collapse all other-data-type format tests together #497
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
karenetheridge
force-pushed
the
ether/streamlined-format-tests
branch
3 times, most recently
from
646849e to
6c9a1b4
Compare
Julian
reviewed
Jul 8, 2021
tests/draft4/format.json
Outdated
| "schema": {"format": "email"}, | ||
| "description": "validation of string formats", | ||
| "schema": { | ||
| "allOf": [ |
There was a problem hiding this comment.
I think I like the previous way more personally if I'm honest.
Or, I don't really have a strong opinion either way on putting these in one place or in each format, but I slightly prefer we don't complicate the schema -- i.e. use allOf.
(I think basically the simpler each schema can be in the test suite, the better -- we don't mean to test interaction between allOf and format in these tests, so it's better if we don't use it, essentially)
Also, previously someone would get one test per format if you map each test case to a test case in your language's testing framework. But now it's one test for all formats, which likely will fail in a more obscure way if one format is broken.
There was a problem hiding this comment.
Consistency is good though -- e.g. it'd be nice certainly to have all of these kinds of tests looking the same way (using the same description, etc.).
karenetheridge
force-pushed
the
ether/streamlined-format-tests
branch
from
6c9a1b4 to
e940330
Compare
|
Ok, I've kept the good bits of this PR - that is, normalizing all the tests, formatting and adding consistent data items. |
karenetheridge
force-pushed
the
ether/streamlined-format-tests
branch
from
e940330 to
e000145
Compare
Julian
reviewed
Jul 9, 2021
karenetheridge
force-pushed
the
ether/streamlined-format-tests
branch
from
e000145 to
dadf73d
Compare
|
Thanks! This looks good to me now, though CI is failing, but it's because of the previous PR: #498 (comment) |
The generating script:
----8< perl >8----
use strict;
use warnings;
use JSON::PP ();
use JSON::MaybeXS;
use Path::Tiny;
my %format_by_draft = (
draft3 => [ qw(email ip-address ipv6 host-name date-time regex date time color uri) ],
draft4 => [ qw(email ipv4 ipv6 hostname date-time uri) ],
draft6 => [ qw(email ipv4 ipv6 hostname date-time json-pointer uri uri-reference uri-template) ],
draft7 => [ qw(email idn-email regex ipv4 ipv6 idn-hostname hostname date date-time time json-pointer relative-json-pointer iri iri-reference uri uri-reference uri-template) ],
'draft2019-09' => [ qw(email idn-email regex ipv4 ipv6 idn-hostname hostname date date-time time json-pointer relative-json-pointer iri iri-reference uri uri-reference uri-template uuid duration) ],
'draft2020-12' => [ qw(email idn-email regex ipv4 ipv6 idn-hostname hostname date date-time time json-pointer relative-json-pointer iri iri-reference uri uri-reference uri-template uuid duration) ],
'draft-future' => [ qw(email idn-email regex ipv4 ipv6 idn-hostname hostname date date-time time json-pointer relative-json-pointer iri iri-reference uri uri-reference uri-template uuid duration) ],
);
my %type_to_data = (
integer => 12,
float => 13.7,
object => {},
array => [],
boolean => JSON::PP::false,
null => undef,
);
my %format_string_data = (
email => '2962',
'ip-address' => '127.0.0.0.1',
ipv6 => '12345::',
'host-name' => '-a-host-name-that-starts-with--',
'date-time' => '1990-02-31T15:59:60.123-08:00',
regex => '^(abc]',
date => '06/19/1963',
time => '08:30:06 PST',
uri => '//foo.bar/?baz=qux#quux',
ipv4 => '127.0.0.0.1',
hostname => '-a-host-name-that-starts-with--',
'json-pointer' => '/foo/bar~',
'uri-reference' => "\\\\WINDOWS\\fileshare",
'uri-template' => 'http://example.com/dictionary/{term:1}/{term',
'idn-email' => '2962',
'idn-hostname' => "〮실례.테스트",
'relative-json-pointer' => '/foo/bar',
iri => 'http://2001:0db8:85a3:0000:0000:8a2e:0370:7334',
'iri-reference' => "\\\\WINDOWS\\filëßåré",
uuid => '2eb8aa08-aa98-11ea-b4aa-73b441d1638',
duration => 'PT1D',
);
my $encoder = JSON::MaybeXS->new(canonical => 1, pretty => 1, indent_length => 4);
foreach my $draft (keys %format_by_draft) {
my $data = [
map +{
description => $_.' format',
schema => { format => $_ },
tests => [
(map +{
description => 'all string formats ignore '.$_.'s',
data => $type_to_data{$_},
valid => JSON::PP::true,
}, qw(integer float object array boolean null)),
$draft !~ /^draft[467]/ && $format_string_data{$_} ? +{
description => 'invalid '.$_.' string is only an annotation by default',
data => $format_string_data{$_},
valid => JSON::PP::true,
} : (),
],
}, @{$format_by_draft{$draft}}
];
my $str = $encoder->encode($data);
$str =~ s/("data" : .+)(\n\s+)("description" : "[^"]+",)/$3$2$1/mg;
$str =~ s/ : /: /g;
$str =~ s/"schema": \{\n\s+("format": "[^"]+")\n\s+/"schema": { $1 /mg;
path('tests', $draft, 'format.json')->spew_raw($str);
}
karenetheridge
force-pushed
the
ether/streamlined-format-tests
branch
from
dadf73d to
ba1f1a7
Compare
|
lgtm now thanks for all the work! |
Julian
added a commit
that referenced
this pull request
Jul 13, 2021
In older (pre-draft7) drafts, implementations are indeed free to treat format as an assertion by default. (ref #497 (comment))
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This doesn't remove any checks, just folds them together consistently.