Add RSPO code detection (#161)

openfoodfacts · Jul 15, 2020 · 9ffebab · 9ffebab
1 parent 38e7acf
commit 9ffebab
Show file tree

Hide file tree

Showing 2 changed files with 40 additions and 13 deletions.
diff --git a/robotoff/insights/ocr/packager_code.py b/robotoff/insights/ocr/packager_code.py
@@ -63,33 +63,41 @@ def process_fsc_match(match) -> str:
         lowercase=True,
         processing_func=process_de_packaging_match,
     ),
+    "rspo": OCRRegex(
+        re.compile(r"(?<!\w)RSPO-\d{7}(?!\d)"),
+        field=OCRField.full_text_contiguous,
+        lowercase=False,
+    ),
 }
 
 
 def find_packager_codes_regex(ocr_result: Union[OCRResult, str]) -> List[RawInsight]:
     results: List[RawInsight] = []
 
     for regex_code, ocr_regex in PACKAGER_CODE.items():
-        text = get_text(ocr_result, ocr_regex)
+        text = get_text(ocr_result, ocr_regex, ocr_regex.lowercase)
 
         if not text:
             continue
 
         for match in ocr_regex.regex.finditer(text):
-            if ocr_regex.processing_func is not None:
+            if ocr_regex.processing_func is None:
+                value = match.group(0)
+            else:
                 value = ocr_regex.processing_func(match)
-                results.append(
-                    RawInsight(
-                        value=value,
-                        data={
-                            "raw": match.group(0),
-                            "type": regex_code,
-                            "notify": ocr_regex.notify,
-                        },
-                        type=InsightType.packager_code,
-                        automatic_processing=True,
-                    )
+
+            results.append(
+                RawInsight(
+                    value=value,
+                    data={
+                        "raw": match.group(0),
+                        "type": regex_code,
+                        "notify": ocr_regex.notify,
+                    },
+                    type=InsightType.packager_code,
+                    automatic_processing=True,
                 )
+            )
 
     return results
 

diff --git a/tests/insights/ocr/test_packager_code.py b/tests/insights/ocr/test_packager_code.py
@@ -0,0 +1,19 @@
+from typing import List
+
+import pytest
+
+from robotoff.insights.ocr.packager_code import find_packager_codes
+
+
+@pytest.mark.parametrize(
+    "text,values",
+    [
+        ("Sustainable palm oil RSPO-5068502 ", ["RSPO-5068502"]),
+        ("RSPO-50685022", []),
+        ("QRSPO-2404885", []),
+    ],
+)
+def test_find_packager_codes(text: str, values: List[str]):
+    insights = find_packager_codes(text)
+    detected_values = set(i.value for i in insights)
+    assert detected_values == set(values)