Document some instances of 0 vs 1-based page numbering, import cleanup

ocrmypdf/exec/qpdf.py

@@ -107,6 +107,15 @@ def split_pages(input_file, work_folder, npages):
         run(args_qpdf, check=True)
 
 
+def extract_page(input_file, output_file, pageno):
+    args_qpdf = [
+        get_program('qpdf'), input_file,
+        '--pages', input_file, '{0}'.format(pageno + 1), '--',
+        output_file
+    ]
+    run(args_qpdf, check=True)
+    
+
 def _merge_inner(input_files, output_file, min_version=None, log=None):
     """Merge the list of input files (all filenames) into the output file.
 

ocrmypdf/helpers.py

@@ -53,6 +53,7 @@ def is_iterable_notstr(thing):
 
 
 def page_number(input_file):
+    "Get one-based page number implied by filename (000002.pdf -> 2)"
     return int(os.path.basename(input_file)[0:6])
 
 

ocrmypdf/pdfinfo.py

@@ -693,6 +693,12 @@ class PdfInfo:
     def has_userunit(self):
         return any(page.userunit != 1.0 for page in self.pages)
 
+    @property
+    def filename(self):
+        if not isinstance(self._infile, (str, Path)):
+            raise NotImplementedError("can't get filename from stream")
+        return self._infile
+
     def __getitem__(self, item):
         return self._pages[item]
 

ocrmypdf/pipeline.py

@@ -2,14 +2,15 @@
 
 from contextlib import suppress
 from shutil import copyfileobj
+from pathlib import Path
 import sys
 import os
 import shutil
 import img2pdf
 import re
 import PyPDF2 as pypdf
-from PIL import Image
 
+from PIL import Image
 from ruffus import formatter, regex, Pipeline, suffix
 
 from .hocrtransform import HocrTransform
@@ -211,6 +212,7 @@ def repair_pdf(
 
 
 def get_pageinfo(input_file, context):
+    "Get zero-based page info implied by filename, e.g. 000002.pdf -> 1"
     pageno = page_number(input_file) - 1
     pageinfo = context.get_pdfinfo()[pageno]
     return pageinfo