diff options
Diffstat (limited to 'lib/kernel')
| -rw-r--r-- | lib/kernel/doc/src/file.xml | 43 | ||||
| -rw-r--r-- | lib/kernel/doc/src/os.xml | 109 | ||||
| -rw-r--r-- | lib/kernel/src/kernel.app.src | 2 | ||||
| -rw-r--r-- | lib/kernel/src/os.erl | 63 | ||||
| -rw-r--r-- | lib/kernel/test/file_name_SUITE.erl | 2 | ||||
| -rw-r--r-- | lib/kernel/test/os_SUITE.erl | 14 |
6 files changed, 204 insertions, 29 deletions
diff --git a/lib/kernel/doc/src/file.xml b/lib/kernel/doc/src/file.xml index b674b3ca93..2ab35b9b05 100644 --- a/lib/kernel/doc/src/file.xml +++ b/lib/kernel/doc/src/file.xml @@ -41,7 +41,7 @@ <p>Regarding filename encoding, the Erlang VM can operate in two modes. The current mode can be queried using function - <seealso marker="#native_name_encoding"><c>native_name_encoding/0</c></seealso>. + <seealso marker="#native_name_encoding/0"><c>native_name_encoding/0</c></seealso>. It returns <c>latin1</c> or <c>utf8</c>.</p> <p>In <c>latin1</c> mode, the Erlang VM does not change the @@ -59,7 +59,7 @@ terminal supports UTF-8, otherwise <c>latin1</c>. The default can be overridden using <c>+fnl</c> (to force <c>latin1</c> mode) or <c>+fnu</c> (to force <c>utf8</c> mode) when starting - <seealso marker="erts:erl"><c>erts:erl</c></seealso>.</p> + <seealso marker="erts:erl"><c>erl</c></seealso>.</p> <p>On operating systems with transparent naming, files can be inconsistently named, for example, some files are encoded in UTF-8 while @@ -81,6 +81,23 @@ <p>See also section <seealso marker="stdlib:unicode_usage#notes-about-raw-filenames">Notes About Raw Filenames</seealso> in the STDLIB User's Guide.</p> + <note><p> + File operations used to accept filenames containing + null characters (integer value zero). This caused + the name to be truncated and in some cases arguments + to primitive operations to be mixed up. Filenames + containing null characters inside the filename + are now <em>rejected</em> and will cause primitive + file operations fail. + </p></note> + <warning><p> + Currently null characters at the end of the filename + will be accepted by primitive file operations. Such + filenames are however still documented as invalid. The + implementation will also change in the future and + reject such filenames. + </p></warning> + </description> <datatypes> @@ -96,9 +113,21 @@ </datatype> <datatype> <name name="filename"/> + <desc> + <p> + See also the documentation of the + <seealso marker="#type-name_all"><c>name_all()</c></seealso> type. + </p> + </desc> </datatype> <datatype> <name name="filename_all"/> + <desc> + <p> + See also the documentation of the + <seealso marker="#type-name_all"><c>name_all()</c></seealso> type. + </p> + </desc> </datatype> <datatype> <name name="io_device"/> @@ -112,21 +141,23 @@ <name name="name"/> <desc> <p>If VM is in Unicode filename mode, <c>string()</c> and <c>char()</c> - are allowed to be > 255. + are allowed to be > 255. See also the documentation of the + <seealso marker="#type-name_all"><c>name_all()</c></seealso> type. </p> </desc> </datatype> <datatype> <name name="name_all"/> <desc> - <p>If VM is in Unicode filename mode, <c>string()</c> and <c>char()</c> + <p>If VM is in Unicode filename mode, characters are allowed to be > 255. <c><anno>RawFilename</anno></c> is a filename not subject to Unicode translation, meaning that it can contain characters not conforming to the Unicode encoding expected from the file system (that is, non-UTF-8 characters although the VM is started - in Unicode filename mode). + in Unicode filename mode). Null characters (integer value zero) + are <em>not</em> allowed in filenames (not even at the end). </p> </desc> </datatype> @@ -1825,7 +1856,7 @@ f.txt: {person, "kalle", 25}. <p>The functions in the module <c>file</c> usually treat binaries as raw filenames, that is, they are passed "as is" even when the encoding of the binary does not agree with - <seealso marker="#native_name_encoding"><c>native_name_encoding()</c></seealso>. + <seealso marker="#native_name_encoding/0"><c>native_name_encoding()</c></seealso>. However, this function expects binaries to be encoded according to the value returned by <c>native_name_encoding()</c>.</p> <p>Typical error reasons are:</p> diff --git a/lib/kernel/doc/src/os.xml b/lib/kernel/doc/src/os.xml index 0e9add4161..0a08e2c78a 100644 --- a/lib/kernel/doc/src/os.xml +++ b/lib/kernel/doc/src/os.xml @@ -36,8 +36,99 @@ only run on a specific platform. On the other hand, with careful use, these functions can be of help in enabling a program to run on most platforms.</p> + + <note> + <p> + File operations used to accept filenames containing + null characters (integer value zero). This caused + the name to be truncated and in some cases arguments + to primitive operations to be mixed up. Filenames + containing null characters inside the filename + are now <em>rejected</em> and will cause primitive + file operations to fail. + </p> + <p> + Also environment variable operations used to accept + names and values of environment variables containing + null characters (integer value zero). This caused + operations to silently produce erroneous results. + Environment variable names and values containing + null characters inside the name or value are now + <em>rejected</em> and will cause environment variable + operations to fail. + </p> + </note> + <warning> + <p> + Currently null characters at the end of filenames, + environment variable names and values will be accepted + by the primitive operations. Such filenames, environment + variable names and values are however still documented as + invalid. The implementation will also change in the + future and reject such filenames, environment variable + names and values. + </p> + </warning> </description> + <datatypes> + <datatype> + <name name="env_var_name"/> + <desc> + <p>A string containing valid characters on the specific + OS for environment variable names using + <seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso> + encoding. Note that specifically null characters (integer + value zero) and <c>$=</c> characters are not allowed. + However, note that not all invalid characters necessarily + will cause the primitiv operations to fail, but may instead + produce invalid results. + </p> + </desc> + </datatype> + <datatype> + <name name="env_var_value"/> + <desc> + <p>A string containing valid characters on the specific + OS for environment variable values using + <seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso> + encoding. Note that specifically null characters (integer + value zero) are not allowed. However, note that not all + invalid characters necessarily will cause the primitiv + operations to fail, but may instead produce invalid results. + </p> + </desc> + </datatype> + <datatype> + <name name="env_var_name_value"/> + <desc> + <p> + Assuming that environment variables has been correctly + set, a strings containing valid characters on the specific + OS for environment variable names and values using + <seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso> + encoding. The first <c>$=</c> characters appearing in + the string separates environment variable name (on the + left) from environment variable value (on the right). + </p> + </desc> + </datatype> + <datatype> + <name name="command_input"/> + <desc> + <p>All characters needs to be valid characters on the + specific OS using + <seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso> + encoding. Note that specifically null characters (integer + value zero) are not allowed. However, note that not all + invalid characters not necessarily will cause + <seealso marker="#cmd/1"><c>os:cmd/1</c></seealso> + to fail, but may instead produce invalid results. + </p> + </desc> + </datatype> + </datatypes> + <funcs> <func> <name name="cmd" arity="1"/> @@ -49,6 +140,15 @@ result as a string. This function is a replacement of the previous function <c>unix:cmd/1</c>; they are equivalent on a Unix platform.</p> + <warning><p>Previous implementation used to allow all characters + as long as they were integer values greater than or equal to zero. + This sometimes lead to unwanted results since null characters + (integer value zero) often are interpreted as string termination. + Current implementation still accepts null characters at the end + of <c><anno>Command</anno></c> even though the documentation + states that no null characters are allowed. This will however + be changed in the future so that no null characters at all will + be accepted.</p></warning> <p><em>Examples:</em></p> <code type="none"> LsOut = os:cmd("ls"), % on unix platform @@ -152,6 +252,15 @@ DirOut = os:cmd("dir"), % on Win32 platform</code> <p>On Unix platforms, the environment is set using UTF-8 encoding if Unicode filename translation is in effect. On Windows, the environment is set using wide character interfaces.</p> + <note> + <p> + <c><anno>VarName</anno></c> is not allowed to contain + an <c>$=</c> character. Previous implementations used + to just let the <c>$=</c> character through which + silently caused erroneous results. Current implementation + will instead throw a <c>badarg</c> exception. + </p> + </note> </desc> </func> diff --git a/lib/kernel/src/kernel.app.src b/lib/kernel/src/kernel.app.src index 2a88cc7e26..080b11fc4d 100644 --- a/lib/kernel/src/kernel.app.src +++ b/lib/kernel/src/kernel.app.src @@ -120,6 +120,6 @@ {applications, []}, {env, [{error_logger, tty}]}, {mod, {kernel, []}}, - {runtime_dependencies, ["erts-9.1", "stdlib-3.4", "sasl-3.0"]} + {runtime_dependencies, ["erts-10.0", "stdlib-3.5", "sasl-3.0"]} ] }. diff --git a/lib/kernel/src/os.erl b/lib/kernel/src/os.erl index 209899d587..b5f19d4b99 100644 --- a/lib/kernel/src/os.erl +++ b/lib/kernel/src/os.erl @@ -25,6 +25,8 @@ -include("file.hrl"). +-export_type([env_var_name/0, env_var_value/0, env_var_name_value/0, command_input/0]). + %%% BIFs -export([getenv/0, getenv/1, getenv/2, getpid/0, @@ -32,21 +34,29 @@ putenv/2, set_signal/2, system_time/0, system_time/1, timestamp/0, unsetenv/1]). --spec getenv() -> [string()]. +-type env_var_name() :: nonempty_string(). + +-type env_var_value() :: string(). + +-type env_var_name_value() :: nonempty_string(). + +-type command_input() :: atom() | io_lib:chars(). + +-spec getenv() -> [env_var_name_value()]. getenv() -> erlang:nif_error(undef). -spec getenv(VarName) -> Value | false when - VarName :: string(), - Value :: string(). + VarName :: env_var_name(), + Value :: env_var_value(). getenv(_) -> erlang:nif_error(undef). -spec getenv(VarName, DefaultValue) -> Value when - VarName :: string(), - DefaultValue :: string(), - Value :: string(). + VarName :: env_var_name(), + DefaultValue :: env_var_value(), + Value :: env_var_value(). getenv(VarName, DefaultValue) -> case os:getenv(VarName) of @@ -75,8 +85,8 @@ perf_counter(Unit) -> erlang:convert_time_unit(os:perf_counter(), perf_counter, Unit). -spec putenv(VarName, Value) -> true when - VarName :: string(), - Value :: string(). + VarName :: env_var_name(), + Value :: env_var_value(). putenv(_, _) -> erlang:nif_error(undef). @@ -99,7 +109,7 @@ timestamp() -> erlang:nif_error(undef). -spec unsetenv(VarName) -> true when - VarName :: string(). + VarName :: env_var_name(). unsetenv(_) -> erlang:nif_error(undef). @@ -232,10 +242,9 @@ extensions() -> %% Executes the given command in the default shell for the operating system. -spec cmd(Command) -> string() when - Command :: atom() | io_lib:chars(). + Command :: os:command_input(). cmd(Cmd) -> - validate(Cmd), - {SpawnCmd, SpawnOpts, SpawnInput, Eot} = mk_cmd(os:type(), Cmd), + {SpawnCmd, SpawnOpts, SpawnInput, Eot} = mk_cmd(os:type(), validate(Cmd)), Port = open_port({spawn, SpawnCmd}, [binary, stderr_to_stdout, stream, in, hide | SpawnOpts]), MonRef = erlang:monitor(port, Port), @@ -255,8 +264,6 @@ mk_cmd({win32,Wtype}, Cmd) -> {Cspec,_} -> lists:concat([Cspec," /c",Cmd]) end, {Command, [], [], <<>>}; -mk_cmd(OsType,Cmd) when is_atom(Cmd) -> - mk_cmd(OsType, atom_to_list(Cmd)); mk_cmd(_,Cmd) -> %% Have to send command in like this in order to make sh commands like %% cd and ulimit available @@ -279,17 +286,33 @@ mk_cmd(_,Cmd) -> <<$\^D>>}. validate(Atom) when is_atom(Atom) -> - ok; + validate(atom_to_list(Atom)); validate(List) when is_list(List) -> - validate1(List). + case validate1(List) of + false -> + List; + true -> + %% Had zeros at end; remove them... + string:trim(List, trailing, [0]) + end. -validate1([C|Rest]) when is_integer(C) -> +validate1([0|Rest]) -> + validate2(Rest); +validate1([C|Rest]) when is_integer(C), C > 0 -> validate1(Rest); validate1([List|Rest]) when is_list(List) -> - validate1(List), - validate1(Rest); + validate1(List) or validate1(Rest); validate1([]) -> - ok. + false. + +%% Ensure that the rest is zero only... +validate2([]) -> + true; +validate2([0|Rest]) -> + validate2(Rest); +validate2([List|Rest]) when is_list(List) -> + validate2(List), + validate2(Rest). get_data(Port, MonRef, Eot, Sofar) -> receive diff --git a/lib/kernel/test/file_name_SUITE.erl b/lib/kernel/test/file_name_SUITE.erl index 899102c908..f23529fec9 100644 --- a/lib/kernel/test/file_name_SUITE.erl +++ b/lib/kernel/test/file_name_SUITE.erl @@ -302,7 +302,9 @@ check_normal(Mod) -> {ok, BC} = Mod:read(FD,1024), ok = file:close(FD) end || {regular,Name,Content} <- NormalDir ], + {error, badarg} = Mod:rename("fil1\0tmp_fil2","tmp_fil1"), Mod:rename("fil1","tmp_fil1"), + {error, badarg} = Mod:read_file("tmp_fil1\0.txt"), {ok, <<"fil1">>} = Mod:read_file("tmp_fil1"), {error,enoent} = Mod:read_file("fil1"), Mod:rename("tmp_fil1","fil1"), diff --git a/lib/kernel/test/os_SUITE.erl b/lib/kernel/test/os_SUITE.erl index 53a9e168ef..8056321448 100644 --- a/lib/kernel/test/os_SUITE.erl +++ b/lib/kernel/test/os_SUITE.erl @@ -22,7 +22,8 @@ -export([all/0, suite/0,groups/0,init_per_suite/1, end_per_suite/1, init_per_group/2,end_per_group/2, init_per_testcase/2,end_per_testcase/2]). --export([space_in_cwd/1, quoting/1, cmd_unicode/1, space_in_name/1, bad_command/1, +-export([space_in_cwd/1, quoting/1, cmd_unicode/1, + null_in_command/1, space_in_name/1, bad_command/1, find_executable/1, unix_comment_in_command/1, deep_list_command/1, large_output_command/1, background_command/0, background_command/1, message_leak/1, close_stdin/0, close_stdin/1, perf_counter_api/1]). @@ -34,7 +35,8 @@ suite() -> {timetrap,{minutes,1}}]. all() -> - [space_in_cwd, quoting, cmd_unicode, space_in_name, bad_command, + [space_in_cwd, quoting, cmd_unicode, null_in_command, + space_in_name, bad_command, find_executable, unix_comment_in_command, deep_list_command, large_output_command, background_command, message_leak, close_stdin, perf_counter_api]. @@ -125,6 +127,14 @@ cmd_unicode(Config) when is_list(Config) -> [] = receive_all(), ok. +null_in_command(Config) -> + {Ok, Error} = case os:type() of + {win32,_} -> {"dir", "di\0r"}; + _ -> {"ls", "l\0s"} + end, + true = is_list(try os:cmd(Ok) catch Class0:_ -> Class0 end), + error = try os:cmd(Error) catch Class1:_ -> Class1 end, + ok. %% Test that program with a space in its name can be executed. space_in_name(Config) when is_list(Config) -> |