diff options
author | Sergio Durigan Junior <sergiodj@redhat.com> | 2017-09-11 00:54:33 -0400 |
---|---|---|
committer | Sergio Durigan Junior <sergiodj@redhat.com> | 2017-09-22 13:28:16 -0400 |
commit | 43573013c9836f2b91b74b9b29dac35fdb41e06b (patch) | |
tree | 01586f996ba29e158508fa6b34960008f7ac8efa /gdb/cli | |
parent | 6ec2e0f5bddc69d55e09fe56792edf7a109ff8ae (diff) | |
download | gdb-43573013c9836f2b91b74b9b29dac35fdb41e06b.zip gdb-43573013c9836f2b91b74b9b29dac35fdb41e06b.tar.gz gdb-43573013c9836f2b91b74b9b29dac35fdb41e06b.tar.bz2 |
Get rid of "gdb_dirbuf" and use "getcwd (NULL, 0)"
Currently we have "current_directory" and "gdb_dirbuf" globals, which
means that we basically have two possible places to consult when we
want to know GDB's current working directory.
This is not ideal and can lead to confusion. Moreover, the way we're
using "gdb_difbuf" along with "getcwd" is problematic because we
declare the buffer with "1024" elements hardcoded, which does not take
into account longer pathnames that are possible in many filesystems.
Using "PATH_MAX" would also not be a solution because of portability
problems. Therefore, the best solution is to rely on the fact that
"getcwd (NULL, 0)" will "do the right thing" and return a
heap-allocated string containing the full path. With the new "getcwd"
module from gnulib, it is now possible to do that without worrying
about breaking some host.
With this patch "current_directory" is now the only place to check for
GDB's cwd.
Reviewed-by: Pedro Alves <palves@redhat.com>
gdb/ChangeLog:
2017-09-22 Sergio Durigan Junior <sergiodj@redhat.com>
* cli/cli-cmds.c (pwd_command): Use "getcwd (NULL, 0)".
(cd_command): Likewise. Free "current_directory" before
assigning to it.
* main.c (captured_main_1): Use "getcwd (NULL, 0)".
* mi/mi-cmd-env.c (mi_cmd_env_pwd): Likewise.
* top.c (gdb_dirbuf): Remove global declaration.
* top.h (gdb_dirbuf): Likewise.
Diffstat (limited to 'gdb/cli')
-rw-r--r-- | gdb/cli/cli-cmds.c | 17 |
1 files changed, 12 insertions, 5 deletions
diff --git a/gdb/cli/cli-cmds.c b/gdb/cli/cli-cmds.c index 260fd3f..cbafb13 100644 --- a/gdb/cli/cli-cmds.c +++ b/gdb/cli/cli-cmds.c @@ -384,13 +384,16 @@ pwd_command (char *args, int from_tty) { if (args) error (_("The \"pwd\" command does not take an argument: %s"), args); - if (! getcwd (gdb_dirbuf, sizeof (gdb_dirbuf))) + + gdb::unique_xmalloc_ptr<char> cwd (getcwd (NULL, 0)); + + if (cwd == NULL) error (_("Error finding name of working directory: %s"), safe_strerror (errno)); - if (strcmp (gdb_dirbuf, current_directory) != 0) + if (strcmp (cwd.get (), current_directory) != 0) printf_unfiltered (_("Working directory %s\n (canonically %s).\n"), - current_directory, gdb_dirbuf); + current_directory, cwd.get ()); else printf_unfiltered (_("Working directory %s.\n"), current_directory); } @@ -418,7 +421,8 @@ cd_command (char *dir, int from_tty) /* There's too much mess with DOSish names like "d:", "d:.", "d:./foo" etc. Instead of having lots of special #ifdef'ed code, simply get the canonicalized name of the current directory. */ - dir = getcwd (gdb_dirbuf, sizeof (gdb_dirbuf)); + gdb::unique_xmalloc_ptr<char> cwd (getcwd (NULL, 0)); + dir = cwd.get (); #endif len = strlen (dir); @@ -436,7 +440,10 @@ cd_command (char *dir, int from_tty) dir_holder.reset (savestring (dir, len)); if (IS_ABSOLUTE_PATH (dir_holder.get ())) - current_directory = dir_holder.release (); + { + xfree (current_directory); + current_directory = dir_holder.release (); + } else { if (IS_DIR_SEPARATOR (current_directory[strlen (current_directory) - 1])) |