Small suggestion for the search regex documentation

[calmeilles]

I was trying to search with the regex "[A-Z][A-Z][A-Z]" — looking for 3 consecutive capital letters but the search operated case-insensitive so I was getting results ABC, AbC, abc etc which came as a considerable surprise.

I believe that this is because my install has LOCALE=en_GB and the collate order for that is case insensitive. I may be wrong, but it's my best guess and actually doesn't matter.*

The Regex documentation page includes how to make a case sensitive [which is what we'd normally expect] ignore case with the "(?i)" syntax but not how to do the reverse. It actually turned out quite a chore tracking down what was required and a note in the documentation I feel would be useful.

What I ended up with was

(?-i:[A-Z]{3})

It's the possibility of -i that's missing and was quite obscure even in the Python docs.

(*I am curious if this is true or something else caused it.
If my guess is right then also mentioning that LOCALE can serious affect your regexes would also be useful.)

[gbm]

Originally Posted by calmeilles

I was trying to search with the regex "[A-Z][A-Z][A-Z]" — looking for 3 consecutive capital letters but the search operated case-insensitive so I was getting results ABC, AbC, abc etc which came as a considerable surprise.

I believe that this is because my install has LOCALE=en_GB and the collate order for that is case insensitive. I may be wrong, but it's my best guess and actually doesn't matter.*

The Regex documentation page includes how to make a case sensitive [which is what we'd normally expect] ignore case with the "(?i)" syntax but not how to do the reverse. It actually turned out quite a chore tracking down what was required and a note in the documentation I feel would be useful.

What I ended up with was

(?-i:[A-Z]{3})

It's the possibility of -i that's missing and was quite obscure even in the Python docs.

(*I am curious if this is true or something else caused it.
If my guess is right then also mentioning that LOCALE can serious affect your regexes would also be useful.)

If you are using the calibre ebook editor then all you needed to do was check the box for case sensitive.

bernie

[calmeilles]

Originally Posted by gbm

If you are using the calibre ebook editor then all you needed to do was check the box for case sensitive.

Er... I'm downright dumbfounded that that check box should affect a REGEX search. It's utility for 'Normal' searches is obvious, but it never even crossed my mind that it would affect a REGEX... that's just... silly!

Even so, I still think noting (?-i) where (?i) is mentioned in the doc would be a nice thing to do.

[jackie_w]

Originally Posted by calmeilles

that's just... silly!

... or perhaps it's designed to be more user-friendly for those with little or no experience of regex flags.

[calmeilles]

Originally Posted by jackie_w

... or perhaps it's designed to be more user-friendly for those with little or no experience of regex flags.

Okay, I've given that some thought.

I don't think that anyone who had encountered regexes before would expect them to ignore case; it's one of the first things you learn about them — they're precise (often in frustrating and unexpected ways, but that's a different subject).

Anyone not familiar with them, trying them out with the help of the documentation would read right at the beginning

You’ll notice, though, that this only matches the exact string "Hello, World!", not e.g. "Hello, wOrld!" or "hello, world!" or any other such variation.

And further down:

Knew you’d ask. Some useful sets are [0-9] matching a single number, [a-z] matching a single lowercase letter, [A-Z] matching a single uppercase letter, [a-zA-Z] matching a single letter and [a-zA-Z0-9] matching a single letter or number.

This is written to show that case matters. Otherwise in the above [a-z] would be sufficient to match a-z and A-Z.

Further on still we come to the part that shows how to flag ignore case

In the beginning, you said there was a way to make a regular expression case insensitive?
Yes, I did, thanks for paying attention and reminding me. You can tell calibre how you want certain things handled by using something called flags. You include flags in your expression by using the special construct (?flags go here) where, obviously, you’d replace “flags go here” with the specific flags you want. For ignoring case, the flag is i, thus you include (?i) in your expression. Thus, (?i)test would match “Test”, “tEst”, “TEst” and any case variation you could think of.

Again, written from the perspective that case sensitivity is the norm.

So I still feel that adding the reverse syntax would be nice.

Mentioning the check-box and its effect at this point could also do users — new and experienced — a favour. The check-box only gets referenced once, right at the end under "Bulk editing metadata" which I skipped as I wasn't bulk editing metadata.

It's not as if I were suggesting a radical change in the application itself; just that a few words be added to documentation.

[kovidgoyal]

In the several years that this feature has existed, yours is the first complaint I've heard about that checkbox. SO I dont think it is quite an un-intuitive as you suggest. However, adding more documentation is always good.

So you are welcome to suggest improvements to the documentation, it is maintained s a simple plain text file, here:

https://github.com/kovidgoyal/calibr...ual/regexp.rst

[Phssthpok]

Originally Posted by calmeilles

Er... I'm downright dumbfounded that that check box should affect a REGEX search. It's utility for 'Normal' searches is obvious, but it never even crossed my mind that it would affect a REGEX... that's just... silly!

It's the same as the "/.../i" option in most regex engines. Or grep's -i option. It saves having to write "[Ff][Ii][Nn][Dd] [Tt][Hh][Ii][Ss]" to find "find this" in a case-insensitive way.