Sublime Text 2 highlighting : how to differenciate variable declaration and variable use? - sublimetext

Is there a way to differenciate variable declaration and variable use by using two different colors with Sublime Text 2 syntax highlighting ?
Here is an example of what I'm trying to do : (screenshot from jsFiddle)
I didn't find anything helpful in any of the Sublime Text 2 color schemes xml files and on the web either. Is what I'm trying to do even possible ?

This is only possible if the syntax definition scopes the variable declarations differently to the uses. The JavaScript syntax that ships with ST doesn't do so.
Therefore, unless you alter the syntax definition, there is no way to change the color scheme to achieve what you want.
If you upgrade to ST3, you could update the syntax so that after it scopes the var keyword, it could push into another context where it would expect an identifier, and scope that uniquely. See the sublime-syntax documentation for details.

Related

How to understand patterns defined in TextMate?

I am learning TextMate in order to define customized keyword for highlighting in a Orion editor(A browser based editor who is using TextMate grammar). I read this link http://manual.macromates.com/en/language_grammars.html#language_rules. And I don't understand how to define a pattern to match a keyword.
Below is an example I tried to define database as a keyword. But it doesn't get recognized by the editor.
{ name = 'keyword.control.db';
match = '\b(database)\b';
}
what is the correct way to do this?

use of roles in asciidoctor macros?

I'm trying to create a simple macro to render a text item in red in asciidoctor.
The following does not work:
:redtext: [red]#some important text in red that occurs a lot#
{redtext}
or for an even simpler example:
:redcross: [red]#✘#
I am not clear on the rules for what can and cannot be substituted by a macro. The asciidoctor manual has a blank space for macros at present (http://asciidoctor.org/docs/user-manual/#macros). The asciidoc manual (http://www.methods.co.nz/asciidoc/chunked/ch21.html) isn't that clear either but may not apply to asciidoctor anyway.
A related unanswered question is Resuable markup fragments with Asciidoctor.
A related question to that suggests using includes which is overkill for this.
What are the limitations of macros?
What you have defined is not a macro, it's an attribute. (When you use it, it's called an attribute reference).
You can perform substitution eagerly in the definition of an attribute using the inline pass macro. In the target position, it accepts a comma-separated list of substitution names (or substitution letters).
In your case, you can write:
:redtext: pass:q[[red]#some important text in red that occurs a lot#]
The relevant part is:
pass:q[...]
See substitutions in an attribute entry for details.
I think includes work well enough. We have a single glossary.asciidoc file to contain all the re-usable snippets, for example:
tag::redtext[]
[red]#some important text in red that occurs a lot#
end::redtext[]
In index.asciidoc you can add a little helper:
:g: glossary.asciidoc
Then wherever you need this snippet:
include::{g}[tag=redtext]

What is the option that makes Sublime Text add whitespace at the left of wrapped lines?

I want Sublime to treat my text files like they are source code files and show whitespace at the left of wrapped lines, like so:
beginning of long line blah blah blah, now it wraps
and it keeps going after an automatic indent
I tried opening the console and setting the option manually:
view.settings().set('indent_subsequent_lines', True)
but nothing changes, any ideas?
Well, that was fun. There's no easily accessible setting for this. But as you indicate, ST decides whether to add an extra indent when soft wrapping, from whether the syntax is considered code-like or plain text-like.
Being one or the other is up to the package defining the syntax to specify. So lacking a global setting from ST, you need to change your text packages. As an example, let's take Text. That contains Plain text.tmLanguage. In that you change
<key>scopeName</key>
<string>text.plain</string>
to
<key>scopeName</key>
<string>source.plain</string>
I'm unsure of whether there'll be ill effects from not keeping .plain.
One easy way to do this is to get the PackageResourceViewerpackage.
After install, do:
cmdshiftp
Type: Open Resource
return
Type: Text
return
Type: Plain text.tmLanguage
Make your edit and save.
PackageResourceViewer will save the modified Text package to your Packages directory. And sublime will display files considered to have Plain text-syntax, like they are code.
The caveat is that you need to do this for every text-syntax you want to be considered code.
As #AdamAL said, it's dependent on both the 'indent_subsequent_lines' setting as well as whether it's considered a "source" or a "text" language ... by default, "markup" languages (such as HTML, CSS, etc.) and plain text, etc., are considered "text", and programming languages such as C++, Java, PHP, etc., are considered "source".
"Source" languages will indent subsequent soft-wrapped lines (if the indent_subsequent_lines is true), whereas "text" languages will only indent up to the same level as the current line.
For each one you want to change, you'll need to edit the settings of the given language. #AdamAL's answer provides a great way to do this using the PackageResourceViewer package:
After install, do:
[Ctrl/Cmd]+[shift]+p, "Open Resource"
Find the name of the language you want to change, and find either the .sublime-syntax file or the .tmLanguage file. .sublime-syntax is supported from build 3084 of ST 3 and appears that it may trump values in the .tmLanguage file in supported versions, if present (when editing the definition of TaskPaper files provided by the "PlainTasks" package, my change didn't take when just editing the PlainTasks.tmLanguage, I had to edit the PlainTasks.sublime-syntax before it took).
In .sublime-syntax (which are YAML files) look for the first scope: line, where the main scope name of the language is identified (there will be lots of other scope: entries further down under contexts:).
In .tmLanguage (which are XML .plist files) look for the <string> following the <key>scopeName</key>.
Sublime Text Syntax Definition Documentation Reference:
scopeName
Name of the topmost scope for this syntax definition. Either source.<lang> or text.<lang>. Use source for programming languages and text for markup and everything else.
The <lang> (without brackets) is just an identifier string for the given syntax/language definition.
I noted that it seems that (in ST 3, anyway), no restart of Sublime Text is needed to get the changes to apply, if the edit is in the right place.
And also note that there may be other effects of changing this in more complex packages -- For example, in PlainTasks, the additional keybindings that it defines depended on it looking for a context that included text.todo, which I changed to source.todo in several places. So in order for the keybindings to work properly again, I also had to update my .sublime-keymap for that package. (This could also be because I changed it in a place besides the .sublime-syntax that I didn't need to. Just sayin' -- YMMV.)

Syntax Highlighting in Sublime Text 2

So I have been trying to figure out how to add syntax highlighting for the name of typedef's in c++ files, in sublime text.
For example, if I have typedef long long integer; I want integer to be highlighted (preferably the same color as the other types: int, bool, etc.). I went looked at the C.tmLanuage file, and tried to add the following regex code ^typedef.*?\s(\w+)\s*; to storage.type.c (line 49), but it didn't work. If I add the word string, it will highlight all instances of the word string. I tried going in the C++.tmLanguage file, and adding the regex code to storage.type.c++, but it still did not work.
Does anybody know how to get typedef's highlighted in sublime text?
Also, is there a way to get syntax highlighting for class name? Let's say I declare a string or vector, I would like either string or vector to be highlighted.
That regex would work (I believe) if you had something along the lines of typedef foo; To get the behavior you want, you will have to create a slightly more complex pattern entry in the tmLanguage file. As the language file is based on TextMates, you will want to have this as a reference (http://manual.macromates.com/en/language_grammars#language_grammars). I would also recommend using PlistJsonConverter (working in JSON is easier for me than working in XML). You will probably need to define begin and end patterns (begin will probably be typedef end will probably be ;). You can then apply whatever patterns you want to that group.
As for the class name highlighting, I would look to see what, if any scopes are being applied. If none are, you will have to come up with a regex to apply the scope to those. You can then add a color entry, or use a defined one from the color scheme.
Edit:
Actually they don't appear to be JSON. I see () rather than []. JSON is pretty simple to understand. You can look for something more in depth, but wikipedia is a good place to start. What you would probably be interested in are the things under the "Rule Keys" section. I did some searching (because I knew there were some better examples out there), and came across http://docs.sublimetext.info/en/latest/extensibility/syntaxdefs.html . It goes over syntax definitions from scratch, but the most relevant section is probably http://docs.sublimetext.info/en/latest/extensibility/syntaxdefs.html#analyzing-patterns. I don't have a regex to find class names, so you would have to come up with one yourself. If you haven't already though, you may want to search around to see if someone else has implemented a language file in a way that works for you.
You will want to start with the built in tmLanguage file and convert that from a Plist to json. You can then edit that file and move it back.

Is it possible to do custom display of members in Flex Builder variables view?

Is it possible to display something other than the definition name in the variables view in Flex Builder for special types?
Basically, what I'd like to do is display the value of a complex member type instead of just showing it's qualified type name. Take a look at the following image:
(source: brightworks.com)
In the second column you can see the value of the member in the first column. What I'd like to to is to somehow make Flex Builder write an arbitrary string instead of foo.bar.MyClass (#0123456), is this possible?
There isn't a configuration setting in FlexBuilder (that I know of) so the only way I can think of to do it would be to replace whatever Java classes in Flex Builder plugin that deal with the variables pane and the debug output with custom ones that you've written to do what you are looking for. Since it's closed source it would have to be a clean room implementation, which would be tough. I've seen some examples of people replacing the java classes in Flex Builder to fix bugs (example) so it's not impossible to do if you have the time and knowledge.

Resources