From c38d3a8b3523b0fff74d9a4456887f22f5087671 Mon Sep 17 00:00:00 2001
From: j <j@mailb.org>
Date: Fri, 1 Feb 2019 17:36:03 +0530
Subject: [PATCH] use pdftotext if available

---
 oml/media/pdf.py | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/oml/media/pdf.py b/oml/media/pdf.py
index bc342d0..75313b2 100644
--- a/oml/media/pdf.py
+++ b/oml/media/pdf.py
@@ -237,10 +237,13 @@ def info(pdf):
 def extract_text(pdf):
     if sys.platform == 'win32':
         pdf = get_short_path_name(pdf)
+    cmd = ['pdftotext', pdf, '-']
     if sys.platform == 'darwin':
-        cmd = ['/usr/bin/mdimport', '-d2', pdf]
-    else:
-        cmd = ['pdftotext', pdf, '-']
+        pdftotext = ['/usr/local/bin/pdftotext', pdf, '-']
+        if os.path.exists(pdftotext[0]):
+            cmd = pdftotext
+        else:
+            cmd = ['/usr/bin/mdimport', '-d2', pdf]
     if sys.platform == 'win32':
         startupinfo = subprocess.STARTUPINFO()
         startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW
@@ -251,7 +254,7 @@ def extract_text(pdf):
     stdout, stderr = p.communicate()
     stdout = stdout.decode()
     stderr = stderr.decode()
-    if sys.platform == 'darwin':
+    if sys.platform == 'darwin' and cmd[0] == '/usr/bin/mdimport':
         if 'kMDItemTextContent' in stderr:
             stdout = stderr.split('kMDItemTextContent = "')[-1].split('\n')[0][:-2]
         else: