Hello Cygwinstas, I spent last night and some of today working on building the gnu-patch utility from the source code downloaded using setup. I desired to do this in order to learn more about the general procedures involved (re-familiarizing myself with Autoconfiscated projects, primarily) , and I plan on applying this acquired knowledge to other builds in the future.
My concerns were raised by how many of the patches I applied (by hand, so to speak) failed in one or more places. I am afraid I can't serve up all the examples (maybe a good thing, would have made this message huge). I closed the terminal I was working in before I decided to compose this message. Here's a list of the patches anyhow: patch.cygport patch-CVE-2018-1000156.patch patch-2.7.6-test-suite-compatibility-fixes.patch patch-2.7.6-switch-from-fork-execlp-to-execute.patch patch-2.7.6-skip-ed-test-when-the-ed-utility-is-not-installed.patch patch-2.7.6-make-debug-output-more-useful.patch patch-2.7.6-improve_support_for_memory_leak_detection.patch patch-2.7.6-fix-segfault-with-mangled-rename-patch.patch patch-2.7.6-fix-korn-shell-incompatibility.patch patch-2.7.6-fix-ed-style-test-failure.patch patch-2.7.6-failed_assertion.patch patch-2.7.6-dont-leak-temporary-file-on-failed-multi-file-ed-style-patch.patch patch-2.7.6-dont-leak-temporary-file-on-failed-ed-style-patch.patch patch-2.7.6-CVE-2019-13638-invoked-ed-directly-instead-of-using-the-shell.patch patch-2.7.6-CVE-2019-13636-symlinks.patch patch-2.7.6-CVE-2018-6952-fix-swapping-fake-lines-in-pch_swap.patch patch-2.7.6-CVE-2018-17942.patch patch-2.7.6-crash-RLIMIT_NOFILE.patch patch-2.7.6-cleanups-in-do_ed_script.patch patch-2.7.6-check-of-return-value-of-fwrite.patch patch-2.7.6-avoid-warnings-gcc8.patch patch-2.7.6-avoid-set_file_attributes-sign-conversion-warnings.patch patch-2.7.6-avoid-invalid-memory-access-in-context-format-diffs.patch patch-2.7.6-allow-input-files-to-be-missing-for-ed-style-patches.patch patch-2.7.6-abort_when_cleaning_up_fails.patch In at least half the cases in the list above, the patch failed; where it succeeded it was sometimes 6 or 9 lines off. Is it unusual to have this happen? I ignored patch.cygport, btw, as it did not seem to apply to the task at hand. 'patch' only failed one test, and a build for Debian of the same release also failed that same test (something like set-mode). I installed the home-built 'patch' and it seems to be sound and reliable. -- Soren 'tino He's earthy, he's pithy, he's ...a bit astringent. Think of him more as a "topical antiseptic" than as a snack. somia...@gmail.com https://instagram.com/samhain08/ -- Problem reports: https://cygwin.com/problems.html FAQ: https://cygwin.com/faq/ Documentation: https://cygwin.com/docs.html Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple
