From d257c835200904331295259ca4e156c566d2eed9 Mon Sep 17 00:00:00 2001 From: "James R. Barlow" Date: Fri, 29 Jul 2016 14:59:17 -0700 Subject: [PATCH] Most tests were failing at split_pages() It seems that ruffus sometimes decides to send a ['inputfile.pdf'] instead of a bare string. --- ocrmypdf/main.py | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/ocrmypdf/main.py b/ocrmypdf/main.py index ab64b9f5..b5e709f3 100755 --- a/ocrmypdf/main.py +++ b/ocrmypdf/main.py @@ -450,8 +450,9 @@ def triage( @transform( input=triage, - filter=formatter('(?i)\.pdf'), - output=os.path.join(work_folder, '{basename[0]}.repaired.pdf'), + filter=suffix('.pdf'), + output='.repaired.pdf', + output_dir=work_folder, extras=[_log, _pdfinfo, _pdfinfo_lock]) def repair_pdf( input_file, @@ -545,12 +546,17 @@ def is_ocr_required(pageinfo, log): os.path.join(work_folder, '*.page.pdf'), extras=[_log, _pdfinfo, _pdfinfo_lock]) def split_pages( - input_file, + input_files, output_files, log, pdfinfo, pdfinfo_lock): + if isinstance(input_files, list): + input_file = input_files[0] + else: + input_file = input_files + for oo in output_files: with suppress(FileNotFoundError): os.unlink(oo)