Repetition Error finding in Emacs (BETA)

This Emacs package helps people who write longer texts to check whether or not they have some repetition errors in there. With repetition errors, we do not just mean those of the form where the same word is repeated repeated right after itself, but those where you use the same word twice in close proximity. After you detect such an error, you can either leave it as is, or consult a thesaurus to find an alternative word to use.

How to Install it

I personally have the file repetition-error.el located in a subfolder of my .emacs.d folder. In my .emacs-file, I simply added the line:

(load "~/.emacs.d/your_subfolder/repetition-error.el")

Execute the line (or restart Emacs), and you are ready to go.

Optionally, one can also test the package to make sure all commands work properly on the installed Emacs version. This can be done by opening the repetition-error.el package and run ert-run-tests-interactively. If there is no error, then the package runs as expected on your system.

How to Use it

1. Write your text
2. Use M-x-find-reperr-whole-buffer to find all repetition errors in the current document. Emacs will switch into an interactive mode, which highlights each finding of a repetition error, and shows the next one when pressing c (continue). This interactive mode can be left -- e.g. in order to correct a found valid repetition error -- by pressing any key but c. Afterwards, one can continue the search from the current point on by using M-x-find-reperr-from-point (this can also be used from the start, if one is only interested in all repetitions after a certain point).
3. When the search for repetition errors is finished, the message "Finished finding repetition errors" is displayed in the minibuffer.

REMARK: There is also a command called M-x find-reperr-latex-whole-buffer resp. M-x find-reperr-latex-from-point, which ignores certain LaTeX regions in the search (e.g. comments, math modes, etc.). This feature is very naively implemented, and might not be able to exclude all possible LaTeX-commands out there. However, taking its outputs with a grain of salt makes it a useful tool (I personally used it for my Ph.D. thesis).

How to Extend/Experiment with it

There are certain parameters which can be set differently, depending on how the user wants it.

• repetition-error-word-block-size: This variable stores the word block size, in which the user does not want any repetitions. The default value is 100.
• repetition-error-min-occurrence: This variable stores how often a value needs to occur in a word-block at least to be detected as a repetition error. The default value is 2.
• min-not-ignore-letters: Some words like "and" will inevitably occur more than once in a word block, and we would like to ignore these. Hence, we introduce this variable to store the minimum length of the word to be considered for a repetition error. Its standard value is 4. This means, that words like and, it, ... will be ignored in the search for repetitions, but words like then, than, they will not.

Other than that, as mentioned above, there is a way to customize the script to ignore commands or other word blocks, given you are writing in a WYSIWYM language like LaTeX. If you want to customize for your language, please study the code and see how this was achieved for the LaTeX part. It is fairly modular, but it requires to study the code in a bit more detail.