fix: parameterized ingest test overwriting (#838)

* sets OVERWRITE_FIXTURES to default to false in test-ingest-local-single-file.sh
* fixes incorrect expected results
* update expected results to properly parse Korean text
* bonus: installs language pack for Korean in CI and ingest fixture workflows

.github/workflows/ci.yml

@@ -194,6 +194,7 @@ jobs:
         sudo apt-get install -y libmagic-dev poppler-utils libreoffice pandoc
         sudo add-apt-repository -y ppa:alex-p/tesseract-ocr5
         sudo apt-get install -y tesseract-ocr
+        sudo apt-get install -y tesseract-ocr-kor
         tesseract --version
         make install-ingest-s3
         make install-ingest-azure

.github/workflows/ingest-test-fixtures-update-pr.yml

@@ -70,6 +70,7 @@ jobs:
           sudo apt-get install -y libmagic-dev poppler-utils libreoffice pandoc
           sudo add-apt-repository -y ppa:alex-p/tesseract-ocr5
           sudo apt-get install -y tesseract-ocr
+          sudo apt-get install -y tesseract-ocr-kor
           tesseract --version
           make install-ingest-s3
           make install-ingest-azure

test_unstructured_ingest/expected-structured-output/parameterized-ingest-output/example-docs/english-and-korean.png.json

@@ -27,13 +27,13 @@
     "text": "Note: Remember to write your own \"OPENING MESSAGE\" before you copy and paste the template. please always include [TREASURE HARUTO] for example:"
   },
   {
-    "type": "UncategorizedText",
-    "element_id": "23a37aa2d5f39d5e2275dec011be76be",
+    "type": "NarrativeText",
+    "element_id": "9854d0df4dceac0f09846f4f272f0703",
     "metadata": {
       "data_source": {},
       "filetype": "image/png"
     },
-    "text": "StS ofAl2, AS|E YGEAS 1B TREASUREMH HARUTOM| 2] BHEYLICH, BHO AY, HARUTO M BE = WSO Hol Wat SSGstsS LRU, O| Wil BS SH ASP ASS AZopO} HAS] TAI St a Bat FSAel GAS HS + U7/S HLICh."
+    "text": "안녕하세요, 저 희 는 YGEAS 그룹 TREASUREMH HARUTOM| 2] 팬 입니다. 팬 으 로서, HARUTO 씨 받 는 대 우 에 대해 의 구 심 과 불 공 평 함 을 LRU, 이 메 일 을 통해 저 희 의 의 혹 을 전 달 하여 귀 사 의 진지한 고 민 과 적극적인 답 변 을 받을 수 있 기 를 바랍니다."
   },
   {
     "type": "NarrativeText",

test_unstructured_ingest/test-ingest-local-single-file.sh

@@ -10,11 +10,14 @@ PYTHONPATH=. ./unstructured/ingest/main.py \
     --local-input-path example-docs/english-and-korean.png \
     --structured-output-dir parameterized-ingest-output \
     --partition-ocr-languages eng+kor \
+    --partition-strategy ocr_only \
     --verbose \
     --reprocess
 
 set +e
 
+OVERWRITE_FIXTURES=${OVERWRITE_FIXTURES:-false}
+
 # to update ingest test fixtures, run scripts/ingest-test-fixtures-update.sh on x86_64
 if [[ "$OVERWRITE_FIXTURES" != "false" ]]; then