feat: support generating pure texts

tests/cases.py

@@ -18,6 +18,12 @@ class ReferenceCase:
     pmid: int | None = None
     arxiv: str | None = None
 
+    @property
+    def expected_text(self) -> str:
+        """Expected text of the reference."""
+        # Currently, all texts can be infered from markdown.
+        return (self.expected_markdown.split(", DOI:")[0] + ".").replace("*", "")
+
 
 TEST_CASES = [
     # mainly from pubmed

tests/test_cli.py

@@ -63,7 +63,7 @@ def test_cli_from_to_stdout_markdown():
     assert out.strip() == case.expected_markdown.strip()
 
 
-def test_cli_from_to_file_markdow():
+def test_cli_from_to_file_markdown():
     """Test wenxian from DOI to a certain file."""
     case = TEST_CASES[0]
     with tempfile.NamedTemporaryFile("w+") as f:
@@ -85,7 +85,7 @@ def test_cli_from_to_file_markdow():
         assert f.read().strip() == case.expected_markdown.strip()
 
 
-def test_cli_from_to_default_file_markdow():
+def test_cli_from_to_default_file_markdown():
     """Test wenxian from DOI to the default file."""
     case = TEST_CASES[0]
     with tempfile.TemporaryDirectory() as tmpdirname:
@@ -108,6 +108,69 @@ def test_cli_from_to_default_file_markdow():
             assert f.read().strip() == case.expected_markdown.strip()
 
 
+def test_cli_from_to_stdout_text():
+    """Test wenxian from DOI to stdout."""
+    case = TEST_CASES[0]
+    out = subprocess.check_output(
+        [
+            sys.executable,
+            "-m",
+            "wenxian",
+            "from",
+            case.reference.doi,
+            "--type",
+            "text",
+        ],
+        text=True,
+    )
+    assert out.strip() == case.expected_text.strip()
+
+
+def test_cli_from_to_file_text():
+    """Test wenxian from DOI to a certain file."""
+    case = TEST_CASES[0]
+    with tempfile.NamedTemporaryFile("w+") as f:
+        subprocess.check_call(
+            [
+                sys.executable,
+                "-m",
+                "wenxian",
+                "from",
+                case.reference.doi,
+                "-o",
+                f.name,
+                "--type",
+                "text",
+            ],
+            text=True,
+        )
+        f.seek(0)
+        assert f.read().strip() == case.expected_text.strip()
+
+
+def test_cli_from_to_default_file_text():
+    """Test wenxian from DOI to the default file."""
+    case = TEST_CASES[0]
+    with tempfile.TemporaryDirectory() as tmpdirname:
+        subprocess.check_call(
+            [
+                sys.executable,
+                "-m",
+                "wenxian",
+                "from",
+                case.reference.doi,
+                "-o",
+                "--type",
+                "text",
+            ],
+            text=True,
+            cwd=tmpdirname,
+        )
+        with open(Path(tmpdirname) / (case.reference.key + ".txt")) as f:
+            # The default file is references.txt
+            assert f.read().strip() == case.expected_text.strip()
+
+
 def test_cli_ignore_errors():
     """Test wenxian from DOI to stdout."""
     # expected success

tests/test_output.py

@@ -26,3 +26,11 @@ def test_bibtex(reference: Reference, expected):
 def test_markdown(reference: Reference, expected):
     """Test generating Markdown from references."""
     assert reference.markdown.strip() == expected
+
+
+@pytest.mark.parametrize(
+    "reference, expected", [(cc.reference, cc.expected_text) for cc in TEST_CASES]
+)
+def test_text(reference: Reference, expected):
+    """Test generating text from references."""
+    assert reference.text.strip() == expected

wenxian/__main__.py

@@ -42,6 +42,8 @@ def cmd_from(
             buff.append(ref.bibtex)
         elif output_type == "markdown":
             buff.append(ref.markdown)
+        elif output_type == "text":
+            buff.append(ref.text)
         else:
             raise ValueError(f"Unknown output type: {output_type}")
     if output is None:
@@ -51,6 +53,7 @@ def cmd_from(
         extension = {
             "bibtex": ".bib",
             "markdown": ".md",
+            "text": ".txt",
         }.get(output_type)
         if extension is None:
             raise ValueError(f"Unknown output type: {output_type}")
@@ -102,6 +105,7 @@ def main_parser() -> argparse.ArgumentParser:
         choices=[
             "bibtex",
             "markdown",
+            "text",
         ],
         default="bibtex",
         help="Output type.",

wenxian/reference.py

@@ -226,6 +226,14 @@ def bibtex(self) -> str:
     @property
     def markdown(self) -> str:
         """Generate a Markdown for this reference."""
+        return self._markdown_or_text(markdown=True)
+
+    @property
+    def text(self) -> str:
+        """Generate a plain text for this reference."""
+        return self._markdown_or_text(markdown=False)
+
+    def _markdown_or_text(self, markdown: bool) -> str:
         if self.author is None:
             author_string = "Unknown Author"
         else:
@@ -244,20 +252,22 @@ def markdown(self) -> str:
                 for ss in (
                     author_string,
                     self.title,
-                    f"*{self.journal_abbr}*" if self.journal_abbr is not None else None,
+                    (f"*{self.journal_abbr}*" if markdown else self.journal_abbr)
+                    if self.journal_abbr is not None
+                    else None,
                     self.year,
                     self.volume,
                     page_string,
                     f"DOI: [{self.doi}](https://doi.org/{self.doi})"
-                    if self.doi is not None
+                    if self.doi is not None and markdown
                     else None,
                 )
                 if ss is not None
             )
             + "."
             + (
                 f" [![Citations](https://citations.njzjz.win/{self.doi})](https://badge.dimensions.ai/details/doi/{self.doi})"
-                if self.doi is not None
+                if self.doi is not None and markdown
                 else ""
             )
             + "\n"