For example, .b._mod should produce <div class="b b_mod">

It would be great to update the Python code to support Python 3.0+. While Python 2 is still widely used, it just feels wrong writing new code in it.

I'll send a pull request if I get anywhere with upgrading it.

When editing UTF-8 files in Notepad++ or Komodo, UTF-8 characters (like Russian letters) in code produces unpredictable results

I would like it to be possible to set the inner text of an element via zen coding.
I can think of 2 good ways of doing this:

1. css like:
   div#res[content=some_text]
   which might have problems with spaces
2. new syntax:
   div#res{some text can go here}

I would try to implement it but it's hard for me to follow up the way ZC works

Hi, can I change something to get :
< div class="carrousel" >< /div >< !-- .carrousel -- >
From : div.carrousel ?

Thank you.

Abbreviation with text node like li{asda} expands to %liasda, there must be a space after tag and before text node in HAML

No errors reported in console. Input box pops up, but nothing happens afterwards.

Chrome version17.0.963.56

Cheers - L

Next version of Zen Coding should allow users to skip obvious tag names from expression. For example, ul.nav>.item should be the same as ul.nav>li.item, tr>.cell as tr>.td.cell and so on. It should be a stand-alone module that can be used in extensions.

Create preferences module that provides general storage for Zen Coding module preferences and convenient methods for listing and modification

I discovered this problem while working on TEA for Espresso with Zen Coding 0.6. Say I wanted to switch to using a different language:

from zencoding import stparser
from zencoding import zen_core as zen
user_settings = {'variables': {'lang': 'ru'}}
zen.update_settings(stparser.get_settings(user_settings))

If I then run an abbreviation that uses the ${lang} variable, I'll still get 'en' (the default value) output. This is because the replace_abbreviation has the vars parameter pre-filled, and modifying the zen_settings variable after the fact doesn't change it. To fix the problem, I modified zen_core.py as follows:

Replacing lines 234-240
def replace_variables(text, vars=None):
"""
Replace variables like ${var} in string
@param text: str
@return: str
"""
if vars is None:
vars = zen_settings['variables']
return re.sub(r'${([\w-]+)}', lambda m: m.group(1) in vars and vars[m.group(1)] or m.group(0), text)

This way, zen_settings['variables'] is always the most recent version, and the user doesn't need to edit zen_settings.py to change them (they can instead override them).

I think it's not good that .block>.-element is counted as .block>.__element by default — there could be people that would want to use this syntax, so the Zen Coding should be less strict on that, however it would nice to have this behavior in settings.

Also, right now .block--mode is expanded to <div class="block--mode"></div>, however there are people that use this as modifier (like Nicolas Gallagher mentioned in his article), so it would be nice if .block--mode was treated like .block.block--mode as .block_mode works now.

However, I think that all those things must be configurable in settings, so someone could say that .-- is an element or the .__ is a modifier.

With this abbreviation: ul>li*>span+b{$#}

Instead of using brackets and nesting, use ^ to ascend one level of the DOM tree.

html>head>title^body>h1

rather than
html>(head>title)+body>h1

fork is here: https://github.com/itaccess/zen-coding/commit/b2dd2a7d0a7b6d36907f39093e7f5fd0b1f61fc4

In properties like -webkit-border-radius the first dash is not respected

In e2, the balance tag (inward) command does exactly the same as the balance tag outward command. Not sure if this is a bug in e or a bug in Zen.

Sometimes (in Eclipse v0.8) a big chunk of code is wrapped with incorrect indentation

The match_pair action function defined in zen_actions.py contains a bug, referencing attributes on a tag with getitem syntax, eg tag['start']. The Tag class in html_matcher.py actually has 'start' as an attribute.

There's also a bug in the the format-css.py code, where the regex will deform the '{%::zen-caret::%}' when trying to put some whitespace between "prop: value"

ps. Thanks for ZC!

In this example: <li><a href="index-03.php"></a></li>

в плагине к TextMate нехватает zencoding внутри .tmbundle/Support/

I've created a test page here: http://david.buser.net/codemirror.html

CodeMirror seems to work OK, but I can't seem to get Zen-Coding commands to work. And I'm seeing corruption of the info window when I click the showInfo() link ...

Any ideas?

Thanks very much!

Komodo Edit plugin uses hardcoded 'xhtml' profile, should be based on current document state

When expanding something like ul>li.item-$*4>a.href-$, the a's $ should refer to parent's number

Example: m0 now outputs margin: 0px;, but margin: 0; is better

I love zencoding, love ST2, but they never seem to get past the first date. He buys her flowers, she says no, etc etc.

Please, help them get married and have lots of high-efficiency text-editing babies. Please.

некорректно раскрывает аббревиатуру span>a.__child:
<span><a href="" class="container container_child"></a></span>

.__child раскрывает правильно:
<a href="" class="container__child"></a>

The zencoding package is placed in one of my plugin folders. Importing zen_actions and zen_core works fine. The problem arises when either of this packages try to import zencoding.filters. So I spent the better part of yesterday trying to figure out what's going in. I latter discover some magic going on in filters.init.py. Don't fully understand it or why it's needed, but it's cause import problems. In fact, I can't import zencoding.filters. Strange and unusual since I don't have this problem with any other python package. I suspect the magic going on in filters.init.py is going to break my python packages.

Also in zen_actions.py, why are you import zencoding to get zen_core when both zen_core and zen_actions are in the same package? You are doing:

from zencoding import zen_core as zen_coding
from zencoding import html_matcher

when you could just do

import zen_core as zen_coding
import html_matcher

Anyway this import problems are cause problem in my application. Do I need to do a global import somewhere to avoid these problems? Am I missing something?

Performing this abbreviation ol>li*>span{$#}+{ }+a[title=$#]{$#}|t on the following:

1. list item one
2. list item two
3. list item three

produces this output:

<ol>
<li>
<span>list item one</span>
<a href="" title="1. list item one">list item one</a>
</li>
<li>
<span>list item two</span>
<a href="" title="2. list item two">list item two</a>
</li>
<li>
<span>list item three</span>
<a href="" title="3. list item three">list item three</a>
</li>
</ol>

Note that the |t (trim) filter did not remove the number marker inside the custom title attribute [].

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

Looks like Komodo can manipulate files, add support for zen_file interface:
http://docs.activestate.com/komodo/4.4/macroapi.html

In Eclipse plugin, when user adds custom snippet, it's not inherited by default inheritance chain

Hi,

Here is my suggestion:

Read abbreviations from css file/css rules.
With this feature users can define one css lib and reuse code.

Like this:

#something{....}
    #something > ul{....}
        #something > ul > li {....}
            #something > ul > li > a{....}

Then i can use:

something

And expand to:

<div id="something">
    <ul>
        <li><a href=""></a></li>
    </ul>
</div>

I know that need more specs about how this works, and don't know if is possible (without years codding).

Upgrade all plugin's replaceValue() method with no_indent argument

Create plugin for CodeMirror: https://github.com/marijnh/CodeMirror/

When caret is inside HTML attribute value, the “Select Next Item” actions matches full attribute, but should match value only

The textmate expansions are throwing an error:

Traceback (most recent call last):
File "/Users/numan/Library/Application Support/TextMate/Pristine Copy/Bundles/Zen Coding.tmbundle/Support/expand_abbreviation.py", line 7, in
import zencoding
ImportError: No module named zencoding

What am I doing wrong?

“Reflect CSS Value” action should be able to handle custom “reflect” delegates, available for extension. Particularly, gradient generator should provide its own delegate.

When you expand el.class1.class2 in HAML filter, it becomes %el.class1 .class2 instead of %el.class1.class2.

When there are unclosed void elements (that is, single tags without a closing slash) inside a tag, the tag balancer doesn't work correctly. If I add the closing slashes, then it finds the closing tag correctly. See the following screencast for a demo:

http://dl.dropbox.com/u/256478/screencasts/Zen_Coding_tag_balancer.swf

Like ths: test $sd $0 $1

title says it all

Hi,

I have an issue when trying to expand my zencode, it writes down in textmate :

Traceback (most recent call last):
Traceback (most recent call last):
File "/Users/franco/Library/Application Support/TextMate/Bundles/Zen Coding.tmbundle/Support/expand_abbreviation.py", line 7, in
import zencoding
ImportError: No module named zencoding

Anybody knows what is wrong ?

It would be great if plugins worked with .scss files in the same way like they work with .css

User reports that textarea plugin has an error (IE?) on the following line:
https://github.com/sergeche/zen-coding/blob/master/plugins/textarea/zen_editor.js#L333

For example, when -prop is expanded, the generated output should have linked tabstops across all values

There is no -ms-box-shadow and no -ms-border-radius so it does not make sense to include those in the snippets for "bsha" and "brad". I would also suggest to use "bxsh+" and "brds+" to make it more consistent with other existing abbreviations and conventions.

The XSL shortcuts need to be extended to include ALL xsl elements.

Here's a few of suggestions that I thought would be a great and helpful addition to the xslt shortcut library...

tex --> xsl:text/xsl:text

com --> xsl:comment/xsl:comment

attr-set --> <xsl:attribute-set name="" use-attribute-sets="">/xsl:attribute-set

imp --> <xsl:import href=""/>

inc --> <xsl:include href=""/>

key --> <xsl:key name="" match="" use=""/>

It'd be great if zen-coding could expand not only HTML etc., but JSON as well.

For example, from something>another{content}+lalala+lololo{blah} you could get

{
    "something":
    [
        {
          "another": "content"
        },
        {
          "lalala": |
        },
        {
            "lololo": "blah"
        }
    ]
}

or something similar.

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

Module that allows developers to easily editor string CSS rules

CSSEX fails to parse data inside braces, especially something like url(data:...)

When wrapping and expanding abbreviations, variables inside text nodes should be expanded

I've set my Aptana to look for updates on every startup with the url http://zen-coding.ru/eclipse/updates/. About 3 weeks ago I received an update through the http://zen-coding.ru/eclipse/updates/ site. After the update it's not working exactly the same anymore. I've only noticed one thing so far. I've setup some custom snippets.

If I type in blank and try to expand it should expand to target="_blank".
This works except for if I try this <a href=""blank|> where my cursor is the pipe. I try to expand and it does nothing.

NOTE: PIPE IS MY CURSOR LOCATION
doesn't work but used to
<a href=""blank|>
<a href="" blank| >
<td blank|>
does work
<a blank|href="">

This worked fine before the update. I can type it anywhere else and it works as long as its not within an html tag.

Below are the details from Aptana from the installation details screen
CURRENT VERSION
Zen Coding for Eclipse 0.8.0.201201232221 ru.zencoding.eclipse.feature.group
VERSION BEFORE UPDATE tried to revert but couldn't because 0.7.0 wasn't available
Zen Coding for Eclipse 0.7.0.201103132329