From df157552f3040773eaac84a2d6dee807e75a1b7e Mon Sep 17 00:00:00 2001
From: "James R. Barlow" <james@purplerock.ca>
Date: Fri, 1 Jan 2021 01:37:09 -0800
Subject: [PATCH] Make ocrmypdf.ocr take a threading lock

---
 docs/api.rst        |  5 +++++
 src/ocrmypdf/api.py | 23 ++++++++++++++++-------
 2 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/docs/api.rst b/docs/api.rst
index a93c4a9c..3459e8b8 100644
--- a/docs/api.rst
+++ b/docs/api.rst
@@ -56,6 +56,11 @@ Programs that call ``ocrmypdf.ocr()`` should also install a SIGBUS signal
 handler (except on Windows), to raise an exception if access to a memory
 mapped file fails. OCRmyPDF may use memory mapping.
 
+``ocrmypdf.ocr()`` will take a threading lock to prevent multiple runs of itself
+in the same Python interpreter process. This is not thread-safe, because of how
+OCRmyPDF's plugins and Python's library import system work. If you need to parallelize
+OCRmyPDF, use processes.
+
 .. warning::
 
     On Windows and macOS, the script that calls ``ocrmypdf.ocr()`` must be
diff --git a/src/ocrmypdf/api.py b/src/ocrmypdf/api.py
index 6eaf5a4b..9a37cad6 100644
--- a/src/ocrmypdf/api.py
+++ b/src/ocrmypdf/api.py
@@ -8,6 +8,7 @@
 import logging
 import os
 import sys
+import threading
 from enum import IntEnum
 from io import IOBase
 from pathlib import Path
@@ -30,6 +31,8 @@ except ModuleNotFoundError:
 StrPath = Union[os.PathLike, AnyStr]
 PathOrIO = Union[BinaryIO, StrPath]
 
+_api_lock = threading.Lock()
+
 
 class Verbosity(IntEnum):
     """Verbosity level for configure_logging."""
@@ -306,12 +309,18 @@ def ocr(  # pylint: disable=unused-argument
 
     parser = get_parser()
     create_options_kwargs['parser'] = parser
-    plugin_manager = get_plugin_manager(plugins)
-    plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
 
-    if 'verbose' in kwargs:
-        warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+    with _api_lock:
+        # We can't allow multiple ocrmypdf.ocr() threads to run in parallel, because
+        # they might install different plugins, and generally speaking we have areas
+        # of code that use global state.
 
-    options = create_options(**create_options_kwargs)
-    check_options(options, plugin_manager)
-    return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)
+        plugin_manager = get_plugin_manager(plugins)
+        plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
+
+        if 'verbose' in kwargs:
+            warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+
+        options = create_options(**create_options_kwargs)
+        check_options(options, plugin_manager)
+        return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)