* Mon Dec 09 2019 David Cantrell <dcantrell@redhat.com> - 0.10-1

- Disable jq usage in the release script for now

- Generate a complete %%changelog block for the spec file

- Force overwrite of generate .tar file in .copr/Makefile

- Run srpm and check targets before koji, pass name of tarball

- Add utils/submit-koji-builds.sh, which drives 'make koji'

- Genericize submit-koji-builds.sh a bit more

- Start a .gitignore file

- Changes to the release and build automation for the RPM changelog

- Handle int32 and int64 size results from Koji over XMLRPC (#61)

- Make --copr mode on utils/mkrpmchangelog.sh generic

- Existing packages that go away in a comparison are VERIFY (#59)

- inspect_elf() RESULT_OK should be NOT_WAIVABLE

- Refactor strprefix()

- Fix a memory leak in init_rpminspect() with the javabytecode array.

- Fix a memory leak in the license inspection.

- Update release and build steps in the README

- test_elf.py tests that verify passing tests are Not Waviable

- Build package download URLs correctly, drop use_volume_name setting

- Clean up rpminspect error messages when downloading.

- For fetch-only mode (-f), default workdir to getcwd()

- The new libmandoc API works, drop explicit 1.14.4 BuildRequires

- Add is_text_file() to librpminspect

- Handle multiline results in run_cmd() correctly (#59)

- Show diff(1) output for changing upstream text sources (#59)

- Adjust run_cmd() calls to match the new API

- Explain how the -w option works with the -f option in the man page

- In REMEDY_LICENSE, explain that valid licenses need to be in licdb

- Fix some memory leaks in the license inspection.

- Add a [products] section to the config file for mapping release

  strings (#68)

- Fix json-c memory leak.

- Small memory leak fixes for the product release string gathering.

- Trying to get the automated release notice posting working on github.

- Use printf to build github release body text

- changelog -> ${CWD}/changelog

- Make upstream spec file include the changelog like downstream.

- Syntax errors fixed in utils/release.sh

- Escape newlines in the release body text.

- Run the release script with -x for now, for debugging.

- Shift things around a bit in the release.sh script

- Use jq(1) in utils/release.sh to generate the JSON data for GitHub.

- Slight changes to the jq(1) commands.

- Even more small changes to the jq(1) command.

- Do not build the spec file or SRPM in the release.sh script

- Correctly generate the git log entry for the release.

- Edit the release on github rather than create a new one.

- Still working out problems with automating releases on github.

- Fix 'Builds have different product release (el7 != el7) (#70)

- Remove unnecessary free()

- Shorten the eu-elfcmp reporting (#71)

- Switch to using a BUFSIZ buffer for getline() in run_cmd()

- Expand the get_product_release() functionality (#72)

- Detect product release strings correctly with underscores (#76)

- Support SRPM files that lack RPMTAG_SOURCE entries (#78)

- Handle subpackages with different version numbers (#77)

- Add support for Koji scratch builds in rpminspect (#60)

- Make get_product_release() even if [products] is missing.

- Drop the extra strdup(), xmlrpc_decompose_value() handles it for us.

- Update the rpminspect(1) man page.

- line-buffer stdout - makes "rpminspect ... | tee" behave better

- Clean up memory leaks with the Koji task handling code.

- Add a SIGABRT handler to rpminspect(1)

- bad free()

- Change the task member of koji_task_entry_t to be 'struct koji_task *'

- Skip shellsyntax inspection on source packages (#79)

- Fix set_worksubdir() TASK_WORKDIR test (#80)

* Wed Nov 20 2019 David Cantrell <dcantrell@redhat.com> - 0.9-3

- Drop explicit versioned BR for libmandoc-devel

* Thu Nov 14 2019 David Cantrell <dcantrell@redhat.com> - 0.9-2

- Fix a %%changelog mishap

* Thu Nov 14 2019 David Cantrell <dcantrell@redhat.com> - 0.9-1

- Complete the tests/test_manpage.py integration tests

- Add xml inspection integration tests.

- disttag inspection failures should be not waivable

- All RESULT_OK results should be NOT_WAIVABLE

- Expand integration test suite to support waiver auth checking

- Improve forbidden_path_prefixes results reporting (#59)

- Add a HISTORY file explaining a bit about the history of rpminspect

- Use lstat(2) in copyfile() so symlinks are correctly handled

- Fix a problem with peer detection when comparing single RPM files.

- Stop setting whichbuild in so many places in build.c

- Two hidden bugs in inspect_elf.c resolved via the integration suite

- Add test_elf.py to the tests/ subdirectory

- Add elf_ipv6_blacklist to the sample rpminspect.conf file

- chmod 0644 fortify.c

- Reformat a line in get_elf_section() to make it more readable.

- Add forbidden IPv6 function use tests to tests/test_elf.py

- Forgot to add tests/data/forbidden-ipv6.c

- Add test_elf.py test cases for DT_TEXTREL on 32-bit architectures

- Use headerGetString() throughout librpminspect

- Add some missing free() calls in run_cmd()

- Use string_list_t ** for user_data in elf_archive_iterate()

- Add losing -fPIC on 32-bit builds test in test_elf.py

- Add get_nevra() to librpminspect to get RPMTAG_NEVRA

- Do not output Waiver Authorization for RESULT_INFO results

- Call rpmFreeRpmrc() from main() before the program exits

- Adjust how and where rpmtdFree() is called

- Use the GNU version of basename(3) and ensure we don't use the libgen

  version.

- Fix forbidden_path_prefixes check in the addedfiles inspection (#59)

- Add missing free() to inspect_desktop.c

- Simplify the is_valid_license() code that concatenates tokens

- Simple Makefile to drive different parts of the build.

- Add the git log to the release notification published to github

- Fix error in the 'make release' target

- If asset ID cannot be found, dump what github returned on stdout

- Use jq(1) to escape strings for JSON