Sublime Forum

Dev Build 2155

#1

Dev Build 2155 is out now, correcting an auto complete regression in 2154, and adding support for completions in HTML.

0 Likes

#2

Need to modify the \ Data \ Packages \ HTML \ html.sublime-snippet
ā€¦
Modify the code:

<! [CDATA [

$ 1 $ 0 ]]> html text.html ................................................................................................................................................................... can continue to modify this ................................................................................................................................................................... **Continue to update to the Sublime Text 2 Build 2156 ** ................................................................................................................................................................... **china:** ................................................................................................................................................................... Can be modified for simplified to: <! [CDATA [ $ 1 $ 0 ]]> html text.html ...................................................................................................................................................................... The next step debugging Sublime Text 2 support for all languages ā€‹ā€‹(such as GB2312) Such as: txt Open garbled.
0 Likes

#3

Thanks John!

Seems to me like there was a font regression this round. Iā€™m using OSX Lion, and havenā€™t previously made any font aliasing adjustments - all defaults. See the attached image. It feels ā€œlighterā€ than before. Thoughts?

0 Likes

#4

jonathan: There havenā€™t been any font rendering changes in the dev build, but going by that screenshot youā€™ve enabled greyscale anti-aliasing (i.e., as opposed to sub-pixel anti-aliasing). You can fix this via sublimetext.com/docs/2/revert.html

0 Likes

#5

Hey thanks for the follow up. I tried that without successā€¦still showing the way it was before. I hadnā€™t previously made any changes to the aliasing settings, it did this all on its own upon the last update.

Also, I tried to ā€œforceā€ the default sub-pixel aliasing (see attached image) and it made no difference as well. Iā€™m curious if anyone else is having this problem?

0 Likes

#6

It works fine for me


0 Likes

#7

There are two small regressions in this build (compared to the current beta):

  • My Erlang plugin defines snippets that start with non-alphanumeric characters.
    Expanding them leaves part of the trigger in the buffer. Example (^ is position where i press TAB)

    -com ^
    Results in:

    --compile([export_all]).

    While it should be

    -compile([export_all]).

    Snippet definition is here
    .
    If ā€œcompileā€ is in the buffer already, the ā€œ-comā€ snippet is ranked lower than word completion, triggering only
    after another press of TAB.

    I have another snippet in my user directory that triggers on %ā€“, using it leaves the whole trigger in the buffer and just inserts
    the snippet content after it.

    Those snippets (starting with non-aphanumeric character) are also not shown in the completion list.

  • Vintage command dd
    repeated with . several times deletes subsequently larger areas of text.
    Example: Using dd deletes 1 line, repeating the first time works as expected, repeating again works as expected,
    repeating the third time deletes 2 lines, a few more repeats and the rest of the buffer is gone.

0 Likes

#8

So update notes say HTML5 support was added, but Iā€™m getting invalid.illegal on

<!doctype html>

Can you please merge this to HTML.tmLanguage

--- C:\Dropbox\Workspace\Projects\HTML5lang\HTML.tmLanguage Tue Dec 27 16:36:49 2011
+++ C:\Dropbox\Workspace\Projects\HTML5lang\HTML5.tmLanguage Wed Dec 21 19:39:58 2011
@@ -15,10 +15,10 @@
        <string>ctp</string>
    </array>
    <key>firstLineMatch</key>
-   <string>&lt;!DOCTYPE|&lt;(?i:html)|&lt;\?(?i:php)</string>
+   <string>&lt;!DOCTYPE|doctype|&lt;(?i:html)|&lt;\?(?i:php)</string>
    <key>foldingStartMarker</key>
    <string>(?x)
-       (&lt;(?i:head|body|table|thead|tbody|tfoot|tr|div|select|fieldset|style|script|ul|ol|li|form|dl)\b.*?&gt;
+       (&lt;(?i:a|article|aside|audio|blockquote|body|canvas|datalist|details|div|dl|fieldset|figcaption|figure|footer|form|head|header|hgroup|li|mark|meter|nav|ol|output|p|progress|rp|rt|ruby|script|section|select|small|style|summary|table|tbody|tfoot|thead|time|tr|ul|video)\b.*?&gt;
        |&lt;!--(?!.*--\s*&gt;)
        |^&lt;!--\ \#tminclude\ (?&gt;.*?--&gt;)$
        |&lt;\?(?:php)?.*\b(if|for(each)?|while)\b.+:
@@ -27,7 +27,7 @@
        )</string>
    <key>foldingStopMarker</key>
    <string>(?x)
-       (&lt;/(?i:head|body|table|thead|tbody|tfoot|tr|div|select|fieldset|style|script|ul|ol|li|form|dl)&gt;
+       (&lt;/(?i:a|article|aside|audio|blockquote|body|canvas|datalist|details|div|dl|fieldset|figcaption|figure|footer|form|head|header|hgroup|li|mark|meter|nav|ol|output|p|progress|rp|rt|ruby|script|section|select|small|style|summary|table|tbody|tfoot|thead|time|tr|ul|video)&gt;
        |^(?!.*?&lt;!--).*?--\s*&gt;
        |^&lt;!--\ end\ tminclude\ --&gt;$
        |&lt;\?(?:php)?.*\bend(if|for(each)?|while)\b
@@ -180,7 +180,7 @@
            <array>
                <dict>
                    <key>begin</key>
-                   <string>(DOCTYPE)</string>
+                   <string>(DOCTYPE|doctype)</string>
                    <key>captures</key>
                    <dict>
                        <key>1</key>

html5lang.diff.zip (820 Bytes)

0 Likes

#9

Just a comment, I think the ā€œamountā€ argument of ā€œmoveā€ command is not working: Try

view.run_command('move', { "by": "characters", "forward": True, "amount": 100 })

You will see that it moves only one character forward. Another thing, how I can see the available arguments for each command?

0 Likes

#10

John, just an update on this. I believe this font aliasing issue Iā€™m experiencing has something to do with the screen resolution. I took my laptop to work and plugged it into my Apple cinema display and the problem went away (see attached screenshot).

Will test at home on my 28" Samsung monitor tonight.

0 Likes

#11

I also get this aliasing problem when connecting my MBP to an external monitor. Resolution on the external monitor is 1920x1080, and my laptop is using 1440x900.

Setting a lower resolution on the external monitor makes all apps blurry though, but with the full hd resolution only sublime text 2 is blurry.

0 Likes

#12

@quarnster, I did some more digging on this and it turns out it has something to do with how OSX renders fonts for certain LCD monitors. I was experiencing this with my MacBook Air when attached to a Samsung LCD monitor. However, there was no issue when just using the Macbook Air on itā€™s own or with my Apple Cinema Display.

Here is what I did to fix it. Run the following code in Terminal:

defaults -currentHost write -globalDomain AppleFontSmoothing -int 2

Then restart your system. Once OSX booted up again, the problem was gone. You can read more about this here:

http://slackrw.wordpress.com/2010/07/31/mac-os-x-10-6-snow-leopard-re-enable-lcd-font-smoothing-for-some-monitors-via-macosxhints-com/

0 Likes

Sublime isn't rendering text properly with dual monitors
#13

Ah, that fixed it. Thanks Jonathan! :smile:

0 Likes

#14

Excellent! My pleasure. :wink:

0 Likes