{"thread":{"id":"53507","subject":"[PATCH 0/3] openbsd: fixes for 2.27.0-RC0","startedAt":"2020-05-20T03:45:04Z","lastAt":"2020-05-20T21:14:35Z","messageCount":11,"participants":["Carlo Marcelo Arenas Belón","Junio C Hamano","Christopher Warrington (CHRISTOPHER)"],"isPatch":true,"patchVersion":1,"patchTotal":3},"messages":[{"id":"398234","messageId":"20200520034444.47932-1-carenas@gmail.com","threadId":"53507","inReplyTo":null,"subject":"[PATCH 0/3] openbsd: fixes for 2.27.0-RC0","fromName":"Carlo Marcelo Arenas Belón","fromEmail":"carenas@gmail.com","sentAt":"2020-05-20T03:44:41Z","receivedAt":"2020-05-20T03:45:04Z","isPatch":true,"sender":{"key":"carenas@gmail.com","avatar":"https://avatars.githubusercontent.com/u/76036?v=4"},"body":"First one is the only one exclusive to OpenBSD (tested with 6.7)\nwhile the other 2 are likely also needed in other POSIX systems.\n\nCarlo Marcelo Arenas Belón (3):\n  t4216: avoid unnecessary subshell in test_bloom_filters_not_used\n  bisect: remove CR characters from revision in replay\n  t5520: avoid alternation in grep's BRE (not POSIX)\n\n git-bisect.sh        | 5 +++--\n t/t4216-log-bloom.sh | 2 +-\n t/t5520-pull.sh      | 2 +-\n 3 files changed, 5 insertions(+), 4 deletions(-)\n\n-- \n2.27.0.rc0.187.gede8c892b8\n\n"},{"id":"398235","messageId":"20200520034444.47932-2-carenas@gmail.com","threadId":"53507","inReplyTo":"20200520034444.47932-1-carenas@gmail.com","subject":"[PATCH 1/3] t4216: avoid unnecessary subshell in test_bloom_filters_not_used","fromName":"Carlo Marcelo Arenas Belón","fromEmail":"carenas@gmail.com","sentAt":"2020-05-20T03:44:42Z","receivedAt":"2020-05-20T03:45:07Z","isPatch":true,"sender":{"key":"carenas@gmail.com","avatar":"https://avatars.githubusercontent.com/u/76036?v=4"},"body":"Seems to trigger a bug in at least OpenBSD's 6.7 sh where it is\ninterpreted as a history lookup and therefore fails 125-126, 128,\n130.\n\nRemove the subshell and get a space between ! and grep, so tests\npass successfully.\n\nSigned-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>\n---\n t/t4216-log-bloom.sh | 2 +-\n 1 file changed, 1 insertion(+), 1 deletion(-)\n\ndiff --git a/t/t4216-log-bloom.sh b/t/t4216-log-bloom.sh\nindex 21b68dd6c8..c855bcd3e7 100755\n--- a/t/t4216-log-bloom.sh\n+++ b/t/t4216-log-bloom.sh\n@@ -70,7 +70,7 @@ test_bloom_filters_used () {\n test_bloom_filters_not_used () {\n \tlog_args=$1\n \tsetup \"$log_args\" &&\n-\t!(grep -q \"statistics:{\\\"filter_not_present\\\":\" \"$TRASH_DIRECTORY/trace.perf\") &&\n+\t! grep -q \"statistics:{\\\"filter_not_present\\\":\" \"$TRASH_DIRECTORY/trace.perf\" &&\n \ttest_cmp log_wo_bloom log_w_bloom\n }\n \n-- \n2.27.0.rc0.187.gede8c892b8\n\n"},{"id":"398236","messageId":"20200520034444.47932-3-carenas@gmail.com","threadId":"53507","inReplyTo":"20200520034444.47932-1-carenas@gmail.com","subject":"[PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Carlo Marcelo Arenas Belón","fromEmail":"carenas@gmail.com","sentAt":"2020-05-20T03:44:43Z","receivedAt":"2020-05-20T03:45:10Z","isPatch":true,"sender":{"key":"carenas@gmail.com","avatar":"https://avatars.githubusercontent.com/u/76036?v=4"},"body":"6c722cbe5a (bisect: allow CRLF line endings in \"git bisect replay\"\ninput, 2020-05-07) includes CR as a field separator, but doesn't\naccount for it being included in the last field, breaking when\nrunning at least under OpenBSD 6.7's sh.\n\nRead the revision into a raw variable and strip it of any possible\nembeded CR characters, before use.\n\nSigned-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>\n---\n git-bisect.sh | 5 +++--\n 1 file changed, 3 insertions(+), 2 deletions(-)\n\ndiff --git a/git-bisect.sh b/git-bisect.sh\nindex 71b367a944..8cd6095a71 100755\n--- a/git-bisect.sh\n+++ b/git-bisect.sh\n@@ -210,14 +210,15 @@ bisect_replay () {\n \ttest -r \"$file\" || die \"$(eval_gettext \"cannot read \\$file for replaying\")\"\n \tgit bisect--helper --bisect-reset || exit\n \toIFS=\"$IFS\" IFS=\"$IFS$(printf '\\015')\"\n-\twhile read git bisect command rev\n+\twhile read git bisect command rawrev\n \tdo\n \t\ttest \"$git $bisect\" = \"git bisect\" || test \"$git\" = \"git-bisect\" || continue\n \t\tif test \"$git\" = \"git-bisect\"\n \t\tthen\n-\t\t\trev=\"$command\"\n+\t\t\trawrev=\"$command\"\n \t\t\tcommand=\"$bisect\"\n \t\tfi\n+\t\trev=$(echo $rawrev | tr -d '\\015')\n \t\tget_terms\n \t\tgit bisect--helper --check-and-set-terms \"$command\" \"$TERM_GOOD\" \"$TERM_BAD\" || exit\n \t\tget_terms\n-- \n2.27.0.rc0.187.gede8c892b8\n\n"},{"id":"398237","messageId":"20200520034444.47932-4-carenas@gmail.com","threadId":"53507","inReplyTo":"20200520034444.47932-1-carenas@gmail.com","subject":"[PATCH 3/3] t5520: avoid alternation in grep's BRE (not POSIX)","fromName":"Carlo Marcelo Arenas Belón","fromEmail":"carenas@gmail.com","sentAt":"2020-05-20T03:44:44Z","receivedAt":"2020-05-20T03:45:10Z","isPatch":true,"sender":{"key":"carenas@gmail.com","avatar":"https://avatars.githubusercontent.com/u/76036?v=4"},"body":"Instead of using a BRE, that broke tests 30-32, 37-39, 42 at least with\nOpenBSD 6.7; use a simpler ERE.\n\nFixes: d9f15d37f1 (pull: pass --autostash to merge, 2020-04-07)\nSigned-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>\n---\n t/t5520-pull.sh | 2 +-\n 1 file changed, 1 insertion(+), 1 deletion(-)\n\ndiff --git a/t/t5520-pull.sh b/t/t5520-pull.sh\nindex 37535d63a9..9fae07cdfa 100755\n--- a/t/t5520-pull.sh\n+++ b/t/t5520-pull.sh\n@@ -28,7 +28,7 @@ test_pull_autostash_fail () {\n \techo dirty >new_file &&\n \tgit add new_file &&\n \ttest_must_fail git pull \"$@\" . copy 2>err &&\n-\ttest_i18ngrep \"\\(uncommitted changes.\\)\\|\\(overwritten by merge:\\)\" err\n+\ttest_i18ngrep -E \"uncommitted changes.|overwritten by merge:\" err\n }\n \n test_expect_success setup '\n-- \n2.27.0.rc0.187.gede8c892b8\n\n"},{"id":"398258","messageId":"xmqqzha2wt7t.fsf@gitster.c.googlers.com","threadId":"53507","inReplyTo":"20200520034444.47932-3-carenas@gmail.com","subject":"Re: [PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-05-20T14:56:22Z","receivedAt":"2020-05-20T14:56:32Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:\n\n> 6c722cbe5a (bisect: allow CRLF line endings in \"git bisect replay\"\n> input, 2020-05-07) includes CR as a field separator, but doesn't\n> account for it being included in the last field, breaking when\n> running at least under OpenBSD 6.7's sh.\n>\n> Read the revision into a raw variable and strip it of any possible\n> embeded CR characters, before use.\n\nThat's quite unsatisfactory, as the whole point of adding CR to IFS\nwas to avoid having to spawn extra processes for this kind of text\nprocessing.  If we were to do the preprocessing, we are better off\njust passing the whole input thru \"tr -d '\\015'\".\n\n>  \toIFS=\"$IFS\" IFS=\"$IFS$(printf '\\015')\"\n> -\twhile read git bisect command rev\n> +\twhile read git bisect command rawrev\n>  \tdo\n>  \t\ttest \"$git $bisect\" = \"git bisect\" || test \"$git\" = \"git-bisect\" || continue\n>  \t\tif test \"$git\" = \"git-bisect\"\n>  \t\tthen\n> -\t\t\trev=\"$command\"\n> +\t\t\trawrev=\"$command\"\n>  \t\t\tcommand=\"$bisect\"\n>  \t\tfi\n> +\t\trev=$(echo $rawrev | tr -d '\\015')\n\nAs we know that \"rev\" ought to consist of just hexadecimal and\ncannot be split into two at $IFS even if we don't tell \"read\" that\n\"everything at the end of line is 'rev'\", can we do\n\n\twhile read git bisect command rev ignored\n\nso that we'll get an empty string after CR in $ignored when reading\nCRLF input, and an empty string because we ran out of the tokens\nwhen reading LF input?\n\nThat is, ...\n\n git-bisect.sh | 2 +-\n 1 file changed, 1 insertion(+), 1 deletion(-)\n\ndiff --git a/git-bisect.sh b/git-bisect.sh\nindex 71b367a944..2a7599b486 100755\n--- a/git-bisect.sh\n+++ b/git-bisect.sh\n@@ -210,7 +210,7 @@ bisect_replay () {\n \ttest -r \"$file\" || die \"$(eval_gettext \"cannot read \\$file for replaying\")\"\n \tgit bisect--helper --bisect-reset || exit\n \toIFS=\"$IFS\" IFS=\"$IFS$(printf '\\015')\"\n-\twhile read git bisect command rev\n+\twhile read git bisect command rev ignored\n \tdo\n \t\ttest \"$git $bisect\" = \"git bisect\" || test \"$git\" = \"git-bisect\" || continue\n \t\tif test \"$git\" = \"git-bisect\"\n"},{"id":"398259","messageId":"xmqqv9kqwt4k.fsf@gitster.c.googlers.com","threadId":"53507","inReplyTo":"20200520034444.47932-4-carenas@gmail.com","subject":"Re: [PATCH 3/3] t5520: avoid alternation in grep's BRE (not POSIX)","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-05-20T14:58:19Z","receivedAt":"2020-05-20T14:58:24Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:\n\n> Instead of using a BRE, that broke tests 30-32, 37-39, 42 at least with\n> OpenBSD 6.7; use a simpler ERE.\n\nThanks.  We already do use \"test_i18ngrep -E\" in another test, so\nthis should be safe.\n\n> Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>\n> ---\n>  t/t5520-pull.sh | 2 +-\n>  1 file changed, 1 insertion(+), 1 deletion(-)\n>\n> diff --git a/t/t5520-pull.sh b/t/t5520-pull.sh\n> index 37535d63a9..9fae07cdfa 100755\n> --- a/t/t5520-pull.sh\n> +++ b/t/t5520-pull.sh\n> @@ -28,7 +28,7 @@ test_pull_autostash_fail () {\n>  \techo dirty >new_file &&\n>  \tgit add new_file &&\n>  \ttest_must_fail git pull \"$@\" . copy 2>err &&\n> -\ttest_i18ngrep \"\\(uncommitted changes.\\)\\|\\(overwritten by merge:\\)\" err\n> +\ttest_i18ngrep -E \"uncommitted changes.|overwritten by merge:\" err\n>  }\n>  \n>  test_expect_success setup '\n"},{"id":"398260","messageId":"xmqqr1vewsug.fsf@gitster.c.googlers.com","threadId":"53507","inReplyTo":"20200520034444.47932-2-carenas@gmail.com","subject":"Re: [PATCH 1/3] t4216: avoid unnecessary subshell in test_bloom_filters_not_used","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-05-20T15:04:23Z","receivedAt":"2020-05-20T15:04:31Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:\n\n> Seems to trigger a bug in at least OpenBSD's 6.7 sh where it is\n> interpreted as a history lookup and therefore fails 125-126, 128,\n> 130.\n>\n> Remove the subshell and get a space between ! and grep, so tests\n> pass successfully.\n\nIt's strange that somebody thinks of doing history lookup in\nnon-interactive use.\n\nBut even more curious is why we have this in a subshell in the first\nplace.  I do not see a reason why we need subshell, nor use of the\ndouble-quote in the outer layer that forces us to use backslashes.\n\n>  test_bloom_filters_not_used () {\n>  \tlog_args=$1\n>  \tsetup \"$log_args\" &&\n> -\t!(grep -q \"statistics:{\\\"filter_not_present\\\":\" \"$TRASH_DIRECTORY/trace.perf\") &&\n> +\t! grep -q \"statistics:{\\\"filter_not_present\\\":\" \"$TRASH_DIRECTORY/trace.perf\" &&\n\nThis is obviously the minimum fix, so I'm willing to take the change\nas-is, but if we were writing it today, perhaps\n\n\t! grep 'statistics:{\"filter_not_present\":' \"$TRASH_DIRECTORY/trace.perf\" &&\n\nis how we write it.  I do not see any reason why we want to use the\n\"-q\" option either.\n\n"},{"id":"398268","messageId":"20200520170843.GC20332@Carlos-MBP","threadId":"53507","inReplyTo":"xmqqzha2wt7t.fsf@gitster.c.googlers.com","subject":"Re: [PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Carlo Marcelo Arenas Belón","fromEmail":"carenas@gmail.com","sentAt":"2020-05-20T17:08:43Z","receivedAt":"2020-05-20T17:08:48Z","isPatch":true,"sender":{"key":"carenas@gmail.com","avatar":"https://avatars.githubusercontent.com/u/76036?v=4"},"body":"On Wed, May 20, 2020 at 07:56:22AM -0700, Junio C Hamano wrote:\n> As we know that \"rev\" ought to consist of just hexadecimal and\n> cannot be split into two at $IFS even if we don't tell \"read\" that\n> \"everything at the end of line is 'rev'\", can we do\n> \n> \twhile read git bisect command rev ignored\n\nthat works (kind of), but will cause test t6030.66 to fail, with a git\nsegfault nonetheless, because we can't handle anymore a line line:\n\ngit bisect start '--term-new' 'term1' '--term-old' 'term2' '32a594a3fdac2d57cf6d02987e30eec68511498c' '88bcdc1839f0ad191ffdd65cae2a2a862d682151'\n\nIMHO it will be probably still cleaner to do `tr -d '\\015'`, even if the\npatch below avoids all current issues from the testsuite.\n\nwill follow up with a fix for the segfault, unless someone else beats me to\nit.\n\nCarlo\n-- >8 --\nSubject: [PATCH v2] bisect: avoid tailing CR characters from revision in\n replay\n\n6c722cbe5a (bisect: allow CRLF line endings in \"git bisect replay\"\ninput, 2020-05-07) includes CR as a field separator, but relies on\nit not being included in the last field, which breaks at least when\nrunning under OpenBSD 6.7's sh.\n\nInstead of just assume the CR will get swallowed, read the rest of\nthe line into an otherwise unused variable and ignore it everywhere\nexcept on the call for git bisect start, where it matters.\n\nSigned-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>\n---\n git-bisect.sh | 4 ++--\n 1 file changed, 2 insertions(+), 2 deletions(-)\n\ndiff --git a/git-bisect.sh b/git-bisect.sh\nindex 71b367a944..08a6ed57dd 100755\n--- a/git-bisect.sh\n+++ b/git-bisect.sh\n@@ -210,7 +210,7 @@ bisect_replay () {\n \ttest -r \"$file\" || die \"$(eval_gettext \"cannot read \\$file for replaying\")\"\n \tgit bisect--helper --bisect-reset || exit\n \toIFS=\"$IFS\" IFS=\"$IFS$(printf '\\015')\"\n-\twhile read git bisect command rev\n+\twhile read git bisect command rev tail\n \tdo\n \t\ttest \"$git $bisect\" = \"git bisect\" || test \"$git\" = \"git-bisect\" || continue\n \t\tif test \"$git\" = \"git-bisect\"\n@@ -223,7 +223,7 @@ bisect_replay () {\n \t\tget_terms\n \t\tcase \"$command\" in\n \t\tstart)\n-\t\t\tcmd=\"bisect_start $rev\"\n+\t\t\tcmd=\"bisect_start $rev $tail\"\n \t\t\teval \"$cmd\" ;;\n \t\t\"$TERM_GOOD\"|\"$TERM_BAD\"|skip)\n \t\t\tgit bisect--helper --bisect-write \"$command\" \"$rev\" \"$TERM_GOOD\" \"$TERM_BAD\" || exit;;\n-- \n2.27.0.rc0.187.gede8c892b8\n\n"},{"id":"398269","messageId":"xmqqeerewmu6.fsf@gitster.c.googlers.com","threadId":"53507","inReplyTo":"20200520170843.GC20332@Carlos-MBP","subject":"Re: [PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-05-20T17:14:09Z","receivedAt":"2020-05-20T17:14:18Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Carlo Marcelo Arenas Belón <carenas@gmail.com> writes:\n\n> that works (kind of), but will cause test t6030.66 to fail, with a git\n> segfault nonetheless, because we can't handle anymore a line line:\n>\n> git bisect start '--term-new' 'term1' '--term-old' 'term2' '32a594a3fdac2d57cf6d02987e30eec68511498c' '88bcdc1839f0ad191ffdd65cae2a2a862d682151'\n\nYuck.  Thanks for noticing.\n\n> will follow up with a fix for the segfault, unless someone else beats me to\n> it.\n\nThanks.\n"},{"id":"398299","messageId":"DM5PR00MB0439A347273E56C8697A587A9BB60@DM5PR00MB0439.namprd00.prod.outlook.com","threadId":"53507","inReplyTo":"20200520170843.GC20332@Carlos-MBP","subject":"RE: [EXTERNAL] Re: [PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Christopher Warrington (CHRISTOPHER)","fromEmail":"christopher.warrington@microsoft.com","sentAt":"2020-05-20T20:59:30Z","receivedAt":"2020-05-20T20:59:35Z","isPatch":true,"sender":{"key":"christopher.warrington@microsoft.com","avatar":null},"body":"On 2020-05-20 at 10:09-07:00, Carlo Marcelo Arenas Belón wrote:\n\n> IMHO it will be probably still cleaner to do `tr -d '\\015'`, even if the\n> patch below avoids all current issues from the testsuite.\n\nMy initial attempt to handle CRLF logs was shaped like this:\n\n\ttr -d '\\r' <\"$file\" | while read ...\n\nThis introduces a subshell, so there were concerns about propagating\nvariables and exits. So, Peff also suggested preprocessing to a file. Around\nthe same time Junio tried using IFS, and that was simpler.\n\nIt would be pretty easy to tr -d '\\r' >\"$GIT_DIR/BISECT_REPLAY_LOG\" (or some\nother name) and then read from that.\n\nTwo things I'm not sure of with such an approach:\n\n1. Is $GIT_DIR the right place to put this? If not, is there a helper to\n   create a temporary file? mktemp may not be available everywhere. I only\n   see git-mergetool.sh using it and only behind the mergetool.writeToTemp\n   variable.\n\n2. How and when does this temp file need to be cleaned up? Looking at using\n   something like trap \"rm -f \\\"$temp_file\\\"\" 0 will conflict with the traps\n   that git-bisect.sh's bisect_start installs, and bisect_replay calls\n   bisect_start. I could introduce a helper like\n   maybe_unlink_temp_replay_log and invoke that from a trap installed in\n   both bisect_replay and bisect_start.\n\n   I looked at whether bisect.c's bisect_clean_state() is the right place to\n   add such clean up: it does not look like the right place. It would result\n   in the temp file getting deleted while it was being read. (git-bisect.sh\n   -> bisect_replay()'s loop -> bisect_start() -> bisect--helper.c's\n   bisect_start() -> bisect.c's bisect_clean_state()) Alas, not all file\n   systems are OK with that happening.\n\nAny guidance?\n\nHere's a sketch that leaves cleanup on non-successful paths unaddressed.\n\nbisect_replay () {\n\tfile=\"$1\"\n\ttest \"$#\" -eq 1 || die \"$(gettext \"No logfile given\")\"\n\ttest -r \"$file\" || die \"$(eval_gettext \"cannot read \\$file for replaying\")\"\n\tgit bisect--helper --bisect-reset || exit\n\n\tscrubbed_file=\"$GIT_DIR/BISECT_CLEANED_LOG\"\n\ttr -d '\\r' <\"$file\" >\"scrubbed_file\" || die \"badness\"\n\n\twhile read git bisect command rev\n\tdo ... done <\"$scrubbed_file\"\n\n\trm -f \"$scrubbed_file\"\n\tbisect_auto_next\n\n--\nChristopher Warrington <chwarr@microsoft.com>\nMicrosoft Corp.\n"},{"id":"398300","messageId":"xmqq1rneux55.fsf@gitster.c.googlers.com","threadId":"53507","inReplyTo":"DM5PR00MB0439A347273E56C8697A587A9BB60@DM5PR00MB0439.namprd00.prod.outlook.com","subject":"Re: [EXTERNAL] Re: [PATCH 2/3] bisect: remove CR characters from revision in replay","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2020-05-20T21:14:30Z","receivedAt":"2020-05-20T21:14:35Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"\"Christopher Warrington (CHRISTOPHER)\"\n<Christopher.Warrington@microsoft.com> writes:\n\n> On 2020-05-20 at 10:09-07:00, Carlo Marcelo Arenas Belón wrote:\n>\n>> IMHO it will be probably still cleaner to do `tr -d '\\015'`, even if the\n>> patch below avoids all current issues from the testsuite.\n>\n> My initial attempt to handle CRLF logs was shaped like this:\n>\n> \ttr -d '\\r' <\"$file\" | while read ...\n>\n> This introduces a subshell, so there were concerns about propagating\n> variables and exits. So, Peff also suggested preprocessing to a file. Around\n> the same time Junio tried using IFS, and that was simpler.\n\nOne thing was that extra processes and temporary files are pure\noverhead when people don't use misbehaving editors, so we could just\nsay \"don't do it then\".  It may not be worth paying the cost of\nbeing pessimistic and preparing for the worst.  The \"$IFS will split\nthe tokens on the line for free for us\" was a no-cost solution and\nthat was why I suggested it.\n\nIn any case, I think there are folks who are rewriting bisect piece\nby piece to C, and at that point it would just be the matter of\nusing strbuf_getline_lf() vs strbuf_getline() to get rid of the\nunwanted CR for free.\n\n\n"}]}