{"thread":{"id":"4666","subject":"[RFC] git --trace: trace command execution","startedAt":"2006-06-25T10:57:09Z","lastAt":"2006-06-29T18:06:21Z","messageCount":10,"participants":["Matthias Lederhofer","Junio C Hamano","Johannes Schindelin","Petr Baudis","Jakub Narebski"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"22543","messageId":"E1FuSIf-0004jK-Tp@moooo.ath.cx","threadId":"4666","inReplyTo":null,"subject":"[RFC] git --trace: trace command execution","fromName":"Matthias Lederhofer","fromEmail":"matled@gmx.net","sentAt":"2006-06-25T10:57:09Z","receivedAt":"2006-06-25T10:57:09Z","isPatch":false,"sender":{"key":"matled@gmx.net","avatar":null},"body":"Show parameters to execve/builtin-cmds before executing them. This\nversion does not yet have a parameter --trace to git to enable this I\njust want to get feedback first :)\n\nI think this is quite useful to debug what is going on since a command\nmay be another program (shell/python/perl/.. script etc) or just an\nalias for a internal command. Before that many commands became\nbuilt-ins this was quite easy to do with strace.\n\nExample:\n% git repo-config alias.showtag\ntrace: exec: /home/matled/local/stow/git/bin/git-repo-config\nalias.showtag\n% git showtag v1.4.1-rc1 > /dev/null\ntrace: exec: /home/matled/local/stow/git/bin/git-showtag v1.4.1-rc1\ntrace: exec failed: No such file or directory\ntrace: built-in command: git cat-file tag v1.4.1-rc1\n% git showtag 'a b \"c\" `d` $e \\f'\ntrace: exec: /home/matled/local/stow/git/bin/git-showtag \"a b \\\"c\\\" \\`d\\` \\$e \\\\f\"\ntrace: exec failed: No such file or directory\ntrace: built-in command: git cat-file tag \"a b \\\"c\\\" \\`d\\` \\$e \\\\f\"\nfatal: Not a valid object name a b \"c\" `d` $e \\f\n% git cat-file tag \"a b \\\"c\\\" \\`d\\` \\$e \\\\f\"\ntrace: built-in command: git cat-file tag \"a b \\\"c\\\" \\`d\\` \\$e \\\\f\"\nfatal: Not a valid object name a b \"c\" `d` $e \\f\n\nprint_shell_escape will escape the arguments to be used as strings in\nthe shell to prevent ambiguity with spaces and other special\ncharacters and make them copy-and-pastable.\n\n---\n exec_cmd.c | 42 ++++++++++++++++++++++++++++++++++++++++++\n exec_cmd.h | 1 +\n git.c | 10 ++++++++++\n 3 files changed, 53 insertions(+), 0 deletions(-)\n\ndiff --git a/exec_cmd.c b/exec_cmd.c\nindex c1539d1..85afbf3 100644\n--- a/exec_cmd.c\n+++ b/exec_cmd.c\n@@ -87,6 +87,16 @@ int execv_git_cmd(const char **argv)\n \t\t\tbreak;\n \t\t}\n \n+\t\tfprintf(stderr, \"trace: exec: \");\n+\t\tprint_shell_escape(stderr, git_command);\n+\t\tconst char **p = argv;\n+\t\twhile (*(++p)) {\n+\t\t\tputc(' ', stderr);\n+\t\t\tprint_shell_escape(stderr, *p);\n+\t\t}\n+\t\tputc('\\n', stderr);\n+\t\tfflush(stderr);\n+\n \t\t/* argv[0] must be the git command, but the argv array\n \t\t * belongs to the caller, and my be reused in\n \t\t * subsequent loop iterations. Save argv[0] and\n@@ -98,6 +108,8 @@ int execv_git_cmd(const char **argv)\n \n \t\t/* execve() can only ever return if it fails */\n \t\texecve(git_command, (char **)argv, environ);\n+\t\tfprintf(stderr, \"trace: exec failed: %s\\n\", strerror(errno));\n+\t\tfflush(stderr);\n \n \t\targv[0] = tmp;\n \t}\n@@ -128,3 +140,33 @@ int execl_git_cmd(const char *cmd,...)\n \targv[argc] = NULL;\n \treturn execv_git_cmd(argv);\n }\n+\n+void print_shell_escape(FILE *stream, const char *s)\n+{\n+\tconst char *c = s;\n+\tshort int quote = 0;\n+\twhile (*c) {\n+\t\tif (*c == '\"' || *c == '`' || *c == '$' || *c == '\\\\' ||\n+\t\t\tisspace(*c))\n+\t\t{\n+\t\t\tquote = 1;\n+\t\t\tbreak;\n+\t\t}\n+\t\t++c;\n+\t}\n+\n+\tif (!quote) {\n+\t\tfputs(s, stream);\n+\t\treturn;\n+\t}\n+\n+\tputc('\"', stream);\n+\tc = s;\n+\twhile (*c) {\n+\t\tif (*c == '\"' || *c == '`' || *c == '$' || *c == '\\\\')\n+\t\t\tputc('\\\\', stream);\n+\t\tputc(*c, stream);\n+\t\t++c;\n+\t}\n+\tputc('\"', stream);\n+}\ndiff --git a/exec_cmd.h b/exec_cmd.h\nindex 989621f..8b237fa 100644\n--- a/exec_cmd.h\n+++ b/exec_cmd.h\n@@ -5,6 +5,7 @@ extern void git_set_exec_path(const char\n extern const char* git_exec_path(void);\n extern int execv_git_cmd(const char **argv); /* NULL terminated */\n extern int execl_git_cmd(const char *cmd, ...);\n+extern void print_shell_escape(FILE *stream, const char *s);\n \n \n #endif /* __GIT_EXEC_CMD_H_ */\ndiff --git a/git.c b/git.c\nindex 94e9a4a..361fb25 100644\n--- a/git.c\n+++ b/git.c\n@@ -198,6 +198,16 @@ static void handle_internal_command(int \n \t\tstruct cmd_struct *p = commands+i;\n \t\tif (strcmp(p->cmd, cmd))\n \t\t\tcontinue;\n+\n+\t\tfprintf(stderr, \"trace: built-in command: git\");\n+\t\tint i;\n+\t\tfor (i = 0; i < argc; ++i) {\n+\t\t\tputc(' ', stderr);\n+\t\t\tprint_shell_escape(stderr, argv[i]);\n+\t\t}\n+\t\tputc('\\n', stderr);\n+\t\tfflush(stderr);\n+\n \t\texit(p->fn(argc, argv, envp));\n \t}\n }\n-- \n1.4.GIT\n"},{"id":"22550","messageId":"7v3bdtv4h3.fsf@assigned-by-dhcp.cox.net","threadId":"4666","inReplyTo":"E1FuSIf-0004jK-Tp@moooo.ath.cx","subject":"Re: [RFC] git --trace: trace command execution","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2006-06-25T11:50:48Z","receivedAt":"2006-06-25T11:50:48Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Matthias Lederhofer writes:\n\n> Show parameters to execve/builtin-cmds before executing them. This\n> version does not yet have a parameter --trace to git to enable this I\n> just want to get feedback first :)\n>\n> I think this is quite useful to debug what is going on since a command\n> may be another program (shell/python/perl/.. script etc) or just an\n> alias for a internal command. Before that many commands became\n> built-ins this was quite easy to do with strace.\n\nInteresting. Debugging one's alias entries would be helped with\nthis I would imagine, and for that you would want something like\nthis:\n\n> Example:\n> % git showtag v1.4.1-rc1 > /dev/null\n> trace: exec: /home/matled/local/stow/git/bin/git-showtag v1.4.1-rc1\n> trace: exec failed: No such file or directory\n* trace: expanded alias \"showtag\" => \"cat-file tag\"\n> trace: built-in command: git cat-file tag v1.4.1-rc1\n\nBy the way \"git cat-file -p\" or \"git verify-tag -v\" might be\nmore pleasant to view a tag since they make the tagger timestamp\nhuman readable.\n\n> print_shell_escape will escape the arguments to be used as strings in\n> the shell to prevent ambiguity with spaces and other special\n> characters and make them copy-and-pastable.\n\nMight be worth reusing quote.c::sq_quote(), perhaps?\n"},{"id":"22554","messageId":"E1FuU5a-0000u3-Bc@moooo.ath.cx","threadId":"4666","inReplyTo":"7v3bdtv4h3.fsf@assigned-by-dhcp.cox.net","subject":"Re: [RFC] git --trace: trace command execution","fromName":"Matthias Lederhofer","fromEmail":"matled@gmx.net","sentAt":"2006-06-25T12:51:46Z","receivedAt":"2006-06-25T12:51:46Z","isPatch":false,"sender":{"key":"matled@gmx.net","avatar":null},"body":"> Interesting. Debugging one's alias entries would be helped with\n> this I would imagine, and for that you would want something like\n> this:\n> \n> > Example:\n> > % git showtag v1.4.1-rc1 > /dev/null\n> > trace: exec: /home/matled/local/stow/git/bin/git-showtag v1.4.1-rc1\n> > trace: exec failed: No such file or directory\n> * trace: expanded alias \"showtag\" => \"cat-file tag\"\nThat's a good idea, I'll integrate that.\n\n> By the way \"git cat-file -p\" or \"git verify-tag -v\" might be\n> more pleasant to view a tag since they make the tagger timestamp\n> human readable.\nOk, yesterday I was searching for something to see the annotation of a\ntag. verify-tag -v looks quite much like that, is there any other way\nto read this? Or in general: how do I work with tags? I want to build\na version tagged as v1.2 so currently I'll do\n> git checkout -b 1.2 v1.2\nand built it. But then I've to type the version number twice (typing\nit once is annoying enough :)) and I've to type it once more to get\nthe tag annotation.\n\n> Might be worth reusing quote.c::sq_quote(), perhaps?\nOh, sure, did not know about this. This would result in a loop of\nmalloc'ing memory for the buffer. Is this ok? Or should I add a\nfunction like sq_quote which takes a stream and writes to it?\n\nSo for the --trace part I think an environment variable GIT_TRACE is\nmore suitable for this because children inherit this. So running git\nstatus will show what internal commands the shell script uses.\nOtherwise I see no way to pass the --trace option down because an\nextern program like git-status, git-annotate etc will not accept\nparameters which can be passed to `git'.\n"},{"id":"22558","messageId":"E1FuV62-0004Jd-Ve@moooo.ath.cx","threadId":"4666","inReplyTo":"7v3bdtv4h3.fsf@assigned-by-dhcp.cox.net","subject":"[PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Matthias Lederhofer","fromEmail":"matled@gmx.net","sentAt":"2006-06-25T13:56:18Z","receivedAt":"2006-06-25T13:56:18Z","isPatch":true,"sender":{"key":"matled@gmx.net","avatar":null},"body":"With the environment variable GIT_TRACE set git will show\n - alias expansion\n - built-in command execution\n - external command execution\non stderr.\n\nSigned-off-by: Matthias Lederhofer \n---\nExamples:\n% GIT_TRACE=1 git status > /dev/null\ntrace: exec: '/home/matled/local/stow/git/bin/git-status'\ntrace: built-in: git 'rev-parse' '--git-dir'\ntrace: built-in: git 'rev-parse' '--show-cdup'\ntrace: built-in: git 'update-index' '-q' '--unmerged' '--refresh'\ntrace: built-in: git 'diff-index' '-M' '--cached' '--name-status' \\\n '--diff-filter=MDTCRA' 'HEAD'\ntrace: built-in: git 'diff-files' '--name-status'\ntrace: built-in: git 'ls-files' '-z' '--others' '--directory' \\\n '--no-empty-directory' '--exclude-from=.git/info/exclude' \\\n '--exclude-per-directory=.gitignore'\n\nThis seems to be quite useful to find out how tools do things and for\ndebugging.\n\n% GIT_TRACE=1 git showtag v1.4.0 > /dev/null\ntrace: exec: '/home/matled/local/stow/git/bin/git-showtag' 'v1.4.0'\ntrace: exec failed: No such file or directory\ntrace: alias expansion: showtag => 'cat-file' 'tag'\ntrace: built-in: git 'cat-file' 'tag' 'v1.4.0'\n\n% GIT_TRACE=1 git l > /dev/null\ntrace: exec: '/home/matled/local/stow/git/bin/git-l'\ntrace: exec failed: No such file or directory\ntrace: alias expansion: l => 'log' '-p'\ntrace: built-in: git 'log' '-p'\n\n Documentation/git.txt | 7 +++++++\n exec_cmd.c | 19 +++++++++++++++++++\n git.c | 25 +++++++++++++++++++++++++\n quote.c | 17 +++++++++++++++++\n quote.h | 1 +\n 5 files changed, 69 insertions(+), 0 deletions(-)\n\ndiff --git a/Documentation/git.txt b/Documentation/git.txt\nindex 51f20c6..603c474 100644\n--- a/Documentation/git.txt\n+++ b/Documentation/git.txt\n@@ -615,6 +615,13 @@ git Diffs\n \tgitlink:git-diff-files[1];\n \tgitlink:git-diff-tree[1]\n \n+other\n+~~~~~\n+'GIT_TRACE'::\n+\tIf this variable is set git will print `trace:` messages on\n+\tstderr telling about alias expansion, built-in command\n+\texecution and external command execution.\n+\n Discussion[[Discussion]]\n ------------------------\n include::README[]\ndiff --git a/exec_cmd.c b/exec_cmd.c\nindex c1539d1..f2133ec 100644\n--- a/exec_cmd.c\n+++ b/exec_cmd.c\n@@ -1,5 +1,6 @@\n #include \"cache.h\"\n #include \"exec_cmd.h\"\n+#include \"quote.h\"\n #define MAX_ARGS\t32\n \n extern char **environ;\n@@ -96,9 +97,27 @@ int execv_git_cmd(const char **argv)\n \t\ttmp = argv[0];\n \t\targv[0] = git_command;\n \n+\t\tif (getenv(\"GIT_TRACE\")) {\n+\t\t\tfputs(\"trace: exec:\", stderr);\n+\t\t\tconst char **p = argv;\n+\t\t\twhile (*p) {\n+\t\t\t\tfputc(' ', stderr);\n+\t\t\t\tsq_quote_print(stderr, *p);\n+\t\t\t\t++p;\n+\t\t\t}\n+\t\t\tputc('\\n', stderr);\n+\t\t\tfflush(stderr);\n+\t\t}\n+\n \t\t/* execve() can only ever return if it fails */\n \t\texecve(git_command, (char **)argv, environ);\n \n+\t\tif (getenv(\"GIT_TRACE\")) {\n+\t\t\tfprintf(stderr, \"trace: exec failed: %s\\n\",\n+\t\t\t\tstrerror(errno));\n+\t\t\tfflush(stderr);\n+\t\t}\n+\n \t\targv[0] = tmp;\n \t}\n \treturn -1;\ndiff --git a/git.c b/git.c\nindex 94e9a4a..8b060e8 100644\n--- a/git.c\n+++ b/git.c\n@@ -11,6 +11,7 @@ #include \n #include \"git-compat-util.h\"\n #include \"exec_cmd.h\"\n #include \"cache.h\"\n+#include \"quote.h\"\n \n #include \"builtin.h\"\n \n@@ -119,6 +120,18 @@ static int handle_alias(int *argcp, cons\n \t\t\tif (!strcmp(alias_command, new_argv[0]))\n \t\t\t\tdie(\"recursive alias: %s\", alias_command);\n \n+\t\t\tif (getenv(\"GIT_TRACE\")) {\n+\t\t\t\tint i;\n+\t\t\t\tfprintf(stderr, \"trace: alias expansion: %s =>\",\n+\t\t\t\t\talias_command);\n+\t\t\t\tfor (i = 0; i < count; ++i) {\n+\t\t\t\t\tfputc(' ', stderr);\n+\t\t\t\t\tsq_quote_print(stderr, new_argv[i]);\n+\t\t\t\t}\n+\t\t\t\tfputc('\\n', stderr);\n+\t\t\t\tfflush(stderr);\n+\t\t\t}\n+\n \t\t\t/* insert after command name */\n \t\t\tif (*argcp > 1) {\n \t\t\t\tnew_argv = realloc(new_argv, sizeof(char*) *\n@@ -198,6 +211,18 @@ static void handle_internal_command(int \n \t\tstruct cmd_struct *p = commands+i;\n \t\tif (strcmp(p->cmd, cmd))\n \t\t\tcontinue;\n+\n+\t\tif (getenv(\"GIT_TRACE\")) {\n+\t\t\tint i;\n+\t\t\tfprintf(stderr, \"trace: built-in: git\");\n+\t\t\tfor (i = 0; i < argc; ++i) {\n+\t\t\t\tfputc(' ', stderr);\n+\t\t\t\tsq_quote_print(stderr, argv[i]);\n+\t\t\t}\n+\t\t\tputc('\\n', stderr);\n+\t\t\tfflush(stderr);\n+\t\t}\n+\n \t\texit(p->fn(argc, argv, envp));\n \t}\n }\ndiff --git a/quote.c b/quote.c\nindex dcc2326..2ca18e8 100644\n--- a/quote.c\n+++ b/quote.c\n@@ -45,6 +45,23 @@ size_t sq_quote_buf(char *dst, size_t n,\n \treturn len;\n }\n \n+void sq_quote_print(FILE *stream, const char *src)\n+{\n+\tchar c;\n+\n+\tfputc('\\'', stream);\n+\twhile ((c = *src++)) {\n+\t\tif (need_bs_quote(c)) {\n+\t\t\tfputs(\"'\\\\\", stream);\n+\t\t\tfputc(c, stream);\n+\t\t\tfputc('\\'', stream);\n+\t\t} else {\n+\t\t\tfputc(c, stream);\n+\t\t}\n+\t}\n+\tfputc('\\'', stream);\n+}\n+\n char *sq_quote(const char *src)\n {\n \tchar *buf;\ndiff --git a/quote.h b/quote.h\nindex c1ab378..fc5481e 100644\n--- a/quote.h\n+++ b/quote.h\n@@ -29,6 +29,7 @@ #include \n */\n \n extern char *sq_quote(const char *src);\n+extern void sq_quote_print(FILE *stream, const char *src);\n extern size_t sq_quote_buf(char *dst, size_t n, const char *src);\n \n /* This unwraps what sq_quote() produces in place, but returns\n-- \n1.4.1.rc1.gfa6c9-dirty\n"},{"id":"22561","messageId":"Pine.LNX.4.63.0606251607090.29667@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"4666","inReplyTo":"E1FuV62-0004Jd-Ve@moooo.ath.cx","subject":"Re: [PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2006-06-25T14:11:48Z","receivedAt":"2006-06-25T14:11:48Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Sun, 25 Jun 2006, Matthias Lederhofer wrote:\n\n> With the environment variable GIT_TRACE set git will show\n> - alias expansion\n> - built-in command execution\n> - external command execution\n> on stderr.\n\nNice.\n\nCiao,\nDscho\n\nP.S.: Now we only have to convert all \"git-\" invocations in the scripts to\n\"git \" invocations so we can benefit from it. But that would mean two \nforks instead of one for the non-builtins. Hmm.\n"},{"id":"22562","messageId":"20060625142203.GF21864@pasky.or.cz","threadId":"4666","inReplyTo":"Pine.LNX.4.63.0606251607090.29667@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: [PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2006-06-25T14:22:03Z","receivedAt":"2006-06-25T14:22:03Z","isPatch":true,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":" Hi,\n\nDear diary, on Sun, Jun 25, 2006 at 04:11:48PM CEST, I got a letter\nwhere Johannes Schindelin said that...\n> P.S.: Now we only have to convert all \"git-\" invocations in the scripts to\n> \"git \" invocations so we can benefit from it. But that would mean two \n> forks instead of one for the non-builtins. Hmm.\n\n actually this is a myth - not two fork()s since 'git' has no reason to\nfork; only two execve()s in line after a single fork().\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nA person is just about as big as the things that make them angry.\n"},{"id":"22570","messageId":"E1FuXBk-0001SG-3n@moooo.ath.cx","threadId":"4666","inReplyTo":"Pine.LNX.4.63.0606251607090.29667@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: [PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Matthias Lederhofer","fromEmail":"matled@gmx.net","sentAt":"2006-06-25T16:10:20Z","receivedAt":"2006-06-25T16:10:20Z","isPatch":true,"sender":{"key":"matled@gmx.net","avatar":null},"body":"> Hi,\n> \n> On Sun, 25 Jun 2006, Matthias Lederhofer wrote:\n> \n> > With the environment variable GIT_TRACE set git will show\n> > - alias expansion\n> > - built-in command execution\n> > - external command execution\n> > on stderr.\n> \n> Nice.\nThanks :)\n\n> P.S.: Now we only have to convert all \"git-\" invocations in the scripts to\n> \"git \" invocations so we can benefit from it. But that would mean two \n> forks instead of one for the non-builtins. Hmm.\n\nWhy do we not use this policy:\n\ngit-* is guaranteed to be the normal command without any strange alias\nexpansion, default parameters or something else a script does not like\nto be changed in the commands. So all scripts use git-*, this will\nprevent a double exec. The path to git-* should be obtained using git\n--exec-path in the beginnig.\n\ngit is to be used by the user if he wants aliases, default\nparameters and other fancy stuff.\n\nUsing this policy the user can always get the normal behaviour and it\nis possible to shadown built-in commands etc.\n"},{"id":"22581","messageId":"Pine.LNX.4.63.0606260129410.29667@wbgn013.biozentrum.uni-wuerzburg.de","threadId":"4666","inReplyTo":"E1FuXBk-0001SG-3n@moooo.ath.cx","subject":"Re: [PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2006-06-25T23:30:42Z","receivedAt":"2006-06-25T23:30:42Z","isPatch":true,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Sun, 25 Jun 2006, Matthias Lederhofer wrote:\n\n> > P.S.: Now we only have to convert all \"git-\" invocations in the scripts to\n> > \"git \" invocations so we can benefit from it. But that would mean two \n> > forks instead of one for the non-builtins. Hmm.\n> \n> Why do we not use this policy:\n> \n> git-* is guaranteed to be the normal command without any strange alias\n> expansion, default parameters or something else a script does not like\n> to be changed in the commands. So all scripts use git-*, this will\n> prevent a double exec. The path to git-* should be obtained using git\n> --exec-path in the beginnig.\n\nThat still leaves my problem: GIT_TRACE=1 on scripts is incomplete.\n\nCiao,\nDscho\n"},{"id":"22748","messageId":"E1Fveg1-0006u8-D7@moooo.ath.cx","threadId":"4666","inReplyTo":"Pine.LNX.4.63.0606260129410.29667@wbgn013.biozentrum.uni-wuerzburg.de","subject":"Re: [PATCH] GIT_TRACE: show which built-in/external commands are executed","fromName":"Matthias Lederhofer","fromEmail":"matled@gmx.net","sentAt":"2006-06-28T18:22:13Z","receivedAt":"2006-06-28T18:22:13Z","isPatch":true,"sender":{"key":"matled@gmx.net","avatar":null},"body":"> That still leaves my problem: GIT_TRACE=1 on scripts is incomplete.\nOk, I see what you mean.\n\nThis actually affects all scripts which are called as git-foo instead\nof git foo (but built-in commands show up anyway). So I'd add a\nwarning to the documentation in which cases GIT_TRACE will not show\nthat a command is executed.\nIn the git repository I found 47 shell scripts (git-* without the\nheader file) which would be affected. Searching for all occurences of\ngit-(one of those shell-scripts) I found 50 lines of code which use\nit. If there is any interest in changing this I can try to change\nthis.\n"},{"id":"22810","messageId":"e814qm$sa$1@sea.gmane.org","threadId":"4666","inReplyTo":"7v3bdtv4h3.fsf@assigned-by-dhcp.cox.net","subject":"Re: [RFC] git --trace: trace command execution","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2006-06-29T18:06:21Z","receivedAt":"2006-06-29T18:06:21Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Junio C Hamano wrote:\n\n> By the way \"git cat-file -p\" or \"git verify-tag -v\" might be\n> more pleasant to view a tag since they make the tagger timestamp\n> human readable.\n\nInteresting, -p makes tagger timestamp human readable, but not author or\ncommiter:\n\n$ git cat-file -p `cat .git/refs/tags/v1.4.0`\ntagger Junio C Hamano Sat Jun 10 12:43:37 2006 -0700\n\n$ git cat-file -p `cat .git/refs/heads/origin`\nauthor Johannes Schindelin 1151491527 +0200\ncommitter Junio C Hamano 1151492136 -0700\n\nIs it intended, or a bug/missing feature (git 1.4.0)?\n\n-- \nJakub Narebski\nWarsaw, Poland\nShadeHawk on #git\n"}]}