)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":1000002,"name":"cron2","display_name":"Gert Doering","email":"gert@greenie.muc.de","username":"cron2"},"change_message_id":"a68253e3a9fe94267be93d79241164c8cb5fc4ab","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":2,"id":"64eb26b9_7739412a","updated":"2024-05-08 13:49:12.000000000","message":"Does not work for me yet.\n\nApplied this to a fresh git checkout on fbsd14, which has pkg-config and lzo2.pc, but it still failed.\n\nRelevant parts from config.log:\n\n```\nconfigure:18781: checking for LZO\nconfigure:18788: $PKG_CONFIG --exists --print-errors \"lzo2\"\nconfigure:18791: $? \u003d 0\nconfigure:18805: $PKG_CONFIG --exists --print-errors \"lzo2\"\nconfigure:18808: $? \u003d 0\nconfigure:18848: result: yes\nconfigure:18850: checking for lzo1x_1_15_compress in -llzo2\nconfigure:18879: cc -o conftest -g -O2 -std\u003dc99   conftest.c -llzo2   -lnv \u003e\u00265\nld: error: unable to find library -llzo2\n```\n\nand from pkgconf output\n```\ngert@fbsd14bb $ pkg-config --cflags lzo2\n-I/usr/local/include/lzo\ngert@fbsd14bb $ pkg-config --libs lzo2\n-L/usr/local/lib -llzo2","commit_id":"defa3e6af7ac2f5366de5f15e716f477b9ff4f78"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"d725fbb944b1503dbecfe172370cd8656298d323","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"6cd27820_e0eccc39","in_reply_to":"3c2239ca_9918530b","updated":"2024-06-21 13:03:12.000000000","message":"Done","commit_id":"defa3e6af7ac2f5366de5f15e716f477b9ff4f78"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"b3d219fc76284b9b65ab7fc97a94268ee88344d9","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":2,"id":"3c2239ca_9918530b","in_reply_to":"64eb26b9_7739412a","updated":"2024-05-08 14:21:10.000000000","message":"Yeah, I tried to include the -llzo check in pkg-config, which just made no sense. And it did not work for the case where LZO_CFLAGS were important anyway. I now decided to just remove it completely.","commit_id":"defa3e6af7ac2f5366de5f15e716f477b9ff4f78"},{"author":{"_account_id":1000002,"name":"cron2","display_name":"Gert Doering","email":"gert@greenie.muc.de","username":"cron2"},"change_message_id":"df3263813f00670986cde21ab226a214e744abd5","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"47c332c0_de2b55fa","updated":"2024-06-02 16:56:06.000000000","message":"close...\n\nfbsd14, master + this patch, \"configure\" with no further arguments\n\n```\nchecking for lzo2... yes\nchecking for lzo/lzoutil.h... no\nchecking for lzoutil.h... no\nconfigure: error: lzoutil.h is missing\n$ locate lzoutil.h\n/usr/local/include/lzo/lzoutil.h\n```\n\nand this might actually be a broken `.pc`, as the failure is\n\n```\nconfigure:18961: result: no\nconfigure:18969: checking for lzoutil.h\nconfigure:18969: cc -c -g -O2 -std\u003dc99 -I/usr/local/include/lzo  conftest.c \u003e\u00265\nIn file included from conftest.c:162:\n/usr/local/include/lzo/lzoutil.h:33:10: fatal error: \u0027lzo/lzoconf.h\u0027 file not fo\nund\n#include \u003clzo/lzoconf.h\u003e\n         ^~~~~~~~~~~~~~~\n```\n\nso we *should* `-I/usr/local/include`.\n\nOn the other side, `lzoutil.h` has this comment\n\n```\n/***********************************************************************\n// LZO-v1 deprecated macros (which were used in the old example programs)\n//\n// THIS FILE IS DEPRECATED. DO NOT USE.\n************************************************************************/\n```\n\nso maybe we shouldn\u0027t check for this one at all?  (we use `lzo_malloc()` and `lzo_free()` though, but those are simple defines to plain malloc/free).\n\nMmmh.","commit_id":"b99a380e81a1195e57627e6941a5de80630b3b07"},{"author":{"_account_id":1000002,"name":"cron2","display_name":"Gert Doering","email":"gert@greenie.muc.de","username":"cron2"},"change_message_id":"5e47ba70cd91ae5e0b1003e35aa328d67ace228d","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"5fdd0545_0f2c87d8","in_reply_to":"47c332c0_de2b55fa","updated":"2024-06-26 12:29:20.000000000","message":"I think the pc is broken... everything LZO itself includes uses \u003clzo/lzoconf.h\u003e... but it\u0027s the same (`Cflags: -I${includedir}/lzo`) on gentoo... where it just happens to work because `/usr/include` is in the include path anyway.  Gah.","commit_id":"b99a380e81a1195e57627e6941a5de80630b3b07"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"c02cbf2187e229d1aacf67ee6fa51d4d00dff2a8","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"eb9fbc67_9cd75928","in_reply_to":"47c332c0_de2b55fa","updated":"2024-06-04 11:09:50.000000000","message":"Okay, I can prepare a separate patch that gets rid of lzoutil.h completely. But as you said, the .pc file seems at fault here.","commit_id":"b99a380e81a1195e57627e6941a5de80630b3b07"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"d8dd4f065e284d7c322f554ee986fde4520ad822","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"b4f67d04_79e89475","in_reply_to":"5fdd0545_0f2c87d8","updated":"2024-06-26 13:51:11.000000000","message":"Done","commit_id":"b99a380e81a1195e57627e6941a5de80630b3b07"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"d725fbb944b1503dbecfe172370cd8656298d323","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"729c8dac_6956a092","in_reply_to":"eb9fbc67_9cd75928","updated":"2024-06-21 13:03:12.000000000","message":"Done","commit_id":"b99a380e81a1195e57627e6941a5de80630b3b07"},{"author":{"_account_id":1000002,"name":"cron2","display_name":"Gert Doering","email":"gert@greenie.muc.de","username":"cron2"},"change_message_id":"5e47ba70cd91ae5e0b1003e35aa328d67ace228d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"cb590536_3257d687","updated":"2024-06-26 12:29:20.000000000","message":"Tested on FreeBSD 14, \"configure with no option\" -\u003e succeeds \u0026 builds fine (LZO in /usr/local/..., make using `-I/usr/local/include/lzo -I/usr/local/include` because that\u0027s where cmocka and lz4 are living).  Disabling LZ4 makes compilation succeed with only `-I/usr/local/include/lzo`.  This is good :-)\n\nAll the buildbots still succeed, which means this hasn\u0027t broken explicitly setting \"whatever we used before\".\n\noi2024 has lzo in `/usr/include/`, so, boring case.\n\nobsd75 has lzo in `/usr/local/include/lzo`, patched configure works, compile works.  With `--disable-lz4` there is no mention of `-I/usr/local/include` anymore, and it still succeeds ;-)\n\nnbsd81 has `/usr/pkg/include/lzo/` and the patched configure finds it fine, and then barfs on\n\n```\nconfigure:18379: checking for lzo1x.h\nconfigure:18379: gcc -c -g -O2 -std\u003dc99 -I/usr/pkg/include/lzo  conftest.c \u003e\u00265\nIn file included from conftest.c:136:0:\n/usr/pkg/include/lzo/lzodefs.h:1035:6: error: #error \"unexpected configuration -\n check your compiler defines\"\n #    error \"unexpected configuration - check your compiler defines\"\n      ^\n```\n\nwhich is something weird about UINT_MAX testing against LZO_0xffffffffL.  Not sure how to fix that.  It does not break on the buildbot as that one has a symlink from `/usr/include/lzo -\u003e /usr/pkg/include/lzo` (to apease older source trees) - with that, the configure change works fine.  Without, it bombs.  I would still tend to progress the patch, unless someone has a good idea how to tackle this...","commit_id":"d36c9b7ea1907b66d68081867b6d62491846d03b"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"d8dd4f065e284d7c322f554ee986fde4520ad822","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"1108e281_4aaa48a3","in_reply_to":"cb590536_3257d687","updated":"2024-06-26 13:51:11.000000000","message":"We found out this is because lzodefs.h relies on lzoconf.h to include limits.h. So we need to imitate that.","commit_id":"d36c9b7ea1907b66d68081867b6d62491846d03b"},{"author":{"_account_id":1000002,"name":"cron2","display_name":"Gert Doering","email":"gert@greenie.muc.de","username":"cron2"},"change_message_id":"66c8e537cac1c6ecf33c06dedc161cdcd59cd954","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":8,"id":"009a0703_db718c26","updated":"2024-06-26 13:51:15.000000000","message":"this should appease all systems :-) - the to-be-expected fails are server-null, so (at least that\u0027s what I expect) not a fault of this patch","commit_id":"9054c7d85f18464c939c64dbb3d4b33109e5d9af"}],"configure.ac":[{"author":{"_account_id":1000003,"name":"plaisthos","display_name":"Arne Schwabe","email":"arne-openvpn@rfc2549.org","username":"plaisthos"},"change_message_id":"96bdbc8e5923a899b532f0a444a07f6d11c7481b","unresolved":true,"context_lines":[{"line_number":1145,"context_line":"AC_ARG_VAR([LZO_LIBS], [linker flags for lzo])"},{"line_number":1146,"context_line":"if test -z \"${LZO_LIBS}\"; then"},{"line_number":1147,"context_line":"    have_lzo\u003d\"no\""},{"line_number":1148,"context_line":"    if test -z \"${LZO_CFLAGS}\" -a -z \"${LZO_LIBS}\"; then"},{"line_number":1149,"context_line":"\t# if the user did not explicitly specify flags, try to autodetect"},{"line_number":1150,"context_line":"\tPKG_CHECK_MODULES([LZO],"},{"line_number":1151,"context_line":"\t\t\t  [lzo2],"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"269f68bb_0cc07361","line":1148,"updated":"2024-05-08 13:49:37.000000000","message":"Isn\u0027t this -z LZO_LIBS always true since it is under the -z LZO_LIBS check anyway?","commit_id":"defa3e6af7ac2f5366de5f15e716f477b9ff4f78"},{"author":{"_account_id":1000001,"name":"flichtenheld","display_name":"Frank Lichtenheld","email":"frank@lichtenheld.com","username":"flichtenheld","status":"OpenVPN Inc."},"change_message_id":"b3d219fc76284b9b65ab7fc97a94268ee88344d9","unresolved":false,"context_lines":[{"line_number":1145,"context_line":"AC_ARG_VAR([LZO_LIBS], [linker flags for lzo])"},{"line_number":1146,"context_line":"if test -z \"${LZO_LIBS}\"; then"},{"line_number":1147,"context_line":"    have_lzo\u003d\"no\""},{"line_number":1148,"context_line":"    if test -z \"${LZO_CFLAGS}\" -a -z \"${LZO_LIBS}\"; then"},{"line_number":1149,"context_line":"\t# if the user did not explicitly specify flags, try to autodetect"},{"line_number":1150,"context_line":"\tPKG_CHECK_MODULES([LZO],"},{"line_number":1151,"context_line":"\t\t\t  [lzo2],"}],"source_content_type":"application/octet-stream","patch_set":2,"id":"1166fc22_a203022f","line":1148,"in_reply_to":"269f68bb_0cc07361","updated":"2024-05-08 14:21:10.000000000","message":"Done","commit_id":"defa3e6af7ac2f5366de5f15e716f477b9ff4f78"}]}