Sublime Text 3 Beta

[xeno.by]

Cmd+Shift+R doesn't display previews, but it looks like it can be taught to do so, analogously to F12. Here's a relevant excerpt from the implementation of F12:

Code: Select all

    def highlight_entry(self, locations, idx):
        fname, display_fname, rowcol = locations[idx]
        row, col = rowcol

        self.window.open_file(fname + ":" + str(row) + ":" + str(col),
            sublime.TRANSIENT | sublime.ENCODED_POSITION)

    self.window.show_quick_panel(
        [self.format_location(l) for l in locations],
        lambda x: self.select_entry(locations, x, v, orig_sel),
        on_highlight = lambda x: self.highlight_entry(locations, x))


[lunixbochs]

As far as I know cmd+shift+R doesn't show previews because you can have multiple definitions of a symbol. If you press enter on a symbol, it gives you a list of their definitions and has previews there.

[atomi]

What would be great is if we could have access to the symbols index for plugins.

[skuroda]

What would be great is if we could have access to the symbols index for plugins.

Something like

Code: Select all

window.lookup_symbol_in_index("Symbol")

[atomi]

What would be great is if we could have access to the symbols index for plugins.

Something like

Code: Select all

window.lookup_symbol_in_index("Symbol")

Hahah Yup I see it now in the Default package symbol.py.

[Anomareh]

@Anomareh

You really like to contradict yourself, don'tcha?

First you argue that Python 3 is too big a change since it will warrant people rewriting plugins (which will take some time but may prominent plugin authors have already given their support and mentioned that they are working on porting over to Python 3). Then you argue that there is not enough changes (in terms of features) but I would argue that the switch to Python 3 is pretty major.

What? Talk about taking something completely out of context just to make an argument. Never did I say there were not enough changes in Python 3 to consider upgrading. I said that the differences between Python 2 and 3 aren't the main reasons the upgrade is a problem.

Well, yeah. 3 > 2 but more importantly, 2 has no where to go. Sure, their might still be some bug patches and stuff in development but other than that: it's dead.

Swapping to Python 3 is not moving forward. It's running across the canyon before the bridge is built.

Just like no one is forcing you to use ST3. Think of it like the same situation between Python 2 and 3. No one is forcing you to upgrade, just don't expect any new features.

Except everyone is as ST2 is EOL and can't even be bought anymore (and no, having the ability to buy ST3 and get a copy of ST2 for free is not having the ability to buy ST2)? ST1 didn't see a single update (bug fix or otherwise) once the ST2 public alpha became available. I don't see the situation between ST2 and ST3 being any different. Looking beyond how insane it is to compare a programming language to a text editor, Python 2 is still being maintained and it's still available.

I suspect most plugin developers will port their projects to ST3 and leave the ST2 versions alone. Also, I suspect that ST3 will eventually have a free trial (just a guess) after more plugins are ported and features are added.

So you can see into the future? What reasons do plugin developers have to update their plugins for ST3? With a split userbase they'll just be choosing one userbase over another. Largely because, as I stated, there aren't enough improvements in ST3 to justify upgrading. Also, I like how you believe ST3 is going to see a bunch of new features added at some point. The majority of the new features of ST2 were available in the public alpha. Only a couple were added afterwards. I don't see any reason for ST3 to be any different.

If he can call it whatever he wants, why do you care if he calls it ST3 instead of ST2.1? Besides, you don't have to pay while it's in beta (and it's probably going to be in beta for a while considering past experience). So why not argue about money when the time comes instead of ranting about the future?

Besides you completely missing the point I was making, you do realize ST3 is the only thing available for purchase right now right?

Umm...? ST3 > ST2(lastest) > ST2(Alpha). That's the only point I was making. And using your logic, ST3 will only get faster... which doesn't seem like a problem to me.

Again, you somehow completely managed to miss the point I was making. Speed isn't a viable reason in the case of ST to increase the major version. As he stated himself, speed has always been a concern and has always seen improvements. ST being faster is not something new to ST. ST's speed has been substantially improved several times throughout the development of ST2.

[C0D312]

@Anomareh well, this conversation was fun at first but now I see you're not even going to read what I'm saying so there's no point...(plus I'd prefer if we stopped mucking up this thread for people with legitimate stuff to post)

I gotta go to bed anyway. I've got computer science class in the morning and we're learning about methods! yay!

[Anomareh]

@Anomareh well, this conversation was fun at first but now I see you're not even going to read what I'm saying so there's no point...(plus I'd prefer if we stopped mucking up this thread for people with legitimate stuff to post)

Funny. First off, you started the discussion by responding to my thoughts on the ST3 beta. Second, you took nearly every single thing I said out of context. Third, I even obliged and directly responded to everything you've said. Also, I don't see how this is off-topic in a thread designated for discussion about the ST3 beta.

[castles_made_of_sand]

as I stated, there aren't enough improvements in ST3 to justify upgrading

Port some plugins before making a final judgement.

I started porting efforts by writing a plugin that opened errors in ST2 where I did all my editing.

I ported an old lxml plugin that parses html buffers in a background thread, making use of the now threadsafe scoping api.

It's constantly building a DOM `as you type`, without ever slowing down. You couldn't really do that in ST2.

There's subtle changes like all plugin modules having absolute path `__file__` properties so you can write exception handlers that navigate to source of errors and navigate the backtrace.

The speed improvements are very much a consequence of changes to the plugin system and plugin loading, breaking changes that couldn't really be `backported` to ST2.

Having said all that, I'm somewhat disappointed there's not more radical changes too, with all this breakage/porting pushing ETA on ST4 well into the future.

[Anomareh]

Subtle isn't really the word you want to see describing a new major release when a price increase comes attached to it :/

"Here's this thing that for the majority of end-users is the same as the last thing except this thing is more expensive and it's going to break nearly everything for a while. Oh, and it's faster. Also, you can't buy that last thing anymore and it's no longer supported."