{"thread":{"id":"55533","subject":"[PATCH] mailinfo: strip CR from base64/quoted-printable email","startedAt":"2021-04-21T01:34:19Z","lastAt":"2021-05-09T17:12:37Z","messageCount":35,"participants":["Đoàn Trần Công Danh","Junio C Hamano","brian m. carlson"],"isPatch":true,"patchVersion":1,"patchTotal":null},"messages":[{"id":"422517","messageId":"20210421013404.17383-1-congdanhqx@gmail.com","threadId":"55533","inReplyTo":null,"subject":"[PATCH] mailinfo: strip CR from base64/quoted-printable email","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-04-21T01:34:04Z","receivedAt":"2021-04-21T01:34:19Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"When an SMTP server receives an 8-bit email message, possibly with only\nLF as line ending, some of those servers decide to change said LF to\nCRLF.\n\nSome other SMTP servers, when receives an 8-bit email message, decide to\nencoding such message in base64 and/or quoted-printable instead.\n\nIf an email is transfered through those 2 email servers in order, the\nfinal recipients will receive an email contains a patch mungled with\nCRLF encoded inside another encoding. Thus, such CR couldn't be dropped\nby mailsplit. Such accidents have been observed in the wild [1].\n\nLet's guess if such CR was added automatically and strip them in\nmailinfo.\n\n[1]: https://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n\n I'm not sure if guessing the heuristic to strip CR is a good approach.\n I think it's better to pass --keep-cr down from git-am.\n Let's say --keep-cr=\n\n mailinfo.c | 20 +++++++++++++++++---\n t/t5100-mailinfo.sh | 5 +++++\n t/t5100/cr-base64.mbox | 22 ++++++++++++++++++++++\n t/t5100/info1000 | 5 +++++\n t/t5100/msg1000 | 2 ++\n t/t5100/patch1000 | 22 ++++++++++++++++++++++\n 6 files changed, 73 insertions(+), 3 deletions(-)\n create mode 100644 t/t5100/cr-base64.mbox\n create mode 100644 t/t5100/info1000\n create mode 100644 t/t5100/msg1000\n create mode 100644 t/t5100/patch1000\n\ndiff --git a/mailinfo.c b/mailinfo.c\nindex 5681d9130d..dbff867f42 100644\n--- a/mailinfo.c\n+++ b/mailinfo.c\n@@ -988,16 +988,27 @@ static int handle_boundary(struct mailinfo *mi, struct strbuf *line)\n }\n \n static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n-\t\t\t\t struct strbuf *prev)\n+\t\t\t\t struct strbuf *prev, int *keep_cr)\n {\n \tsize_t len = line->len;\n \tconst char *rest;\n \n \tif (!mi->format_flowed) {\n+\t\tif (*keep_cr == -1 && len >= 2)\n+\t\t\t*keep_cr = !(line->buf[len - 2] == '\\r' &&\n+\t\t\t\t line->buf[len - 1] == '\\n');\n+\t\tif (!*keep_cr && len >= 2 &&\n+\t\t line->buf[len - 2] == '\\r' &&\n+\t\t line->buf[len - 1] == '\\n') {\n+\t\t\tstrbuf_setlen(line, len - 2);\n+\t\t\tstrbuf_addch(line, '\\n');\n+\t\t\tlen--;\n+\t\t}\n \t\thandle_filter(mi, line);\n \t\treturn;\n \t}\n \n+\t*keep_cr = 1;\n \tif (line->buf[len - 1] == '\\n') {\n \t\tlen--;\n \t\tif (len && line->buf[len - 1] == '\\r')\n@@ -1036,6 +1047,7 @@ static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n static void handle_body(struct mailinfo *mi, struct strbuf *line)\n {\n \tstruct strbuf prev = STRBUF_INIT;\n+\tint keep_cr = -1;\n \n \t/* Skip up to the first boundary */\n \tif (*(mi->content_top)) {\n@@ -1081,7 +1093,7 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \t\t\t\t\t\tstrbuf_addbuf(&prev, sb);\n \t\t\t\t\t\tbreak;\n \t\t\t\t\t}\n-\t\t\t\thandle_filter_flowed(mi, sb, &prev);\n+\t\t\t\thandle_filter_flowed(mi, sb, &prev, &keep_cr);\n \t\t\t}\n \t\t\t/*\n \t\t\t * The partial chunk is saved in \"prev\" and will be\n@@ -1091,7 +1103,9 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \t\t\tbreak;\n \t\t}\n \t\tdefault:\n-\t\t\thandle_filter_flowed(mi, line, &prev);\n+\t\t\t/* CR in plain message was processed in mailsplit */\n+\t\t\tkeep_cr = 1;\n+\t\t\thandle_filter_flowed(mi, line, &prev, &keep_cr);\n \t\t}\n \n \t\tif (mi->input_error)\ndiff --git a/t/t5100-mailinfo.sh b/t/t5100-mailinfo.sh\nindex 147e616533..9ccc11d16a 100755\n--- a/t/t5100-mailinfo.sh\n+++ b/t/t5100-mailinfo.sh\n@@ -228,4 +228,9 @@ test_expect_success 'mailinfo handles unusual header whitespace' '\n \ttest_cmp expect actual\n '\n \n+test_expect_success 'mailinfo strip CR after decode base64' '\n+\tcp $DATA/cr-base64.mbox 1000 &&\n+\tcheck_mailinfo 1000 \"\"\n+'\n+\n test_done\ndiff --git a/t/t5100/cr-base64.mbox b/t/t5100/cr-base64.mbox\nnew file mode 100644\nindex 0000000000..6ea9806a6b\n--- /dev/null\n+++ b/t/t5100/cr-base64.mbox\n@@ -0,0 +1,22 @@\n+From: A U Thor \n+To: list@example.org\n+Subject: [PATCH v2] sample\n+Date: Mon, 3 Aug 2020 22:40:55 +0700\n+Message-Id: \n+Content-Type: text/plain; charset=\"utf-8\"\n+Content-Transfer-Encoding: base64\n+\n+T24gZGlmZmVyZW50IGRpc3RybywgcHl0ZXN0IGlzIHN1ZmZpeGVkIHdpdGggZGlmZmVyZW50IHBh\n+dHRlcm5zLg0KDQotLS0NCiBjb25maWd1cmUgfCAyICstDQogMSBmaWxlIGNoYW5nZWQsIDEgaW5z\n+ZXJ0aW9uKCspLCAxIGRlbGV0aW9uKC0pDQoNCmRpZmYgLS1naXQgYS9jb25maWd1cmUgYi9jb25m\n+aWd1cmUNCmluZGV4IGRiMzUzOGIzLi5mN2MxYzA5NSAxMDA3NTUNCi0tLSBhL2NvbmZpZ3VyZQ0K\n+KysrIGIvY29uZmlndXJlDQpAQCAtODE0LDcgKzgxNCw3IEBAIGlmIFsgJGhhdmVfcHl0aG9uMyAt\n+ZXEgMSBdOyB0aGVuDQogICAgIHByaW50ZiAiQ2hlY2tpbmcgZm9yIHB5dGhvbjMgcHl0ZXN0ICg+\n+PSAzLjApLi4uICINCiAgICAgY29uZj0kKG1rdGVtcCkNCiAgICAgcHJpbnRmICJbcHl0ZXN0XVxu\n+bWludmVyc2lvbj0zLjBcbiIgPiAkY29uZg0KLSAgICBpZiBweXRlc3QtMyAtYyAkY29uZiAtLXZl\n+cnNpb24gPi9kZXYvbnVsbCAyPiYxOyB0aGVuDQorICAgIGlmICIkcHl0aG9uIiAtbSBweXRlc3Qg\n+LWMgJGNvbmYgLS12ZXJzaW9uID4vZGV2L251bGwgMj4mMTsgdGhlbg0KICAgICAgICAgcHJpbnRm\n+ICJZZXMuXG4iDQogICAgICAgICBoYXZlX3B5dGhvbjNfcHl0ZXN0PTENCiAgICAgZWxzZQ0KLS0g\n+DQoyLjI4LjANCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f\n+CmV4YW1wbGUgbWFpbGluZyBsaXN0IC0tIGxpc3RAZXhhbXBsZS5vcmcKVG8gdW5zdWJzY3JpYmUg\n+c2VuZCBhbiBlbWFpbCB0byBsaXN0LWxlYXZlQGV4YW1wbGUub3JnCg==\ndiff --git a/t/t5100/info1000 b/t/t5100/info1000\nnew file mode 100644\nindex 0000000000..dab2228b70\n--- /dev/null\n+++ b/t/t5100/info1000\n@@ -0,0 +1,5 @@\n+Author: A U Thor\n+Email: mail@example.com\n+Subject: sample\n+Date: Mon, 3 Aug 2020 22:40:55 +0700\n+\ndiff --git a/t/t5100/msg1000 b/t/t5100/msg1000\nnew file mode 100644\nindex 0000000000..5e8e860aae\n--- /dev/null\n+++ b/t/t5100/msg1000\n@@ -0,0 +1,2 @@\n+On different distro, pytest is suffixed with different patterns.\n+\ndiff --git a/t/t5100/patch1000 b/t/t5100/patch1000\nnew file mode 100644\nindex 0000000000..51c4fb4cb5\n--- /dev/null\n+++ b/t/t5100/patch1000\n@@ -0,0 +1,22 @@\n+---\n+ configure | 2 +-\n+ 1 file changed, 1 insertion(+), 1 deletion(-)\n+\n+diff --git a/configure b/configure\n+index db3538b3..f7c1c095 100755\n+--- a/configure\n++++ b/configure\n+@@ -814,7 +814,7 @@ if [ $have_python3 -eq 1 ]; then\n+ printf \"Checking for python3 pytest (>= 3.0)... \"\n+ conf=$(mktemp)\n+ printf \"[pytest]\\nminversion=3.0\\n\" > $conf\n+- if pytest-3 -c $conf --version >/dev/null 2>&1; then\n++ if \"$python\" -m pytest -c $conf --version >/dev/null 2>&1; then\n+ printf \"Yes.\\n\"\n+ have_python3_pytest=1\n+ else\n+-- \n+2.28.0\n+_______________________________________________\n+example mailing list -- list@example.org\n+To unsubscribe send an email to list-leave@example.org\n-- \n2.31.1.192.g0881477623\n\n"},{"id":"422518","messageId":"xmqqy2dcz8ve.fsf@gitster.g","threadId":"55533","inReplyTo":"20210421013404.17383-1-congdanhqx@gmail.com","subject":"Re: [PATCH] mailinfo: strip CR from base64/quoted-printable email","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2021-04-21T02:09:25Z","receivedAt":"2021-04-21T02:09:29Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Đoàn Trần Công Danh writes:\n\n> When an SMTP server receives an 8-bit email message, possibly with only\n> LF as line ending, some of those servers decide to change said LF to\n> CRLF.\n>\n> Some other SMTP servers, when receives an 8-bit email message, decide to\n> encoding such message in base64 and/or quoted-printable instead.\n\nencoding -> encode\n\n>\n> If an email is transfered through those 2 email servers in order, the\n> final recipients will receive an email contains a patch mungled with\n> CRLF encoded inside another encoding. Thus, such CR couldn't be dropped\n> by mailsplit. Such accidents have been observed in the wild [1].\n>\n> Let's guess if such CR was added automatically and strip them in\n> mailinfo.\n>\n> [1]: https://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n>\n> Signed-off-by: Đoàn Trần Công Danh \n> ---\n>\n> I'm not sure if guessing the heuristic to strip CR is a good approach.\n> I think it's better to pass --keep-cr down from git-am.\n> Let's say --keep-cr=\n\nIt matches my instinct to tie this with the existing --keep-cr\noption, even though I admit that I haven't thought things through.\n\n.\n"},{"id":"422521","messageId":"YH+ct4haFn4q5qNB@camp.crustytoothpaste.net","threadId":"55533","inReplyTo":"20210421013404.17383-1-congdanhqx@gmail.com","subject":"Re: [PATCH] mailinfo: strip CR from base64/quoted-printable email","fromName":"brian m. carlson","fromEmail":"sandals@crustytoothpaste.net","sentAt":"2021-04-21T03:32:07Z","receivedAt":"2021-04-21T03:32:45Z","isPatch":true,"sender":{"key":"sandals@crustytoothpaste.net","avatar":"https://avatars.githubusercontent.com/u/497054?v=4"},"body":"On 2021-04-21 at 01:34:04, Đoàn Trần Công Danh wrote:\n> When an SMTP server receives an 8-bit email message, possibly with only\n> LF as line ending, some of those servers decide to change said LF to\n> CRLF.\n> \n> Some other SMTP servers, when receives an 8-bit email message, decide to\n> encoding such message in base64 and/or quoted-printable instead.\n\nThis really isn't an SMTP server. It's mailing list software, namely\nmailman, and I would argue it's a bug, even though we may want to work\naround it. For example, re-encoding the message breaks DKIM signatures,\nwhich means that mailman is likely to cause mail to be needlessly\nrejected.\n\n8BITMIME is now so common with SMTP that I'd argue that we should just\nwrite off servers that don't support it (especially in the context of\nSMTPUTF8 existing), but this isn't the case of an SMTP server being\nstuck in the last century. Can we say more accurately that this is\nmailing list software (or just call it out by name)?\n\n> If an email is transfered through those 2 email servers in order, the\n> final recipients will receive an email contains a patch mungled with\n> CRLF encoded inside another encoding. Thus, such CR couldn't be dropped\n> by mailsplit. Such accidents have been observed in the wild [1].\n> \n> Let's guess if such CR was added automatically and strip them in\n> mailinfo.\n>\n> [1]: https://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n> \n> Signed-off-by: Đoàn Trần Công Danh \n> ---\n> \n> I'm not sure if guessing the heuristic to strip CR is a good approach.\n> I think it's better to pass --keep-cr down from git-am.\n> Let's say --keep-cr=\n\nI think we may want a separate option here. When I send a 7bit or 8bit\nbody, I expect text canonicalization on the line endings. However, when\nI send a base64 or quoted-printable body, I don't expect my data to be\nmodified at all, and absent a compelling reason, doing so is incorrect.\nIn most cases, using base64 or quoted-printable is going to mean that\nthe sender knew that the body shouldn't be modified, not that mailman\nmodified it, so we should make line munging in this case opt-in.\n-- \nbrian m. carlson (he/him or they/them)\nHouston, Texas, US\n"},{"id":"422576","messageId":"YIAVjpdwQhPJbVbl@danh.dev","threadId":"55533","inReplyTo":"YH+ct4haFn4q5qNB@camp.crustytoothpaste.net","subject":"Re: [PATCH] mailinfo: strip CR from base64/quoted-printable email","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-04-21T12:07:42Z","receivedAt":"2021-04-21T12:07:46Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"On 2021-04-21 03:32:07+0000, \"brian m. carlson\" wrote:\n> On 2021-04-21 at 01:34:04, Đoàn Trần Công Danh wrote:\n> > When an SMTP server receives an 8-bit email message, possibly with only\n> > LF as line ending, some of those servers decide to change said LF to\n> > CRLF.\n> > \n> > Some other SMTP servers, when receives an 8-bit email message, decide to\n> > encoding such message in base64 and/or quoted-printable instead.\n> \n> This really isn't an SMTP server. It's mailing list software, namely\n> mailman, and I would argue it's a bug, even though we may want to work\n> around it. For example, re-encoding the message breaks DKIM signatures,\n> which means that mailman is likely to cause mail to be needlessly\n> rejected.\n> \n> 8BITMIME is now so common with SMTP that I'd argue that we should just\n> write off servers that don't support it (especially in the context of\n> SMTPUTF8 existing), but this isn't the case of an SMTP server being\n> stuck in the last century. Can we say more accurately that this is\n> mailing list software (or just call it out by name)?\n\nI think replace \"SMTP servers\" with \"mailing list managers\" is\ncorrect. I don't feel comfortable to call it out, since I don't know\nif other managers do it that way or not.\n\n> \n> > If an email is transfered through those 2 email servers in order, the\n> > final recipients will receive an email contains a patch mungled with\n> > CRLF encoded inside another encoding. Thus, such CR couldn't be dropped\n> > by mailsplit. Such accidents have been observed in the wild [1].\n> > \n> > Let's guess if such CR was added automatically and strip them in\n> > mailinfo.\n> >\n> > [1]: https://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n> > \n> > Signed-off-by: Đoàn Trần Công Danh \n> > ---\n> > \n> > I'm not sure if guessing the heuristic to strip CR is a good approach.\n> > I think it's better to pass --keep-cr down from git-am.\n> > Let's say --keep-cr=\n> \n> I think we may want a separate option here. When I send a 7bit or 8bit\n> body, I expect text canonicalization on the line endings. However, when\n> I send a base64 or quoted-printable body, I don't expect my data to be\n> modified at all, and absent a compelling reason, doing so is incorrect.\n> In most cases, using base64 or quoted-printable is going to mean that\n> the sender knew that the body shouldn't be modified, not that mailman\n> modified it, so we should make line munging in this case opt-in.\n\nMake sense, this patch was sent mostly for some discussion first.\nWould you mind suggest something for the option.\n\nI'm thinking about --quoted-cr=, mimicking the\n--whitespace option.\n\n-- \nDanh\n"},{"id":"422657","messageId":"YIDNDjJh8vQujkFU@camp.crustytoothpaste.net","threadId":"55533","inReplyTo":"YIAVjpdwQhPJbVbl@danh.dev","subject":"Re: [PATCH] mailinfo: strip CR from base64/quoted-printable email","fromName":"brian m. carlson","fromEmail":"sandals@crustytoothpaste.net","sentAt":"2021-04-22T01:10:38Z","receivedAt":"2021-04-22T01:11:16Z","isPatch":true,"sender":{"key":"sandals@crustytoothpaste.net","avatar":"https://avatars.githubusercontent.com/u/497054?v=4"},"body":"On 2021-04-21 at 12:07:42, Đoàn Trần Công Danh wrote:\n> I think replace \"SMTP servers\" with \"mailing list managers\" is\n> correct. I don't feel comfortable to call it out, since I don't know\n> if other managers do it that way or not.\n\nI think that's fair. I would hope not for the reasons I mentioned, but\nit also would not be surprising to me if others did nevertheless.\n\n> Make sense, this patch was sent mostly for some discussion first.\n> Would you mind suggest something for the option.\n> \n> I'm thinking about --quoted-cr=, mimicking the\n> --whitespace option.\n\nI think that sounds like a great name.\n-- \nbrian m. carlson (he/him or they/them)\nHouston, Texas, US\n"},{"id":"423591","messageId":"cover.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"20210421013404.17383-1-congdanhqx@gmail.com","subject":"[PATCH v2 0/5] Teach am/mailinfo to process quoted CR","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:19:57Z","receivedAt":"2021-05-04T17:20:13Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"When an SMTP server receives an 8-bit email message, possibly with only\nLF as line ending, some of those servers decide to change said LF to\nCRLF.\n\nSome mailing list software, when receives an 8-bit email message,\ndecide to encoding such message in base64 or quoted-printable.\n\nThis series try to help users of such softwares deal with such patches.\n\nThis series is a complete rewrite of v1.\nHence, no {inter,range}-diff.\n\nSorry for the long delay, life carries me away from Git.\n\n[1]:\nhttps://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n\nĐoàn Trần Công Danh (5):\n mailinfo: avoid magic number in option parsing\n mailinfo: warn if CR found in base64/quoted-printable email\n mailinfo: skip quoted CR on user's wish\n mailinfo: strip quoted CR on users' wish\n am: learn to process quoted lines that ends with CRLF\n\n Documentation/git-am.txt | 4 +++\n Documentation/git-mailinfo.txt | 19 +++++++++++-\n builtin/am.c | 56 ++++++++++++++++++++++++++++++++++\n builtin/mailinfo.c | 13 +++++---\n mailinfo.c | 47 ++++++++++++++++++++++++++--\n mailinfo.h | 10 ++++++\n t/t4258-am-quoted-cr.sh | 37 ++++++++++++++++++++++\n t/t4258/mbox | 12 ++++++++\n t/t5100-mailinfo.sh | 21 +++++++++++++\n t/t5100/quoted-cr-info | 5 +++\n t/t5100/quoted-cr-msg | 2 ++\n t/t5100/quoted-cr-patch | 22 +++++++++++++\n t/t5100/quoted-cr.mbox | 22 +++++++++++++\n 13 files changed, 262 insertions(+), 8 deletions(-)\n create mode 100755 t/t4258-am-quoted-cr.sh\n create mode 100644 t/t4258/mbox\n create mode 100644 t/t5100/quoted-cr-info\n create mode 100644 t/t5100/quoted-cr-msg\n create mode 100644 t/t5100/quoted-cr-patch\n create mode 100644 t/t5100/quoted-cr.mbox\n\n-- \n2.31.1.500.gbc6bbdd36b\n\n"},{"id":"423592","messageId":"e1e0d8196197d0eb3a39454a48b112a5843d33d7.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"cover.1620148732.git.congdanhqx@gmail.com","subject":"[PATCH v2 1/5] mailinfo: avoid magic number in option parsing","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:19:58Z","receivedAt":"2021-05-04T17:20:17Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"While current magic number is perfectly correct, in a later change,\nwe would like to parse for another option, namely \"--quoted-cr\".\nLet's refactor a bit to remove that magic number.\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n builtin/mailinfo.c | 5 +++--\n 1 file changed, 3 insertions(+), 2 deletions(-)\n\ndiff --git a/builtin/mailinfo.c b/builtin/mailinfo.c\nindex cfb667a594..b309badce5 100644\n--- a/builtin/mailinfo.c\n+++ b/builtin/mailinfo.c\n@@ -24,6 +24,7 @@ int cmd_mailinfo(int argc, const char **argv, const char *prefix)\n \tmi.metainfo_charset = def_charset;\n \n \twhile (1 < argc && argv[1][0] == '-') {\n+\t\tconst char *str;\n \t\tif (!strcmp(argv[1], \"-k\"))\n \t\t\tmi.keep_subject = 1;\n \t\telse if (!strcmp(argv[1], \"-b\"))\n@@ -34,8 +35,8 @@ int cmd_mailinfo(int argc, const char **argv, const char *prefix)\n \t\t\tmi.metainfo_charset = def_charset;\n \t\telse if (!strcmp(argv[1], \"-n\"))\n \t\t\tmi.metainfo_charset = NULL;\n-\t\telse if (starts_with(argv[1], \"--encoding=\"))\n-\t\t\tmi.metainfo_charset = argv[1] + 11;\n+\t\telse if (skip_prefix(argv[1], \"--encoding=\", &str))\n+\t\t\tmi.metainfo_charset = str;\n \t\telse if (!strcmp(argv[1], \"--scissors\"))\n \t\t\tmi.use_scissors = 1;\n \t\telse if (!strcmp(argv[1], \"--no-scissors\"))\n-- \n2.31.1.500.gbc6bbdd36b\n\n"},{"id":"423594","messageId":"45d2c4ab58c4b0c6f0c7790890bbf75eb373f999.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"cover.1620148732.git.congdanhqx@gmail.com","subject":"[PATCH v2 2/5] mailinfo: warn if CR found in base64/quoted-printable email","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:19:59Z","receivedAt":"2021-05-04T17:20:22Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"When an SMTP server receives an 8-bit email message, possibly with only\nLF as line ending, some of those servers decide to change said LF to\nCRLF.\n\nSome mailing list softwares, when receives an 8-bit email message,\ndecide to encoding such message in base64 or quoted-printable.\n\nIf an email is transfered through above mail servers, then distributed\nby such mailing list softwares, the recipients will receive an email\ncontains a patch mungled with CRLF encoded inside another encoding.\nThus, such CR couldn't be dropped by mailsplit. Hence, the mailed patch\ncouldn't be applied cleanly. Such accidents have been observed in the wild [1].\n\nLet's give our users some warnings if such CR is found.\n\n[1]: https://nmbug.notmuchmail.org/nmweb/show/m2lf9ejegj.fsf%40guru.guru-group.fi\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n mailinfo.c | 21 ++++++++++++++++++---\n t/t5100-mailinfo.sh | 15 +++++++++++++++\n t/t5100/quoted-cr-info | 5 +++++\n t/t5100/quoted-cr-msg | 2 ++\n t/t5100/quoted-cr-patch | 22 ++++++++++++++++++++++\n t/t5100/quoted-cr.mbox | 22 ++++++++++++++++++++++\n 6 files changed, 84 insertions(+), 3 deletions(-)\n create mode 100644 t/t5100/quoted-cr-info\n create mode 100644 t/t5100/quoted-cr-msg\n create mode 100644 t/t5100/quoted-cr-patch\n create mode 100644 t/t5100/quoted-cr.mbox\n\ndiff --git a/mailinfo.c b/mailinfo.c\nindex 5681d9130d..713567f84b 100644\n--- a/mailinfo.c\n+++ b/mailinfo.c\n@@ -988,12 +988,17 @@ static int handle_boundary(struct mailinfo *mi, struct strbuf *line)\n }\n \n static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n-\t\t\t\t struct strbuf *prev)\n+\t\t\t\t struct strbuf *prev, int *have_quoted_cr)\n {\n \tsize_t len = line->len;\n \tconst char *rest;\n \n \tif (!mi->format_flowed) {\n+\t\tif (len >= 2 &&\n+\t\t line->buf[len - 2] == '\\r' &&\n+\t\t line->buf[len - 1] == '\\n') {\n+\t\t\t*have_quoted_cr = 1;\n+\t\t}\n \t\thandle_filter(mi, line);\n \t\treturn;\n \t}\n@@ -1033,9 +1038,16 @@ static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n \thandle_filter(mi, line);\n }\n \n+static void summarize_quoted_cr(struct mailinfo *mi, int have_quoted_cr)\n+{\n+\tif (have_quoted_cr)\n+\t\twarning(\"quoted CR detected\");\n+}\n+\n static void handle_body(struct mailinfo *mi, struct strbuf *line)\n {\n \tstruct strbuf prev = STRBUF_INIT;\n+\tint have_quoted_cr = 0;\n \n \t/* Skip up to the first boundary */\n \tif (*(mi->content_top)) {\n@@ -1051,6 +1063,8 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \t\t\t\thandle_filter(mi, &prev);\n \t\t\t\tstrbuf_reset(&prev);\n \t\t\t}\n+\t\t\tsummarize_quoted_cr(mi, have_quoted_cr);\n+\t\t\thave_quoted_cr = 0;\n \t\t\tif (!handle_boundary(mi, line))\n \t\t\t\tgoto handle_body_out;\n \t\t}\n@@ -1081,7 +1095,7 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \t\t\t\t\t\tstrbuf_addbuf(&prev, sb);\n \t\t\t\t\t\tbreak;\n \t\t\t\t\t}\n-\t\t\t\thandle_filter_flowed(mi, sb, &prev);\n+\t\t\t\thandle_filter_flowed(mi, sb, &prev, &have_quoted_cr);\n \t\t\t}\n \t\t\t/*\n \t\t\t * The partial chunk is saved in \"prev\" and will be\n@@ -1091,7 +1105,7 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \t\t\tbreak;\n \t\t}\n \t\tdefault:\n-\t\t\thandle_filter_flowed(mi, line, &prev);\n+\t\t\thandle_filter_flowed(mi, line, &prev, &have_quoted_cr);\n \t\t}\n \n \t\tif (mi->input_error)\n@@ -1100,6 +1114,7 @@ static void handle_body(struct mailinfo *mi, struct strbuf *line)\n \n \tif (prev.len)\n \t\thandle_filter(mi, &prev);\n+\tsummarize_quoted_cr(mi, have_quoted_cr);\n \n \tflush_inbody_header_accum(mi);\n \ndiff --git a/t/t5100-mailinfo.sh b/t/t5100-mailinfo.sh\nindex 147e616533..d8fdda6bea 100755\n--- a/t/t5100-mailinfo.sh\n+++ b/t/t5100-mailinfo.sh\n@@ -228,4 +228,19 @@ test_expect_success 'mailinfo handles unusual header whitespace' '\n \ttest_cmp expect actual\n '\n \n+check_quoted_cr_mail() {\n+\tgit mailinfo -u \"$@\" quoted-cr-msg quoted-cr-patch \\\n+\t\t<\"$DATA/quoted-cr.mbox\" >quoted-cr-info 2>quoted-cr-err &&\n+\ttest_cmp \"expect-cr-msg\" quoted-cr-msg &&\n+\ttest_cmp \"expect-cr-patch\" quoted-cr-patch &&\n+\ttest_cmp \"$DATA/quoted-cr-info\" quoted-cr-info\n+}\n+\n+test_expect_success 'mailinfo warn CR in base64 encoded email' '\n+\tsed \"s/%%/$(printf \\\\015)/\" \"$DATA/quoted-cr-msg\" >expect-cr-msg &&\n+\tsed \"s/%%/$(printf \\\\015)/\" \"$DATA/quoted-cr-patch\" >expect-cr-patch &&\n+\tcheck_quoted_cr_mail &&\n+\tgrep \"quoted CR detected\" quoted-cr-err\n+'\n+\n test_done\ndiff --git a/t/t5100/quoted-cr-info b/t/t5100/quoted-cr-info\nnew file mode 100644\nindex 0000000000..dab2228b70\n--- /dev/null\n+++ b/t/t5100/quoted-cr-info\n@@ -0,0 +1,5 @@\n+Author: A U Thor\n+Email: mail@example.com\n+Subject: sample\n+Date: Mon, 3 Aug 2020 22:40:55 +0700\n+\ndiff --git a/t/t5100/quoted-cr-msg b/t/t5100/quoted-cr-msg\nnew file mode 100644\nindex 0000000000..a148bc7e26\n--- /dev/null\n+++ b/t/t5100/quoted-cr-msg\n@@ -0,0 +1,2 @@\n+On different distro, pytest is suffixed with different patterns.%%\n+%%\ndiff --git a/t/t5100/quoted-cr-patch b/t/t5100/quoted-cr-patch\nnew file mode 100644\nindex 0000000000..580e2bddb8\n--- /dev/null\n+++ b/t/t5100/quoted-cr-patch\n@@ -0,0 +1,22 @@\n+---%%\n+ configure | 2 +-%%\n+ 1 file changed, 1 insertion(+), 1 deletion(-)%%\n+%%\n+diff --git a/configure b/configure%%\n+index db3538b3..f7c1c095 100755%%\n+--- a/configure%%\n++++ b/configure%%\n+@@ -814,7 +814,7 @@ if [ $have_python3 -eq 1 ]; then%%\n+ printf \"Checking for python3 pytest (>= 3.0)... \"%%\n+ conf=$(mktemp)%%\n+ printf \"[pytest]\\nminversion=3.0\\n\" > $conf%%\n+- if pytest-3 -c $conf --version >/dev/null 2>&1; then%%\n++ if \"$python\" -m pytest -c $conf --version >/dev/null 2>&1; then%%\n+ printf \"Yes.\\n\"%%\n+ have_python3_pytest=1%%\n+ else%%\n+-- %%\n+2.28.0%%\n+_______________________________________________\n+example mailing list -- list@example.org\n+To unsubscribe send an email to list-leave@example.org\ndiff --git a/t/t5100/quoted-cr.mbox b/t/t5100/quoted-cr.mbox\nnew file mode 100644\nindex 0000000000..6ea9806a6b\n--- /dev/null\n+++ b/t/t5100/quoted-cr.mbox\n@@ -0,0 +1,22 @@\n+From: A U Thor \n+To: list@example.org\n+Subject: [PATCH v2] sample\n+Date: Mon, 3 Aug 2020 22:40:55 +0700\n+Message-Id: \n+Content-Type: text/plain; charset=\"utf-8\"\n+Content-Transfer-Encoding: base64\n+\n+T24gZGlmZmVyZW50IGRpc3RybywgcHl0ZXN0IGlzIHN1ZmZpeGVkIHdpdGggZGlmZmVyZW50IHBh\n+dHRlcm5zLg0KDQotLS0NCiBjb25maWd1cmUgfCAyICstDQogMSBmaWxlIGNoYW5nZWQsIDEgaW5z\n+ZXJ0aW9uKCspLCAxIGRlbGV0aW9uKC0pDQoNCmRpZmYgLS1naXQgYS9jb25maWd1cmUgYi9jb25m\n+aWd1cmUNCmluZGV4IGRiMzUzOGIzLi5mN2MxYzA5NSAxMDA3NTUNCi0tLSBhL2NvbmZpZ3VyZQ0K\n+KysrIGIvY29uZmlndXJlDQpAQCAtODE0LDcgKzgxNCw3IEBAIGlmIFsgJGhhdmVfcHl0aG9uMyAt\n+ZXEgMSBdOyB0aGVuDQogICAgIHByaW50ZiAiQ2hlY2tpbmcgZm9yIHB5dGhvbjMgcHl0ZXN0ICg+\n+PSAzLjApLi4uICINCiAgICAgY29uZj0kKG1rdGVtcCkNCiAgICAgcHJpbnRmICJbcHl0ZXN0XVxu\n+bWludmVyc2lvbj0zLjBcbiIgPiAkY29uZg0KLSAgICBpZiBweXRlc3QtMyAtYyAkY29uZiAtLXZl\n+cnNpb24gPi9kZXYvbnVsbCAyPiYxOyB0aGVuDQorICAgIGlmICIkcHl0aG9uIiAtbSBweXRlc3Qg\n+LWMgJGNvbmYgLS12ZXJzaW9uID4vZGV2L251bGwgMj4mMTsgdGhlbg0KICAgICAgICAgcHJpbnRm\n+ICJZZXMuXG4iDQogICAgICAgICBoYXZlX3B5dGhvbjNfcHl0ZXN0PTENCiAgICAgZWxzZQ0KLS0g\n+DQoyLjI4LjANCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f\n+CmV4YW1wbGUgbWFpbGluZyBsaXN0IC0tIGxpc3RAZXhhbXBsZS5vcmcKVG8gdW5zdWJzY3JpYmUg\n+c2VuZCBhbiBlbWFpbCB0byBsaXN0LWxlYXZlQGV4YW1wbGUub3JnCg==\n-- \n2.31.1.500.gbc6bbdd36b\n\n"},{"id":"423593","messageId":"3215ea95cf869c8495d95cfd774973c330c14d1d.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"cover.1620148732.git.congdanhqx@gmail.com","subject":"[PATCH v2 3/5] mailinfo: skip quoted CR on user's wish","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:20:00Z","receivedAt":"2021-05-04T17:20:23Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"In previous change, we've turned on warning for quoted CR in base64\nencoded email. Despite those warnings are usually helpful for our users,\nthey may expect quoted CR in their emails.\n\nLet's give them an option to turn off the warning completely.\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n Documentation/git-mailinfo.txt | 18 +++++++++++++++++-\n builtin/mailinfo.c | 8 ++++++--\n mailinfo.c | 21 ++++++++++++++++++++-\n mailinfo.h | 8 ++++++++\n t/t5100-mailinfo.sh | 6 ++++--\n 5 files changed, 55 insertions(+), 6 deletions(-)\n\ndiff --git a/Documentation/git-mailinfo.txt b/Documentation/git-mailinfo.txt\nindex d343f040f5..c776b27515 100644\n--- a/Documentation/git-mailinfo.txt\n+++ b/Documentation/git-mailinfo.txt\n@@ -9,7 +9,7 @@ git-mailinfo - Extracts patch and authorship from a single e-mail message\n SYNOPSIS\n --------\n [verse]\n-'git mailinfo' [-k|-b] [-u | --encoding= | -n] [--[no-]scissors] \n+'git mailinfo' [-k|-b] [-u | --encoding= | -n] [--[no-]scissors] [--quoted-cr=] \n \n \n DESCRIPTION\n@@ -89,6 +89,22 @@ This can be enabled by default with the configuration option mailinfo.scissors.\n --no-scissors::\n \tIgnore scissors lines. Useful for overriding mailinfo.scissors settings.\n \n+--quoted-cr=::\n+\tAction when processes email messages sent with base64 or\n+\tquoted-printable encoding, and the decoded lines end with CR-LF\n+\tinstead of a simple LF.\n++\n+The valid actions are:\n++\n+--\n+*\t`nowarn`: Git will do nothing with this action.\n+*\t`warn`: Git will issue a warning for each message if such CR-LF is\n+\tfound.\n+--\n++\n+The default action could be set by configuration option `mailinfo.quotedCR`.\n+If no such configuration option has been set, `warn` will be used.\n+\n ::\n \tThe commit log message extracted from e-mail, usually\n \texcept the title line which comes from e-mail Subject.\ndiff --git a/builtin/mailinfo.c b/builtin/mailinfo.c\nindex b309badce5..1d600263cb 100644\n--- a/builtin/mailinfo.c\n+++ b/builtin/mailinfo.c\n@@ -9,7 +9,7 @@\n #include \"mailinfo.h\"\n \n static const char mailinfo_usage[] =\n-\t\"git mailinfo [-k | -b] [-m | --message-id] [-u | --encoding= | -n] [--scissors | --no-scissors] < mail >info\";\n+\t\"git mailinfo [-k | -b] [-m | --message-id] [-u | --encoding= | -n] [--scissors | --no-scissors] [--quoted-cr=] < mail >info\";\n \n int cmd_mailinfo(int argc, const char **argv, const char *prefix)\n {\n@@ -43,7 +43,11 @@ int cmd_mailinfo(int argc, const char **argv, const char *prefix)\n \t\t\tmi.use_scissors = 0;\n \t\telse if (!strcmp(argv[1], \"--no-inbody-headers\"))\n \t\t\tmi.use_inbody_headers = 0;\n-\t\telse\n+\t\telse if (skip_prefix(argv[1], \"--quoted-cr=\", &str)) {\n+\t\t\tmi.quoted_cr = mailinfo_parse_quoted_cr_action(str);\n+\t\t\tif (mi.quoted_cr == quoted_cr_invalid_action)\n+\t\t\t\tusage(mailinfo_usage);\n+\t\t} else\n \t\t\tusage(mailinfo_usage);\n \t\targc--; argv++;\n \t}\ndiff --git a/mailinfo.c b/mailinfo.c\nindex 713567f84b..fe7ffd01d0 100644\n--- a/mailinfo.c\n+++ b/mailinfo.c\n@@ -1040,7 +1040,8 @@ static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n \n static void summarize_quoted_cr(struct mailinfo *mi, int have_quoted_cr)\n {\n-\tif (have_quoted_cr)\n+\tif (have_quoted_cr\n+\t && mi->quoted_cr == quoted_cr_warn)\n \t\twarning(\"quoted CR detected\");\n }\n \n@@ -1221,9 +1222,19 @@ int mailinfo(struct mailinfo *mi, const char *msg, const char *patch)\n \treturn mi->input_error;\n }\n \n+enum quoted_cr_action mailinfo_parse_quoted_cr_action(const char *action)\n+{\n+\tif (!strcmp(action, \"nowarn\"))\n+\t\treturn quoted_cr_nowarn;\n+\telse if (!strcmp(action, \"warn\"))\n+\t\treturn quoted_cr_warn;\n+\treturn quoted_cr_invalid_action;\n+}\n+\n static int git_mailinfo_config(const char *var, const char *value, void *mi_)\n {\n \tstruct mailinfo *mi = mi_;\n+\tconst char *str;\n \n \tif (!starts_with(var, \"mailinfo.\"))\n \t\treturn git_default_config(var, value, NULL);\n@@ -1231,6 +1242,13 @@ static int git_mailinfo_config(const char *var, const char *value, void *mi_)\n \t\tmi->use_scissors = git_config_bool(var, value);\n \t\treturn 0;\n \t}\n+\tif (!strcmp(var, \"mailinfo.quotedcr\")) {\n+\t\tgit_config_string(&str, var, value);\n+\t\tmi->quoted_cr = mailinfo_parse_quoted_cr_action(str);\n+\t\tif (mi->quoted_cr == quoted_cr_invalid_action)\n+\t\t\tdie(_(\"bad action '%s' for '%s'\"), str, var);\n+\t\tfree((void *)str);\n+\t}\n \t/* perhaps others here */\n \treturn 0;\n }\n@@ -1243,6 +1261,7 @@ void setup_mailinfo(struct mailinfo *mi)\n \tstrbuf_init(&mi->charset, 0);\n \tstrbuf_init(&mi->log_message, 0);\n \tstrbuf_init(&mi->inbody_header_accum, 0);\n+\tmi->quoted_cr = quoted_cr_warn;\n \tmi->header_stage = 1;\n \tmi->use_inbody_headers = 1;\n \tmi->content_top = mi->content;\ndiff --git a/mailinfo.h b/mailinfo.h\nindex 79b1d6774e..1bcef5a6f3 100644\n--- a/mailinfo.h\n+++ b/mailinfo.h\n@@ -5,6 +5,12 @@\n \n #define MAX_BOUNDARIES 5\n \n+enum quoted_cr_action {\n+\tquoted_cr_nowarn,\n+\tquoted_cr_warn,\n+\tquoted_cr_invalid_action\n+};\n+\n struct mailinfo {\n \tFILE *input;\n \tFILE *output;\n@@ -14,6 +20,7 @@ struct mailinfo {\n \tstruct strbuf email;\n \tint keep_subject;\n \tint keep_non_patch_brackets_in_subject;\n+\tenum quoted_cr_action quoted_cr;\n \tint add_message_id;\n \tint use_scissors;\n \tint use_inbody_headers;\n@@ -39,6 +46,7 @@ struct mailinfo {\n \tint input_error;\n };\n \n+enum quoted_cr_action mailinfo_parse_quoted_cr_action(const char *action);\n void setup_mailinfo(struct mailinfo *);\n int mailinfo(struct mailinfo *, const char *msg, const char *patch);\n void clear_mailinfo(struct mailinfo *);\ndiff --git a/t/t5100-mailinfo.sh b/t/t5100-mailinfo.sh\nindex d8fdda6bea..57b8fc8104 100755\n--- a/t/t5100-mailinfo.sh\n+++ b/t/t5100-mailinfo.sh\n@@ -236,11 +236,13 @@ check_quoted_cr_mail() {\n \ttest_cmp \"$DATA/quoted-cr-info\" quoted-cr-info\n }\n \n-test_expect_success 'mailinfo warn CR in base64 encoded email' '\n+test_expect_success 'mailinfo handle CR in base64 encoded email' '\n \tsed \"s/%%/$(printf \\\\015)/\" \"$DATA/quoted-cr-msg\" >expect-cr-msg &&\n \tsed \"s/%%/$(printf \\\\015)/\" \"$DATA/quoted-cr-patch\" >expect-cr-patch &&\n \tcheck_quoted_cr_mail &&\n-\tgrep \"quoted CR detected\" quoted-cr-err\n+\tgrep \"quoted CR detected\" quoted-cr-err &&\n+\tcheck_quoted_cr_mail --quoted-cr=nowarn &&\n+\ttest_must_be_empty quoted-cr-err\n '\n \n test_done\n-- \n2.31.1.500.gbc6bbdd36b\n\n"},{"id":"423595","messageId":"01d4a4853b8a0b6911a2d7773f836620566b4293.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"cover.1620148732.git.congdanhqx@gmail.com","subject":"[PATCH v2 4/5] mailinfo: strip quoted CR on users' wish","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:20:01Z","receivedAt":"2021-05-04T17:20:25Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"In previous changes, we've turned on warning for quoted CR in base64 or\nquoted-printable email messages. Some projects sees those quoted CR a lot\nand they know that it happens most of the time.\n\nThose projects in question usually fall back to use other tools to handle\npatches when receiving such patches.\n\nLet's help those projects handle those patches by stripping those\nexcessive CR-s.\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n Documentation/git-mailinfo.txt | 1 +\n mailinfo.c | 7 +++++++\n mailinfo.h | 1 +\n t/t5100-mailinfo.sh | 4 ++++\n 4 files changed, 13 insertions(+)\n\ndiff --git a/Documentation/git-mailinfo.txt b/Documentation/git-mailinfo.txt\nindex c776b27515..d700929a46 100644\n--- a/Documentation/git-mailinfo.txt\n+++ b/Documentation/git-mailinfo.txt\n@@ -100,6 +100,7 @@ The valid actions are:\n *\t`nowarn`: Git will do nothing with this action.\n *\t`warn`: Git will issue a warning for each message if such CR-LF is\n \tfound.\n+*\t`strip`: Git will convert those CR-LF to LF.\n --\n +\n The default action could be set by configuration option `mailinfo.quotedCR`.\ndiff --git a/mailinfo.c b/mailinfo.c\nindex fe7ffd01d0..68f4eba72a 100644\n--- a/mailinfo.c\n+++ b/mailinfo.c\n@@ -998,6 +998,11 @@ static void handle_filter_flowed(struct mailinfo *mi, struct strbuf *line,\n \t\t line->buf[len - 2] == '\\r' &&\n \t\t line->buf[len - 1] == '\\n') {\n \t\t\t*have_quoted_cr = 1;\n+\t\t\tif (mi->quoted_cr == quoted_cr_strip) {\n+\t\t\t\tstrbuf_setlen(line, len - 2);\n+\t\t\t\tstrbuf_addch(line, '\\n');\n+\t\t\t\tlen--;\n+\t\t\t}\n \t\t}\n \t\thandle_filter(mi, line);\n \t\treturn;\n@@ -1228,6 +1233,8 @@ enum quoted_cr_action mailinfo_parse_quoted_cr_action(const char *action)\n \t\treturn quoted_cr_nowarn;\n \telse if (!strcmp(action, \"warn\"))\n \t\treturn quoted_cr_warn;\n+\telse if (!strcmp(action, \"strip\"))\n+\t\treturn quoted_cr_strip;\n \treturn quoted_cr_invalid_action;\n }\n \ndiff --git a/mailinfo.h b/mailinfo.h\nindex 1bcef5a6f3..e0e094c311 100644\n--- a/mailinfo.h\n+++ b/mailinfo.h\n@@ -8,6 +8,7 @@\n enum quoted_cr_action {\n \tquoted_cr_nowarn,\n \tquoted_cr_warn,\n+\tquoted_cr_strip,\n \tquoted_cr_invalid_action\n };\n \ndiff --git a/t/t5100-mailinfo.sh b/t/t5100-mailinfo.sh\nindex 57b8fc8104..7559c922c6 100755\n--- a/t/t5100-mailinfo.sh\n+++ b/t/t5100-mailinfo.sh\n@@ -242,6 +242,10 @@ test_expect_success 'mailinfo handle CR in base64 encoded email' '\n \tcheck_quoted_cr_mail &&\n \tgrep \"quoted CR detected\" quoted-cr-err &&\n \tcheck_quoted_cr_mail --quoted-cr=nowarn &&\n+\ttest_must_be_empty quoted-cr-err &&\n+\tsed \"s/%%//\" \"$DATA/quoted-cr-msg\" >expect-cr-msg &&\n+\tsed \"s/%%//\" \"$DATA/quoted-cr-patch\" >expect-cr-patch &&\n+\tcheck_quoted_cr_mail --quoted-cr=strip &&\n \ttest_must_be_empty quoted-cr-err\n '\n \n-- \n2.31.1.500.gbc6bbdd36b\n\n"},{"id":"423596","messageId":"d2052bcea925ec9e7d780e99e5b5e3bb1b9bd056.1620148732.git.congdanhqx@gmail.com","threadId":"55533","inReplyTo":"cover.1620148732.git.congdanhqx@gmail.com","subject":"[PATCH v2 5/5] am: learn to process quoted lines that ends with CRLF","fromName":"Đoàn Trần Công Danh","fromEmail":"congdanhqx@gmail.com","sentAt":"2021-05-04T17:20:02Z","receivedAt":"2021-05-04T17:20:26Z","isPatch":true,"sender":{"key":"congdanhqx@gmail.com","avatar":"https://avatars.githubusercontent.com/u/42673067?v=4"},"body":"In previous changes, mailinfo has learnt to process lines that decoded\nfrom base64 or quoted-printable and ends with CRLF.\n\nLet's teach \"am\" that new option, too.\n\nSigned-off-by: Đoàn Trần Công Danh \n---\n Documentation/git-am.txt | 4 +++\n builtin/am.c | 56 ++++++++++++++++++++++++++++++++++++++++\n mailinfo.h | 1 +\n t/t4258-am-quoted-cr.sh | 37 ++++++++++++++++++++++++++\n t/t4258/mbox | 12 +++++++++\n 5 files changed, 110 insertions(+)\n create mode 100755 t/t4258-am-quoted-cr.sh\n create mode 100644 t/t4258/mbox\n\ndiff --git a/Documentation/git-am.txt b/Documentation/git-am.txt\nindex decd8ae122..8714dfcb76 100644\n--- a/Documentation/git-am.txt\n+++ b/Documentation/git-am.txt\n@@ -15,6 +15,7 @@ SYNOPSIS\n \t [--whitespace=