MySQL, which is a sort of abbreviation for the casual term “My Sequel” is what is called a “relational database management system” that may be a part of the operating system of your website. In fact, MySQL is used in a number of the Internet’s most popular and most frequently visited websites, such as Google and Facebook, among others. As noted on the MySQL Wikipedia page (incidentally, the Wikipedia website uses MySQL as well), MySQL exists as a server that allows multiple users access to a number of databases. There are different levels or versions of MySQL available, but, for those interested, it is worth knowing that at least a basic version of the MySQL server may be in use when you create a personal website through a popular hosting company.

This is just one of a number of comprehensive computer programs, coding languages, complex servers, etc. that may be a part of the functionality of your website. Ultimately, it is not necessarily required of you (depending on how your domain host works) to understand this side of things, or to be able to make any sense of it. However, for those with more computer experience, or those who simply have an interest in how things work, and not simply what they result in, these sorts of things can be quite interesting. There is, after all, much to learn about the computer science behind all of the creations you conceive of and the changes you implement while designing your own website. In the end, an increased understanding of this computer science will likely help you to be able to understand how to make a better website as well.

Display results vertically to avoid line-wrapping

If you do a select * from a table with a lot of columns, you will often end up with the results being wrapped across your terminal, making them just about impossible to read. To display the results vertically (i.e. <column>: <value>), end your query with \G instead of ;

select * from users \G

Clear the screen

If you type help it says that \c is the clear command - this is actually something else (I will cover it later). To clear your screen of output and put a fresh mysql> at the top, type: \! clear (or on Windows, \! cls).

Cancel the current command and clear the input

How often do you type select * fom uesrs or similar? If you are used to Linux, your instinctive response is to press Ctrl-C - unfortunately in older versions of MySQL this will quit the shell and dump you back to bash (or equivalent). Try this instead:

slect * fom typo\c

The input will be cleared and MySQL won’t display any errors or warnings.

Edit your last query

Often you will select * when you don’t need to, and if it’s a multi-line query that you have pasted into the shell, it is laborious to fix. Try \e;.

This will open your last run query in whatever text editor you have defined in $EDITOR (vim/nano, etc.). You can then edit your query and then save and exit, and it will execute with the changes.

pbcopy takes the standard input and places it in the specified paste-board. If no pasteboard is specified, the general pasteboard will be used by default.

pbpaste removes the data from the pasteboard and writes it to the standard output.

For example:

user@host:~/Projects/some/dir/that/is/long$ pwd | pbcopy

The working directory is now in your system clipboard, available for you to paste somewhere.

Additional links

1. pbcopy / pbpaste in OS X
2. Clipboard access from shell (utf-8)

Here’s how to add automatic paging and highlight changes in bright colours.

Install colordiff on Ubuntu:

apt-get install colordiff

OS X:

port install colordiff

Make a new file called svndiff with the following inside:

#!/bin/bash
svn diff $* | colordiff | less -r

Make this file executable (chmod +x svndiff) and copy it somewhere in your path (I put it in /usr/local/bin).

The $* bash argument passes any arguments you give the script to the svn diff command, so you can still do things like:

svndiff file1 file2

Here’s the wise and generally cool Rafiq with some more info.

Coolest Guy on the Internet

--->  Building py30-readline
Error: Target org.macports.build returned: shell command " \
cd "/opt/local/var/macports/build/_opt_local_var_ \
macports_sources_rsync.macports.org_release_ports \
_python_py30-readline/work/Python-3.0/Modules" && \
/opt/local/Library/Frameworks/Python.framework/ \
Versions/3.0/bin/python3.0 setup.py build " returned \
error 1
Command output: readline.c:396: warning: passing \
argument 1 of 'free' discards qualifiers from pointer \
target type... 

I fixed it with the following:

sudo port sync
sudo port install readline
sudo port install py30-readline

The last command will cause the error, but will leave the package unarchived in the following directory:

/opt/local/var/macports/build \
/_opt_local_var_macports_sources_rsync. \
macports.org_release_ports_python_py30- \
readline/work/Python-3.0/Modules

Change to this directory and edit setup.py:

Add /opt/local/include and /opt/local/lib to the respective directory arrays at the top of the file, like so:

lib_dirs = [prefix + '/lib', "/opt/local/lib"]

and the same for the include directory. Save and exit the editor and run the port install command again. It should work now. I still got the following error after installation:

Error: Target org.macports.activate returned: Not a directory
Error: Status 1 encountered during processing.

but it didn’t seem to affect anything.

cd project
mvim project_file.php

MacVim supports tabs, but unfortunately calling mvim multiple times from the command-line results in multiple separate windows opening, instead of multiple tabs in one window. I made the following modifications to the mvim script to correct this.

Add the following line to the top of the file, below the commented section:

tabs=true

Replace the if structure at the bottom of the file with the following:

# Last step:  fire up vim.
if [ "$gui" ]; then
  if $tabs && [[ `$binary --serverlist` = "VIM" ]]; then
    exec "$binary" -g $opts --remote-tab-silent ${1:+"$@"}
  else
    exec "$binary" -g $opts ${1:+"$@"}
  fi
else
  exec "$binary" $opts ${1:+"$@"}
fi

Subversion 1.1 (and later) has the ability to put a symlink under version control, via the usual svn add command.

Details: the Subversion repository has no internal concept of a symlink. It stores a “versioned symlink” as an ordinary file with an ’svn:special’ property attached. The svn client (on unix) sees the property and translates the file into a symlink in the working copy. Win32 has no symlinks, so a win32 client won’t do any such translation: the object appears as a normal file.

So I tested this with:

cd propsite/tools/library
ln -s ../../config/database.yml .
svn add database.yml
svn ci -m "Symlink test"

And then on the test server:

cd propsite2
svn up
cd tools/library
ls -la

The symlink is there, and correct.

1. Open TextMate
2. Go to TextMate > Preferences > Advanced
3. Click Shell Variables
4. Add a new variable called TM_SVN and set it to the path of the binary you want to use

Close the preference pane and open the bundle editor.

Bundles > Bundle Editor > Show Bundle Editor

Expand the GetBundle entry and in Install Bundle and Update Installed Bundles change:

SVN="${TM_BUNDLE_SUPPORT}/bin/svn"

to

SVN=${TM_SVN:-"${TM_BUNDLE_SUPPORT}/bin/svn"}

~/Documents/src/subversion-1.4.4$ make uninstall
make: *** No rule to make target `uninstall'.  Stop.

Checking out a couple forums I found this had been brought up before and people felt that maintaining an uninstall target in the makefile was unwieldly and wouldn’t be done any time soon. GNU Stow was suggested as a package watchdog, but this cannot be applied retroactively obviously.

The best solution came in the form of using gfind (GNU find), which I first had to install:

sudo port install findutils

Then run:

gfind /usr/local -printf '%T@ ' -ls | sort -n > tfiles

This will put a list of all files in /usr/local ordered by their time-modified fields into svnfiles. Opening this file I found the first mention of apr which had a bunch of entries right before neon which came right before libsvn and svn. I extracted all lines referencing the above 3 packages in the same approximate time-frame and saved them to a new file, svnfilelist.

more svnfilelist | awk '{print $12;}' > svnfilenames
more svnfilenames | xargs sudo rm -rf

Do not do this without editing the file list by hand, otherwise you will delete a lot of things you don’t want to delete. It’s not my fault, etc.