HTTPS clone URL
Subversion checkout URL
Conformance testing data for the twitter-text-* repositories
accented_hashtags add_autolink_json add_middle_dot_test allow_pipe allow_zwnj_in_hashtag arabic_hebrew_thai_hashtags autolink_tco cashtag_more_tests cashtag domain_in_space dont_extract_bz2 dont_extract_in_hash double_word_unicode emailaddress expand_languages extract_mention_strict extract_url_slash_no_protocol fix_extract_hashtag fix_flickr_urls fix_for_new_gtld_foo flickr_urls_are_broken hash_tag_can_contain_period_but_not_at_end hashbang_autolink hashtag_latin_char hashtag_mention_before_newline hashtag_more_char https hyphenated_domains i18n_url link_at_symbol master mention_after_rt preceding-dollar-sign prep_v1.4.0 prep_142 proposed_1_3_2 punc_before_url subdomain_underscores tfw-extract-urls tweet_length update_bundle_2011_08_05 validation_tests yaml_unicode
Nothing to show
Fetching latest commit...
Cannot retrieve the latest commit at this time.
|Failed to load latest commit information.|
== Purpose This conformance package provides a cross-platform definition of the test cases for auto linking, extracting and hit highlighting of Tweets. The primary use for this is the twitter-text-* libraries; both those managed by Twitter and those created by the community. The reason for this conformance suite is to provide a way to keep the various implementations of Twitter text handling working in a consistent and interoperable way. While anyone can feel free to implement this logic however they choose the recommendation to developers is to use libraries which pass this conformance suite. == Format The test cases are stored in YAML files. There is one YAML file for each major operation type, and within those files there is one section for each publicly accessible API. Each test case is defined by: * description: This provides a meaningful name for the test case, for use as an error message if a test fails. * text: The input text of the Tweet. * expected: What results are expected for this input text == Guidelines for use If you are creating a new twitter-text library in a different programming language please follow these few guidelines: 1. Create a test which reads these files and executes the test cases. 1.a. Do not convert these files to test cases statically. These test cases will change over time. 2. Be sure to implement all of the publicly accessible APIs (the keys to the YAML file) 3. Only expose the public API method and not the underlying regular expressions 3.a. If your language or environment does not allow for this please make a comment to the effect 3.b. This prevent breakage when regular expressions need to change in fundamental ways == Submitting new conformance tests * You can fork the github.com repository at https://github.com/twitter/twitter-text-conformance * Add your new tests and send a pull request. * You can open an issue on github.com at https://github.com/twitter/twitter-text-conformance/issues * Please be sure to provide example input and output as well as a brief description of the problem. == Changelog * v1.3.1 - 2010-12-03 - [ Git tag v1.3.1 ] * [DOC] Updated README with Changelog section * [FIX] Autolink URLs with paths ending in + and - * [FIX] Extract URLs with paths ending in + and - * v1.3.0 - 2010-12-03 - [ Git tag v1.3.0 ] * [NOTE] First tagged version (sorry) * [DOC] Updated README file with guidelines for use and format information * [FIX] Do not autolink URLs without protocols * [FIX] Do not extract URLs without protocols * v1.0.0 - 2010-01-21 - [ Git tag v1.0.0 (retroactively) ] * Initial version