Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
branch: master
Commits on May 4, 2013
  1. Merge pull request #18 from inkarkat/enh-multiline-tags

    authored
    Conflicts:
    	ftplugin/html.vim
Commits on Apr 27, 2013
  1. @inkarkat
  2. @inkarkat

    ENH: Support tags that span multiple lines.

    inkarkat authored
    Some code styles spread tags that have many attributes over multiple lines, e.g.
    <foo bar="1"
         baz="2">
    
    This patch extends the highlighting to those tags. In s:GetCurrentCursorTag(), when the current line does not contain a full tag, as a fallback the start of a tag is accepted. (It does not verify that the tag is actually properly closed in following lines.)
    In s:SearchForMatchingTag(), the patterns are modified to also match newlines (while ensuring that when searching for the start tag, the current end tag is not accidentally matched). (The patterns already include the fix for subset tag names from bc576b5.)
  3. Merge pull request #16 from inkarkat/also-skip-xml-comments

    authored
    FIX: Also skip tags inside XML comments.
  4. Merge pull request #17 from inkarkat/fix-avoid-tag-subset-matches

    authored
    FIX: Avoid matching tag names that contain the current tag name.
Commits on Apr 16, 2013
  1. @inkarkat

    FIX: Avoid matching tag names that contain the current tag name.

    inkarkat authored
    The .\{-} match for the remainder is too lax; it can pick up intermediate tags where the current tag name is a subset of, e.g. the GSP <g:else> and <g:elseif> tags.
    If there's content after the tag name, it must be separated by whitespace.
    Also, as no attributes are allowed on the end tag, the pattern can be restricted to only optional whitespace.
  2. @inkarkat

    FIX: Also skip tags inside XML comments.

    inkarkat authored
    The check only contained the syntax group names of HTML comments; there are similar syntax groups for XML. As it's hard to derive the names from the actually used filetype (it could be a custom one derived from either HTML or XML), always match both.
Commits on Mar 16, 2013
  1. Merge pull request #15 from inkarkat/fix-regexp-error

    authored
    Fix regexp error and make it more robust.
Commits on Mar 8, 2013
  1. @inkarkat

    FIX: Properly escape literal tag name in matching.

    inkarkat authored
    Even with the restriction to valid tag names, certain allowed characters (like ".") have special meaning in the searchpairpos() regexp, potentially causing wrong matches (e.g. with <foo> vs. <f.o>).
    Change the regexp to "very nomagic" /\V/ and escape any remaining backslashes as a precaution.
  2. @inkarkat

    FIX: Only pick up valid tag names to avoid pattern errors.

    inkarkat authored
    When inserting (unescaped, invalid) text into the buffer, for example:
        :noremap <F4> :%s/\([\.!?]\<Bar[\.!?]"\)\s\s/\1text /g<CR>
    navigating over the text (here: <Bar) causes script errors (here: "E55: Unmatched \)") on every cursor move.
    Refine the pattern that picks up HTML / XML tag names (as close to the standard as feasible in Vim) to avoid that invalid tag names are picked up. Even invalid buffer contents should not cause script errors.
Commits on Dec 2, 2011
  1. Merge pull request #6 from mattsa/timeout-support

    authored
    Add support for vim > 7.0 by making timeout optional
Commits on Dec 1, 2011
  1. @mattsacks
  2. Merge pull request #5 from Twinside/master

    authored
    Match tag works with XML files too
  3. @mattsacks
Commits on Nov 28, 2011
  1. @Twinside
Commits on Nov 23, 2011
  1. I hate NTFS.

    authored
Commits on Nov 22, 2011
  1. Added more info to readme.

    authored
  2. Added readme.

    authored
  3. Initial commit.

    authored
Something went wrong with that request. Please try again.