refactor(cli): refactor output formats to -o argument

roedoejet · joanise · commit f34f4268ae7f · 2021-11-04T13:44:13.000-04:00
diff --git a/readalongs/align.py b/readalongs/align.py
@@ -473,14 +473,11 @@ def save_readalong(  # noqa C901
     output_dir: str,
     output_basename: str,
     config=None,
-    text_grid: bool = False,
-    closed_captioning: bool = False,
-    output_xhtml: bool = False,
     audiofile: str,
     audiosegment: AudioSegment = None,
-    html: bool = False,
+    output_formats=[],
 ):
-    """Save the results from align_audio() into the otuput files required for a
+    """Save the results from align_audio() into the output files required for a
         readalong
 
     Args:
@@ -489,10 +486,8 @@ def save_readalong(  # noqa C901
             output_dir should already exist, files it contains may be overwritten
         output_basename (str): basename of the files to save in output_dir
         config ([type TODO], optional): alignment configuration loaded from the json
-        text_grid (bool, optional): if True, also save in Praat TextGrid and ELAN EAF formats
-        closed_captioning (bool, optional): if True, also save in .vtt and .srt subtitle formats
-        output_xhtml (bool, optional): if True, convert XML into XHTML format before writing
         audiofile (str): path to the audio file passed to align_audio()
+        output_formats (List[str], optional): list of desired output formats
         audiosegment (AudioSegment): a pydub.AudioSegment object of processed audio.
                               if None, then original audio will be saved at `audiofile`
 
@@ -510,30 +505,41 @@ def save_readalong(  # noqa C901
 
     output_base = os.path.join(output_dir, output_basename)
 
-    if text_grid:
+    # Create textgrid object if outputting to TextGrid or eaf
+    if "TextGrid" in output_formats or "eaf" in output_formats:
         audio = read_audio_from_file(audiofile)
         duration = audio.frame_count() / audio.frame_rate
         words, sentences = return_words_and_sentences(align_results)
         textgrid = write_to_text_grid(words, sentences, duration)
+
+    if "TextGrid" in output_formats:
         textgrid.to_file(output_base + ".TextGrid")
+
+    if "eaf" in output_formats:
         textgrid.to_eaf().to_file(output_base + ".eaf")
 
-    if closed_captioning:
+    # Create webvtt object if outputting to vtt or srt
+    if "srt" in output_formats or "vtt" in output_formats:
         words, sentences = return_words_and_sentences(align_results)
-        webvtt_sentences = write_to_subtitles(sentences)
-        webvtt_sentences.save(output_base + "_sentences.vtt")
-        webvtt_sentences.save_as_srt(output_base + "_sentences.srt")
-        webvtt_words = write_to_subtitles(words)
-        webvtt_words.save(output_base + "_words.vtt")
-        webvtt_words.save_as_srt(output_base + "_words.srt")
-
-    if output_xhtml:
-        convert_to_xhtml(align_results["tokenized"])
-        tokenized_xml_path = output_base + ".xhtml"
-    else:
-        tokenized_xml_path = output_base + ".xml"
+        cc_sentences = write_to_subtitles(sentences)
+        cc_words = write_to_subtitles(words)
+
+    if "srt" in output_formats:
+        cc_sentences.save_as_srt(output_base + "_sentences.srt")
+        cc_words.save_as_srt(output_base + "_words.srt")
+
+    if "vtt" in output_formats:
+        cc_words.save(output_base + "_words.vtt")
+        cc_sentences.save(output_base + "_sentences.vtt")
+
+    tokenized_xml_path = output_base + ".xml"
     save_xml(tokenized_xml_path, align_results["tokenized"])
 
+    if "xhtml" in output_formats:
+        convert_to_xhtml(align_results["tokenized"])
+        tokenized_xhtml_path = output_base + ".xhtml"
+        save_xml(tokenized_xhtml_path, align_results["tokenized"])
+
     _, audio_ext = os.path.splitext(audiofile)
     audio_path = output_base + audio_ext
     audio_format = audio_ext[1:]
@@ -562,7 +568,7 @@ def save_readalong(  # noqa C901
     )
     save_txt(smil_path, smil)
 
-    if html:
+    if "html" in output_formats:
         html_out_path = output_base + ".html"
         html_out = create_web_component_html(tokenized_xml_path, smil_path, audio_path)
         with open(html_out_path, "w") as f:
diff --git a/readalongs/cli.py b/readalongs/cli.py
@@ -45,6 +45,19 @@
 LANGS, _ = getLangs()
 ensure_using_supported_python_version()
 
+SUPPORTED_OUTPUT_FORMATS = {
+    "eaf": "ELAN file",
+    "html": "Single-file, offline HTML",
+    "srt": "SRT subtitle",
+    "TextGrid": "Praat TextGrid",
+    "vtt": "WebVTT subtitle",
+    "xhtml": "Simple XHTML",
+}
+
+SUPPORTED_OUTPUT_FORMATS_DESC = "\n\n".join(
+    k + " - " + v for k, v in SUPPORTED_OUTPUT_FORMATS.items()
+)
+
 
 def create_app():
     """Returns the app"""
@@ -99,10 +112,12 @@ def cli():
     help="Use ReadAlong-Studio configuration file (in JSON format)",
 )
 @click.option(
-    "-C",
-    "--closed-captioning",
-    is_flag=True,
-    help="Export sentences to WebVTT and SRT files",
+    "-o",
+    "--output-formats",
+    type=click.Choice(SUPPORTED_OUTPUT_FORMATS.keys()),
+    multiple=True,
+    help="The output file formats to export to. By default the text is exported as XML and alignments are exported as SMIL. "
+    + f"Other formats include:\b \n\n{SUPPORTED_OUTPUT_FORMATS_DESC}",
 )
 @click.option("-d", "--debug", is_flag=True, help="Add debugging messages to logger")
 @click.option(
@@ -126,13 +141,6 @@ def cli():
     is_flag=True,
     help="Save intermediate stages of processing and temporary files (dictionary, FSG, tokenization, etc)",
 )
-@click.option(
-    "-t", "--text-grid", is_flag=True, help="Export to Praat TextGrid & ELAN eaf file"
-)
-@click.option("-H", "--html", is_flag=True, help="Export to a single-file HTML format")
-@click.option(
-    "-x", "--output-xhtml", is_flag=True, help="Output simple XHTML instead of XML"
-)
 @click.option(
     "--g2p-fallback",
     default=None,
@@ -259,17 +267,17 @@ def align(**kwargs):  # noqa: C901
     except RuntimeError as e:
         LOGGER.error(e)
         sys.exit(1)
+
+    output_formats = kwargs["output_formats"]
+
     save_readalong(
         align_results=results,
         output_dir=output_dir,
         output_basename=output_basename,
         config=config,
-        text_grid=kwargs["text_grid"],
-        closed_captioning=kwargs["closed_captioning"],
-        output_xhtml=kwargs["output_xhtml"],
         audiofile=kwargs["audiofile"],
         audiosegment=results["audio"],
-        html=kwargs["html"],
+        output_formats=output_formats,
     )
 
 
diff --git a/test/test_align_cli.py b/test/test_align_cli.py
@@ -29,8 +29,14 @@ def test_invoke_align(self):
             [
                 "-i",
                 "-s",
-                "-C",
-                "-t",
+                "-o",
+                "vtt",
+                "-o",
+                "srt",
+                "-o",
+                "TextGrid",
+                "-o",
+                "eaf",
                 "-l",
                 "fra",
                 "--config",
@@ -87,7 +93,8 @@ def test_invoke_align(self):
         results_dna = self.runner.invoke(
             align,
             [
-                "-x",
+                "-o",
+                "xhtml",
                 "-s",
                 "--config",
                 join(self.data_dir, "sample-config.json"),
@@ -104,7 +111,7 @@ def test_invoke_align(self):
         )
         self.assertTrue(
             exists(join(output, "output.xhtml")),
-            "successful alignment with -x should have created output.xhtml",
+            "successful alignment with -o xhtml should have created output.xhtml",
         )
         self.assertIn("Please copy image-for-page1.jpg to ", results_dna.stdout)
         self.assertFalse(
@@ -114,7 +121,7 @@ def test_invoke_align(self):
 
         # Functionally the same as self.assertTrue(filecmp.cmp(f1, f2)), but show where
         # the differences are if the files are not identical
-        # Since f2 was created using -x, we need to substitute .xhtml back to .xml during
+        # Since f2 was created using -o xhtml, we need to substitute .xhtml back to .xml during
         # the comparison of the contents of the .smil files.
         with open(join(output1, "output.smil"), encoding="utf8") as f1, open(
             join(output, "output.smil"), encoding="utf8"
@@ -153,7 +160,7 @@ def test_invoke_align(self):
         )
 
     def test_align_with_package(self):
-        """Test creating a single-file package, with --html"""
+        """Test creating a single-file package, with -o html"""
 
         output = join(self.tempdir, "html")
         with SoundSwallowerStub("t0b0d0p0s0w0:920:1620", "t0b0d0p0s1w0:1620:1690"):
@@ -163,7 +170,8 @@ def test_align_with_package(self):
                     join(self.data_dir, "ej-fra-package.xml"),
                     join(self.data_dir, "ej-fra.m4a"),
                     output,
-                    "--html",
+                    "-o",
+                    "html",
                 ],
             )
         # print(results_html.output)
diff --git a/test/test_silence.py b/test/test_silence.py
@@ -23,8 +23,14 @@ def test_basic_silence_insertion(self):
             align,
             [
                 "-s",
-                "-C",
-                "-t",
+                "-o",
+                "srt",
+                "-o",
+                "vtt",
+                "-o",
+                "TextGrid",
+                "-o",
+                "eaf",
                 "-l",
                 "fra",
                 os.path.join(self.data_dir, "ej-fra-silence.xml"),
@@ -61,8 +67,14 @@ def test_bad_silence(self):
             align,
             [
                 "-s",
-                "-C",
-                "-t",
+                "-o",
+                "srt",
+                "-o",
+                "vtt",
+                "-o",
+                "TextGrid",
+                "-o",
+                "eaf",
                 "-l",
                 "fra",
                 os.path.join(self.data_dir, "ej-fra-silence-bad.xml"),
