From b77ca5914f114978921ce22fd65a1c8a560f04d8 Mon Sep 17 00:00:00 2001
From: Stephan Porada <sporada@uni-bielefeld.de>
Date: Wed, 10 Jun 2020 11:48:58 +0200
Subject: [PATCH] Set relative file paths in hocr

---
 ocr | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/ocr b/ocr
index 73b24eb..b261a96 100755
--- a/ocr
+++ b/ocr
@@ -256,6 +256,19 @@ class OCRPipeline(WorkflowRunner):
                            os.listdir(input_dir))
             files.sort(key=lambda x: int(re.search(r'\d+', x).group(0)))
             files = map(lambda x: os.path.join(input_dir, x), files)
+            # set relative file paths into hocr
+            relative_files = map(lambda x: os.path.join('..',
+                                                        'tiff',
+                                                        os.path.basename(x).replace('.hocr', '.tif')),  # noqa
+                                 files)
+            for file, relative_file in zip(files, relative_files):
+                with open(file, 'r+') as f:
+                    html = f.read()
+                    html = html.replace(file.replace('.hocr', '.tif'),
+                                        relative_file)
+                    f.seek(0)
+                    f.truncate(0)  # deletes content of file to write new html
+                    f.write(html)
             output_path_base = os.path.join(job.output_dir, 'PoCo')
             output_path = os.path.join(output_path_base, 'hocr')
             cmd = 'cp "{}" "{}"'.format('" "'.join(files), output_path)
@@ -398,7 +411,6 @@ class OCRPipeline(WorkflowRunner):
             zip_jobs.append(self.addTask(command=cmd, dependencies=deps,
                                          label=lbl))
         # zip PoCo files
-        # TODO: Fix relative paths before?
             poco_paths = []
             poco_names = []
             for i, job in enumerate(self.jobs):