From e6fbbfec41e85af362d870709d5aa17ad54f4af7 Mon Sep 17 00:00:00 2001
From: Dan Helfman <dan@luminotes.com>
Date: Fri, 20 Feb 2009 15:22:23 -0800
Subject: [PATCH] Several minor improvements to CSV importing and exporting.

---
 NEWS                                       |  1 +
 plugins/export_csv/__init__.py             | 16 ++++++++++++++--
 plugins/export_csv/test/Test_export_csv.py | 19 +++++++++++++++++--
 plugins/export_html/Html_file.py           |  2 ++
 4 files changed, 34 insertions(+), 4 deletions(-)

diff --git a/NEWS b/NEWS
index 25596f7..3fee715 100644
--- a/NEWS
+++ b/NEWS
@@ -2,6 +2,7 @@
  * When you export your notebook as an HTML or CSV file, the saved filename is
    now based on the name of your notebook, for instance "my-to-do-list.html"
    or "ideas-for-my-novel.csv".
+ * Several minor improvements to CSV importing and exporting.
  * Converted the existing HTML and CSV export features to work as separate
    export plugins. This means that a new export format can be implemented as
    a new plugin.
diff --git a/plugins/export_csv/__init__.py b/plugins/export_csv/__init__.py
index 7a971d8..ff8ca76 100644
--- a/plugins/export_csv/__init__.py
+++ b/plugins/export_csv/__init__.py
@@ -1,18 +1,30 @@
+import re
 import csv
 from cStringIO import StringIO
 from model.User import User
 
 
+FILE_LINK_PATTERN = re.compile( u'<a\s+href="[^"]*\/files\/download\?file_id=[^"]+"[^>]*>', re.IGNORECASE )
+IMAGE_PATTERN = re.compile( u'<img [^>]* ?/?>', re.IGNORECASE )
+NEWLINE_PATTERN = re.compile( u'[\r\n]+' )
+
+
 def export( database, notebook, notes, response_headers ):
   """
   Format the given notes as a CSV file and return it as a streaming generator.
   """
   buffer = StringIO()
-  writer = csv.writer( buffer )
+  writer = csv.writer( buffer, quoting = csv.QUOTE_NONNUMERIC )
 
   response_headers[ u"Content-Disposition" ] = u"attachment; filename=%s.csv" % notebook.friendly_id
   response_headers[ u"Content-Type" ] = u"text/csv;charset=utf-8"
 
+  def prepare_contents( contents ):
+    contents = FILE_LINK_PATTERN.sub( '<a>', contents )
+    contents = IMAGE_PATTERN.sub( '', contents )
+    contents = NEWLINE_PATTERN.sub( ' ', contents )
+    return contents.strip()
+
   def stream():
     writer.writerow( ( u"contents", u"title", u"note_id", u"startup", u"username", u"revision_date" ) )
     yield buffer.getvalue()
@@ -24,7 +36,7 @@ def export( database, notebook, notes, response_headers ):
         user = database.load( User, note.user_id )
 
       writer.writerow( (
-        note.contents and note.contents.strip().encode( "utf8" ) or None,
+        note.contents and prepare_contents( note.contents ).encode( "utf8" ) or None,
         note.title and note.title.strip().encode( "utf8" ) or None,
         note.object_id,
         note.startup and 1 or 0,
diff --git a/plugins/export_csv/test/Test_export_csv.py b/plugins/export_csv/test/Test_export_csv.py
index 9110d49..7491d1f 100644
--- a/plugins/export_csv/test/Test_export_csv.py
+++ b/plugins/export_csv/test/Test_export_csv.py
@@ -41,7 +41,7 @@ class Test_export_csv( object ):
     self.note2 = Note.create( note_id, u"<h3>other title</h3>whee", notebook_id = self.notebook.object_id, user_id = self.user.object_id )
     self.database.save( self.note2, commit = False )
 
-  def test_export_csv( self, note_contents = None ):
+  def test_export_csv( self, note_contents = None, expected_contents = None ):
     if not note_contents:
       note_contents = u"<h3>blah</h3>foo"
 
@@ -87,7 +87,10 @@ class Test_export_csv( object ):
       expected_note = expected_notes[ note_count ]
 
       assert expected_note
-      assert contents.decode( "utf8" ) == expected_note.contents.strip()
+      if expected_contents and note_id == note3.object_id:
+        assert contents.decode( "utf8" ) == expected_contents.replace( "\n", " " ).strip()
+      else:
+        assert contents.decode( "utf8" ) == expected_note.contents.replace( "\n", " " ).strip()
 
       if expected_note.title:
         assert title.decode( "utf8" ) == expected_note.title.strip()
@@ -115,6 +118,18 @@ class Test_export_csv( object ):
   def test_export_csv_with_trailing_newline_in_contents( self ):
     self.test_export_csv( note_contents = u"<h3>blah</h3>foo\n" )
 
+  def test_export_csv_with_file_attachment_in_contents( self ):
+    self.test_export_csv(
+      note_contents = u"<h3>blah</h3>foo<a href=\"/files/download?file_id=blah&quote_filename=False\">file</a>",
+      expected_contents = "<h3>blah</h3>foo<a>file</a>",
+    )
+
+  def test_export_csv_with_image_in_contents( self ):
+    self.test_export_csv(
+      note_contents = u"<h3>blah</h3>foo<a href=\"/files/download?file_id=blah&quote_filename=False\"><img src=\"whee.png\" /></a>",
+      expected_contents = "<h3>blah</h3>foo<a></a>",
+    )
+
   def test_export_csv_with_blank_username( self ):
     self.user._User__username = None
     self.database.save( self.user )
diff --git a/plugins/export_html/Html_file.py b/plugins/export_html/Html_file.py
index b160d48..041899e 100644
--- a/plugins/export_html/Html_file.py
+++ b/plugins/export_html/Html_file.py
@@ -5,6 +5,7 @@ from view.Tags import Html, Head, Title, Style, Meta, Body, H1, Div, Span, Hr, A
 
 class Html_file( Html ):
   NOTE_LINK_PATTERN = re.compile( u'<a\s+href="[^"]*(?:\/notebooks\/)?[^>]+[?&]note_id=([a-z0-9]*)"[^>]*>', re.IGNORECASE )
+  FILE_LINK_PATTERN = re.compile( u'<a\s+href="[^"]*\/files\/download\?file_id=[^"]+"[^>]*>', re.IGNORECASE )
   IMAGE_PATTERN = re.compile( u'<img [^>]* ?/?>', re.IGNORECASE )
 
   def __init__( self, notebook, notes ):
@@ -14,6 +15,7 @@ class Html_file( Html ):
     # images since they're not presently included with the download
     for note in notes:
       contents = self.NOTE_LINK_PATTERN.sub( r'<a href="#note_\1">', note.contents )
+      contents = self.FILE_LINK_PATTERN.sub( '<a>', contents )
       contents = self.IMAGE_PATTERN.sub( '', contents )
       relinked_notes[ note.object_id ] = contents