Legacy versions of Corpus Presenter
Release notes for Version 11 of Corpus Presenter
Version 11 of Corpus Presenter provides a number of additional features which further increase the flexibility of the program. For instance, there is now a function Search via word list extract in the Search menu. Via this option you can create a customised extract of forms from a fuller list which you generate yourself – probably via the Make word list option. You can use this extract for a search across those texts from which the fuller list was generated. This guarantees that the forms are found and returned.
Another new feature concerns the exporting of returns to a Microsoft Excel table for chart generation. This feature now includes the option of generating not just raw returns from a search but also percentages reflecting the relative frequencies of strings in texts. Along with this there is an option of producting the frequency of finds per thousand words.
The interface for the Quick search function has been simplified considerably so that only the essential options for a search are now visible with the extra options available in an additional window.
In addition, some smaller changes have been made to existing functions and options.
The supplied word processor – CP WordPro – has been greatly expanded and can now load and save Word 2003 and 2007 files directly. Other programs, notably CP File, CP Text Tool and CP Find Text have also been expanded considerably.
Raymond Hickey
February 2009
Release notes for Version 10 of Corpus Presenter
The major change in Version 10, compared to Version 9, is the option on all search levels of storing returns as a database which can then be used for chart generation in Microsoft Excel. The way this works is as follows: assume that you search for a certain structure across a set of 100 texts and that a percentage of these show the structure, but to varying extents. The database which Corpus Presenter stores to disk contains the names of those texts in which the structure was found and the number of times it was found in each text. Such a database can then be loaded into Microsoft Excel and via the command Insert, Chart (after selecting all the rows and columns of the database with Ctrl-A) you can generate a chart in which the occurrences of the structure you searched for are shown in chart form.
Corpus Presenter can furthermore gather up to six sets of returns for a group of texts and transfer these to a database (for later chart generation). The advantage of this is you can see whether the occurrences of more than one structure run parallel across a set of texts, either ascending or descending. You can also check whether two or more structures run in opposite directions, e.g. whether the increase of one structure correlates with the decrease of another.
Some other options have been added, especially those allowing users to store any set of returns on disk and re-load these for editing in a later work session.
More on chart generation in Version 11 of Corpus Presenter
There is also a new utility CP Make Tree which makes the design and editing of tree structures for Corpus Presenter even easier, see the description of this programm (with screen shots).
The latest version is included in the ZIP file which can be downloaded from this website (goto section Downloads).
Raymond Hickey
January 2007
Release notes for Version 9 of Corpus Presenter
Four major changes have been made to the main program Corpus Presenter in the latest version:
1) The directory interface has been improved and you now see all the drives on your computer in a tree on the left. Other components on this level have also been improved, e.g. the file list on the right can be made much faster by simplifying the display. Toggle to the quick display mode (hotkey: Shift-Ctrl-F5) and you will find that it only requires a split second to load and display a few thousand files. Two additional features should be highlighted:
(i) You can now generate a DataSet file (the small control file which Corpus Presenter uses to display texts in a tree on the left of the main screen) by just selecting files on the directory lister level and clicking on the button Make data set at the bottom of the screen. Once you do this, the selected files are entered into a new data set file. This is now shown and can be selected for processing by double clicking it and all the retrieval options of Corpus Presenter are at your disposal.
(ii) You can now convert files on this level. Click on the button Convert files and a window opens offering a range of options and showing the files from the current directory which are eligible for conversion. You could, for instance, convert XML files to text files and use them straight away for search tasks.
2) The simple search function, accessed by pressing Ctrl-F, has been enhanced. You can now specify whether the program is to search through all loaded texts, starting at the current one and moving downwards in the list. You can also say whether the search is to be carried out automatically. The results of a search are deposited in the Windows clipboard and can be retrieved via Ctrl-V or Paste in any text processing software.
3) On both the Quick Locate and the Text Retrieval levels you can now store any returns to disk in a format which allows re-loading later (this was retricted to multi-line grid returns on the Text Retrieval level up to this. There is no limit to the sets of returns you can save to disk. If you re-load a set of returns with the corpus of text files from which they were generated (this will normally be the case for most users), then the GoTo function allows you to jump to the point in the relevant texts where finds were made when the set of returns was originally generated.
4) You can now automatically comb through returns on both the Quick Locate and the Text Retrieval levels and have the results stored to the Windows clipboard. If you have a very large set of returns, say several hundred, it might be useful to search for text within such a set, rather than going through it manually. The function gains additional usefulness in combination with the previous option, so that you can search through sets of returns which were generated in a previous work session.
Some small changes have also been made to display in various parts of the program (it now remembers user-specified sizes of returns tables). There are also one or two changes in file handling (so that the program remembers every alteration the user might make to any file). On the retrieval level you can now look for two strings with zero intervening items, i.e. for strings which are adjacent to each other in a text. Please note that you can now choose to load HTML files as plain text (click on button "Setting" on the directory listing level). This leads to quick retrieval of text and may be a solution to problems loading complex HTML files in their native format.
In the following table the major changes have been summarised.
| Improvements to Corpus Presenter |
| The option of saving the results of an operation – string search, text retrieval, word list, text statistics– in HTML format, i.e. as an .HTM file, has been extended to all levels of the programme. |
| On the text retrieval level (Corpus Presenter only) you can output the results of any grid to an RTF table. There is now a supplied utility CP_RtfTableEditor for editing these tables directly (includes option of rearranging columns and rows). |
| You can now start Corpus Presenter and have it go straight to the directory listing level. Click on the button Settings here to specify whether should happen each time you load either programme. |
| On the directory listing level you can now generate a Word List or Text Statistics using a number of input files (previously this applied only to the current file). All you need to do is select files by holding down the Ctrl-key and clicking on file names with the left-mouse button. |
| There is a popup menu on the directory level which contains most of the commands on this level for your convenience. |
Raymond Hickey
March 2006
Release notes for Version 8 of Corpus Presenter
This version is the first to be released after the publication of the book Corpus Presenter. Software for Language Anaylsis by John Benjamins in 2003. All functions of the program are described in the online help and in the accompanying book just mentioned. This version runs faster than the previous one and the Quick Locate, Word List and Text Retrieval modules have been considerably improved. Many of the additional programs have also been updated and run more efficiently.
Note that this version, 8.0, will only work on recent Windows platforms, that is on Windows 2000, on Windows XP and later versions. It will not run on Windows 98 (either edition) or on Windows 95.
Raymond Hickey
December 2004
History of Corpus Presenter
If you are a newcomer to Corpus Presenter you will be using version 11 and you need not be concerned about this section of the website.
For those who have used previous versions or followed the history of the software the following remarks are offered by way of explanation. The Corpus Presenter software has been many years in the making. During that time much valuable feedback was gained in conversations with colleagues and, of course, I reflected continuously on what such software should do. When it came to publishing the book with the CD-ROM, the software was at version 7. In 2002 when the software was being prepared for publication I had to make a decision regarding operating systems: most colleagues worked at universities where the computers are not always the newest with the latest version of the operating system, to say the least. Corpus Presenter had to work with earlier versions of Windows and so the cut-off point was set at Windows 98 Second Edition because that was the least one could expect colleagues to be using. This was a compromise and one which caused difficulties as many of the problems which arose were due to the inadequacies of earlier versions of Windows.
In 2004 I decided to go for Windows XP only as it was obvious by then that this was the dominant version of Windows and, importantly, that colleagues in university departments around the world were changing to this. The decision was a happy one as many of the problems, like unexplainable crashes due to system memory failure or due to conflicting modules within Windows, melted away.
In addition to advances on the operating system front, I was able to improve the software and I am grateful to many colleagues, especially in Sweden and Finland, who were forthcoming with many sensible suggestions which I did my best to implement. In late 2004 version 8 was distributed.
During 2005 a number of further ideas were implemented, especially with regard to types of files which can be processed (including file conversion and XML preparation). The start on the directory level was re-arranged so that users could basically carry out a whole range of functions without even loading files. Some other embellishments were added to the retrieval modules, including a simple but powerful search facility, and the search engine behind the whole program was examined and re-examined to ensure absolute stability.
These developments, in the years since the software was prepared for publication with the book, have meant that Corpus Presenter has been steadily improved and that anyone who has used previous versions should most definitely upgrade to version 12 to gain maximum benefit from the program suite.
Please ensure that you only install version 12 on computers running under Windows XP (Service Pack 3) or Windows Vista / Windows 7. That way the software is completely stable and fully functional.
