Ubuntu QA:
BlogBrainstormPackage status
Log in
Ubuntu QA
The Ubuntu community has contributed 13963 ideas, 66846 comments, 1291785 votes

Idea #10694: Firefox should allow regular expressions natively



up
-8
down
Written by bert.ubuntu the 4 Jul 08 at 06:08. Category: Others.
Related to: Firefox. Status: New
Description
There are lots of situations in which regular expressions are useful to search some text more than "word*".

Some advantages:
- Search pictures *.jpg
- Search audio *.[omw][apg][3gv]
- Find your name in a list (John Nobody = Jo.*dy)
- Multiselection on incremental search would be possible!
- If someone doesn't know R.E it would work the same way as before. So it makes nobody's life harder.

Furthermore, as Firefox updates, lots of addons simply stop working. So if it have this feature natively (which is simple do add, and not heavy to process) it won't be a problem as well.

Attachments
No attachments.


Duplicates


Comments
Auzy wrote on the 4 Jul 08 at 06:28
Couldn't this be done via an addon?

Auzy wrote on the 4 Jul 08 at 07:36
Ahh ok, you mentioned that. Not sure how useful it would be though

Ssdg wrote on the 4 Jul 08 at 08:07
Auzy> natively? it won't be usefull for the Firefox (especially for Tristant Nitot's grandmother ;) (Mr Nitot is mozilla europe's CEO and his moto is: If my grandmother can't use it then we failed, and I agree)).

I think it should be an addon too. people who know regexps syntax should know how to install an addon.

Sidney wrote on the 4 Jul 08 at 12:42
Sorry, but what you describe aren't regular expressions. Those are more complex.

And while I quite like regexp searches, I can't find any advantage for Firefox. Use case?

Auzy wrote on the 4 Jul 08 at 12:51
well, either way, it couldn't be a default anyway, but possibly an obscure option in about:config.

Because "." in a regexp expression doesn't mean full stop, it means any character

bert.ubuntu wrote on the 4 Jul 08 at 17:50
The regular expressions are TRANSPARENT to the users who don't know how to use it !

---------
@Nitot: Please remove the addons' feature from firefox. I am pretty sure your grandma is not able to search and install one by herself.

Also, I think she don't need the Creative Commons Database, so please remove that too.

Btw, does she use the simple find ? (I know what I am talking about, I know people above 70 using firefox, they even use youtube, but don't try to bother than with "complicate" things like tabs).

When firefox crashes and it doesn't want to run again because "another instance is running" (but you don't see anything because it's tricked by the lock file) how does your grandma solve it ?

Oh, Nitot's grandma can solve ? How about the rest of the world ?
------------------------

Again:
1- The native support to RE would avoid problems in the update, keeping this useful feature always working
2- If you don't know RE, just keep typing as you did before, nothing will change for you. Voting positive will help a lot.
3- Before the existence of tabs most of the people didn't miss it. What about now ?

I hope you consider my arguments. It's not a problem at all having that ! It will just help more advanced users with no harm!

Mr.elderman wrote on the 4 Jul 08 at 18:01
I get your point.5k more in the code would do no harm in the face of the extra benefits.

Transparency is the key that makes it possible to keep it native. I just hate when something stops working after an update!

+1

zooounds wrote on the 4 Jul 08 at 20:34
Write a plugin, don't force it on people

Auzy wrote on the 4 Jul 08 at 23:26
Actually bert.. That's not true..

If I do a search for ...

I'm not searching for DOT, DOT DOT. A regex would treat that as searching for: Any character, ANY CHARACTER, ANY CHARACTER. (translation: As long as the page has more then 3 characters, just about the entire page matches).

To look up ... using a regex, is actually \.\.\.

People don't expect that behavior. I guess we could always require it to be placed in // syntax though. But then realtime searching will be inconsistant in some operations.

Like I said, if you include it, definately not by default. Despite what you claim Bert, something will break. Unless my logic is wrong here (feel free to correct me if I'm wrong)

Sidney wrote on the 5 Jul 08 at 11:11
Auzy, your logic is correct, but bert.ubuntu still talks about wildcard searches with "*" (like his example "*.mp3").

But even if this won't hurt, I can't see any advantage? I can already search for ".mp3" and find any link. The wildcard doesn't help. Again: Use case?
Because I'm not for including any feature without an idea why it could be good.

Auzy wrote on the 5 Jul 08 at 14:50
Well, then shouldn't this idea be marked invalid anyway? Because everything except the examples in the idea suggests he is talking about actual Perl-like regex's.

Mr.elderman wrote on the 10 Jul 08 at 16:28
@Sidney: Don't you see applications for that ? So that is your reason to not to have it ? Because *you* cannot use it ? What a shame ... but don't take it wrong. I am just saying you should think about the rest of the world sometimes.

@Auzy: You are right. It makes the task act non transparently anymore. So it cannot be included without braking anything. Off course we could make exceptions, but noone would be in the mood to predict that (even me, proposing that).

Sidney wrote on the 11 Jul 08 at 12:12
Are you now attacking me because I can't see any real use for this idea? That's quite ridiculous as I asked for use cases in both of my comments here. And still, the only thing you put forward is not a use case but a personal attack because I can't find one.

I'm not against ideas that I don't need, but I'm against ideas nobody needs. And as long as nobody gives a decent use case, I don't see how anybody would need that. So back your opinion up instead of attacking me for questioning you.


Post your comment