revise word replace

[fuwako]

Description

This is a collection of changes to the Word Replace system.

Case Sensitivity

Word Replace will become dependent on capitalization.

For example, Apple and apple will be treated as two different entries.

This is semantically important.

For example, the user may only want to transform the capitalization of a word.

• The current system is case-insensitive and forcibly writes replacement keys in lowercase, such as pizza➔pizza. This implicitly means "any capitalization of pizza, Pizza, pIzza, piZza, and etc. that will be replaced by a user-specified replacement value." This example shows how the change in capitalization is not visually communicated.
• With case-sensitivity, we can have a visually meaningful display of the change (e.g., Pizza➔pizza). This visually communicates what is being changed.

This enables users to specifically target common nouns and proper nouns.

However, case-sensitivity is more explicit, so it may be more difficult to use. When using speech-to-text, the user will have to declare their replacement keys according to the output of the Web Speech API.

Sanitization

When text is being processed for replacement, regex metacharacters in replacement keys will be automatically escaped to prevent the accidental injection of regular expressions.

The user's entries will not be modified and will remain as the user has declared them.

This fixes a crash that required a refresh of the running instance.

Additionally, users can now replace words/phrases that contain special characters (e.g., words with many asterisks).

Replacement Prioritization

Longer replacement keys will have higher priority than shorter replacement keys.

For example, we may have 2 entries. Hello world➔apple Hello➔banana

Hello world is the longer key; therefore, it has higher priority. The transcript Hello world will become apple instead of banana world.

(Other)

This has only been tested for the English language. I haven't fully determined whether it's functional and stable with other languages.

What is the purpose of this pull request?

• [X] Bug fix
• [X] New Feature
• [ ] Documentation update
• [ ] Other

[naeruru]

I like the pr, but I think that having an option for case insensitivity is still important. Due to the nature of speech-to-text providers being all inherently different, capitalization of words can vary based on sentence structure and proper nouns and a user who is not aware of this could have a more confusing time. Capitalization is also different between browsers. For example Edge will capitalize and punctuate sentences. this means someone might always have to include two different entries for words they didn't care for the case to be, because there is a non zero chance that the word can be at the beginning of the sentence.

[fuwako]

I've implemented "Match whole word only" and "Match case" as options, instead. If you don't want them, I would be okay with stripping down the code and removing them. This pull request unexpectedly became complicated.

Before	After

[naeruru]

awesome, give me some time to look over it^^