ivi sorting and counting

Sorting

Command: SOrt

Purpose: Arrange the corefile lines or records in lexical order

Format: SO <digit>|<number>|<number>|<number>

Parameters: The four parameters work as follows:

In TEXT view, the sort-key begins at the cursor and extends for n characters, where n is the key-length specified by the second parameter. If the key-length is negative, the sort-key runs from the cursor backwards (i.e. right- to-left).

In either of RECORD views, the position of the cursor selects a specific field of a record, and the sort-key is the corresponding field in every record.

Examples:

SO |5|*|* - sort the starred block in ascending order with key of length 5.

SO 1|-8 - sort the entire corefile in descending order. Sort-keys are compared from right to left with a key-length of 8.

Note: if a column of words is aligned by the right-hand end, this "reverse sort" brings together words with a similar ending:

      walked
      edited
       spoke
    speaking
     walking
     editing
       speak
        walk
      speaks
       walks
       edits
        edit

Sorting accented characters

The basic ivisort uses the ISO-8859-1 character mapping to sort characters. Unfortunately, this means that accented characters, like the e acute, are placed at the end of the alphabet. To remedy this, the SKey command has been defined. In a nutshell, it permits a sequence of sorts based on either the letter ("e", for example), or the diacritic (the acute, grave, etc. accent).

Command: SKey

Purpose: Sort characters by base characters or diacritics

Format: SO <digit>

Parameters: There are three possible parameters:

To use the SKey command, type SK on the command line. You will see the prompt:

0: ASCII; 1: French unaccented; 2: accents

First, go to the top of the portion of the text to be sorted and type: SK 2 and then SO. The characters with accents will be sorted together in the appropriate order. Then, go back to the start of the block to be sorted and type SK 1 and then SO. Base characters will be sorted, with accented characters in the appropriate place with respect to their unaccented versions.

Example:

Suppose an initial text like this:

abbesse
était
elle
êtes
Esther
femme

A regular sort will produce:

abbesse
elle
Esther
était
êtes
femme

On the other hand, typing SK2 followed by SO will produce:

Esther
elle
femme
abbesse
était
êtes

and SK1 followed by SO will give the final product of

abbesse
elle
Esther
était
êtes
femme

Word and Sentence Length

This operation was provided for a class of humanities students exploring the use of computers to extract information from literary texts.

Operation: Length tables

Purpose: Construct word and sentence length tables for the text in the corefile

Default Key: <ESC> N

Operation Number: 105

The operation scans the text in the current corefile starting at the cursor and continuing to the end. Word and sentence length tables are appended to corefile 7.