BatchalignHK 0.7.18.post6__tar.gz → 0.7.18.post8__tar.gz

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/BatchalignHK.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: BatchalignHK
-Version: 0.7.18.post6
+Version: 0.7.18.post8
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: BatchalignHK
-Version: 0.7.18.post6
+Version: 0.7.18.post8
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/batchalign/cli/cli.py

@@ -230,6 +230,8 @@ def translate(ctx, in_dir, out_dir, **kwargs):
 @common_options
 @click.option("--retokenize/--keeptokens",
               default=False, help="Retokenize the main line to fit the UD tokenizations.")
+@click.option("--skipmultilang/--multilang",
+              default=False, help="skip code switching")
 @click.option("--lexicon",
               type=click.Path(exists=True,
                               file_okay=True, dir_okay=False),
@@ -253,7 +255,9 @@ def morphotag(ctx, in_dir, out_dir, **kwargs):
             doc.ba_special_["special_mor_notation"] = True
         return (
             doc,
-            {"retokenize": kwargs["retokenize"], "mwt": mwt}
+            {"retokenize": kwargs["retokenize"],
+             "skipmultilang": kwargs["skipmultilang"],
+             "mwt": mwt}
         )
 
     def writer(doc, output):

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/batchalign/document.py

@@ -153,6 +153,7 @@ class Utterance(BaseModel):
     tier: Tier = Field(default=Tier())
     content: Sentence
     text: Optional[str] = Field(default=None)
+    override_lang: Optional[str] = Field(default=None)
     translation: Optional[str] = Field(default=None)
     time: Optional[Tuple[int,int]] = Field(default=None)
     custom_dependencies: List[CustomLine]  = Field(default=[])

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/batchalign/formats/chat/parser.py

@@ -300,6 +300,7 @@ def chat_parse_doc(lines, special_mor=False):
                                                         content=line.strip()))
 
                 # parse the actual utterance
+                multilingual = re.findall(r"^\[- (\w+)\]", text)
                 parsed, delim = chat_parse_utterance(text, mor, gra, wor, additional)
 
                 # get the timing of the utterance
@@ -313,7 +314,8 @@ def chat_parse_doc(lines, special_mor=False):
                     "text": text,
                     "delim": delim,
                     "custom_dependencies": additional,
-                    "translation": translation
+                    "translation": translation,
+                    "override_lang": None if len(multilingual) == 0 else multilingual[0]
                 })
 
                 timing = re.findall(rf"\x15(\d+)_(\d+)\x15", text)

{batchalignhk-0.7.18.post6 → batchalignhk-0.7.18.post8}/batchalign/pipelines/morphosyntax/ud.py

@@ -462,7 +462,10 @@ def parse_sentence(sentence, delimiter=".", special_forms=[], lang="$nospecial$"
             # specivl forms: recall the special form marker is xbxxx
             if "xbxxx" in word.text.strip():
                 form = special_forms.pop(0)
-                mor.append(f"{form[1].strip()}|{form[0].strip().replace(',', 'cm')}")
+                if form[1][0] == "s":
+                    mor.append("L2|xxx")
+                else:
+                    mor.append(f"{form[1].strip()}|{form[0].strip().replace(',', 'cm')}")
                 special_form_ids.append(word.id)
             else:
                 mor.append(mor_word)
@@ -707,7 +710,7 @@ def adlist_postprocessor(i, lang, adlist):
     return cpy
 
 ######
-def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, **kwargs):
+def morphoanalyze(doc: Document, retokenize:bool, skipmultilang:bool, status_hook:callable = None, **kwargs):
     L.debug("Starting Stanza...")
     inputs = []
 
@@ -785,6 +788,8 @@ def morphoanalyze(doc: Document, retokenize:bool, status_hook:callable = None, *
         L.info(f"Stanza processing utterance {indx+1}/{len(doc.content)}")
         if not isinstance(i, Utterance):
             continue
+        if i.override_lang and skipmultilang:
+            continue
 
         # generate simplified version of the line
         line = str(i)
@@ -1011,4 +1016,4 @@ class StanzaEngine(BatchalignEngine):
         self.status_hook = status_hook
 
     def process(self, doc, **kwargs):
-        return morphoanalyze(doc, retokenize=kwargs.get("retokenize", False), status_hook=self.status_hook, mwt=kwargs.get("mwt", {}))
+        return morphoanalyze(doc, retokenize=kwargs.get("retokenize", False), skipmultilang=kwargs.get("skipmultilang", False), status_hook=self.status_hook, mwt=kwargs.get("mwt", {}))

batchalignhk-0.7.18.post8/batchalign/version

@@ -0,0 +1,3 @@
+0.7.18-post.8
+April 27th, 2025
+skip multilang

batchalignhk-0.7.18.post6/batchalign/version

@@ -1,3 +0,0 @@
-0.7.18-post.6
-April 26th, 2025
-Fix translation tabs