v11.4.4 release notes

Our method of getting data from pdfminer would silently consume a StopIteration
if pdfminer returned no processed pages, leading to odd error message.

We improve an error from pdfminer properly, and returning a more
descriptive error of our own.

It would be possible for ocrmypdf to repair the file before sending it to
pdfminer, but this seems to be rare enough that we won't do that yet.

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

docs/release_notes.rst

@@ -12,6 +12,15 @@ may be unreliable. Use the API to depend on precise behavior.
 The public API may be useful in scripts that launch OCRmyPDF processes or that
 wish to use some of its features for working with PDFs.
 
+v11.4.4
+=======
+
+-  Fixed ``AttributeError: 'NoneType' object has no attribute 'userunit'``, issue #700,
+   related to OCRmyPDF not properly forwarded an error message from pdfminer.six.
+-  Adjusted typing of some arguments.
+-  ``ocrmypdf.ocr`` now takes a ``threading.Lock`` for reasons outlined in the
+   documentation.
+
 v11.4.3
 =======
 

src/ocrmypdf/_concurrent.py

@@ -109,15 +109,11 @@ def exec_progress_pool(
         )
         try:
             results = pool.imap_unordered(task, task_arguments)
-            while True:
+            for result in results:
-                try:
+                if task_finished:
-                    result = results.next()
+                    task_finished(result, pbar)
-                    if task_finished:
+                else:
-                        task_finished(result, pbar)
+                    pbar.update()
-                    else:
-                        pbar.update()
-                except StopIteration:
-                    break
         except KeyboardInterrupt:
             # Terminate pool so we exit instantly
             pool.terminate()

src/ocrmypdf/_exec/tesseract.py

@@ -14,7 +14,7 @@ from collections import namedtuple
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import List
+from typing import List, Optional
 
 from PIL import Image
 
@@ -118,7 +118,7 @@ def get_languages():
     return set(lang.strip() for lang in rest)
 
 
-def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
+def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
     args = ['tesseract']
     if langs:
         args.extend(['-l', '+'.join(langs)])
@@ -127,7 +127,7 @@ def tess_base_args(langs: List[str], engine_mode: int) -> List[str]:
     return args
 
 
-def get_orientation(input_file: Path, engine_mode: int, timeout: float):
+def get_orientation(input_file: Path, engine_mode: Optional[int], timeout: float):
     args_tesseract = tess_base_args(['osd'], engine_mode) + [
         '--psm',
         '0',

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:
+    with _api_lock:
-        warn("ocrmypdf.ocr(verbose=) is ignored. Use ocrmypdf.configure_logging().")
+        # 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)
+        plugin_manager = get_plugin_manager(plugins)
-    check_options(options, plugin_manager)
+        plugin_manager.hook.add_options(parser=parser)  # pylint: disable=no-member
-    return run_pipeline(options=options, plugin_manager=plugin_manager, api=True)
+
+        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)

src/ocrmypdf/pdfinfo/info.py

@@ -22,7 +22,7 @@ import pikepdf
 from pikepdf import Object, Pdf, PdfMatrix
 
 from ocrmypdf._concurrent import exec_progress_pool
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 from ocrmypdf.helpers import Resolution, available_cpu_count, pikepdf_enable_mmap
 from ocrmypdf.pdfinfo.layout import get_page_analysis, get_text_boxes
 
@@ -598,6 +598,8 @@ def _pdf_pageinfo_concurrent(
 
     def update_pageinfo(result, pbar):
         page = result
+        if not page:
+            raise InputFileError("Could read a page in the PDF")
         pages[page.pageno] = page
         pbar.update()
 

src/ocrmypdf/pdfinfo/layout.py

@@ -21,7 +21,7 @@ from pdfminer.pdffont import PDFSimpleFont, PDFUnicodeNotDefined
 from pdfminer.pdfpage import PDFPage
 from pdfminer.utils import bbox2str, matrix2str
 
-from ocrmypdf.exceptions import EncryptedPdfError
+from ocrmypdf.exceptions import EncryptedPdfError, InputFileError
 
 STRIP_NAME = re.compile(r'[0-9]+')
 
@@ -236,8 +236,13 @@ def get_page_analysis(infile, pageno, pscript5_mode):
 
     try:
         with Path(infile).open('rb') as f:
-            page = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
+            page_iter = PDFPage.get_pages(f, pagenos=[pageno], maxpages=0)
-            interp.process_page(next(page))
+            page = next(page_iter, None)
+            if page is None:
+                raise InputFileError(
+                    f"pdfminer could not process page {pageno} (counting from 0)."
+                )
+            interp.process_page(page)
     except PDFTextExtractionNotAllowed as e:
         raise EncryptedPdfError() from e
     finally:

tests/test_pdfinfo.py

@@ -15,7 +15,11 @@ from PIL import Image
 from reportlab.pdfgen.canvas import Canvas
 
 from ocrmypdf import pdfinfo
+from ocrmypdf.exceptions import InputFileError
 from ocrmypdf.pdfinfo import Colorspace, Encoding
+from ocrmypdf.pdfinfo.layout import PDFPage
+
+run_ocrmypdf_api = pytest.helpers.run_ocrmypdf_api
 
 # pylint: disable=protected-access
 
@@ -179,3 +183,18 @@ def test_stack_abuse():
     with pytest.warns(None):
         with pytest.raises(RuntimeError):
             pdfinfo.info._interpret_contents(stream)
+
+
+def test_pages_issue700(monkeypatch, resources):
+    def get_no_pages(*args, **kwargs):
+        return iter([])
+
+    monkeypatch.setattr(PDFPage, 'get_pages', get_no_pages)
+
+    with pytest.raises(InputFileError, match="pdfminer"):
+        pdfinfo.PdfInfo(
+            resources / 'cardinal.pdf',
+            detailed_analysis=True,
+            progbar=False,
+            max_workers=1,
+        )