Jeff King [Fri, 19 Aug 2022 08:50:54 +0000 (04:50 -0400)]
match_pathname(): drop unused "flags" parameter
This field has not been used since the function was introduced in b559263216 (exclude: split pathname matching code into a separate
function, 2012-10-15), though there was a brief period where it was
erroneously used and then reverted in ed4958477b (dir: fix pattern
matching on dirs, 2021-09-24) and 5ceb663e92 (dir: fix
directory-matching bug, 2021-11-02).
It's possible we'd eventually add a flag that makes it useful here, but
there are only a handful of callers. It would be easy to add back if
necessary, and in the meantime this makes the function interface less
misleading.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 08:50:33 +0000 (04:50 -0400)]
log-tree: drop unused commit param in remerge_diff()
This function has never used its "commit" parameter since it was added
in db757e8b8d (show, log: provide a --remerge-diff capability,
2022-02-02).
This makes sense; we already have separate parameters for the parents
(which lets us redo the merge) and the oid of the result tree (which we
can then diff against the remerge result).
Let's drop the unused parameter in the name of clarity.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 08:49:41 +0000 (04:49 -0400)]
xdiff: drop unused mmfile parameters from xdl_do_histogram_diff()
These are no longer used since 9df0fc3d57 (xdiff: fix a memory leak,
2022-02-16), as the caller is expected to call xdl_prepare_env() itself.
After that change the histogram code only examines the prepared
xdfenv_t, not the original buffers.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:54 +0000 (06:08 -0400)]
is_path_owned_by_current_uid(): mark "report" parameter as unused
In the non-Windows version of this function, we never have any errors to
report, and thus the "report" parameter is unused. But we can't drop it,
because we have to maintain function call compatibility with the version
in compat/mingw.h, which does use this parameter.
Note that there's an extra level of indirection here; the common
function is actually is_path_owned_by_current_user, which is a macro
pointing to "by_current_uid" or "by_current_sid", depending on the
platform. So an alternative here is to eat the unused parameter in the
macro, since -Wunused-parameter doesn't complain about macros. But I
think the UNUSED() annotation is less obfuscated for somebody reading
the code later.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:52 +0000 (06:08 -0400)]
run-command: mark unused async callback parameters
The start_async(), etc, functions need a "proc" callback that conforms
to a particular interface. Not every callback needs every parameter
(e.g., the caller might not even ask to open an input descriptor, in
which case there is no point in the callback looking at it). Let's mark
these for -Wunused-parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:46 +0000 (06:08 -0400)]
hashmap: mark unused callback parameters
Hashmap comparison functions must conform to a particular callback
interface, but many don't use all of their parameters. Especially the
void cmp_data pointer, but some do not use keydata either (because they
can easily form a full struct to pass when doing lookups). Let's mark
these to make -Wunused-parameter happy.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:44 +0000 (06:08 -0400)]
config: mark unused callback parameters
The callback passed to git_config() must conform to a particular
interface. But most callbacks don't actually look at the extra "void
*data" parameter. Let's mark the unused parameters to make
-Wunused-parameter happy.
Note there's one unusual case here in get_remote_default() where we
actually ignore the "value" parameter. That's because it's only checking
whether the option is found at all, and not parsing its value.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:42 +0000 (06:08 -0400)]
streaming: mark unused virtual method parameters
Streaming "open" functions need to conform to the same virtual function
interface, but not every implementation needs every parameter. Mark the
unused ones as such to appease -Wunused-parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:39 +0000 (06:08 -0400)]
transport: mark bundle transport_options as unused
get_refs_from_bundle() is a virtual function which must match the
signature of other transports, but it doesn't look at its
transport_options at all. This isn't a bug, because not all transports
necessarily support all options. Let's mark it as unused to appease
-Wunused-parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:37 +0000 (06:08 -0400)]
refs: mark unused virtual method parameters
The refs code uses various polymorphic types (e.g., loose vs packed
ref_stores, abstracted iterators). Not every virtual function or
callback needs all of its parameters. Let's mark the unused ones to
quiet -Wunused-parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:35 +0000 (06:08 -0400)]
refs: mark unused reflog callback parameters
Functions used with for_each_reflog_ent() need to conform to a
particular interface, but not every function needs all of the
parameters. Mark the unused ones to make -Wunused-parameter happy.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:32 +0000 (06:08 -0400)]
refs: mark unused each_ref_fn parameters
Functions used with for_each_ref(), etc, need to conform to the
each_ref_fn interface. But most of them don't need every parameter;
let's annotate the unused ones to quiet -Wunused-parameter.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Fri, 19 Aug 2022 10:08:27 +0000 (06:08 -0400)]
git-compat-util: add UNUSED macro
In preparation for compiling with -Wunused-parameter, we'd like to be
able to annotate some function parameters as false positives (e.g.,
parameters which must exist to conform to a callback interface).
Ideally our annotation will:
- be portable, turning into nothing on platforms which don't support
it
- be easy to read, without looking too syntactically odd or taking
attention away from the rest of the parameters
- help us notice when a parameter marked as unused is actually used,
which keeps our annotations accurate. In theory a compiler could
tell us this easily, but gcc has no such warning. Clang has
-Wused-but-marked-unused, but it triggers false positives with our
MAYBE_UNUSED annotation (e.g., for commit-slab functions)
This patch introduces an UNUSED() macro which takes the parameter name
as an argument. That lets us tweak the name in such a way that we'll
notice if somebody tries to use it. It looks like this in use:
int some_ref_cb(const char *refname,
const struct object_id *UNUSED(oid),
int UNUSED(flags),
void *UNUSED(data))
{
printf("got refname %s", refname);
return 0;
}
Because the unused parameter names are rewritten behind the scenes to
UNUSED_oid, etc, adding code like:
printf("oid is %s", oid_to_hex(oid));
will fail compilation with "oid undeclared". Sadly, the "did you mean"
feature of modern compilers is not generally smart enough to suggest the
"unused" name. If we used a very short prefix like U_oid, that does
convince gcc to say "did you mean", but since the "U_" in the suggestion
isn't much of a hint, it doesn't really help. In practice, a look at the
function definition usually makes the problem pretty obvious.
Note that we have to put the definition of UNUSED early in
git-compat-util.h, because it will eventually be used for some compat
functions themselves (both directly here and in mingw.h).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:11 +0000 (18:04 +0200)]
builtin/worktree.c: let parse-options parse subcommands
'git worktree' parses its subcommands with a long list of if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:10 +0000 (18:04 +0200)]
builtin/stash.c: let parse-options parse subcommands
'git stash' parses its subcommands with a long list of if-else if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
and listing subcommands for Bash completion.
Note that the push_stash() function implementing the 'push' subcommand
accepts an extra flag parameter to indicate whether push was assumed,
so add a wrapper function with the standard subcommand function
signature.
Note also that this change "hides" the '-h' option in 'git stash push
-h' from the parse_option() call in cmd_stash(), as it comes after the
subcommand. Consequently, from now on it will emit the usage of the
'push' subcommand instead of the usage of 'git stash'. We had a
failing test for this case, which can now be flipped to expect
success.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:09 +0000 (18:04 +0200)]
builtin/sparse-checkout.c: let parse-options parse subcommands
'git sparse-checkout' parses its subcommands with a couple of if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
Note that some of the functions implementing each subcommand only
accept the 'argc' and '**argv' parameters, so add a (unused) '*prefix'
parameter to make them match the type expected by parse-options, and
thus avoid casting function pointers.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:08 +0000 (18:04 +0200)]
builtin/remote.c: let parse-options parse subcommands
'git remote' parses its subcommands with a long list of if-else if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling unknown subcommands, and listing subcommands for Bash
completion. Make sure that the default operation mode doesn't accept
any arguments; and while at it remove the capitalization of the error
message and adjust the test checking it accordingly.
Note that 'git remote' has both 'remove' and 'rm' subcommands, and the
former is preferred [1], so hide the latter for completion.
Note also that the functions implementing each subcommand only accept
the 'argc' and '**argv' parameters, so add a (unused) '*prefix'
parameter to make them match the type expected by parse-options, and
thus avoid casting a bunch of function pointers.
[1] e17dba8fe1 (remote: prefer subcommand name 'remove' to 'rm',
2012-09-06)
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:07 +0000 (18:04 +0200)]
builtin/reflog.c: let parse-options parse subcommands
'git reflog' parses its subcommands with a couple of if-else if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
and listing subcommands for Bash completion.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:06 +0000 (18:04 +0200)]
builtin/notes.c: let parse-options parse subcommands
'git notes' parses its subcommands with a long list of if-else if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling unknown subcommands, and listing subcommands for Bash
completion. Make sure that the default operation mode doesn't accept
any arguments.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:05 +0000 (18:04 +0200)]
builtin/multi-pack-index.c: let parse-options parse subcommands
'git multi-pack-index' parses its subcommands with a couple of if-else
if statements. parse-options has just learned to parse subcommands,
so let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
Note that the functions implementing each subcommand only accept the
'argc' and '**argv' parameters, so add a (unused) '*prefix' parameter
to make them match the type expected by parse-options, and thus avoid
casting function pointers.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:04 +0000 (18:04 +0200)]
builtin/hook.c: let parse-options parse subcommands
'git hook' parses its currently only subcommand with an if statement.
parse-options has just learned to parse subcommands, so let's use that
facility instead, with the benefits of shorter code, handling missing
or unknown subcommands, and listing subcommands for Bash completion.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:03 +0000 (18:04 +0200)]
builtin/gc.c: let parse-options parse 'git maintenance's subcommands
'git maintenanze' parses its subcommands with a couple of if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
This change makes 'git maintenance' consistent with other commands in
that the help text shown for '-h' goes to standard output, not error,
in the exit code and error message on unknown subcommand, and the
error message on missing subcommand. There is a test checking these,
which is now updated accordingly.
Note that some of the functions implementing each subcommand don't
accept any parameters, so add the (unused) 'argc', '**argv' and
'*prefix' parameters to make them match the type expected by
parse-options, and thus avoid casting function pointers.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:02 +0000 (18:04 +0200)]
builtin/commit-graph.c: let parse-options parse subcommands
'git commit-graph' parses its subcommands with an if-else if
statement. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
Note that the functions implementing each subcommand only accept the
'argc' and '**argv' parameters, so add a (unused) '*prefix' parameter
to make them match the type expected by parse-options, and thus avoid
casting function pointers.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:01 +0000 (18:04 +0200)]
builtin/bundle.c: let parse-options parse subcommands
'git bundle' parses its subcommands with a couple of if-else if
statements. parse-options has just learned to parse subcommands, so
let's use that facility instead, with the benefits of shorter code,
handling missing or unknown subcommands, and listing subcommands for
Bash completion.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:04:00 +0000 (18:04 +0200)]
parse-options: add support for parsing subcommands
Several Git commands have subcommands to implement mutually exclusive
"operation modes", and they usually parse their subcommand argument
with a bunch of if-else if statements.
Teach parse-options to handle subcommands as well, which will result
in shorter and simpler code with consistent error handling and error
messages on unknown or missing subcommand, and it will also make
possible for our Bash completion script to handle subcommands
programmatically.
The approach is guided by the following observations:
- Most subcommands [1] are implemented in dedicated functions, and
most of those functions [2] either have a signature matching the
'int cmd_foo(int argc, const char **argc, const char *prefix)'
signature of builtin commands or can be trivially converted to
that signature, because they miss only that last prefix parameter
or have no parameters at all.
- Subcommand arguments only have long form, and they have no double
dash prefix, no negated form, and no description, and they don't
take any arguments, and can't be abbreviated.
- There must be exactly one subcommand among the arguments, or zero
if the command has a default operation mode.
- All arguments following the subcommand are considered to be
arguments of the subcommand, and, conversely, arguments meant for
the subcommand may not preceed the subcommand.
So in the end subcommand declaration and parsing would look something
like this:
Here each OPT_SUBCOMMAND specifies the name of the subcommand and the
function implementing it, and the address of the same 'fn' subcommand
function pointer. parse_options() then processes the arguments until
it finds the first argument matching one of the subcommands, sets 'fn'
to the function associated with that subcommand, and returns, leaving
the rest of the arguments unprocessed. If none of the listed
subcommands is found among the arguments, parse_options() will show
usage and abort.
If a command has a default operation mode, 'fn' should be initialized
to the function implementing that mode, and parse_options() should be
invoked with the PARSE_OPT_SUBCOMMAND_OPTIONAL flag. In this case
parse_options() won't error out when not finding any subcommands, but
will return leaving 'fn' unchanged. Note that if that default
operation mode has any --options, then the PARSE_OPT_KEEP_UNKNOWN_OPT
flag is necessary as well (otherwise parse_options() would error out
upon seeing the unknown option meant to the default operation mode).
Some thoughts about the implementation:
- The same pointer to 'fn' must be specified as 'value' for each
OPT_SUBCOMMAND, because there can be only one set of mutually
exclusive subcommands; parse_options() will BUG() otherwise.
There are other ways to tell parse_options() where to put the
function associated with the subcommand given on the command line,
but I didn't like them:
- Change parse_options()'s signature by adding a pointer to
subcommand function to be set to the function associated with
the given subcommand, affecting all callsites, even those that
don't have subcommands.
- Introduce a specific parse_options_and_subcommand() variant
with that extra funcion parameter.
- I decided against automatically calling the subcommand function
from within parse_options(), because:
- There are commands that have to perform additional actions
after option parsing but before calling the function
implementing the specified subcommand.
- The return code of the subcommand is usually the return code
of the git command, but preserving the return code of the
automatically called subcommand function would have made the
API awkward.
- Also add a OPT_SUBCOMMAND_F() variant to allow specifying an
option flag: we have two subcommands that are purposefully
excluded from completion ('git remote rm' and 'git stash save'),
so they'll have to be specified with the PARSE_OPT_NOCOMPLETE
flag.
- Some of the 'parse_opt_flags' don't make sense with subcommands,
and using them is probably just an oversight or misunderstanding.
Therefore parse_options() will BUG() when invoked with any of the
following flags while the options array contains at least one
OPT_SUBCOMMAND:
- PARSE_OPT_KEEP_DASHDASH: parse_options() stops parsing
arguments when encountering a "--" argument, so it doesn't
make sense to expect and keep one before a subcommand, because
it would prevent the parsing of the subcommand.
However, this flag is allowed in combination with the
PARSE_OPT_SUBCOMMAND_OPTIONAL flag, because the double dash
might be meaningful for the command's default operation mode,
e.g. to disambiguate refs and pathspecs.
- PARSE_OPT_STOP_AT_NON_OPTION: As its name suggests, this flag
tells parse_options() to stop as soon as it encouners a
non-option argument, but subcommands are by definition not
options... so how could they be parsed, then?!
- PARSE_OPT_KEEP_UNKNOWN: This flag can be used to collect any
unknown --options and then pass them to a different command or
subsystem. Surely if a command has subcommands, then this
functionality should rather be delegated to one of those
subcommands, and not performed by the command itself.
However, this flag is allowed in combination with the
PARSE_OPT_SUBCOMMAND_OPTIONAL flag, making possible to pass
--options to the default operation mode.
- If the command with subcommands has a default operation mode, then
all arguments to the command must preceed the arguments of the
subcommand.
AFAICT we don't have any commands where this makes a difference,
because in those commands either only the command accepts any
arguments ('notes' and 'remote'), or only the default subcommand
('reflog' and 'stash'), but never both.
- The 'argv' array passed to subcommand functions currently starts
with the name of the subcommand. Keep this behavior. AFAICT no
subcommand functions depend on the actual content of 'argv[0]',
but the parse_options() call handling their options expects that
the options start at argv[1].
- To support handling subcommands programmatically in our Bash
completion script, 'git cmd --git-completion-helper' will now list
both subcommands and regular --options, if any. This means that
the completion script will have to separate subcommands (i.e.
words without a double dash prefix) from --options on its own, but
that's rather easy to do, and it's not much work either, because
the number of subcommands a command might have is rather low, and
those commands accept only a single --option or none at all. An
alternative would be to introduce a separate option that lists
only subcommands, but then the completion script would need not
one but two git invocations and command substitutions for commands
with subcommands.
Note that this change doesn't affect the behavior of our Bash
completion script, because when completing the --option of a
command with subcommands, e.g. for 'git notes --<TAB>', then all
subcommands will be filtered out anyway, as none of them will
match the word to be completed starting with that double dash
prefix.
[1] Except 'git rerere', because many of its subcommands are
implemented in the bodies of the if-else if statements parsing the
command's subcommand argument.
[2] Except 'credential', 'credential-store' and 'fsmonitor--daemon',
because some of the functions implementing their subcommands take
special parameters.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
This doesn't matter for the completion script, because field splitting
discards that space anyway.
However, later patches in this series will teach parse-options to
handle subcommands, and subcommands will be included in the completion
helper output as well. This will make the loop printing options (and
subcommands) a tad more complex, so I wanted to test the result. The
test would have to account for the presence of that leading space,
which bugged my OCD, so let's get rid of it.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:58 +0000 (18:03 +0200)]
parse-options: clarify the limitations of PARSE_OPT_NODASH
Update the comment documenting 'struct option' to clarify that
PARSE_OPT_NODASH can only be an argumentless short option; see 51a9949eda (parseopt: add PARSE_OPT_NODASH, 2009-05-07).
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:57 +0000 (18:03 +0200)]
parse-options: PARSE_OPT_KEEP_UNKNOWN only applies to --options
The description of 'PARSE_OPT_KEEP_UNKNOWN' starts with "Keep unknown
arguments instead of erroring out". This is a bit misleading, as this
flag only applies to unknown --options, while non-option arguments are
kept even without this flag.
Update the description to clarify this, and rename the flag to
PARSE_OPTIONS_KEEP_UNKNOWN_OPT to make this obvious just by looking at
the flag name.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:56 +0000 (18:03 +0200)]
api-parse-options.txt: fix description of OPT_CMDMODE
The description of the 'OPT_CMDMODE' macro states that "enum_val is
set to int_var when ...", but it's the other way around, 'int_var' is
set to 'enum_val'. Fix this.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:55 +0000 (18:03 +0200)]
t0040-parse-options: test parse_options() with various 'parse_opt_flags'
In 't0040-parse-options.sh' we thoroughly test the parsing of all
types and forms of options, but in all those tests parse_options() is
always invoked with a 0 flags parameter.
Add a few tests to demonstrate how various 'enum parse_opt_flags'
values are supposed to influence option parsing.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:54 +0000 (18:03 +0200)]
t5505-remote.sh: check the behavior without a subcommand
'git remote' without a subcommand defaults to listing all remotes and
doesn't accept any arguments except the '-v|--verbose' option.
We are about to teach parse-options to handle subcommands, and update
'git remote' to make use of that new feature. So let's add some tests
to make sure that the upcoming changes don't inadvertently change the
behavior in these cases.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:53 +0000 (18:03 +0200)]
t3301-notes.sh: check that default operation mode doesn't take arguments
'git notes' without a subcommand defaults to listing all notes and
doesn't accept any arguments.
We are about to teach parse-options to handle subcommands, and update
'git notes' to make use of that new feature. So let's add a test to
make sure that the upcoming changes don't inadvertenly change the
behavior in this corner case.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
SZEDER Gábor [Fri, 19 Aug 2022 16:03:52 +0000 (18:03 +0200)]
git.c: update NO_PARSEOPT markings
Our Bash completion script can complete --options for commands using
parse-options even when that command doesn't have a dedicated
completion function, but to do so the completion script must know
which commands use parse-options and which don't. Therefore, commands
not using parse-options are marked in 'git.c's command list with the
NO_PARSEOPT flag.
Update this list, and remove this flag from the commands that by now
use parse-options.
After this change we can TAB complete --options of the plumbing
commands 'commit-tree', 'mailinfo' and 'mktag'.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Justin Donnelly [Wed, 17 Aug 2022 00:18:12 +0000 (00:18 +0000)]
git-prompt: show presence of unresolved conflicts at command prompt
If GIT_PS1_SHOWCONFLICTSTATE is set to "yes", show the word "CONFLICT"
on the command prompt when there are unresolved conflicts.
Example prompt: (main|CONFLICT)
Signed-off-by: Justin Donnelly <justinrdonnelly@gmail.com> Acked-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Elijah Newren [Fri, 19 Aug 2022 04:28:10 +0000 (04:28 +0000)]
revision: allow --ancestry-path to take an argument
We have long allowed users to run e.g.
git log --ancestry-path master..seen
which shows all commits which satisfy all three of these criteria:
* are an ancestor of seen
* are not an ancestor of master
* have master as an ancestor
This commit allows another variant:
git log --ancestry-path=$TOPIC master..seen
which shows all commits which satisfy all of these criteria:
* are an ancestor of seen
* are not an ancestor of master
* have $TOPIC in their ancestry-path
that last bullet can be defined as commits meeting any of these
criteria:
* are an ancestor of $TOPIC
* have $TOPIC as an ancestor
* are $TOPIC
This also allows multiple --ancestry-path arguments, which can be
used to find commits with any of the given topics in their ancestry
path.
Signed-off-by: Elijah Newren <newren@gmail.com> Acked-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Derrick Stolee [Fri, 19 Aug 2022 04:28:09 +0000 (04:28 +0000)]
t6019: modernize tests with helper
The tests in t6019 are repetitive, so create a helper that greatly
simplifies the test script.
In addition, update the common pattern that places 'git rev-list' on the
left side of a pipe, which can hide some exit codes. Send the output to
a 'raw' file that is then consumed by other tools so the Git exit code
is verified as zero. And since we're using --format anyway, switch to
`git log`, so that we get the desired format and can avoid using sed.
Signed-off-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Elijah Newren <newren@gmail.com> Acked-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Elijah Newren [Fri, 19 Aug 2022 04:45:55 +0000 (04:45 +0000)]
merge-ort: remove code obsoleted by other changes
Commit 66b209b86a ("merge-ort: implement CE_SKIP_WORKTREE handling with
conflicted entries", 2021-03-20) added some code for merge-ort to handle
conflicted and skip_worktree entries in general. Included in this was
an ugly hack for dealing with present-despite-skipped entries and a
testcase (t6428.2) specific to that hack, since at that time users could
accidentally get files into that state when using a sparse checkout.
However, with the merging of 82386b4496 ("Merge branch
'en/present-despite-skipped'", 2022-03-09), that class of problems was
addressed globally and in a much cleaner way. As such, the
present-despite-skipped hack in merge-ort is no longer needed and can
simply be removed.
No additional testcase is needed here; t6428.2 was written to test the
necessary functionality and is being kept. The fact that this test
continues to pass despite the code being removed shows that the extra
code is no longer necessary.
Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:53 +0000 (21:40 +0000)]
scalar: update technical doc roadmap with FSMonitor support
Update the Scalar roadmap to reflect completion of enabling the built-in
FSMonitor in Scalar.
Note that implementation of 'scalar help' was moved to the final set of
changes to move Scalar out of 'contrib/'. This is due to a dependency on
changes to 'git help', as all changes to the main Git tree *exclusively*
implemented to support Scalar are part of that series.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Especially on Windows, we will need to stop that daemon, just in case
that the directory needs to be removed (the daemon would otherwise hold
a handle to that directory, preventing it from being deleted).
Helped-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Using the built-in FSMonitor makes many common commands quite a bit
faster. So let's teach the `scalar register` command to enable the
built-in FSMonitor and kick-start the fsmonitor--daemon process (for
convenience).
For simplicity, we only support the built-in FSMonitor (and no external
file system monitor such as e.g. Watchman).
Helped-by: Junio C Hamano <gitster@pobox.com> Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Matthew John Cheetham <mjcheetham@outlook.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:50 +0000 (21:40 +0000)]
scalar: move config setting logic into its own function
Create function 'set_scalar_config()' to contain the logic used in setting
Scalar-defined Git config settings, including how to handle reconfiguring &
overwriting existing values. This function allows future patches to set
config values in parts of 'scalar.c' other than 'set_recommended_config()'.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:49 +0000 (21:40 +0000)]
scalar-delete: do not 'die()' in 'delete_enlistment()'
Rather than exiting with 'die()' when 'delete_enlistment()' encounters an
error, return an error code with the appropriate message. There's no need
for an abrupt exit with 'die()' in 'delete_enlistment()' because its only
caller ('cmd_delete()') properly cleans up allocated resources and returns
the 'delete_enlistment()' return value as its own exit code.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:48 +0000 (21:40 +0000)]
scalar-[un]register: clearly indicate source of error
When a step in 'register_dir()' or 'unregister_dir()' fails, indicate which
step failed with an error message, rather than silently assigning a nonzero
return code.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:47 +0000 (21:40 +0000)]
scalar-unregister: handle error codes greater than 0
When 'scalar unregister' tries to disable maintenance and remove an
enlistment, ensure that the return value is nonzero if either operation
produces *any* nonzero return value, not just when they return a value less
than 0.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Thu, 18 Aug 2022 21:40:46 +0000 (21:40 +0000)]
scalar: constrain enlistment search
Make the search for repository and enlistment root in
'setup_enlistment_directory()' more constrained to simplify behavior and
adhere to 'GIT_CEILING_DIRECTORIES'.
Previously, 'setup_enlistment_directory()' would check whether the provided
path (or current working directory) '<dir>' or its subdirectory '<dir>/src'
was a repository root. If not, the process would repeat on the parent of
'<dir>' until the repository was found or it reached the root of the
filesystem. This meant that a user could specify a path *anywhere* inside an
enlistment (including paths not in the repository contained within the
enlistment) and it would be found.
The downside to this process is that the search would not account for
'GIT_CEILING_DIRECTORIES', so the upward search could result in modifying
repository contents past 'GIT_CEILING_DIRECTORIES'. Similarly, operations
like 'scalar delete' could end up unintentionally deleting the parent of a
repo if its root was named 'src'.
To make this 'setup_enlistment_directory()' both adhere to
'GIT_CEILING_DIRECTORIES' and avoid unwanted deletions, the search for an
enlistment directory is simplified to:
- if '<dir>/src' is a repository root, '<dir>' is the enlistment root
- if '<dir>' is either the repository root or contained within a repository,
the repository root is the enlistment root
Now, only 'setup_git_directory()' (called by 'setup_enlistment_directory()')
searches upwards from the 'scalar' specified path, enforcing
'GIT_CEILING_DIRECTORIES' in the process. Additionally, 'scalar delete
<dir>/src' will not delete '<dir>' (if users would like to delete it, they
can still specify the enlistment root with 'scalar delete <dir>'). This is
true of any 'scalar' operation; users can invoke 'scalar' on the enlistment
root, but paths must otherwise be inside the repository to be valid.
To help clarify the updated behavior, new tests are added to
't9099-scalar.sh'.
Finally, this change leaves 'strbuf_parent_directory()' with only a single,
WIN32-specific caller in 'delete_enlistment()'. Rather than wrap
'strbuf_parent_directory()' in '#ifdef WIN32' to avoid the "unused function"
compiler error, move the contents of 'strbuf_parent_directory()' into
'delete_enlistment()' and remove the function.
Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano [Thu, 18 Aug 2022 20:07:05 +0000 (13:07 -0700)]
Merge branch 'sy/sparse-rm'
"git rm" has become more aware of the sparse-index feature.
* sy/sparse-rm:
rm: integrate with sparse-index
rm: expand the index only when necessary
pathspec.h: move pathspec_needs_expanded_index() from reset.c to here
t1092: add tests for `git-rm`
Junio C Hamano [Thu, 18 Aug 2022 20:07:04 +0000 (13:07 -0700)]
Merge branch 'jk/fsck-tree-mode-bits-fix'
"git fsck" reads mode from tree objects but canonicalizes the mode
before passing it to the logic to check object sanity, which has
hid broken tree objects from the checking logic. This has been
corrected, but to help exiting projects with broken tree objects
that they cannot fix retroactively, the severity of anomalies this
code detects has been demoted to "info" for now.
* jk/fsck-tree-mode-bits-fix:
fsck: downgrade tree badFilemode to "info"
fsck: actually detect bad file modes in trees
tree-walk: add a mechanism for getting non-canonicalized modes
Elijah Newren [Thu, 18 Aug 2022 07:15:27 +0000 (07:15 +0000)]
merge-ort: provide helpful submodule update message when possible
In commit 4057523a40 ("submodule merge: update conflict error message",
2022-08-04), a more detailed message was provided when submodules
conflict, in order to help users know how to resolve those conflicts.
There were a couple situations for which a different message would be
more appropriate, but that commit left handling those for future work.
Unfortunately, that commit would check if any submodules were of the
type that it didn't know how to explain, and, if so, would avoid
providing the more detailed explanation even for the submodules it did
know how to explain.
Change this to have the code print the helpful messages for the subset
of submodules it knows how to explain.
Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Elijah Newren [Thu, 18 Aug 2022 07:15:26 +0000 (07:15 +0000)]
merge-ort: avoid surprise with new sub_flag variable
Commit 4057523a40 ("submodule merge: update conflict error message",
2022-08-04) added a sub_flag variable that is used to store a value from
enum conflict_and_info_types, but initializes it with a value of -1 that
does not correspond to any of the conflict_and_info_types. The code may
never set it to a valid value and yet still use it, which can be
surprising when reading over the code at first. Initialize it instead
to the generic CONFLICT_SUBMODULE_FAILED_TO_MERGE value, which is still
distinct from the two values we need to special case.
Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Elijah Newren [Thu, 18 Aug 2022 07:15:25 +0000 (07:15 +0000)]
merge-ort: remove translator lego in new "submodule conflict suggestion"
In commit 4057523a40 ("submodule merge: update conflict error message",
2022-08-04), the new "submodule conflict suggestion" code was
translating 6 different pieces of the new message and then used
carefully crafted logic to allow stitching it back together with special
formatting. Keep the components of the message together as much as
possible, so that:
* we reduce the number of things translators have to translate
* translators have more control over the format of the output
* the code is much easier for developers to understand too
Also, reformat some comments running beyond the 80th column while at it.
Signed-off-by: Elijah Newren <newren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Wed, 17 Aug 2022 06:10:22 +0000 (02:10 -0400)]
pipe_command(): mark stdin descriptor as non-blocking
Our pipe_command() helper lets you both write to and read from a child
process on its stdin/stdout. It's supposed to work without deadlocks
because we use poll() to check when descriptors are ready for reading or
writing. But there's a bug: if both the data to be written and the data
to be read back exceed the pipe buffer, we'll deadlock.
The issue is that the code assumes that if you have, say, a 2MB buffer
to write and poll() tells you that the pipe descriptor is ready for
writing, that calling:
write(cmd->in, buf, 2*1024*1024);
will do a partial write, filling the pipe buffer and then returning what
it did write. And that is what it would do on a socket, but not for a
pipe. When writing to a pipe, at least on Linux, it will block waiting
for the child process to read() more. And now we have a potential
deadlock, because the child may be writing back to us, waiting for us to
read() ourselves.
The diff against HEAD~200 will be big, and the filter wants to write all
of it back to us (obviously this is a dummy filter, but in the real
world something like diff-highlight would similarly stream back a big
output).
If you set add.interactive.useBuiltin to false, the problem goes away,
because now we're not using pipe_command() anymore (instead, that part
happens in perl). But this isn't a bug in the interactive code at all.
It's the underlying pipe_command() code which is broken, and has been
all along.
We presumably didn't notice because most calls only do input _or_
output, not both. And the few that do both, like gpg calls, may have
large inputs or outputs, but never both at the same time (e.g., consider
signing, which has a large payload but a small signature comes back).
The obvious fix is to put the descriptor into non-blocking mode, and
indeed, that makes the problem go away. Callers shouldn't need to
care, because they never see the descriptor (they hand us a buffer to
feed into it).
The included test fails reliably on Linux without this patch. Curiously,
it doesn't fail in our Windows CI environment, but has been reported to
do so for individual developers. It should pass in any environment after
this patch (courtesy of the compat/ layers added in the last few
commits).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Wed, 17 Aug 2022 06:09:42 +0000 (02:09 -0400)]
pipe_command(): handle ENOSPC when writing to a pipe
When write() to a non-blocking pipe fails because the buffer is full,
POSIX says we should see EAGAIN. But our mingw_write() compat layer on
Windows actually returns ENOSPC for this case. This is probably
something we want to correct, but given that we don't plan to use
non-blocking descriptors in a lot of places, we can work around it by
just catching ENOSPC alongside EAGAIN. If we ever do fix mingw_write(),
then this patch can be reverted.
We don't actually use a non-blocking pipe yet, so this is still just
preparation.
Helped-by: René Scharfe <l.s.r@web.de> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Wed, 17 Aug 2022 06:08:06 +0000 (02:08 -0400)]
pipe_command(): avoid xwrite() for writing to pipe
If xwrite() sees an EAGAIN response, it will loop forever until the
write succeeds (or encounters a real error). This is due to ef1cf0167a
(xwrite: poll on non-blocking FDs, 2016-06-26), with the idea that we
won't be surprised by a descriptor unexpectedly set as non-blocking.
But that will make things awkward when we do want a non-blocking
descriptor, and a future patch will switch pipe_command() to using one.
In that case, looping on EAGAIN is bad, because the process on the other
end of the pipe may be waiting on us before doing another read() on the
pipe, which would mean we deadlock.
In practice we're not supposed to ever see EAGAIN here, since poll()
will have just told us the descriptor is ready for writing. But our
Windows emulation of poll() will always return "ready" for writing to a
pipe descriptor! This is due to 94f4d01932 (mingw: workaround for hangs
when sending STDIN, 2020-02-17).
Our best bet in that case is to keep handling other descriptors, as any
read() we do may allow the child command to make forward progress (i.e.,
its write() finishes, and then it read()s from its stdin, freeing up
space in the pipe buffer). This means we might busy-loop between poll()
and write() on Windows if the child command is slow to read our input,
but it's much better than the alternative of deadlocking.
In practice, this busy-looping should be rare:
- for small inputs, we'll just write the whole thing in a single
write() anyway, non-blocking or not
- for larger inputs where the child reads input and then processes it
before writing (e.g., gpg verifying a signature), we may make a few
extra write() calls that get EAGAIN during the initial write, but
once it has taken in the whole input, we'll correctly block waiting
to read back the data.
- for larger inputs where the child process is streaming output back
(like a diff filter), we'll likewise see some extra EAGAINs, but
most of them will be followed immediately by a read(), which will
let the child command make forward progress.
Of course it won't happen at all for now, since we don't yet use a
non-blocking pipe. This is just preparation for when we do.
Helped-by: René Scharfe <l.s.r@web.de> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Wed, 17 Aug 2022 06:06:58 +0000 (02:06 -0400)]
git-compat-util: make MAX_IO_SIZE define globally available
We define MAX_IO_SIZE within wrapper.c, but it's useful for any code
that wants to do a raw write() for whatever reason (say, because they
want different EAGAIN handling). Let's make it available everywhere.
The alternative would be adding xwrite_foo() variants to give callers
more options. But there's really no reason MAX_IO_SIZE needs to be
abstracted away, so this give callers the most flexibility.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
René Scharfe [Wed, 17 Aug 2022 06:05:25 +0000 (02:05 -0400)]
nonblock: support Windows
Implement enable_pipe_nonblock() using the Windows API. This works only
for pipes, but that is sufficient for this limited interface. Despite
the API calls used, it handles both "named" and anonymous pipes from our
pipe() emulation.
Signed-off-by: René Scharfe <l.s.r@web.de> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Wed, 17 Aug 2022 06:04:55 +0000 (02:04 -0400)]
compat: add function to enable nonblocking pipes
We'd like to be able to make some of our pipes nonblocking so that
poll() can be used effectively, but O_NONBLOCK isn't portable. Let's
introduce a compat wrapper so this can be abstracted for each platform.
The interface is as narrow as possible to let platforms do what's
natural there (rather than having to implement fcntl() and a fake
O_NONBLOCK for example, or having to handle other types of descriptors).
The next commit will add Windows support, at which point we should be
covering all platforms in practice. But if we do find some other
platform without O_NONBLOCK, we'll return ENOSYS. Arguably we could just
trigger a build-time #error in this case, which would catch the problem
earlier. But since we're not planning to use this compat wrapper in many
code paths, a seldom-seen runtime error may be friendlier for such a
platform than blocking compilation completely. Our test suite would
still notice it.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Josh Steadmon [Tue, 2 Aug 2022 22:04:05 +0000 (15:04 -0700)]
fetch-pack: add tracing for negotiation rounds
Currently, negotiation for V0/V1/V2 fetch have trace2 regions covering
the entire negotiation process. However, we'd like additional data, such
as timing for each round of negotiation or the number of "haves" in each
round. Additionally, "independent negotiation" (AKA push negotiation)
has no tracing at all. Having this data would allow us to compare the
performance of the various negotation implementations, and to debug
unexpectedly slow fetch & push sessions.
Add per-round trace2 regions for all negotiation implementations (V0+V1,
V2, and independent negotiation), as well as an overall region for
independent negotiation. Add trace2 data logging for the number of haves
and "in vain" objects for each round, and for the total number of rounds
once negotiation completes. Finally, add a few checks into various
tests to verify that the number of rounds is logged as expected.
Signed-off-by: Josh Steadmon <steadmon@google.com> Acked-by: Jeff Hostetler <jeffhost@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano [Mon, 15 Aug 2022 06:19:28 +0000 (23:19 -0700)]
Merge branch 'js/safe-directory-plus'
Platform-specific code that determines if a directory is OK to use
as a repository has been taught to report more details, especially
on Windows.
* js/safe-directory-plus:
mingw: handle a file owned by the Administrators group correctly
mingw: be more informative when ownership check fails on FAT32
mingw: provide details about unsafe directories' ownership
setup: prepare for more detailed "dubious ownership" messages
setup: fix some formatting
Junio C Hamano [Mon, 15 Aug 2022 06:19:27 +0000 (23:19 -0700)]
Merge branch 'ab/tech-docs-to-help'
Expose a lot of "tech docs" via "git help" interface.
* ab/tech-docs-to-help:
docs: move http-protocol docs to man section 5
docs: move cruft pack docs to gitformat-pack
docs: move pack format docs to man section 5
docs: move signature docs to man section 5
docs: move index format docs to man section 5
docs: move protocol-related docs to man section 5
docs: move commit-graph format docs to man section 5
git docs: add a category for file formats, protocols and interfaces
git docs: add a category for user-facing file, repo and command UX
git help doc: use "<doc>" instead of "<guide>"
help.c: remove common category behavior from drop_prefix() behavior
help.c: refactor drop_prefix() to use a "switch" statement"
Matheus Tavares [Mon, 15 Aug 2022 01:06:38 +0000 (22:06 -0300)]
tests: use the new C rot13-filter helper to avoid PERL prereq
The previous commit implemented a C version of the t0021/rot13-filter.pl
script. Let's use this new C helper to eliminate the PERL prereq from
various tests, and also remove the superseded Perl script.
Signed-off-by: Matheus Tavares <matheus.bernardino@usp.br> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Matheus Tavares [Mon, 15 Aug 2022 01:06:37 +0000 (22:06 -0300)]
t0021: implementation the rot13-filter.pl script in C
This script is currently used by three test files: t0021-conversion.sh,
t2080-parallel-checkout-basics.sh, and
t2082-parallel-checkout-attributes.sh. To avoid the need for the PERL
dependency at these tests, let's convert the script to a C test-tool
command. The following commit will take care of actually modifying the
said tests to use the new C helper and removing the Perl script.
The Perl script flushes the log file handler after each write. As
commented in [1], this seems to be an early design decision that was
later reconsidered, but possibly ended up being left in the code by
accident:
Maybe, but autoflush is not explicitly enabled. I will
enable it again (I disabled it because of Eric's comment
but I re-read the comment and he is only talking about
pipes).
Anyways, this behavior is not really needed for the tests and the
flush() calls make the code slightly larger, so let's avoid them
altogether in the new C version.
Helped-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Matheus Tavares <matheus.bernardino@usp.br> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Matheus Tavares [Mon, 15 Aug 2022 01:06:36 +0000 (22:06 -0300)]
t0021: avoid grepping for a Perl-specific string at filter output
This test sets the t0021/rot13-filter.pl script as a long-running
process filter for a git checkout command. It then expects the filter to
fail producing a specific error message at stderr. In the following
commits we are going to replace the script with a C test-tool helper,
but the test currently expects the error message in a Perl-specific
format. That is, when you call `die <msg>` in Perl, it emits
"<msg> at - line 1." In preparation for the conversion, let's avoid the
Perl-specific part and only grep for <msg> itself.
Signed-off-by: Matheus Tavares <matheus.bernardino@usp.br> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King [Sun, 14 Aug 2022 06:29:15 +0000 (02:29 -0400)]
is_promisor_object(): fix use-after-free of tree buffer
Since commit fcc07e980b (is_promisor_object(): free tree buffer after
parsing, 2021-04-13), we'll always free the buffers attached to a
"struct tree" after searching them for promisor links. But there's an
important case where we don't want to do so: if somebody else is already
using the tree!
This can happen during a "rev-list --missing=allow-promisor" traversal
in a partial clone that is missing one or more trees or blobs. The
backtrace for the free looks like this:
We're in the middle of walking through the entries of a tree object via
process_tree_contents(). We see a blob (or it could even be another tree
entry) that we don't have, so we call is_promisor_object() to check it.
That function loops over all of the objects in the promisor packfile,
including the tree we're currently walking. When we're done with it
there, we free the tree buffer. But as we return to the walk in
process_tree_contents(), it's still holding on to a pointer to that
buffer, via its tree_desc iterator, and it accesses the freed memory.
Even a trivial use of "--missing=allow-promisor" triggers this problem,
as the included test demonstrates (it's just a vanilla --blob:none
clone).
We can detect this case by only freeing the tree buffer if it was
allocated on our behalf. This is a little tricky since that happens
inside parse_object(), and it doesn't tell us whether the object was
already parsed, or whether it allocated the buffer itself. But by
checking for an already-parsed tree beforehand, we can distinguish the
two cases.
That feels a little hacky, and does incur an extra lookup in the
object-hash table. But that cost is fairly minimal compared to actually
loading objects (and since we're iterating the whole pack here, we're
likely to be loading most objects, rather than reusing cached results).
It may also be a good direction for this function in general, as there
are other possible optimizations that rely on doing some analysis before
parsing:
- we could detect blobs and avoid reading their contents; they can't
link to other objects, but parse_object() doesn't know that we don't
care about checking their hashes.
- we could avoid allocating object structs entirely for most objects
(since we really only need them in the oidset), which would save
some memory.
- promisor commits could use the commit-graph rather than loading the
object from disk
This commit doesn't do any of those optimizations, but I think it argues
that this direction is reasonable, rather than relying on parse_object()
and trying to teach it to give us more information about whether it
parsed.
The included test fails reliably under SANITIZE=address just when
running "rev-list --missing=allow-promisor". Checking the output isn't
strictly necessary to detect the bug, but it seems like a reasonable
addition given the general lack of coverage for "allow-promisor" in the
test suite.
Reported-by: Andrew Olsen <andrew.olsen@koordinates.com> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:18 +0000 (20:10 +0000)]
scalar-diagnose: use 'git diagnose --mode=all'
Replace implementation of 'scalar diagnose' with an internal invocation of
'git diagnose --mode=all'. This simplifies the implementation of
'cmd_diagnose' by making it a direct alias of 'git diagnose' and removes
some code in 'scalar.c' that is duplicated in 'builtin/diagnose.c'. The
simplicity of the alias also sets up a clean deprecation path for 'scalar
diagnose' (in favor of 'git diagnose'), if that is desired in the future.
This introduces one minor change to the output of 'scalar diagnose', which
is that the prefix of the created zip archive is changed from 'scalar_' to
'git-diagnostics-'.
Helped-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:17 +0000 (20:10 +0000)]
builtin/bugreport.c: create '--diagnose' option
Create a '--diagnose' option for 'git bugreport' to collect additional
information about the repository and write it to a zipped archive.
The '--diagnose' option behaves effectively as an alias for simultaneously
running 'git bugreport' and 'git diagnose'. In the documentation, users are
explicitly recommended to attach the diagnostics alongside a bug report to
provide additional context to readers, ideally reducing some back-and-forth
between reporters and those debugging the issue.
Note that '--diagnose' may take an optional string arg (either 'stats' or
'all'). If specified without the arg, the behavior corresponds to running
'git diagnose' without '--mode'. As with 'git diagnose', this default is
intended to help reduce unintentional leaking of sensitive information).
Users can also explicitly specify '--diagnose=(stats|all)' to generate the
respective archive created by 'git diagnose --mode=(stats|all)'.
Suggested-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:16 +0000 (20:10 +0000)]
builtin/diagnose.c: add '--mode' option
Create '--mode=<mode>' option in 'git diagnose' to allow users to optionally
select non-default diagnostic information to include in the output archive.
Additionally, document the currently-available modes, emphasizing the
importance of not sharing a '--mode=all' archive publicly due to the
presence of sensitive information.
Note that the option parsing callback - 'option_parse_diagnose()' - is added
to 'diagnose.c' rather than 'builtin/diagnose.c' so that it may be reused in
future callers configuring a diagnostics archive.
Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:15 +0000 (20:10 +0000)]
builtin/diagnose.c: create 'git diagnose' builtin
Create a 'git diagnose' builtin to generate a standalone zip archive of
repository diagnostics.
The "diagnose" functionality was originally implemented for Scalar in aa5c79a331 (scalar: implement `scalar diagnose`, 2022-05-28). However, the
diagnostics gathered are not specific to Scalar-cloned repositories and
can be useful when diagnosing issues in any Git repository.
Helped-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:14 +0000 (20:10 +0000)]
diagnose.c: add option to configure archive contents
Update 'create_diagnostics_archive()' to take an argument 'mode'. When
archiving diagnostics for a repository, 'mode' is used to selectively
include/exclude information based on its value. The initial options for
'mode' are:
* DIAGNOSE_NONE: do not collect any diagnostics or create an archive
(no-op).
* DIAGNOSE_STATS: collect basic repository metadata (Git version, repo path,
filesystem available space) as well as sizing and count statistics for the
repository's objects and packfiles.
* DIAGNOSE_ALL: collect basic repository metadata, sizing/count statistics,
and copies of the '.git', '.git/hooks', '.git/info', '.git/logs', and
'.git/objects/info' directories.
These modes are introduced to provide users the option to collect
diagnostics without the sensitive information included in copies of '.git'
dir contents. At the moment, only 'scalar diagnose' uses
'create_diagnostics_archive()' (with a hardcoded 'DIAGNOSE_ALL' mode to
match existing functionality), but more callers will be introduced in
subsequent patches.
Finally, refactor from a hardcoded set of 'add_directory_to_archiver()'
calls to iterative invocations gated by 'DIAGNOSE_ALL'. This allows for
easier future modification of the set of directories to archive and improves
error reporting when 'add_directory_to_archiver()' fails.
Helped-by: Derrick Stolee <derrickstolee@github.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:13 +0000 (20:10 +0000)]
scalar-diagnose: move functionality to common location
Move the core functionality of 'scalar diagnose' into a new 'diagnose.[c,h]'
library to prepare for new callers in the main Git tree generating
diagnostic archives. These callers will be introduced in subsequent patches.
While this patch appears large, it is mostly made up of moving code out of
'scalar.c' and into 'diagnose.c'. Specifically, the functions
are all copied verbatim from 'scalar.c'. The 'create_diagnostics_archive()'
function is a mostly identical (partial) copy of 'cmd_diagnose()', with the
primary changes being that 'zip_path' is an input and "Enlistment root" is
corrected to "Repository root" in the archiver log.
Helped-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:12 +0000 (20:10 +0000)]
scalar-diagnose: move 'get_disk_info()' to 'compat/'
Move 'get_disk_info()' function into 'compat/'. Although Scalar-specific
code is generally not part of the main Git tree, 'get_disk_info()' will be
used in subsequent patches by additional callers beyond 'scalar diagnose'.
This patch prepares for that change, at which point this platform-specific
code should be part of 'compat/' as a matter of convention.
The function is copied *mostly* verbatim, with two exceptions:
* '#ifdef WIN32' is replaced with '#ifdef GIT_WINDOWS_NATIVE' to allow
'statvfs' to be used with Cygwin.
* the 'struct strbuf buf' and 'int res' (as well as their corresponding
cleanup & return) are moved outside of the '#ifdef' block.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:11 +0000 (20:10 +0000)]
scalar-diagnose: add directory to archiver more gently
If a directory added to the 'scalar diagnose' archiver does not exist, warn
and return 0 from 'add_directory_to_archiver()' rather than failing with a
fatal error. This handles a failure edge case where the '.git/logs' has not
yet been created when running 'scalar diagnose', but extends to any
situation where a directory may be missing in the '.git' dir.
Now, when a directory is missing a warning is captured in the diagnostic
logs. This provides a user with more complete information than if 'scalar
diagnose' simply failed with an error.
Helped-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:10 +0000 (20:10 +0000)]
scalar-diagnose: avoid 32-bit overflow of size_t
Avoid 32-bit size_t overflow when reporting the available disk space in
'get_disk_info' by casting the block size and available block count to
'off_t' before multiplying them. Without this change, 'st_mult' would
(correctly) report a size_t overflow on 32-bit systems at or exceeding 2^32
bytes of available space.
Note that 'off_t' is a 64-bit integer even on 32-bit systems due to the
inclusion of '#define _FILE_OFFSET_BITS 64' in 'git-compat-util.h' (see b97e911643 (Support for large files on 32bit systems., 2007-02-17)).
Helped-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Victoria Dye [Fri, 12 Aug 2022 20:10:09 +0000 (20:10 +0000)]
scalar-diagnose: use "$GIT_UNZIP" in test
Use the "$GIT_UNZIP" test variable rather than verbatim 'unzip' to unzip the
'scalar diagnose' archive. Using "$GIT_UNZIP" is needed to run the Scalar
tests on systems where 'unzip' is not in the system path.
Signed-off-by: Victoria Dye <vdye@github.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano [Fri, 12 Aug 2022 20:19:08 +0000 (13:19 -0700)]
Merge branch 'ab/plug-revisions-leak'
Plug a bit more leaks in the revisions API.
* ab/plug-revisions-leak:
revisions API: don't leak memory on argv elements that need free()-ing
bisect.c: partially fix bisect_rev_setup() memory leak
log: refactor "rev.pending" code in cmd_show()
log: fix a memory leak in "git show <revision>..."
test-fast-rebase helper: use release_revisions() (again)
bisect.c: add missing "goto" for release_revisions()
Junio C Hamano [Fri, 12 Aug 2022 20:19:08 +0000 (13:19 -0700)]
Merge branch 'ab/leak-check'
Extend SANITIZE=leak checking and declare more tests "currently leak-free".
* ab/leak-check:
CI: use "GIT_TEST_SANITIZE_LEAK_LOG=true" in linux-leaks
upload-pack: fix a memory leak in create_pack_file()
leak tests: mark passing SANITIZE=leak tests as leak-free
leak tests: don't skip some tests under SANITIZE=leak
test-lib: have the "check" mode for SANITIZE=leak consider leak logs
test-lib: add a GIT_TEST_PASSING_SANITIZE_LEAK=check mode
test-lib: simplify by removing test_external
tests: move copy/pasted PERL + Test::More checks to a lib-perl.sh
t/Makefile: don't remove test-results in "clean-except-prove-cache"
test-lib: add a SANITIZE=leak logging mode
t/README: reword the "GIT_TEST_PASSING_SANITIZE_LEAK" description
test-lib: add a --invert-exit-code switch
test-lib: fix GIT_EXIT_OK logic errors, use BAIL_OUT
test-lib: don't set GIT_EXIT_OK before calling test_atexit_handler
test-lib: use $1, not $@ in test_known_broken_{ok,failure}_
Teng Long [Fri, 12 Aug 2022 02:56:46 +0000 (10:56 +0800)]
tr2: shows scope unconditionally in addition to key-value pair
When we specify GIT_TRACE2_CONFIG_PARAMS or trace2.configparams,
trace2 will prints "interesting" config values to log. Sometimes,
when a config set in multiple scope files, the following output
looks like (the irrelevant fields are omitted here as "..."):
As the log shows, even each config in different scope is dumped, but
we don't know which scope it comes from. Therefore, it's better to
add the scope names as well to make them be more recognizable. For
example, when execute:
Teng Long [Fri, 12 Aug 2022 02:56:45 +0000 (10:56 +0800)]
api-trace2.txt: print config key-value pair
It's supported to print "interesting" config key-value paire
to tr2 log by setting "GIT_TRACE2_CONFIG_PARAMS" environment
variable and the "trace2.configparam" config, let's add the
related docs in Documentaion/technical/api-trace2.txt.
Signed-off-by: Teng Long <dyroneteng@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Eric DeCosta [Thu, 11 Aug 2022 23:57:11 +0000 (23:57 +0000)]
fsmonitor: option to allow fsmonitor to run against network-mounted repos
Though perhaps not common, there are use cases where users have large,
network-mounted repos. Having the ability to run fsmonitor against
network paths would benefit those users.
Most modern Samba-based filers have the necessary support to enable
fsmonitor on network-mounted repos. As a first step towards enabling
fsmonitor to work against network-mounted repos, introduce a
configuration option, 'fsmonitor.allowRemote'. Setting this option to
true will override the default behavior (erroring-out) when a
network-mounted repo is detected by fsmonitor.
Signed-off-by: Eric DeCosta <edecosta@mathworks.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Li Linchao [Thu, 11 Aug 2022 04:47:54 +0000 (04:47 +0000)]
rev-list: support human-readable output for `--disk-usage`
The '--disk-usage' option for git-rev-list was introduced in 16950f8384
(rev-list: add --disk-usage option for calculating disk usage, 2021-02-09).
This is very useful for people inspect their git repo's objects usage
infomation, but the resulting number is quit hard for a human to read.
Teach git rev-list to output a human readable result when using
'--disk-usage'.
Signed-off-by: Li Linchao <lilinchao@oschina.cn> Signed-off-by: Junio C Hamano <gitster@pobox.com>