From d323e5f131d19c52d9cb9a2db7fc5ef6c8d57d50 Mon Sep 17 00:00:00 2001
From: Stephan Gabler <stephan.gabler@gmail.com>
Date: Wed, 18 Jul 2012 10:38:25 +0200
Subject: [PATCH] test for file path parsing

---
 jrnl/Journal.py    | 22 +++++++++++++++++++++-
 tests/empty.txt    |  0
 tests/test_jrnl.py | 42 ++++++++++++++++++++++++++++++++++++++++++
 tests/url_test.txt |  6 ++++++
 4 files changed, 69 insertions(+), 1 deletion(-)
 create mode 100644 tests/empty.txt
 create mode 100644 tests/url_test.txt

diff --git a/jrnl/Journal.py b/jrnl/Journal.py
index d88cd3f9..a00dc502 100644
--- a/jrnl/Journal.py
+++ b/jrnl/Journal.py
@@ -15,7 +15,7 @@ import readline, glob
 try:
     from Crypto.Cipher import AES
     from Crypto.Random import random, atfork
-except ImportError: 
+except ImportError:
     pass
 import hashlib
 import getpass
@@ -23,11 +23,13 @@ try:
     import clint
 except ImportError:
     clint = None
+import shutil
 
 class Journal:
     def __init__(self, config, **kwargs):
         config.update(kwargs)
         self.config = config
+        self.path_search = re.compile('[\[\(]?(([A-Z]:|/)\S*?\.(tif|tiff|jpg|jpeg|gif|png))[\[\)]?')
 
         # Set up date parser
         consts = pdc.Constants()
@@ -233,6 +235,23 @@ class Journal:
 
         return date
 
+    def parse_for_images(self, body):
+        print 'in parse for images'
+        # TODO: parse also for name (like used in markdown [name](url))
+        path_start = re.compile('[\[\(]?([A-Z]:|/).*?\.(tif|tiff|jpg|jpeg|gif|png)[\[\)]?')
+        for word in body.split():
+            res = path_start.match(word)
+            if res and len(res.groups()) == 2 and os.path.exists(word.strip('()[]')):
+                word = word.strip('()[]')
+                print word
+                save_path = os.path.splitext(self.config['journal'])[0] + '_data'
+                if not os.path.exists(save_path):
+                    os.mkdir(save_path)
+                new_path = os.path.join(save_path, os.path.basename(word))
+                # TODO: replace name by hash
+                shutil.copyfile(word, new_path)
+        return body
+
     def new_entry(self, raw, date=None):
         """Constructs a new entry from some raw text input.
         If a date is given, it will parse and use this, otherwise scan for a date in the input first."""
@@ -247,6 +266,7 @@ class Journal:
                 title_end = sep_pos
         title = raw[:title_end+1]
         body = raw[title_end+1:].strip()
+        body = self.parse_for_images(body)
         if not date:
             if title.find(":") > 0:
                 date = self.parse_date(title[:title.find(":")])
diff --git a/tests/empty.txt b/tests/empty.txt
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/test_jrnl.py b/tests/test_jrnl.py
index 8280fe6e..bca77bb0 100644
--- a/tests/test_jrnl.py
+++ b/tests/test_jrnl.py
@@ -2,6 +2,8 @@
 # encoding: utf-8
 
 import unittest
+import os
+from jrnl.Journal import Journal
 
 class TestClasses(unittest.TestCase):
     """Test the behavior of the classes.
@@ -11,12 +13,52 @@ class TestClasses(unittest.TestCase):
     """
 
     def setUp(self):
+        self.data_path = os.path.dirname(os.path.abspath(__file__))
+        self.config =   {
+                            "timeformat": "%Y-%m-%d %H:%M",
+                            "encrypt": False,
+                            "tagsymbols": "@",
+                            "journal": ""
+                        }
+
+    def tearDown(self):
+        # TODO: delete copied file, etc
         pass
 
     def test_colon_in_textbody(self):
         """colons should not cause problems in the text body"""
         pass
 
+    # def test_file_detection(self):
+    #     self.config['journal'] = os.path.join(self.data_path, 'empty.txt')
+    #     journal = Journal(config=self.config)
+    #     with open(os.path.join(self.data_path, 'url_test.txt')) as f:
+    #         journal.new_entry(f.read())
+    #     # TODO: check that copied file exists
+
+    def test_rendering_md(self):
+        pass
+
+    def test_rendering_htm(self):
+        pass
+
+    def test_open_in_browser(self):
+        pass
+
+    def test_pathsearch_regex(self):
+        true_positive = ['/Volumes/dedan/bla.png',
+                         '/Users/dedan/projects/jrnl/tests/golden.jpg',
+                         '/Volumes/dedan/test.png']
+        false_positive = ['/Volumes/dedan/bla.blub',
+                          'http://en.wikipedia.org/wiki/Generative_model']
+        self.config['journal'] = os.path.join(self.data_path, 'empty.txt')
+        journal = Journal(config=self.config)
+        with open(os.path.join(self.data_path, 'url_test.txt')) as f:
+            results = [res[0] for res in journal.path_search.findall(f.read())]
+            for tp in true_positive:
+                self.assertIn(tp, results)
+            for fp in false_positive:
+                self.assertNotIn(fp, results)
 
 class TestCLI(unittest.TestCase):
     """test the command-line interaction part of the program"""
diff --git a/tests/url_test.txt b/tests/url_test.txt
new file mode 100644
index 00000000..222ecd1b
--- /dev/null
+++ b/tests/url_test.txt
@@ -0,0 +1,6 @@
+ein normalser satz.
+
+Lorem ipsum dolor sit amet, consectetur adipisicing elit, (/Volumes/dedan/bla.png
+) sed do eiusmod tempor incididunt ut (http://en.wikipedia.org/wiki/Generative_model) labore et dolore magna aliqua. Ut enim ad minim veniam, (/Users/dedan/projects/jrnl/tests/golden.jpg)
+quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo
+consequat. Duis aute irure dolor http://www.dict.cc/?s=descendant in reprehenderit in voluptate velit esse /Volumes/dedan/test.png cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non (/Volumes/dedan/bla.blub) proident, sunt in culpa qui officia deserunt mollit anim id est laborum.