bf44e357e4
Drop support for Rubinius
...
Rubinius hasn't been tested on for years, nor is it really relevant as a
Ruby implementation these days. While Oga probably still works fine on
Rubinius, I don't want to claim we support something when we officially
don't.
2019-12-03 01:34:26 +01:00
82373d164f
Bump MRI requirements to 2.3.0
...
See https://gitlab.com/yorickpeterse/oga/issues/196 for more
information.
2019-12-03 01:33:52 +01:00
e413165afd
Release 2.17
2019-12-02 19:32:43 +01:00
95da93949b
Fix XPath queries using default XML namespace
...
This fixes https://gitlab.com/yorickpeterse/oga/issues/195
2019-12-02 17:40:11 +00:00
d492a775bf
Release 2.16
2019-11-29 15:43:39 +01:00
977bd594c8
Add support for XPath namespace aliases
...
This fixes https://gitlab.com/yorickpeterse/oga/issues/176
2019-11-29 14:21:45 +00:00
da9721cb34
Suppress deprecation warning on RDoc
...
> NOTE: Gem::Specification#has_rdoc= is deprecated with no replacement. It
> will be removed on or after 2018-12-01.
2019-11-05 16:12:31 +00:00
d9e7346b60
Fix older option for gem command
2019-11-05 12:38:17 +00:00
e086515e59
Release 2.15
2018-04-11 21:42:08 +02:00
8ac0055e42
Allow "th" to occur in thead, tbody, and tfoot
...
Fixes https://gitlab.com/yorickpeterse/oga/issues/190
2018-04-11 21:32:30 +02:00
6c10f41446
Release 2.14
2018-01-30 22:58:43 +01:00
bc87711f9c
Return an Enumerator from each* methods when no block is given
2018-01-29 13:12:42 -05:00
0b7b54119b
Release 2.13
2018-01-05 10:29:42 +01:00
886a160c6a
Strip leading/trailing whitespace from CSS exprs
...
When tokenising CSS expressions we now strip leading and trailing
whitespace from the input string. This is performed without any checks
as a check + `String#strip` ended up being slower compared to just
running `String#strip`. On top of that we cache expressions anyway, so
the overhead of `String#strip` is very small.
Fixes https://gitlab.com/yorickpeterse/oga/issues/187
2018-01-04 22:02:50 +01:00
d1336e760a
Disable appveyor Email notifications
2018-01-02 13:23:22 +01:00
952779da39
Use msys64 for installing Ragel
2018-01-02 13:14:55 +01:00
a5cb9887b0
Try adding vcpkg bin/ to the PATH
2018-01-02 13:06:06 +01:00
5d693134d7
Try using vcpkg for installing Ragel
2018-01-02 12:44:07 +01:00
84748a8d85
Fix typo in gemspec homepage URL
2018-01-02 16:20:52 +05:30
e40bd38384
Fixed CHANGELOG typo
2017-12-29 20:42:06 +01:00
db00fcdd55
Release 2.12
2017-12-29 20:40:24 +01:00
f574197ea6
Ignore nested element start tags
...
This ensures that Oga is able to tokenize input such as the following:
<script<script>foo</script>
Oga will now treat this as:
<script>foo</script>
This is based on libxml behaviour, which seems to differ a bit from
Chromium which treats the node as a text node. This however would
require complex look-ahead logic (as far as I can tell) that I really
don't want to implement in Oga.
Fixes #186
2017-12-28 16:12:20 +01:00
1e002de527
Update CI links
...
[ci skip]
2017-11-02 14:31:43 +01:00
e6eaff1a28
Fix YAML for JRuby builds
2017-11-02 14:13:18 +01:00
11e83f911b
Try to install openjdk for JRuby
...
This is necessary so "javac" is available.
2017-11-02 14:07:51 +01:00
dea9bafee1
Use Ragel without a version
...
Apparently on some platforms (e.g. the JRuby build) 6.9 breaks things.
2017-11-02 13:24:13 +01:00
9cfd628b55
Install build-base on Alpine
...
This is necessary to get programs such as "make".
2017-11-02 13:13:57 +01:00
54e3115607
Bundle without nproc on CI
...
nproc is in coreutils which isn't installed by default. Since the
Gemfile is so small there's no real benefit to using -jX anyway.
2017-11-02 12:56:53 +01:00
794291990e
Use ragel 6.9-r0
...
Maybe this will work, because apparently just using "ragel=6.9" refuses
to install.
2017-11-02 12:41:46 +01:00
de2166eb40
Don't use sudo for Alpine
2017-11-02 12:29:47 +01:00
b248cc7c0f
Updated Appveyor config
2017-11-02 12:21:02 +01:00
e811b511b0
Fixed the GitLab CI YAML
2017-11-02 02:39:40 +01:00
5d1d7fd1d8
Move to GitLab and GitLab CI
2017-11-02 02:13:00 +01:00
f85869ecab
Release 2.11
2017-09-07 00:11:24 +02:00
39bf7ffaeb
Silence method redefinition warnings
...
As the community progressively moves to a useful practice of enabling
ruby warnings on tests, knowingly redefining a method produces a
distracting warning that has to be special-cased when running automated
tests. We thus skip dynamic definitions of methods we know will be
redefined right after.
2017-09-07 00:03:36 +02:00
151788abad
Silence uninitialized variable warnings
...
As the community progressively moves to a useful practice of enabling
ruby warnings on tests, assigning an instance variable before use becomes
a necessary practice. Here we set some variables at initialization that
were previously lazily or conditionally set:
- `decoded` is assigned false which seems to make more semantic sense
than than using nil
- `namespace` is assigned nil, its value being lazily computed later
- `available_namespaces` is assigned nil so as to respect the cache
invalidation mechanism
2017-09-07 00:03:36 +02:00
ef1b8d2a28
Fix typo in README
2017-07-11 13:03:52 +02:00
d1f46e289c
Reworked the contributing guide a bit
...
A lot of the stuff in it was rather blunt.
2017-07-07 00:03:48 +02:00
2710976e48
Added note about wanting more patches
2017-07-03 17:58:37 +02:00
b5848b07a9
Rename COC.md to make GitHub happy
...
GitHub only detects CODE_OF_CONDUCT.md.
2017-06-20 17:01:18 +02:00
6f747656b6
Use RSpec 3 expect syntax for tests
...
This should make it a little bit easier for others to contribute.
2017-06-17 13:52:43 +02:00
8282325569
Don't warn for implicit fallthroughs
...
This is the result of Ragel output which we can't control.
2017-06-17 13:46:59 +02:00
8dc2318020
Release 2.10
2017-04-18 12:56:24 +02:00
84c4db3e9f
Clean up changes from PR #174
2017-04-18 12:55:11 +02:00
4250033ed5
Update CHANGELOG
...
[ci skip]
2017-04-18 12:51:34 +02:00
21b5eeec4b
Fix using symbol on Element#attribute alwas getting nil
2017-04-18 12:51:34 +02:00
e9953d4212
Release 2.9
2017-02-10 15:59:15 +01:00
673f4a29db
Use HTML5 style closing tags for void elements
...
This ensures that element tags such as <img> tags don't use a closing />
when documents are parsed as HTML documents.
Fixes #170
2017-02-10 15:24:41 +01:00
131fba7aed
Doctype inherits from Node
...
This makes it possible to parse documents where a doctype resides in a
node, instead of being located at the root.
Fixes #169
2017-02-10 15:10:30 +01:00
b13cfdfea5
Release 2.8
2017-01-04 12:28:37 +01:00
Yorick Peterse
KitaitiMakoto
Yorick Peterse
David Cornu
Vidur Murali
Loic Nageleisen
hinamiyagk
PikachuEXE