• Preliminaries
• Installing CorpusSearch
• Running CorpusSearch

Preliminaries

• If you are using a Unix-derived system, including Linux and MacOS, you will need to learn something about shells. The instructions below assume that you are using the T shell (tcsh), but MacOS now installs the z shell (zsh) by default, which is an extension of the Bourne shell (bash). The basic installation is the same across shells, but each shell has a slightly different syntax for path specification, aliases, and so on. Please should consult local documentation for details.
• Changes introduced with MacOS 10.15 (Catalina) cause certain Java applications to fail, but as of late 2023, CorpusSearch runs under MacOS 13 (Ventura) and likely also under MacOS 14 (Sonoma).

• The following item may be outdated. We leave it up for the reader's convenience. If you are using Windows, we recommend downloading a copy of the Java Virtual Machine (version 1.4 or later) from the Sun java site (http://www.java.com/) rather than running Microsoft's Java.

  Here's how to check which virtual machine you are running under Windows XP.

  • Search for the Command Prompt by clicking on the "Run" icon on the Start menu.
  • Enter "cmd" in the dialog box that comes up.
  • Enter "C:\Documents and Settings\username>java -version" in the Command Prompt dialog box that comes up.

• Download the latest version of CorpusSearch from https://corpussearch.sourceforge.net/, or use the version that comes with the LDC release of the Penn Parsed Corpora of Historical English. In what follows, we assume that this is version CS_2.003.04.jar.

• Move the .jar file to a convenient location - for example, the "/Applications" folder under MacOS or the "Program Files" folder under Windows.

• Unix/Linux users (including MacOS):

  • Open a window that allows you to run commands - the Terminal program under MacOS or an xterm window under Unix or Linux.

  • Assuming that you have moved CS_2.003.04.jar to the "/Applications" folder, enter the following line at the prompt in the terminal window.

    java -classpath /Applications/CS_2.003.04.jar csearch/CorpusSearch
    

    If you have moved the .jar file to some other location than /Applications, edit the path accordingly.

    Please be aware that if your operating system displays paths to you in a anguage other than English, it almost certainly still uses the English pathnames underlyingly. So a path that appears to you as "/Applicazioni" or "/Utenti/beatrice/Scrivania" will likely need to be specified by the English "/Applications" or "/Users/beatrice/Desktop".

  • CorpusSearch should now prompt you for input. If not, check the path to the .jar file. Once you're prompted for input, follow the instructions to quit the search. All you are doing at this point is checking that the path to the program is set up correctly.

  • For convenience, you can add an alias for the above java command to your shell initialization file (for instance, .cshrc, .zshrc, .bashrc). The following line will create such an alias in a .cshrc file:

    alias CS 'java -classpath /Applications/CS_2.003.04.jar csearch/CorpusSearch'
    

    Use the appropriate syntax for other shells. Now you can invoke CorpusSearch with just "CS".

    Again, if your operating system displays pathnames in a language other than English, the alias in your shell initialization file almost certainly needs to be specified in terms of the English directories.

• Windows users:

  • Launch the Command Prompt via Start -> Run and enter "cmd" in the box ("command" under Windows 98).

  • Enter the following command into the Command Prompt dialog box:

    cd C:\
    

  • Enter the following line at the resulting "C:\>" prompt:

    java -classpath "C:\Program Files\CS.jar" csearch/CorpusSearch
    

    Be sure to include the quotation marks in the first argument after "-classpath". Also note the different direction of the slashes in the first and the second argument (backward and forward, respectively).

Before initiating a search in the background, make sure there is no .out file corresponding to the query you intend to run. If there is, your search will stall.