How to ignore French accents during Search

[OldGrantonian]

My question relates to the use of Ctrl+F to find a string in a web page.

In Chrome, I can search for a French word such as "près" simply by typing "pres". Chrome will display the results for both "près" and "pres". I would like FF to do the same thing.

But the same search in Firefox only finds "pres", but not "près"

How can I force FF to ignore accents when searching?

Maybe something in about:config?

Thanks

[elist74]

They talk about this https://support.mozilla.org/en-US/questions/1192995

[therube]

Can't say I've ever thought about that before, but surprised there is no ignore diacritics option.

WONTFIX: Bug 202251 Find/FindAsYouType will not find text if entered with diacritics ("nikud") in Hebrew, or accented characters in other languages

(At least there is some somewhat recent conversation in that bug.)

[scott967]

Just a user here, but I think this is a "deep" issue, not easy to solve.

First, you should consider how a text has been rendered on the screen in FF. Are strings that render visually identically really the same (or is it a limitation in the renderer/font in use?)

Second, what does it mean for two text strings to be "the same" for purposes of search? I don't know that any two users would have the same opinion. (I personally have a problem in that I would like to enter a Korean jamo and find all Hangul composed syllables containing the jamo in the primary position).

Unicode standard attempts to assist in this by providing means to build string compare algorithms to implement the desired functionality. There are "normalization" options that allow conversion (normalization) of strings. Then there is the idea of primary, secondary, and tertiary differences in codepoints that allow for a granularity in string compares (I think for Latin "accented" characters, the accent is considered secondary). This allows for all letters "a" (including upper and lower case and with / without accents) to be "binned" during search. The strict ordering is determined by the user's locale collation (sort) selection. Some locales have dictionary and phonebook collation differences. Here's a graphic representation: http://unicode.org/charts/collation/

Recently I have been looking at the problem of emoji and emoticon text rendering. I suppose in general emoji should be searchable, right? If so, how to deal with alternate presentation options (text or emoji based on unicode variation selector)?

Of course USians, generally content with code page 437, don't always see the potential issues.

scott s.
.

[OldGrantonian]

.
Thanks for all the help.

The post by @therube gave me a good search term - "ignore diacritics". Someone should write a "search term generator" that works using voice info, such as Cortana or Siri. I might do that while I'm waiting for the Manafort verdict.

If I type "Firefox ignore diacritics", I get dozens of entries, which proves that this is a long-standing complaint with FF.

Here's a 7-year-old bug report:

https://bugzilla.mozilla.org/show_bug.cgi?id=652434

That report specifically says "Chrome does it" - as I mentioned in my OP.

Thanks.

[elist74]

That would be great if firefox having that ability / feature (my talking language is french).