Add check for poppler installation

pdelfin/birrpipeline.py

@@ -29,7 +29,7 @@
 from pdelfin.prompts import build_finetuning_prompt, PageResponse
 from pdelfin.prompts.anchor import get_anchor_text
 from pdelfin.s3_utils import parse_custom_id, expand_s3_glob, get_s3_bytes, parse_s3_path
-
+from pdelfin.check import check_poppler_version
 
 # Initialize logger
 logger = logging.getLogger(__name__)
@@ -699,6 +699,8 @@ def get_current_round(s3_workspace: str) -> int:
     current_round = get_current_round(args.workspace)
     logger.info(f"Current round is {current_round}")
 
+    check_poppler_version()
+
     # One shared executor to rule them all
     executor = ProcessPoolExecutor(max_workers=args.workers)
 

pdelfin/check.py

@@ -0,0 +1,18 @@
+import sys
+import subprocess
+import logging
+
+logger = logging.getLogger(__name__)
+
+def check_poppler_version():
+    try:
+        result = subprocess.run(['pdftoppm', '-h'], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+        if result.returncode == 0 and "pdftoppm" in result.stdout:
+            logger.info("pdftoppm is installed and working.")
+        else:
+            logger.error("pdftoppm is installed but returned an error.")
+            sys.exit(1)
+    except FileNotFoundError:
+        logger.error("pdftoppm is not installed.")
+        logger.error("Check the README in the https://github.com/allenai/pdelfin/blob/main/README.md for installation instructions")
+        sys.exit(1)