dennisdoomen / mockly / 33954005085

85%

push

github

Add ImportFromCurl to bootstrap a mock from an existing curl command (#135)

* Add HttpMock.ImportFromCurl to build mocks from curl commands (closes #124)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* Address review comments on ImportFromCurl

- Match the imported curl URL literally instead of as a Mockly wildcard
  pattern: percent-encoded query values now compare against the decoded
  request query, a literal '*' in the URL no longer broadens the match,
  and a scheme-less URL now defaults to http (curl's default) instead of
  Mockly's https default.
- Support unquoted backslash escapes so common DevTools-style commands
  using the '\'' idiom to embed an apostrophe parse correctly.
- Implement the curl encoding --data-urlencode performs (content,
  =content and name=content forms) instead of treating it like raw data;
  reject the @filename/name@filename forms that read from a file.
- Compare multi-token header values (e.g. User-Agent) using their
  original separator instead of always joining with ', ', which broke
  matching for headers such as 'Mozilla/5.0 Foo/1.0'.
- Represent curl's 'Name:' (suppress header) and 'Name;' (explicit empty
  value) syntaxes correctly instead of treating an empty header value as
  a required-but-empty header.
- Fix InspectCode findings: remove the redundant IndexOfChar helper,
  convert CurlTokenizer to a primary constructor, drop the curlCommand
  parameter that was only used to capture its name, and avoid multiple
  enumeration in the net472 header-matching fallback.
- Rename Can_import_* tests to state the fact per naming guidance.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Simplify WebUtility.UrlEncode calls (never-null result)

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

568 of 779 branches covered (72.91%)

Branch coverage included in aggregate %.

345 of 400 new or added lines in 2 files covered. (86.25%)

1761 of 1973 relevant lines covered (89.25%)

309.92 hits per line