A developer's blog

The community documentation can be of great help, namely the following section

1. If you use encrypted filenames (standard in Ubuntu >= 9.04) you have to do the following first:
   •  sudo ecryptfs-add-passphrase --fnek 
   •  Passphrase:  (Enter the mount passphrase you recorded when you setup the mount–this passphrase is different from your login passphrase.)
   • You should now get two lines looking like this:
   •  Inserted auth tok with sig [9986ad986f986af7] into the user session keyring 
   •  Inserted auth tok with sig [76a9f69af69a86fa] into the user session keyring  (write down the second value in the square brackets)
2. Mount using sudo:
   •  sudo mkdir -p /home/username/Private  
   •  sudo mount -t ecryptfs /home/username/.Private /home/username/Private 
   •  Selection: 3  (use a passphrase key type)
   •  Passphrase:  (Enter the mount passphrase you recorded when you setup the mount–this passphrase is different from your login passphrase.)
   •  Selection: aes  (use the aes cipher)
   •  Selection: 16  (use a 16 byte key)
   •  Enable plaintext passthrough: n 
   •  Enable filename encryption: y  (This and the following options only apply if you are using filename encryption)
   •  Filename Encryption Key (FNEK) Signature:  (the value you wrote down from the second line above)

Problem is, like some folks at the Ubuntu forums, I encounted the following error when trying sudo mount -t encryptfs /source /destination:

Attempting to mount with the following options:
  ecryptfs_unlink_sigs
  ecryptfs_fnek_sig=xxx
  ecryptfs_key_bytes=16
  ecryptfs_cipher=aes
  ecryptfs_sig=xxx
Error mounting eCryptfs: [-2] No such file or directory
Check your system logs; visit

After fiddling around for a whole day, I finally figured out the problem: the blind confidence in Linux developers’ ability. You see, I used to believe Linux’s symlink always point to the same directory on the drive event after it was mounted somewhere else. So say if a link from your old drive, say ~/etc points to /etc, when you mount the root filesystem on another machine, it should now point to /media/mount/etc right? Wrong!

Problem is, the hidden .Private in the encrypted folder is only one such symlink, and when you attempted to mount it in another system, the symlink is broken. If you do a ls -l, you can see the symlink points to /home/.ecryptfs/<yourusername>/.Private relative to your old directory structure, so to successfully mount your encrypted folder, you shouldn’t mount /home/<yourusername>/.Private like mentioned in the tutorial, but instead

sudo mount -t ecryptfs /media/[old drive mount point]/home/.ecryptfs/[your old username]/.Private /home/[your new username]/[new mount point]

Problem solved!

Some of you might have landed on this article searching for the book instead, here it is :)

I have been trying to improve my typing by learning Dvorak and so far I’ve been to be able to type more accurately (I’m not talking about speed here :p). I have been using the Grass Soft touch typing program to train on Windows, it is pretty useful except for the limited word list. They claimed that they have included 500 most used word in English. That just doesn’t seem enough to me!

Besides, I have also started to use Linux and I needed a program that will keep me busy on the new platform. Of course in the open source world there’s always a lot of option too choose from, most of them are free too. I have chosen KTouch as it seems to be under active development and also endorsed by KDE.

KTouch with my customized lesson

So far, it’s a great program! It automatically detects your keyboard layout and select the appropriate starting lesson. It have two lesson for Dvorak: ABCD and computer generated. Despite the name, the computer generated lesson does not change as you take it because it’s just a file pre-rendered by the developer. I hit the same obstacle I did with the previous touch typing program: no variety.

Fortunately KTouch allows you to load external lesson files (they are just XML files) and text file; but due to poor application design, the text file won’t let you type the whole file but only the first few sentences. Which means if you want to practice with a long text, you’ll have to split them manually into multiple files.

<?xml version="1.0" encoding="utf-8"?><KTouchLecture>
<Title>Puzo, Mario - The Godfather.txt</Title>
<Comment>This is a lession created from Puzo, Mario - The Godfather.txt</Comment>
<FontSuggestions>Courier 10 Pitch</FontSuggestions>
<Levels>
<Level>
<LevelComment>level 1</LevelComment>
<NewCharacters>a lot :)</NewCharacters>
<Line>Chapter 1</Line>
<Line>Amerigo Bonasera sat in New York Criminal Court Number 3 and waited for</Line>
<Line>justice; vengeance on the men who had so cruelly hurt his daughter, who had</Line>
<Line>tried to dishonor her.</Line>
<Line>The judge, a formidably heavy-featured man, rolled up the sleeves of his black</Line>
<Line>robe as if to physically chastise the two young men standing before the bench.</Line>
</Level>
</Levels></KTouchLecture>

Structure of XML lesson files

So I made a program with C# to make KTouch lesson out of text files. Thanks to C#’s ability to process strings and XML, this has been a fairly easy task, here’s the code:

            const int LinePerLevel = 6;
            const int CharactersPerLine = 80;
            FileInfo Info = new FileInfo(Filename);
            string[] RawData = File.ReadAllLines(Filename, Encoding.Default);
            XmlTextWriter Writer = new XmlTextWriter(Filename + ".xml", Encoding.UTF8);
            Writer.WriteStartDocument();
            Writer.WriteStartElement("KTouchLecture");
            Writer.WriteRaw("\r\n");
            {
                Writer.WriteElementString("Title", Info.Name);
                Writer.WriteRaw("\r\n");
                Writer.WriteElementString("Comment", "This is a lession created from " + Info.Name);
                Writer.WriteRaw("\r\n");
                Writer.WriteElementString("FontSuggestions", "Courier 10 Pitch");
                Writer.WriteRaw("\r\n");

                Writer.WriteStartElement("Levels");
                Writer.WriteRaw("\r\n");
                {
                    int LevelCount = 1;
                    int LineCount = 0;
                    string[] LevelLines = new string[LinePerLevel];
                    string Buffer = "";
                    for (int i = 0; i < RawData.Length; i++)
                    {
                        string TrimmedLine = RawData[i].Trim();
                        if (string.IsNullOrEmpty(TrimmedLine))
                            continue;

                        TrimmedLine = TrimmedLine.Replace('“', '"');
                        TrimmedLine = TrimmedLine.Replace('”', '"');
                        TrimmedLine = TrimmedLine.Replace('’', '\'');
                        TrimmedLine = TrimmedLine.Replace('‘', '\'');
                        TrimmedLine = TrimmedLine.Replace("\t", "");

                        if (Buffer.Length > 0 && (Buffer[Buffer.Length - 1] == '.' || Buffer[Buffer.Length - 1] == '?' || Buffer[Buffer.Length - 1] == '!'))
                            Buffer = Buffer + " " + TrimmedLine;
                        else
                            Buffer = Buffer + TrimmedLine;

                        while (LineCount < LinePerLevel && Buffer.Length > CharactersPerLine)
                        {
                            int CutOffSpacePos = Buffer.LastIndexOf(" ", CharactersPerLine - 1);
                            LevelLines[LineCount++] = Buffer.Substring(0, CutOffSpacePos);
                            Buffer = Buffer.Substring(CutOffSpacePos + 1);
                        }

                        if (Buffer.Length < = CharactersPerLine && LineCount < LinePerLevel)
                        {
                            LevelLines[LineCount++] = Buffer;
                            Buffer = "";
                        }

                        if (LineCount >= LinePerLevel)
                        {
                            Writer.WriteStartElement("Level");
                            Writer.WriteRaw("\r\n");
                            {
                                Writer.WriteElementString("LevelComment", "level " + LevelCount.ToString());
                                Writer.WriteRaw("\r\n");
                                Writer.WriteElementString("NewCharacters", "a lot :)");
                                Writer.WriteRaw("\r\n");
                                for (int j = 0; j < LevelLines.Length; j++)
                                {
                                    Writer.WriteElementString("Line", LevelLines[j]);
                                    Writer.WriteRaw("\r\n");
                                }
                            }
                            Writer.WriteEndElement();
                            LevelCount++;
                            LineCount = 0;
                        }
                    }

                    while (!string.IsNullOrEmpty(Buffer))
                    {
                        while (LineCount < LinePerLevel && Buffer.Length > CharactersPerLine)
                        {
                            int CutOffSpacePos = Buffer.LastIndexOf(" ", CharactersPerLine - 1);
                            LevelLines[LineCount++] = Buffer.Substring(0, CutOffSpacePos);
                            Buffer = Buffer.Substring(CutOffSpacePos + 1);
                        }

                        if (Buffer.Length < = CharactersPerLine && LineCount < LinePerLevel)
                        {
                            LevelLines[LineCount++] = Buffer;
                            Buffer = "";
                        }

                        if (LineCount >= LinePerLevel)
                        {
                            Writer.WriteStartElement("Level");
                            Writer.WriteRaw("\r\n");
                            {
                                Writer.WriteElementString("LevelComment", "level " + LevelCount.ToString());
                                Writer.WriteRaw("\r\n");
                                Writer.WriteElementString("NewCharacters", "a lot :)");
                                Writer.WriteRaw("\r\n");
                                for (int j = 0; j < LevelLines.Length; j++)
                                {
                                    Writer.WriteElementString("Line", LevelLines[j]);
                                    Writer.WriteRaw("\r\n");
                                }
                            }
                            Writer.WriteEndElement();
                            LevelCount++;
                            LineCount = 0;
                        }
                    }
                }
                Writer.WriteEndElement();
            }
            Writer.WriteEndElement();
            Writer.WriteEndDocument();
            Writer.Close();

It does:

• Convert Unicode punctuations to their ordinary counterpart. For example “ and ” will be converted to “
• Trim and combine different lines from different paragraphs to they’ll fit into the line format of KTouch. KTouch does not work well with long lines (you can’t see much toward the end of the line) so by default the line is trimmed at 80 characters

For a text source to practice, I decided to drop by project Gutenberg (FYI they digitize out-of-copyright books). They offer a range of about 3000 books in a dozen of languages to choose from. The books are mostly in plain text format, which is just perfect for this purpose.

The first novel I have chosen to type is Godfather by Mario Puzo :) If you want to have a ride too then here’s the lesson maker’s source and here’s the lesson file for Godfather (note that due to bugs in KTouch, some characters toward the end of each lesson will disappear). The lesson file has over 2000 levels available for practicing.