keep original text length (#247)

* keep original text length, skip analysis if not translated

* bump version, remove debug

ammico/test/test_text.py

@@ -159,9 +159,11 @@ def test_truncate_text(accepted):
     test_obj.subdict["text"] = "I like cats and dogs."
     test_obj._truncate_text()
     assert test_obj.subdict["text"] == "I like cats and dogs."
+    assert "text_truncated" not in test_obj.subdict
     test_obj.subdict["text"] = 20000 * "m"
     test_obj._truncate_text()
-    assert test_obj.subdict["text"] == 5000 * "m"
+    assert test_obj.subdict["text_truncated"] == 5000 * "m"
+    assert test_obj.subdict["text"] == 20000 * "m"
 
 
 @pytest.mark.gcv
@@ -173,6 +175,14 @@ def test_analyse_image(set_testdict, set_environ, accepted):
             set_testdict[item], analyse_text=True, accept_privacy=accepted
         )
         test_obj.analyse_image()
+    testdict = {}
+    testdict["text"] = 20000 * "m"
+    test_obj = tt.TextDetector(
+        testdict, skip_extraction=True, analyse_text=True, accept_privacy=accepted
+    )
+    test_obj.analyse_image()
+    assert test_obj.subdict["text_truncated"] == 5000 * "m"
+    assert test_obj.subdict["text"] == 20000 * "m"
 
 
 @pytest.mark.gcv

ammico/text.py

@@ -263,7 +263,7 @@ class TextDetector(AnalysisMethod):
         """Truncate the text if it is too long for googletrans."""
         if self.subdict["text"] and len(self.subdict["text"]) > max_length:
             print("Text is too long - truncating to {} characters.".format(max_length))
-            self.subdict["text"] = self.subdict["text"][:max_length]
+            self.subdict["text_truncated"] = self.subdict["text"][:max_length]
 
     def analyse_image(self) -> dict:
         """Perform text extraction and analysis of the text.
@@ -283,7 +283,7 @@ class TextDetector(AnalysisMethod):
             self._truncate_text()
             self.translate_text()
             self.remove_linebreaks()
-            if self.analyse_text:
+            if self.analyse_text and self.subdict["text_english"]:
                 self._run_spacy()
                 self.clean_text()
                 self.text_summary()
@@ -336,8 +336,13 @@ class TextDetector(AnalysisMethod):
             raise ValueError(
                 "Privacy disclosure not accepted - skipping text translation."
             )
+        text_to_translate = (
+            self.subdict["text_truncated"]
+            if "text_truncated" in self.subdict
+            else self.subdict["text"]
+        )
         try:
-            translated = self.translator.translate(self.subdict["text"])
+            translated = self.translator.translate(text_to_translate)
         except Exception:
             print("Could not translate the text with error {}.".format(Exception))
             translated = None

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "ammico"
-version = "0.2.5"
+version = "0.2.6"
 description = "AI Media and Misinformation Content Analysis Tool"
 readme = "README.md"
 maintainers = [