expected_lines = open(
+        os.path.join("tests", "test_files", "count_minimal_ref.txt")
+    ).read()
+    assert result.stdout == expected_lines
+
+
 def test_count_BIO_twofiles() -> None:
     runner = CliRunner()
     result = runner.invoke(
@@ -48,6 +66,7 @@ def test_count_BIO_twofiles() -> None:
             "BIO",
             os.path.join("tests", "conll_annotation", "minimal.bio"),
             os.path.join("tests", "conll_annotation", "minimal2.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -66,9 +85,10 @@ def test_count_BIO_tab1() -> None:
         [
             "--labels",
             "BIO",
-            "--delim",
+            "--output-delim",
             "\t",  # Actual tab
             os.path.join("tests", "conll_annotation", "minimal.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -86,9 +106,10 @@ def test_count_BIO_tab2() -> None:
         [
             "--labels",
             "BIO",
-            "--delim",
+            "--output-delim",
             r"\t",  # Backlash and t
             os.path.join("tests", "conll_annotation", "minimal.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -106,9 +127,10 @@ def test_count_BIO_tab3() -> None:
         [
             "--labels",
             "BIO",
-            "--delim",
+            "--output-delim",
             "tab",  # Tab spelled out
             os.path.join("tests", "conll_annotation", "minimal.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -126,9 +148,10 @@ def test_count_BIO_comma() -> None:
         [
             "--labels",
             "BIO",
-            "--delim",
+            "--output-delim",
             ",",
             os.path.join("tests", "conll_annotation", "minimal.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_out.txt"),
         ],
     )
@@ -149,6 +172,7 @@ def test_count_BIOES() -> None:
             "--repair-method",
             "none",
             os.path.join("tests", "conll_annotation", "minimal.bioes"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIOES_out.txt"),
         ],
     )
@@ -169,6 +193,7 @@ def test_count_IO() -> None:
             "--repair-method",
             "none",
             os.path.join("tests", "conll_annotation", "minimal.io"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_IO_out.txt"),
         ],
     )
@@ -189,6 +214,7 @@ def test_count_BIO_invalid_conlleval() -> None:
             "--repair-method",
             "conlleval",
             os.path.join("tests", "conll_annotation", "invalid1.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_conlleval_out.txt"),
         ],
     )
@@ -209,6 +235,7 @@ def test_count_BIO_invalid_discard() -> None:
             "--repair-method",
             "discard",
             os.path.join("tests", "conll_annotation", "invalid1.bio"),
+            "--output-file",
             os.path.join(TMP_DIR.name, "count_BIO_discard_out.txt"),
         ],
     )