From 44b36a2f210c651f1491d14a6eb0fadc32a37136 Mon Sep 17 00:00:00 2001
From: Gabo <gabriel.piles.glez@gmail.com>
Date: Thu, 11 Jul 2024 14:24:00 +0200
Subject: [PATCH] Fix race condition

---
 Makefile                |  2 +-
 src/QueueProcessor.py   |  1 -
 src/extract_segments.py | 36 ++++++++++++++++++------------------
 3 files changed, 19 insertions(+), 20 deletions(-)

diff --git a/Makefile b/Makefile
index 7ae904b..18903fb 100644
--- a/Makefile
+++ b/Makefile
@@ -19,7 +19,7 @@ start:
 	docker compose up --build
 
 stop:
-	docker compose stop ; docker compose stop -f docker-compose-test.yml
+	docker compose stop ; docker compose -f docker-compose-test.yml stop
 
 start-test:
 	docker compose  -f docker-compose-test.yml up --attach api-pdf-layout --attach queue-processor-pdf-layout --attach worker-pdf-layout --build
diff --git a/src/QueueProcessor.py b/src/QueueProcessor.py
index 795dc2e..5178526 100644
--- a/src/QueueProcessor.py
+++ b/src/QueueProcessor.py
@@ -9,7 +9,6 @@
 from sentry_sdk.integrations.redis import RedisIntegration
 import sentry_sdk
 
-from PdfFile import PdfFile
 from configuration import (
     MONGO_HOST,
     MONGO_PORT,
diff --git a/src/extract_segments.py b/src/extract_segments.py
index 1c46246..1eaccaa 100644
--- a/src/extract_segments.py
+++ b/src/extract_segments.py
@@ -1,4 +1,6 @@
-from configuration import DOCUMENT_LAYOUT_ANALYSIS_URL
+from time import sleep
+
+from configuration import DOCUMENT_LAYOUT_ANALYSIS_URL, service_logger
 from data_model.SegmentBox import SegmentBox
 from PdfFile import PdfFile
 from data_model.ExtractionData import ExtractionData
@@ -10,11 +12,24 @@ def get_xml_name(task: Task) -> str:
     return f"{task.tenant}__{task.params.filename.lower().replace('.pdf', '.xml')}"
 
 
+def exists_file(tenant: str, file_name: str) -> bool:
+    for i in range(5):
+        pdf_file = PdfFile(tenant)
+        if pdf_file.get_path(file_name).exists():
+            return True
+
+        service_logger.info(f"File {pdf_file.get_path(file_name)} exists")
+        sleep(1)
+
+    return False
+
+
 def extract_segments(task: Task, xml_file_name: str = "") -> ExtractionData:
-    pdf_file = PdfFile(task.tenant)
-    if not pdf_file.get_path(task.params.filename).exists():
+    if not exists_file(task.tenant, task.params.filename):
         raise FileNotFoundError
 
+    pdf_file = PdfFile(task.tenant)
+
     with open(pdf_file.get_path(task.params.filename), "rb") as stream:
         files = {"file": stream}
 
@@ -34,18 +49,3 @@ def extract_segments(task: Task, xml_file_name: str = "") -> ExtractionData:
         page_height=0 if not segments else segments[0].page_height,
         page_width=0 if not segments else segments[0].page_width,
     )
-
-
-if __name__ == "__main__":
-    a = {
-        "left": 1,
-        "top": 1,
-        "width": 1,
-        "height": 1,
-        "page_number": 1,
-        "page_width": 1,
-        "page_height": 1,
-        "text": "",
-        "type": "Section_Header",
-    }
-    print(SegmentBox(**a))