From eefd8d53d015ce5a8a3c8bfbee0d79fef9fd4862 Mon Sep 17 00:00:00 2001
From: Patrick Jentsch <p.jentsch@uni-bielefeld.de>
Date: Fri, 11 Jan 2019 15:54:15 +0100
Subject: [PATCH] shlexed all cmds!

---
 ocr_pyflow | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/ocr_pyflow b/ocr_pyflow
index f815c26..074b2ee 100755
--- a/ocr_pyflow
+++ b/ocr_pyflow
@@ -25,6 +25,7 @@ __status__ = "Development"
 import argparse
 import multiprocessing
 import os
+import shlex
 import sys
 import unicodedata
 from pyflow import WorkflowRunner
@@ -114,6 +115,7 @@ class OCRWorkflow(WorkflowRunner):
                 os.path.join(job["output_dir"], "tmp", "ocropus-nlbin"),
                 os.path.join(job["output_dir"], "tmp", "tesseract"),
                 os.path.join(job["output_dir"], "tmp", "tiff_files"))
+            cmd = shlex.escape(cmd);
             mkdir_jobs.append(self.addTask(label="mkdir_job_-_%i" % (mkdir_job_number), command=cmd))
 
 
@@ -138,12 +140,14 @@ class OCRWorkflow(WorkflowRunner):
                 os.path.join(job["output_dir"], "tmp", "tiff_files", os.path.basename(job["path"]).rsplit(".", 1)[0] + ".pdf"),
                 os.path.join(job["output_dir"], "tmp", "tiff_files", os.path.basename(job["path"]).rsplit(".", 1)[0]),
                 os.path.join(job["output_dir"], "tmp", "tiff_files", os.path.basename(job["path"]).rsplit(".", 1)[0] + ".pdf"))
+            cmd = shlex.escape(cmd);
             split_jobs.append(self.addTask(label="split_job_-_%i" % (split_job_number), command=cmd, dependencies=mkdir_jobs))
         for job in self.jobs["pdf"]:
             split_job_number += 1
             cmd = "pdftoppm %s %s -tiff -r 300 -tiffcompression lzw -cropbox" % (
                 job["path"],
                 os.path.join(job["output_dir"], "tmp", "tiff_files", os.path.basename(job["path"]).rsplit(".", 1)[0]))
+            cmd = shlex.escape(cmd);
             split_jobs.append(self.addTask(label="split_job_-_%i" % (split_job_number), command=cmd, dependencies=mkdir_jobs))
 
 
@@ -158,6 +162,7 @@ class OCRWorkflow(WorkflowRunner):
             cmd = "ocropus-nlbin -o %s %s" % (
                 os.path.join(job["output_dir"], "tmp", "ocropus-nlbin"),
                 os.path.join(job["output_dir"], "tmp", "tiff_files", os.path.basename(job["path"]).rsplit(".", 1)[0] + "-*.tif"))
+            cmd = shlex.escape(cmd);
             ocropusnlbin_jobs.append(self.addTask(label="ocropusnlbin_job_-_%i" % (ocropusnlbin_job_number), command=cmd, dependencies=split_jobs))
 
 
@@ -177,6 +182,7 @@ class OCRWorkflow(WorkflowRunner):
                     os.path.join(job["output_dir"], "tmp", "tesseract", file.rsplit(".", 2)[0]),
                     self.lang,
                     "pdf" if self.pdf else "")
+                cmd = shlex.escape(cmd);
                 tesseract_jobs.append(self.addTask(label="tesseract_job_-_%i" % (tesseract_job_number), command=cmd, dependencies=ocropusnlbin_jobs, nCores=min(4, self.nCores)))
 
 
@@ -192,6 +198,7 @@ class OCRWorkflow(WorkflowRunner):
                 cmd = "pdftk %s cat output %s" % (
                     os.path.join(job["output_dir"], "tmp", "tesseract", "*.pdf"),
                     os.path.join(job["output_dir"], os.path.basename(job["path"].rsplit(".", 1)[0] + ".pdf")))
+                cmd = shlex.escape(cmd);
                 pdf_merge_jobs.append(self.addTask(label="pdf_merge_job_-_%i" % (pdf_merge_job_number), command=cmd, dependencies=tesseract_jobs))
 
 
@@ -206,6 +213,7 @@ class OCRWorkflow(WorkflowRunner):
                 pdf_to_txt_job_number += 1
                 cmd = "pdftotext -raw %s" % (
                     os.path.join(job["output_dir"], os.path.basename(job["path"].rsplit(".", 1)[0] + ".pdf")))
+                cmd = shlex.escape(cmd);
                 pdf_merge_jobs.append(self.addTask(label="pdf_to_txt_job_-_%i" % (pdf_to_txt_job_number), command=cmd, dependencies=pdf_merge_jobs))
 
 
@@ -220,6 +228,7 @@ class OCRWorkflow(WorkflowRunner):
             cmd = "mv %s %s" % (
                 os.path.join(job["output_dir"], "tmp", "tesseract", "*.hocr"),
                 os.path.join(job["output_dir"], "hocr_files"))
+            cmd = shlex.escape(cmd);
             move_hocr_jobs.append(self.addTask(label="move_hocr_job_-_%i" % (move_hocr_job_number), command=cmd, dependencies=tesseract_jobs))
 
 
@@ -234,6 +243,7 @@ class OCRWorkflow(WorkflowRunner):
             cmd = "parse_hocr %s %s" % (
                 os.path.join(job["output_dir"], "hocr_files"),
                 os.path.join(os.path.join(job["output_dir"], os.path.basename(job["path"]).rsplit(".", 1)[0] + ".xml")))
+            cmd = shlex.escape(cmd);
             hocr_to_teip5_jobs.append(self.addTask(label="hocr_to_teip5_job_-_%i" % (hocr_to_teip5_job_number), command=cmd, dependencies=move_hocr_jobs))
 
 
@@ -248,6 +258,7 @@ class OCRWorkflow(WorkflowRunner):
             for job in self.jobs["images"] + self.jobs["pdf"]:
                 cleanup_job_counter += 1
                 cmd = "rm -r %s" % (os.path.join(job["output_dir"], "tmp"))
+                cmd = shlex.escape(cmd);
                 cleanup_jobs.append(self.addTask(label="cleanup_job_-_%i" % (cleanup_job_counter), command=cmd))