Add support for generating + enable analysis without disambiguation

conmorfeusz/conmorfeusz/service/analyzer.py

@@ -1,51 +1,23 @@
 import morfeusz2
-from conmorfeusz import concraft
-from typing import Dict, List
-
-_TAG_SPECS: Dict[str, List[str]] = {
-    "adv": ["degree"],
-    "imps": ["aspect"],
-    "inf": ["aspect"],
-    "pant": ["aspect"],
-    "pcon": ["aspect"],
-    "qub": ["vocalicity"],
-    "prep": ["case", "vocalicity"],
-    "siebie": ["case"],
-    "subst": ["number", "case", "gender"],
-    "depr": ["number", "case", "gender"],
-    "ger": ["number", "case", "gender", "aspect", "negation"],
-    "ppron12": ["number", "case", "gender", "person", "accentability"],
-    "ppron3": [
-        "number",
-        "case",
-        "gender",
-        "person",
-        "accentability",
-        "postprepositionality",
-    ],
-    "num": ["number", "case", "gender", "accommodability"],
-    "numcol": ["number", "case", "gender", "accommodability"],
-    "adj": ["number", "case", "gender", "degree"],
-    "pact": ["number", "case", "gender", "aspect", "negation"],
-    "ppas": ["number", "case", "gender", "aspect", "negation"],
-    "winien": ["number", "gender", "aspect"],
-    "praet": ["number", "gender", "aspect", "agglutination"],
-    "bedzie": ["number", "person", "aspect"],
-    "fin": ["number", "person", "aspect"],
-    "impt": ["number", "person", "aspect"],
-    "aglt": ["number", "person", "aspect", "vocalicity"],
-}
+from conmorfeusz import concraft, utils
 
 
-def parse_tags(tags: str) -> Dict[str, str | None]:
-    type_, *rest = tags.split(":")
-    fields = _TAG_SPECS.get(type_, [])
-    padded_values = rest + [None] * (len(fields) - len(rest))
-    parsed = {field: value for field, value in zip(fields, padded_values)}
+def analysis_entry_to_dict(entry):
+  start, end, morf = entry
+  segment, lemma, tags, frequency, qualifiers = morf
 
-    return {"type": type_, **parsed}
+  return {
+    "start": start,
+    "end": end,
+    "segment": segment,
+    "lemma": lemma,
+    "tags": utils.parse_tags(tags),
+    "frequency": frequency,
+    "qualifiers": qualifiers,
+  }
 
-def entry_to_dict(entry):
+
+def disamb_entry_to_dict(entry):
   start, end, morf, prob, eos, disamb = entry
   segment, lemma, tags, frequency, qualifiers = morf
 
@@ -54,7 +26,7 @@ def entry_to_dict(entry):
     "end": end,
     "segment": segment,
     "lemma": lemma,
-    "tags": parse_tags(tags),
+    "tags": utils.parse_tags(tags),
     "frequency": frequency,
     "qualifiers": qualifiers,
     "prob": float(prob),
@@ -63,8 +35,13 @@ def entry_to_dict(entry):
   }
 
 
-def analyse(text, port):
+def analyse(text, disamb, port):
   morf = morfeusz2.Morfeusz(expand_tags=True)
-  conc = concraft.Concraft(port=port)
   analysis = morf.analyse(text)  
-  return [entry_to_dict(e) for e in conc.disamb(analysis) if len(e) == 6]
+
+  if disamb == False:
+    return [analysis_entry_to_dict(e) for e in analysis if len(e) == 3]
+
+  conc = concraft.Concraft(port=port)
+  
+  return [disamb_entry_to_dict(e) for e in conc.disamb(analysis) if len(e) == 6]

conmorfeusz/conmorfeusz/service/generator.py

@@ -0,0 +1,18 @@
+import morfeusz2
+from conmorfeusz import utils
+
+def entry_to_dict(entry):  
+  segment, lemma, tags, frequency, qualifiers = entry
+
+  return {
+    "segment": segment,
+    "lemma": lemma,
+    "tags": utils.parse_tags(tags),
+    "frequency": frequency,
+    "qualifiers": qualifiers,
+  }
+
+
+def generate(lemma):
+  morf = morfeusz2.Morfeusz(expand_tags=True)
+  return [entry_to_dict(e) for e in morf.generate(lemma) if len(e) > 0]

conmorfeusz/conmorfeusz/utils/__init__.py

@@ -0,0 +1,44 @@
+from typing import Dict, List
+
+_TAG_SPECS: Dict[str, List[str]] = {
+    "adv": ["degree"],
+    "imps": ["aspect"],
+    "inf": ["aspect"],
+    "pant": ["aspect"],
+    "pcon": ["aspect"],
+    "qub": ["vocalicity"],
+    "prep": ["case", "vocalicity"],
+    "siebie": ["case"],
+    "subst": ["number", "case", "gender"],
+    "depr": ["number", "case", "gender"],
+    "ger": ["number", "case", "gender", "aspect", "negation"],
+    "ppron12": ["number", "case", "gender", "person", "accentability"],
+    "ppron3": [
+        "number",
+        "case",
+        "gender",
+        "person",
+        "accentability",
+        "postprepositionality",
+    ],
+    "num": ["number", "case", "gender", "accommodability"],
+    "numcol": ["number", "case", "gender", "accommodability"],
+    "adj": ["number", "case", "gender", "degree"],
+    "pact": ["number", "case", "gender", "aspect", "negation"],
+    "ppas": ["number", "case", "gender", "aspect", "negation"],
+    "winien": ["number", "gender", "aspect"],
+    "praet": ["number", "gender", "aspect", "agglutination"],
+    "bedzie": ["number", "person", "aspect"],
+    "fin": ["number", "person", "aspect"],
+    "impt": ["number", "person", "aspect"],
+    "aglt": ["number", "person", "aspect", "vocalicity"],
+}
+
+
+def parse_tags(tags: str) -> Dict[str, str | None]:
+    type_, *rest = tags.split(":")
+    fields = _TAG_SPECS.get(type_, [])
+    padded_values = rest + [None] * (len(fields) - len(rest))
+    parsed = {field: value for field, value in zip(fields, padded_values)}
+
+    return {"type": type_, **parsed}

conmorfeusz/conmorfeusz/web/__init__.py

@@ -1,5 +1,5 @@
 from flask import Flask
-from . import analyzer 
+from . import analyzer, generator
 
 def create_app(config=None):
   app = Flask(__name__, instance_relative_config=True)
@@ -13,5 +13,6 @@ def create_app(config=None):
     app.config.update(config)
 
   app.register_blueprint(analyzer.bp)
+  app.register_blueprint(generator.bp)
 
   return app

conmorfeusz/conmorfeusz/web/analyzer.py

@@ -8,5 +8,5 @@ def analyze():
   data = request.get_json()  
   
   return {
-    "analysis": service.analyse(data["text"], current_app.config['CONCRAFT_PL_PORT'])
+    "analysis": service.analyse(data["text"], request.args.get('disamb', False), current_app.config['CONCRAFT_PL_PORT'])
   }

conmorfeusz/conmorfeusz/web/generator.py

@@ -0,0 +1,12 @@
+from flask import Blueprint, request, current_app
+from conmorfeusz.service import generator as service
+
+bp = Blueprint('generator', __name__, url_prefix='/generator')
+
+@bp.post("/generate")
+def analyze():
+  data = request.get_json()  
+  
+  return {
+    "output": service.generate(data["lemma"])
+  }