diff options
Diffstat (limited to 'perl')
| -rw-r--r-- | perl/.gitignore | 5 | ||||
| -rw-r--r-- | perl/Git.pm | 1362 | ||||
| -rw-r--r-- | perl/Makefile | 47 | ||||
| -rw-r--r-- | perl/Makefile.PL | 38 | ||||
| -rw-r--r-- | perl/private-Error.pm | 827 |
5 files changed, 2279 insertions, 0 deletions
diff --git a/perl/.gitignore b/perl/.gitignore new file mode 100644 index 0000000000..98b24772c7 --- /dev/null +++ b/perl/.gitignore @@ -0,0 +1,5 @@ +perl.mak +perl.mak.old +blib +blibdirs +pm_to_blib diff --git a/perl/Git.pm b/perl/Git.pm new file mode 100644 index 0000000000..e8df55d2f2 --- /dev/null +++ b/perl/Git.pm @@ -0,0 +1,1362 @@ +=head1 NAME + +Git - Perl interface to the Git version control system + +=cut + + +package Git; + +use strict; + + +BEGIN { + +our ($VERSION, @ISA, @EXPORT, @EXPORT_OK); + +# Totally unstable API. +$VERSION = '0.01'; + + +=head1 SYNOPSIS + + use Git; + + my $version = Git::command_oneline('version'); + + git_cmd_try { Git::command_noisy('update-server-info') } + '%s failed w/ code %d'; + + my $repo = Git->repository (Directory => '/srv/git/cogito.git'); + + + my @revs = $repo->command('rev-list', '--since=last monday', '--all'); + + my ($fh, $c) = $repo->command_output_pipe('rev-list', '--since=last monday', '--all'); + my $lastrev = <$fh>; chomp $lastrev; + $repo->command_close_pipe($fh, $c); + + my $lastrev = $repo->command_oneline( [ 'rev-list', '--all' ], + STDERR => 0 ); + + my $sha1 = $repo->hash_and_insert_object('file.txt'); + my $tempfile = tempfile(); + my $size = $repo->cat_blob($sha1, $tempfile); + +=cut + + +require Exporter; + +@ISA = qw(Exporter); + +@EXPORT = qw(git_cmd_try); + +# Methods which can be called as standalone functions as well: +@EXPORT_OK = qw(command command_oneline command_noisy + command_output_pipe command_input_pipe command_close_pipe + command_bidi_pipe command_close_bidi_pipe + version exec_path html_path hash_object git_cmd_try + remote_refs + temp_acquire temp_release temp_reset temp_path); + + +=head1 DESCRIPTION + +This module provides Perl scripts easy way to interface the Git version control +system. The modules have an easy and well-tested way to call arbitrary Git +commands; in the future, the interface will also provide specialized methods +for doing easily operations which are not totally trivial to do over +the generic command interface. + +While some commands can be executed outside of any context (e.g. 'version' +or 'init'), most operations require a repository context, which in practice +means getting an instance of the Git object using the repository() constructor. +(In the future, we will also get a new_repository() constructor.) All commands +called as methods of the object are then executed in the context of the +repository. + +Part of the "repository state" is also information about path to the attached +working copy (unless you work with a bare repository). You can also navigate +inside of the working copy using the C<wc_chdir()> method. (Note that +the repository object is self-contained and will not change working directory +of your process.) + +TODO: In the future, we might also do + + my $remoterepo = $repo->remote_repository (Name => 'cogito', Branch => 'master'); + $remoterepo ||= Git->remote_repository ('http://git.or.cz/cogito.git/'); + my @refs = $remoterepo->refs(); + +Currently, the module merely wraps calls to external Git tools. In the future, +it will provide a much faster way to interact with Git by linking directly +to libgit. This should be completely opaque to the user, though (performance +increase notwithstanding). + +=cut + + +use Carp qw(carp croak); # but croak is bad - throw instead +use Error qw(:try); +use Cwd qw(abs_path); +use IPC::Open2 qw(open2); +use Fcntl qw(SEEK_SET SEEK_CUR); +} + + +=head1 CONSTRUCTORS + +=over 4 + +=item repository ( OPTIONS ) + +=item repository ( DIRECTORY ) + +=item repository () + +Construct a new repository object. +C<OPTIONS> are passed in a hash like fashion, using key and value pairs. +Possible options are: + +B<Repository> - Path to the Git repository. + +B<WorkingCopy> - Path to the associated working copy; not strictly required +as many commands will happily crunch on a bare repository. + +B<WorkingSubdir> - Subdirectory in the working copy to work inside. +Just left undefined if you do not want to limit the scope of operations. + +B<Directory> - Path to the Git working directory in its usual setup. +The C<.git> directory is searched in the directory and all the parent +directories; if found, C<WorkingCopy> is set to the directory containing +it and C<Repository> to the C<.git> directory itself. If no C<.git> +directory was found, the C<Directory> is assumed to be a bare repository, +C<Repository> is set to point at it and C<WorkingCopy> is left undefined. +If the C<$GIT_DIR> environment variable is set, things behave as expected +as well. + +You should not use both C<Directory> and either of C<Repository> and +C<WorkingCopy> - the results of that are undefined. + +Alternatively, a directory path may be passed as a single scalar argument +to the constructor; it is equivalent to setting only the C<Directory> option +field. + +Calling the constructor with no options whatsoever is equivalent to +calling it with C<< Directory => '.' >>. In general, if you are building +a standard porcelain command, simply doing C<< Git->repository() >> should +do the right thing and setup the object to reflect exactly where the user +is right now. + +=cut + +sub repository { + my $class = shift; + my @args = @_; + my %opts = (); + my $self; + + if (defined $args[0]) { + if ($#args % 2 != 1) { + # Not a hash. + $#args == 0 or throw Error::Simple("bad usage"); + %opts = ( Directory => $args[0] ); + } else { + %opts = @args; + } + } + + if (not defined $opts{Repository} and not defined $opts{WorkingCopy} + and not defined $opts{Directory}) { + $opts{Directory} = '.'; + } + + if (defined $opts{Directory}) { + -d $opts{Directory} or throw Error::Simple("Directory not found: $!"); + + my $search = Git->repository(WorkingCopy => $opts{Directory}); + my $dir; + try { + $dir = $search->command_oneline(['rev-parse', '--git-dir'], + STDERR => 0); + } catch Git::Error::Command with { + $dir = undef; + }; + + if ($dir) { + $dir =~ m#^/# or $dir = $opts{Directory} . '/' . $dir; + $opts{Repository} = abs_path($dir); + + # If --git-dir went ok, this shouldn't die either. + my $prefix = $search->command_oneline('rev-parse', '--show-prefix'); + $dir = abs_path($opts{Directory}) . '/'; + if ($prefix) { + if (substr($dir, -length($prefix)) ne $prefix) { + throw Error::Simple("rev-parse confused me - $dir does not have trailing $prefix"); + } + substr($dir, -length($prefix)) = ''; + } + $opts{WorkingCopy} = $dir; + $opts{WorkingSubdir} = $prefix; + + } else { + # A bare repository? Let's see... + $dir = $opts{Directory}; + + unless (-d "$dir/refs" and -d "$dir/objects" and -e "$dir/HEAD") { + # Mimick git-rev-parse --git-dir error message: + throw Error::Simple("fatal: Not a git repository: $dir"); + } + my $search = Git->repository(Repository => $dir); + try { + $search->command('symbolic-ref', 'HEAD'); + } catch Git::Error::Command with { + # Mimick git-rev-parse --git-dir error message: + throw Error::Simple("fatal: Not a git repository: $dir"); + } + + $opts{Repository} = abs_path($dir); + } + + delete $opts{Directory}; + } + + $self = { opts => \%opts }; + bless $self, $class; +} + +=back + +=head1 METHODS + +=over 4 + +=item command ( COMMAND [, ARGUMENTS... ] ) + +=item command ( [ COMMAND, ARGUMENTS... ], { Opt => Val ... } ) + +Execute the given Git C<COMMAND> (specify it without the 'git-' +prefix), optionally with the specified extra C<ARGUMENTS>. + +The second more elaborate form can be used if you want to further adjust +the command execution. Currently, only one option is supported: + +B<STDERR> - How to deal with the command's error output. By default (C<undef>) +it is delivered to the caller's C<STDERR>. A false value (0 or '') will cause +it to be thrown away. If you want to process it, you can get it in a filehandle +you specify, but you must be extremely careful; if the error output is not +very short and you want to read it in the same process as where you called +C<command()>, you are set up for a nice deadlock! + +The method can be called without any instance or on a specified Git repository +(in that case the command will be run in the repository context). + +In scalar context, it returns all the command output in a single string +(verbatim). + +In array context, it returns an array containing lines printed to the +command's stdout (without trailing newlines). + +In both cases, the command's stdin and stderr are the same as the caller's. + +=cut + +sub command { + my ($fh, $ctx) = command_output_pipe(@_); + + if (not defined wantarray) { + # Nothing to pepper the possible exception with. + _cmd_close($fh, $ctx); + + } elsif (not wantarray) { + local $/; + my $text = <$fh>; + try { + _cmd_close($fh, $ctx); + } catch Git::Error::Command with { + # Pepper with the output: + my $E = shift; + $E->{'-outputref'} = \$text; + throw $E; + }; + return $text; + + } else { + my @lines = <$fh>; + defined and chomp for @lines; + try { + _cmd_close($fh, $ctx); + } catch Git::Error::Command with { + my $E = shift; + $E->{'-outputref'} = \@lines; + throw $E; + }; + return @lines; + } +} + + +=item command_oneline ( COMMAND [, ARGUMENTS... ] ) + +=item command_oneline ( [ COMMAND, ARGUMENTS... ], { Opt => Val ... } ) + +Execute the given C<COMMAND> in the same way as command() +does but always return a scalar string containing the first line +of the command's standard output. + +=cut + +sub command_oneline { + my ($fh, $ctx) = command_output_pipe(@_); + + my $line = <$fh>; + defined $line and chomp $line; + try { + _cmd_close($fh, $ctx); + } catch Git::Error::Command with { + # Pepper with the output: + my $E = shift; + $E->{'-outputref'} = \$line; + throw $E; + }; + return $line; +} + + +=item command_output_pipe ( COMMAND [, ARGUMENTS... ] ) + +=item command_output_pipe ( [ COMMAND, ARGUMENTS... ], { Opt => Val ... } ) + +Execute the given C<COMMAND> in the same way as command() +does but return a pipe filehandle from which the command output can be +read. + +The function can return C<($pipe, $ctx)> in array context. +See C<command_close_pipe()> for details. + +=cut + +sub command_output_pipe { + _command_common_pipe('-|', @_); +} + + +=item command_input_pipe ( COMMAND [, ARGUMENTS... ] ) + +=item command_input_pipe ( [ COMMAND, ARGUMENTS... ], { Opt => Val ... } ) + +Execute the given C<COMMAND> in the same way as command_output_pipe() +does but return an input pipe filehandle instead; the command output +is not captured. + +The function can return C<($pipe, $ctx)> in array context. +See C<command_close_pipe()> for details. + +=cut + +sub command_input_pipe { + _command_common_pipe('|-', @_); +} + + +=item command_close_pipe ( PIPE [, CTX ] ) + +Close the C<PIPE> as returned from C<command_*_pipe()>, checking +whether the command finished successfully. The optional C<CTX> argument +is required if you want to see the command name in the error message, +and it is the second value returned by C<command_*_pipe()> when +called in array context. The call idiom is: + + my ($fh, $ctx) = $r->command_output_pipe('status'); + while (<$fh>) { ... } + $r->command_close_pipe($fh, $ctx); + +Note that you should not rely on whatever actually is in C<CTX>; +currently it is simply the command name but in future the context might +have more complicated structure. + +=cut + +sub command_close_pipe { + my ($self, $fh, $ctx) = _maybe_self(@_); + $ctx ||= '<unknown>'; + _cmd_close($fh, $ctx); +} + +=item command_bidi_pipe ( COMMAND [, ARGUMENTS... ] ) + +Execute the given C<COMMAND> in the same way as command_output_pipe() +does but return both an input pipe filehandle and an output pipe filehandle. + +The function will return return C<($pid, $pipe_in, $pipe_out, $ctx)>. +See C<command_close_bidi_pipe()> for details. + +=cut + +sub command_bidi_pipe { + my ($pid, $in, $out); + $pid = open2($in, $out, 'git', @_); + return ($pid, $in, $out, join(' ', @_)); +} + +=item command_close_bidi_pipe ( PID, PIPE_IN, PIPE_OUT [, CTX] ) + +Close the C<PIPE_IN> and C<PIPE_OUT> as returned from C<command_bidi_pipe()>, +checking whether the command finished successfully. The optional C<CTX> +argument is required if you want to see the command name in the error message, +and it is the fourth value returned by C<command_bidi_pipe()>. The call idiom +is: + + my ($pid, $in, $out, $ctx) = $r->command_bidi_pipe('cat-file --batch-check'); + print "000000000\n" $out; + while (<$in>) { ... } + $r->command_close_bidi_pipe($pid, $in, $out, $ctx); + +Note that you should not rely on whatever actually is in C<CTX>; +currently it is simply the command name but in future the context might +have more complicated structure. + +=cut + +sub command_close_bidi_pipe { + local $?; + my ($pid, $in, $out, $ctx) = @_; + foreach my $fh ($in, $out) { + unless (close $fh) { + if ($!) { + carp "error closing pipe: $!"; + } elsif ($? >> 8) { + throw Git::Error::Command($ctx, $? >>8); + } + } + } + + waitpid $pid, 0; + + if ($? >> 8) { + throw Git::Error::Command($ctx, $? >>8); + } +} + + +=item command_noisy ( COMMAND [, ARGUMENTS... ] ) + +Execute the given C<COMMAND> in the same way as command() does but do not +capture the command output - the standard output is not redirected and goes +to the standard output of the caller application. + +While the method is called command_noisy(), you might want to as well use +it for the most silent Git commands which you know will never pollute your +stdout but you want to avoid the overhead of the pipe setup when calling them. + +The function returns only after the command has finished running. + +=cut + +sub command_noisy { + my ($self, $cmd, @args) = _maybe_self(@_); + _check_valid_cmd($cmd); + + my $pid = fork; + if (not defined $pid) { + throw Error::Simple("fork failed: $!"); + } elsif ($pid == 0) { + _cmd_exec($self, $cmd, @args); + } + if (waitpid($pid, 0) > 0 and $?>>8 != 0) { + throw Git::Error::Command(join(' ', $cmd, @args), $? >> 8); + } +} + + +=item version () + +Return the Git version in use. + +=cut + +sub version { + my $verstr = command_oneline('--version'); + $verstr =~ s/^git version //; + $verstr; +} + + +=item exec_path () + +Return path to the Git sub-command executables (the same as +C<git --exec-path>). Useful mostly only internally. + +=cut + +sub exec_path { command_oneline('--exec-path') } + + +=item html_path () + +Return path to the Git html documentation (the same as +C<git --html-path>). Useful mostly only internally. + +=cut + +sub html_path { command_oneline('--html-path') } + + +=item repo_path () + +Return path to the git repository. Must be called on a repository instance. + +=cut + +sub repo_path { $_[0]->{opts}->{Repository} } + + +=item wc_path () + +Return path to the working copy. Must be called on a repository instance. + +=cut + +sub wc_path { $_[0]->{opts}->{WorkingCopy} } + + +=item wc_subdir () + +Return path to the subdirectory inside of a working copy. Must be called +on a repository instance. + +=cut + +sub wc_subdir { $_[0]->{opts}->{WorkingSubdir} ||= '' } + + +=item wc_chdir ( SUBDIR ) + +Change the working copy subdirectory to work within. The C<SUBDIR> is +relative to the working copy root directory (not the current subdirectory). +Must be called on a repository instance attached to a working copy +and the directory must exist. + +=cut + +sub wc_chdir { + my ($self, $subdir) = @_; + $self->wc_path() + or throw Error::Simple("bare repository"); + + -d $self->wc_path().'/'.$subdir + or throw Error::Simple("subdir not found: $!"); + # Of course we will not "hold" the subdirectory so anyone + # can delete it now and we will never know. But at least we tried. + + $self->{opts}->{WorkingSubdir} = $subdir; +} + + +=item config ( VARIABLE ) + +Retrieve the configuration C<VARIABLE> in the same manner as C<config> +does. In scalar context requires the variable to be set only one time +(exception is thrown otherwise), in array context returns allows the +variable to be set multiple times and returns all the values. + +This currently wraps command('config') so it is not so fast. + +=cut + +sub config { + my ($self, $var) = _maybe_self(@_); + + try { + my @cmd = ('config'); + unshift @cmd, $self if $self; + if (wantarray) { + return command(@cmd, '--get-all', $var); + } else { + return command_oneline(@cmd, '--get', $var); + } + } catch Git::Error::Command with { + my $E = shift; + if ($E->value() == 1) { + # Key not found. + return; + } else { + throw $E; + } + }; +} + + +=item config_bool ( VARIABLE ) + +Retrieve the bool configuration C<VARIABLE>. The return value +is usable as a boolean in perl (and C<undef> if it's not defined, +of course). + +This currently wraps command('config') so it is not so fast. + +=cut + +sub config_bool { + my ($self, $var) = _maybe_self(@_); + + try { + my @cmd = ('config', '--bool', '--get', $var); + unshift @cmd, $self if $self; + my $val = command_oneline(@cmd); + return undef unless defined $val; + return $val eq 'true'; + } catch Git::Error::Command with { + my $E = shift; + if ($E->value() == 1) { + # Key not found. + return undef; + } else { + throw $E; + } + }; +} + +=item config_int ( VARIABLE ) + +Retrieve the integer configuration C<VARIABLE>. The return value +is simple decimal number. An optional value suffix of 'k', 'm', +or 'g' in the config file will cause the value to be multiplied +by 1024, 1048576 (1024^2), or 1073741824 (1024^3) prior to output. +It would return C<undef> if configuration variable is not defined, + +This currently wraps command('config') so it is not so fast. + +=cut + +sub config_int { + my ($self, $var) = _maybe_self(@_); + + try { + my @cmd = ('config', '--int', '--get', $var); + unshift @cmd, $self if $self; + return command_oneline(@cmd); + } catch Git::Error::Command with { + my $E = shift; + if ($E->value() == 1) { + # Key not found. + return undef; + } else { + throw $E; + } + }; +} + +=item get_colorbool ( NAME ) + +Finds if color should be used for NAMEd operation from the configuration, +and returns boolean (true for "use color", false for "do not use color"). + +=cut + +sub get_colorbool { + my ($self, $var) = @_; + my $stdout_to_tty = (-t STDOUT) ? "true" : "false"; + my $use_color = $self->command_oneline('config', '--get-colorbool', + $var, $stdout_to_tty); + return ($use_color eq 'true'); +} + +=item get_color ( SLOT, COLOR ) + +Finds color for SLOT from the configuration, while defaulting to COLOR, +and returns the ANSI color escape sequence: + + print $repo->get_color("color.interactive.prompt", "underline blue white"); + print "some text"; + print $repo->get_color("", "normal"); + +=cut + +sub get_color { + my ($self, $slot, $default) = @_; + my $color = $self->command_oneline('config', '--get-color', $slot, $default); + if (!defined $color) { + $color = ""; + } + return $color; +} + +=item remote_refs ( REPOSITORY [, GROUPS [, REFGLOBS ] ] ) + +This function returns a hashref of refs stored in a given remote repository. +The hash is in the format C<refname =\> hash>. For tags, the C<refname> entry +contains the tag object while a C<refname^{}> entry gives the tagged objects. + +C<REPOSITORY> has the same meaning as the appropriate C<git-ls-remote> +argument; either an URL or a remote name (if called on a repository instance). +C<GROUPS> is an optional arrayref that can contain 'tags' to return all the +tags and/or 'heads' to return all the heads. C<REFGLOB> is an optional array +of strings containing a shell-like glob to further limit the refs returned in +the hash; the meaning is again the same as the appropriate C<git-ls-remote> +argument. + +This function may or may not be called on a repository instance. In the former +case, remote names as defined in the repository are recognized as repository +specifiers. + +=cut + +sub remote_refs { + my ($self, $repo, $groups, $refglobs) = _maybe_self(@_); + my @args; + if (ref $groups eq 'ARRAY') { + foreach (@$groups) { + if ($_ eq 'heads') { + push (@args, '--heads'); + } elsif ($_ eq 'tags') { + push (@args, '--tags'); + } else { + # Ignore unknown groups for future + # compatibility + } + } + } + push (@args, $repo); + if (ref $refglobs eq 'ARRAY') { + push (@args, @$refglobs); + } + + my @self = $self ? ($self) : (); # Ultra trickery + my ($fh, $ctx) = Git::command_output_pipe(@self, 'ls-remote', @args); + my %refs; + while (<$fh>) { + chomp; + my ($hash, $ref) = split(/\t/, $_, 2); + $refs{$ref} = $hash; + } + Git::command_close_pipe(@self, $fh, $ctx); + return \%refs; +} + + +=item ident ( TYPE | IDENTSTR ) + +=item ident_person ( TYPE | IDENTSTR | IDENTARRAY ) + +This suite of functions retrieves and parses ident information, as stored +in the commit and tag objects or produced by C<var GIT_type_IDENT> (thus +C<TYPE> can be either I<author> or I<committer>; case is insignificant). + +The C<ident> method retrieves the ident information from C<git var> +and either returns it as a scalar string or as an array with the fields parsed. +Alternatively, it can take a prepared ident string (e.g. from the commit +object) and just parse it. + +C<ident_person> returns the person part of the ident - name and email; +it can take the same arguments as C<ident> or the array returned by C<ident>. + +The synopsis is like: + + my ($name, $email, $time_tz) = ident('author'); + "$name <$email>" eq ident_person('author'); + "$name <$email>" eq ident_person($name); + $time_tz =~ /^\d+ [+-]\d{4}$/; + +=cut + +sub ident { + my ($self, $type) = _maybe_self(@_); + my $identstr; + if (lc $type eq lc 'committer' or lc $type eq lc 'author') { + my @cmd = ('var', 'GIT_'.uc($type).'_IDENT'); + unshift @cmd, $self if $self; + $identstr = command_oneline(@cmd); + } else { + $identstr = $type; + } + if (wantarray) { + return $identstr =~ /^(.*) <(.*)> (\d+ [+-]\d{4})$/; + } else { + return $identstr; + } +} + +sub ident_person { + my ($self, @ident) = _maybe_self(@_); + $#ident == 0 and @ident = $self ? $self->ident($ident[0]) : ident($ident[0]); + return "$ident[0] <$ident[1]>"; +} + + +=item hash_object ( TYPE, FILENAME ) + +Compute the SHA1 object id of the given C<FILENAME> considering it is +of the C<TYPE> object type (C<blob>, C<commit>, C<tree>). + +The method can be called without any instance or on a specified Git repository, +it makes zero difference. + +The function returns the SHA1 hash. + +=cut + +# TODO: Support for passing FILEHANDLE instead of FILENAME +sub hash_object { + my ($self, $type, $file) = _maybe_self(@_); + command_oneline('hash-object', '-t', $type, $file); +} + + +=item hash_and_insert_object ( FILENAME ) + +Compute the SHA1 object id of the given C<FILENAME> and add the object to the +object database. + +The function returns the SHA1 hash. + +=cut + +# TODO: Support for passing FILEHANDLE instead of FILENAME +sub hash_and_insert_object { + my ($self, $filename) = @_; + + carp "Bad filename \"$filename\"" if $filename =~ /[\r\n]/; + + $self->_open_hash_and_insert_object_if_needed(); + my ($in, $out) = ($self->{hash_object_in}, $self->{hash_object_out}); + + unless (print $out $filename, "\n") { + $self->_close_hash_and_insert_object(); + throw Error::Simple("out pipe went bad"); + } + + chomp(my $hash = <$in>); + unless (defined($hash)) { + $self->_close_hash_and_insert_object(); + throw Error::Simple("in pipe went bad"); + } + + return $hash; +} + +sub _open_hash_and_insert_object_if_needed { + my ($self) = @_; + + return if defined($self->{hash_object_pid}); + + ($self->{hash_object_pid}, $self->{hash_object_in}, + $self->{hash_object_out}, $self->{hash_object_ctx}) = + command_bidi_pipe(qw(hash-object -w --stdin-paths)); +} + +sub _close_hash_and_insert_object { + my ($self) = @_; + + return unless defined($self->{hash_object_pid}); + + my @vars = map { 'hash_object_' . $_ } qw(pid in out ctx); + + command_close_bidi_pipe(@$self{@vars}); + delete @$self{@vars}; +} + +=item cat_blob ( SHA1, FILEHANDLE ) + +Prints the contents of the blob identified by C<SHA1> to C<FILEHANDLE> and +returns the number of bytes printed. + +=cut + +sub cat_blob { + my ($self, $sha1, $fh) = @_; + + $self->_open_cat_blob_if_needed(); + my ($in, $out) = ($self->{cat_blob_in}, $self->{cat_blob_out}); + + unless (print $out $sha1, "\n") { + $self->_close_cat_blob(); + throw Error::Simple("out pipe went bad"); + } + + my $description = <$in>; + if ($description =~ / missing$/) { + carp "$sha1 doesn't exist in the repository"; + return -1; + } + + if ($description !~ /^[0-9a-fA-F]{40} \S+ (\d+)$/) { + carp "Unexpected result returned from git cat-file"; + return -1; + } + + my $size = $1; + + my $blob; + my $bytesRead = 0; + + while (1) { + my $bytesLeft = $size - $bytesRead; + last unless $bytesLeft; + + my $bytesToRead = $bytesLeft < 1024 ? $bytesLeft : 1024; + my $read = read($in, $blob, $bytesToRead, $bytesRead); + unless (defined($read)) { + $self->_close_cat_blob(); + throw Error::Simple("in pipe went bad"); + } + + $bytesRead += $read; + } + + # Skip past the trailing newline. + my $newline; + my $read = read($in, $newline, 1); + unless (defined($read)) { + $self->_close_cat_blob(); + throw Error::Simple("in pipe went bad"); + } + unless ($read == 1 && $newline eq "\n") { + $self->_close_cat_blob(); + throw Error::Simple("didn't find newline after blob"); + } + + unless (print $fh $blob) { + $self->_close_cat_blob(); + throw Error::Simple("couldn't write to passed in filehandle"); + } + + return $size; +} + +sub _open_cat_blob_if_needed { + my ($self) = @_; + + return if defined($self->{cat_blob_pid}); + + ($self->{cat_blob_pid}, $self->{cat_blob_in}, + $self->{cat_blob_out}, $self->{cat_blob_ctx}) = + command_bidi_pipe(qw(cat-file --batch)); +} + +sub _close_cat_blob { + my ($self) = @_; + + return unless defined($self->{cat_blob_pid}); + + my @vars = map { 'cat_blob_' . $_ } qw(pid in out ctx); + + command_close_bidi_pipe(@$self{@vars}); + delete @$self{@vars}; +} + + +{ # %TEMP_* Lexical Context + +my (%TEMP_FILEMAP, %TEMP_FILES); + +=item temp_acquire ( NAME ) + +Attempts to retreive the temporary file mapped to the string C<NAME>. If an +associated temp file has not been created this session or was closed, it is +created, cached, and set for autoflush and binmode. + +Internally locks the file mapped to C<NAME>. This lock must be released with +C<temp_release()> when the temp file is no longer needed. Subsequent attempts +to retrieve temporary files mapped to the same C<NAME> while still locked will +cause an error. This locking mechanism provides a weak guarantee and is not +threadsafe. It does provide some error checking to help prevent temp file refs +writing over one another. + +In general, the L<File::Handle> returned should not be closed by consumers as +it defeats the purpose of this caching mechanism. If you need to close the temp +file handle, then you should use L<File::Temp> or another temp file faculty +directly. If a handle is closed and then requested again, then a warning will +issue. + +=cut + +sub temp_acquire { + my $temp_fd = _temp_cache(@_); + + $TEMP_FILES{$temp_fd}{locked} = 1; + $temp_fd; +} + +=item temp_release ( NAME ) + +=item temp_release ( FILEHANDLE ) + +Releases a lock acquired through C<temp_acquire()>. Can be called either with +the C<NAME> mapping used when acquiring the temp file or with the C<FILEHANDLE> +referencing a locked temp file. + +Warns if an attempt is made to release a file that is not locked. + +The temp file will be truncated before being released. This can help to reduce +disk I/O where the system is smart enough to detect the truncation while data +is in the output buffers. Beware that after the temp file is released and +truncated, any operations on that file may fail miserably until it is +re-acquired. All contents are lost between each release and acquire mapped to +the same string. + +=cut + +sub temp_release { + my ($self, $temp_fd, $trunc) = _maybe_self(@_); + + if (exists $TEMP_FILEMAP{$temp_fd}) { + $temp_fd = $TEMP_FILES{$temp_fd}; + } + unless ($TEMP_FILES{$temp_fd}{locked}) { + carp "Attempt to release temp file '", + $temp_fd, "' that has not been locked"; + } + temp_reset($temp_fd) if $trunc and $temp_fd->opened; + + $TEMP_FILES{$temp_fd}{locked} = 0; + undef; +} + +sub _temp_cache { + my ($self, $name) = _maybe_self(@_); + + _verify_require(); + + my $temp_fd = \$TEMP_FILEMAP{$name}; + if (defined $$temp_fd and $$temp_fd->opened) { + if ($TEMP_FILES{$$temp_fd}{locked}) { + throw Error::Simple("Temp file with moniker '" . + $name . "' already in use"); + } + } else { + if (defined $$temp_fd) { + # then we're here because of a closed handle. + carp "Temp file '", $name, + "' was closed. Opening replacement."; + } + my $fname; + + my $tmpdir; + if (defined $self) { + $tmpdir = $self->repo_path(); + } + + ($$temp_fd, $fname) = File::Temp->tempfile( + 'Git_XXXXXX', UNLINK => 1, DIR => $tmpdir, + ) or throw Error::Simple("couldn't open new temp file"); + + $$temp_fd->autoflush; + binmode $$temp_fd; + $TEMP_FILES{$$temp_fd}{fname} = $fname; + } + $$temp_fd; +} + +sub _verify_require { + eval { require File::Temp; require File::Spec; }; + $@ and throw Error::Simple($@); +} + +=item temp_reset ( FILEHANDLE ) + +Truncates and resets the position of the C<FILEHANDLE>. + +=cut + +sub temp_reset { + my ($self, $temp_fd) = _maybe_self(@_); + + truncate $temp_fd, 0 + or throw Error::Simple("couldn't truncate file"); + sysseek($temp_fd, 0, SEEK_SET) and seek($temp_fd, 0, SEEK_SET) + or throw Error::Simple("couldn't seek to beginning of file"); + sysseek($temp_fd, 0, SEEK_CUR) == 0 and tell($temp_fd) == 0 + or throw Error::Simple("expected file position to be reset"); +} + +=item temp_path ( NAME ) + +=item temp_path ( FILEHANDLE ) + +Returns the filename associated with the given tempfile. + +=cut + +sub temp_path { + my ($self, $temp_fd) = _maybe_self(@_); + + if (exists $TEMP_FILEMAP{$temp_fd}) { + $temp_fd = $TEMP_FILEMAP{$temp_fd}; + } + $TEMP_FILES{$temp_fd}{fname}; +} + +sub END { + unlink values %TEMP_FILEMAP if %TEMP_FILEMAP; +} + +} # %TEMP_* Lexical Context + +=back + +=head1 ERROR HANDLING + +All functions are supposed to throw Perl exceptions in case of errors. +See the L<Error> module on how to catch those. Most exceptions are mere +L<Error::Simple> instances. + +However, the C<command()>, C<command_oneline()> and C<command_noisy()> +functions suite can throw C<Git::Error::Command> exceptions as well: those are +thrown when the external command returns an error code and contain the error +code as well as access to the captured command's output. The exception class +provides the usual C<stringify> and C<value> (command's exit code) methods and +in addition also a C<cmd_output> method that returns either an array or a +string with the captured command output (depending on the original function +call context; C<command_noisy()> returns C<undef>) and $<cmdline> which +returns the command and its arguments (but without proper quoting). + +Note that the C<command_*_pipe()> functions cannot throw this exception since +it has no idea whether the command failed or not. You will only find out +at the time you C<close> the pipe; if you want to have that automated, +use C<command_close_pipe()>, which can throw the exception. + +=cut + +{ + package Git::Error::Command; + + @Git::Error::Command::ISA = qw(Error); + + sub new { + my $self = shift; + my $cmdline = '' . shift; + my $value = 0 + shift; + my $outputref = shift; + my(@args) = (); + + local $Error::Depth = $Error::Depth + 1; + + push(@args, '-cmdline', $cmdline); + push(@args, '-value', $value); + push(@args, '-outputref', $outputref); + + $self->SUPER::new(-text => 'command returned error', @args); + } + + sub stringify { + my $self = shift; + my $text = $self->SUPER::stringify; + $self->cmdline() . ': ' . $text . ': ' . $self->value() . "\n"; + } + + sub cmdline { + my $self = shift; + $self->{'-cmdline'}; + } + + sub cmd_output { + my $self = shift; + my $ref = $self->{'-outputref'}; + defined $ref or undef; + if (ref $ref eq 'ARRAY') { + return @$ref; + } else { # SCALAR + return $$ref; + } + } +} + +=over 4 + +=item git_cmd_try { CODE } ERRMSG + +This magical statement will automatically catch any C<Git::Error::Command> +exceptions thrown by C<CODE> and make your program die with C<ERRMSG> +on its lips; the message will have %s substituted for the command line +and %d for the exit status. This statement is useful mostly for producing +more user-friendly error messages. + +In case of no exception caught the statement returns C<CODE>'s return value. + +Note that this is the only auto-exported function. + +=cut + +sub git_cmd_try(&$) { + my ($code, $errmsg) = @_; + my @result; + my $err; + my $array = wantarray; + try { + if ($array) { + @result = &$code; + } else { + $result[0] = &$code; + } + } catch Git::Error::Command with { + my $E = shift; + $err = $errmsg; + $err =~ s/\%s/$E->cmdline()/ge; + $err =~ s/\%d/$E->value()/ge; + # We can't croak here since Error.pm would mangle + # that to Error::Simple. + }; + $err and croak $err; + return $array ? @result : $result[0]; +} + + +=back + +=head1 COPYRIGHT + +Copyright 2006 by Petr Baudis E<lt>pasky@suse.czE<gt>. + +This module is free software; it may be used, copied, modified +and distributed under the terms of the GNU General Public Licence, +either version 2, or (at your option) any later version. + +=cut + + +# Take raw method argument list and return ($obj, @args) in case +# the method was called upon an instance and (undef, @args) if +# it was called directly. +sub _maybe_self { + UNIVERSAL::isa($_[0], 'Git') ? @_ : (undef, @_); +} + +# Check if the command id is something reasonable. +sub _check_valid_cmd { + my ($cmd) = @_; + $cmd =~ /^[a-z0-9A-Z_-]+$/ or throw Error::Simple("bad command: $cmd"); +} + +# Common backend for the pipe creators. +sub _command_common_pipe { + my $direction = shift; + my ($self, @p) = _maybe_self(@_); + my (%opts, $cmd, @args); + if (ref $p[0]) { + ($cmd, @args) = @{shift @p}; + %opts = ref $p[0] ? %{$p[0]} : @p; + } else { + ($cmd, @args) = @p; + } + _check_valid_cmd($cmd); + + my $fh; + if ($^O eq 'MSWin32') { + # ActiveState Perl + #defined $opts{STDERR} and + # warn 'ignoring STDERR option - running w/ ActiveState'; + $direction eq '-|' or + die 'input pipe for ActiveState not implemented'; + # the strange construction with *ACPIPE is just to + # explain the tie below that we want to bind to + # a handle class, not scalar. It is not known if + # it is something specific to ActiveState Perl or + # just a Perl quirk. + tie (*ACPIPE, 'Git::activestate_pipe', $cmd, @args); + $fh = *ACPIPE; + + } else { + my $pid = open($fh, $direction); + if (not defined $pid) { + throw Error::Simple("open failed: $!"); + } elsif ($pid == 0) { + if (defined $opts{STDERR}) { + close STDERR; + } + if ($opts{STDERR}) { + open (STDERR, '>&', $opts{STDERR}) + or die "dup failed: $!"; + } + _cmd_exec($self, $cmd, @args); + } + } + return wantarray ? ($fh, join(' ', $cmd, @args)) : $fh; +} + +# When already in the subprocess, set up the appropriate state +# for the given repository and execute the git command. +sub _cmd_exec { + my ($self, @args) = @_; + if ($self) { + $self->repo_path() and $ENV{'GIT_DIR'} = $self->repo_path(); + $self->repo_path() and $self->wc_path() + and $ENV{'GIT_WORK_TREE'} = $self->wc_path(); + $self->wc_path() and chdir($self->wc_path()); + $self->wc_subdir() and chdir($self->wc_subdir()); + } + _execv_git_cmd(@args); + die qq[exec "@args" failed: $!]; +} + +# Execute the given Git command ($_[0]) with arguments ($_[1..]) +# by searching for it at proper places. +sub _execv_git_cmd { exec('git', @_); } + +# Close pipe to a subprocess. +sub _cmd_close { + my ($fh, $ctx) = @_; + if (not close $fh) { + if ($!) { + # It's just close, no point in fatalities + carp "error closing pipe: $!"; + } elsif ($? >> 8) { + # The caller should pepper this. + throw Git::Error::Command($ctx, $? >> 8); + } + # else we might e.g. closed a live stream; the command + # dying of SIGPIPE would drive us here. + } +} + + +sub DESTROY { + my ($self) = @_; + $self->_close_hash_and_insert_object(); + $self->_close_cat_blob(); +} + + +# Pipe implementation for ActiveState Perl. + +package Git::activestate_pipe; +use strict; + +sub TIEHANDLE { + my ($class, @params) = @_; + # FIXME: This is probably horrible idea and the thing will explode + # at the moment you give it arguments that require some quoting, + # but I have no ActiveState clue... --pasky + # Let's just hope ActiveState Perl does at least the quoting + # correctly. + my @data = qx{git @params}; + bless { i => 0, data => \@data }, $class; +} + +sub READLINE { + my $self = shift; + if ($self->{i} >= scalar @{$self->{data}}) { + return undef; + } + my $i = $self->{i}; + if (wantarray) { + $self->{i} = $#{$self->{'data'}} + 1; + return splice(@{$self->{'data'}}, $i); + } + $self->{i} = $i + 1; + return $self->{'data'}->[ $i ]; +} + +sub CLOSE { + my $self = shift; + delete $self->{data}; + delete $self->{i}; +} + +sub EOF { + my $self = shift; + return ($self->{i} >= scalar @{$self->{data}}); +} + + +1; # Famous last words diff --git a/perl/Makefile b/perl/Makefile new file mode 100644 index 0000000000..4ab21d61b8 --- /dev/null +++ b/perl/Makefile @@ -0,0 +1,47 @@ +# +# Makefile for perl support modules and routine +# +makfile:=perl.mak + +PERL_PATH_SQ = $(subst ','\'',$(PERL_PATH)) +prefix_SQ = $(subst ','\'',$(prefix)) + +ifndef V + QUIET = @ +endif + +all install instlibdir: $(makfile) + $(QUIET)$(MAKE) -f $(makfile) $@ + +clean: + $(QUIET)test -f $(makfile) && $(MAKE) -f $(makfile) $@ || exit 0 + $(RM) ppport.h + $(RM) $(makfile) + $(RM) $(makfile).old + +ifdef NO_PERL_MAKEMAKER +instdir_SQ = $(subst ','\'',$(prefix)/lib) +$(makfile): ../GIT-CFLAGS Makefile + echo all: private-Error.pm Git.pm > $@ + echo ' mkdir -p blib/lib' >> $@ + echo ' $(RM) blib/lib/Git.pm; cp Git.pm blib/lib/' >> $@ + echo ' $(RM) blib/lib/Error.pm' >> $@ + '$(PERL_PATH_SQ)' -MError -e 'exit($$Error::VERSION < 0.15009)' || \ + echo ' cp private-Error.pm blib/lib/Error.pm' >> $@ + echo install: >> $@ + echo ' mkdir -p "$$(DESTDIR)$(instdir_SQ)"' >> $@ + echo ' $(RM) "$$(DESTDIR)$(instdir_SQ)/Git.pm"; cp Git.pm "$$(DESTDIR)$(instdir_SQ)"' >> $@ + echo ' $(RM) "$$(DESTDIR)$(instdir_SQ)/Error.pm"' >> $@ + '$(PERL_PATH_SQ)' -MError -e 'exit($$Error::VERSION < 0.15009)' || \ + echo ' cp private-Error.pm "$$(DESTDIR)$(instdir_SQ)/Error.pm"' >> $@ + echo instlibdir: >> $@ + echo ' echo $(instdir_SQ)' >> $@ +else +$(makfile): Makefile.PL ../GIT-CFLAGS + $(PERL_PATH) $< PREFIX='$(prefix_SQ)' +endif + +# this is just added comfort for calling make directly in perl dir +# (even though GIT-CFLAGS aren't used yet. If ever) +../GIT-CFLAGS: + $(MAKE) -C .. GIT-CFLAGS diff --git a/perl/Makefile.PL b/perl/Makefile.PL new file mode 100644 index 0000000000..0b9deca2cc --- /dev/null +++ b/perl/Makefile.PL @@ -0,0 +1,38 @@ +use ExtUtils::MakeMaker; + +sub MY::postamble { + return <<'MAKE_FRAG'; +instlibdir: + @echo '$(INSTALLSITELIB)' + +ifneq (,$(DESTDIR)) +ifeq (0,$(shell expr '$(MM_VERSION)' '>' 6.10)) +$(error ExtUtils::MakeMaker version "$(MM_VERSION)" is older than 6.11 and so \ + is likely incompatible with the DESTDIR mechanism. Try setting \ + NO_PERL_MAKEMAKER=1 instead) +endif +endif + +MAKE_FRAG +} + +my %pm = ('Git.pm' => '$(INST_LIBDIR)/Git.pm'); + +# We come with our own bundled Error.pm. It's not in the set of default +# Perl modules so install it if it's not available on the system yet. +eval { require Error }; +if ($@ || $Error::VERSION < 0.15009) { + $pm{'private-Error.pm'} = '$(INST_LIBDIR)/Error.pm'; +} + +# redirect stdout, otherwise the message "Writing perl.mak for Git" +# disrupts the output for the target 'instlibdir' +open STDOUT, ">&STDERR"; + +WriteMakefile( + NAME => 'Git', + VERSION_FROM => 'Git.pm', + PM => \%pm, + MAKEFILE => 'perl.mak', + INSTALLSITEMAN3DIR => '$(SITEPREFIX)/share/man/man3' +); diff --git a/perl/private-Error.pm b/perl/private-Error.pm new file mode 100644 index 0000000000..11e9cd9a02 --- /dev/null +++ b/perl/private-Error.pm @@ -0,0 +1,827 @@ +# Error.pm +# +# Copyright (c) 1997-8 Graham Barr <gbarr@ti.com>. All rights reserved. +# This program is free software; you can redistribute it and/or +# modify it under the same terms as Perl itself. +# +# Based on my original Error.pm, and Exceptions.pm by Peter Seibel +# <peter@weblogic.com> and adapted by Jesse Glick <jglick@sig.bsh.com>. +# +# but modified ***significantly*** + +package Error; + +use strict; +use vars qw($VERSION); +use 5.004; + +$VERSION = "0.15009"; + +use overload ( + '""' => 'stringify', + '0+' => 'value', + 'bool' => sub { return 1; }, + 'fallback' => 1 +); + +$Error::Depth = 0; # Depth to pass to caller() +$Error::Debug = 0; # Generate verbose stack traces +@Error::STACK = (); # Clause stack for try +$Error::THROWN = undef; # last error thrown, a workaround until die $ref works + +my $LAST; # Last error created +my %ERROR; # Last error associated with package + +sub throw_Error_Simple +{ + my $args = shift; + return Error::Simple->new($args->{'text'}); +} + +$Error::ObjectifyCallback = \&throw_Error_Simple; + + +# Exported subs are defined in Error::subs + +sub import { + shift; + local $Exporter::ExportLevel = $Exporter::ExportLevel + 1; + Error::subs->import(@_); +} + +# I really want to use last for the name of this method, but it is a keyword +# which prevent the syntax last Error + +sub prior { + shift; # ignore + + return $LAST unless @_; + + my $pkg = shift; + return exists $ERROR{$pkg} ? $ERROR{$pkg} : undef + unless ref($pkg); + + my $obj = $pkg; + my $err = undef; + if($obj->isa('HASH')) { + $err = $obj->{'__Error__'} + if exists $obj->{'__Error__'}; + } + elsif($obj->isa('GLOB')) { + $err = ${*$obj}{'__Error__'} + if exists ${*$obj}{'__Error__'}; + } + + $err; +} + +sub flush { + shift; #ignore + + unless (@_) { + $LAST = undef; + return; + } + + my $pkg = shift; + return unless ref($pkg); + + undef $ERROR{$pkg} if defined $ERROR{$pkg}; +} + +# Return as much information as possible about where the error +# happened. The -stacktrace element only exists if $Error::DEBUG +# was set when the error was created + +sub stacktrace { + my $self = shift; + + return $self->{'-stacktrace'} + if exists $self->{'-stacktrace'}; + + my $text = exists $self->{'-text'} ? $self->{'-text'} : "Died"; + + $text .= sprintf(" at %s line %d.\n", $self->file, $self->line) + unless($text =~ /\n$/s); + + $text; +} + +# Allow error propagation, ie +# +# $ber->encode(...) or +# return Error->prior($ber)->associate($ldap); + +sub associate { + my $err = shift; + my $obj = shift; + + return unless ref($obj); + + if($obj->isa('HASH')) { + $obj->{'__Error__'} = $err; + } + elsif($obj->isa('GLOB')) { + ${*$obj}{'__Error__'} = $err; + } + $obj = ref($obj); + $ERROR{ ref($obj) } = $err; + + return; +} + +sub new { + my $self = shift; + my($pkg,$file,$line) = caller($Error::Depth); + + my $err = bless { + '-package' => $pkg, + '-file' => $file, + '-line' => $line, + @_ + }, $self; + + $err->associate($err->{'-object'}) + if(exists $err->{'-object'}); + + # To always create a stacktrace would be very inefficient, so + # we only do it if $Error::Debug is set + + if($Error::Debug) { + require Carp; + local $Carp::CarpLevel = $Error::Depth; + my $text = defined($err->{'-text'}) ? $err->{'-text'} : "Error"; + my $trace = Carp::longmess($text); + # Remove try calls from the trace + $trace =~ s/(\n\s+\S+__ANON__[^\n]+)?\n\s+eval[^\n]+\n\s+Error::subs::try[^\n]+(?=\n)//sog; + $trace =~ s/(\n\s+\S+__ANON__[^\n]+)?\n\s+eval[^\n]+\n\s+Error::subs::run_clauses[^\n]+\n\s+Error::subs::try[^\n]+(?=\n)//sog; + $err->{'-stacktrace'} = $trace + } + + $@ = $LAST = $ERROR{$pkg} = $err; +} + +# Throw an error. this contains some very gory code. + +sub throw { + my $self = shift; + local $Error::Depth = $Error::Depth + 1; + + # if we are not rethrow-ing then create the object to throw + $self = $self->new(@_) unless ref($self); + + die $Error::THROWN = $self; +} + +# syntactic sugar for +# +# die with Error( ... ); + +sub with { + my $self = shift; + local $Error::Depth = $Error::Depth + 1; + + $self->new(@_); +} + +# syntactic sugar for +# +# record Error( ... ) and return; + +sub record { + my $self = shift; + local $Error::Depth = $Error::Depth + 1; + + $self->new(@_); +} + +# catch clause for +# +# try { ... } catch CLASS with { ... } + +sub catch { + my $pkg = shift; + my $code = shift; + my $clauses = shift || {}; + my $catch = $clauses->{'catch'} ||= []; + + unshift @$catch, $pkg, $code; + + $clauses; +} + +# Object query methods + +sub object { + my $self = shift; + exists $self->{'-object'} ? $self->{'-object'} : undef; +} + +sub file { + my $self = shift; + exists $self->{'-file'} ? $self->{'-file'} : undef; +} + +sub line { + my $self = shift; + exists $self->{'-line'} ? $self->{'-line'} : undef; +} + +sub text { + my $self = shift; + exists $self->{'-text'} ? $self->{'-text'} : undef; +} + +# overload methods + +sub stringify { + my $self = shift; + defined $self->{'-text'} ? $self->{'-text'} : "Died"; +} + +sub value { + my $self = shift; + exists $self->{'-value'} ? $self->{'-value'} : undef; +} + +package Error::Simple; + +@Error::Simple::ISA = qw(Error); + +sub new { + my $self = shift; + my $text = "" . shift; + my $value = shift; + my(@args) = (); + + local $Error::Depth = $Error::Depth + 1; + + @args = ( -file => $1, -line => $2) + if($text =~ s/\s+at\s+(\S+)\s+line\s+(\d+)(?:,\s*<[^>]*>\s+line\s+\d+)?\.?\n?$//s); + push(@args, '-value', 0 + $value) + if defined($value); + + $self->SUPER::new(-text => $text, @args); +} + +sub stringify { + my $self = shift; + my $text = $self->SUPER::stringify; + $text .= sprintf(" at %s line %d.\n", $self->file, $self->line) + unless($text =~ /\n$/s); + $text; +} + +########################################################################## +########################################################################## + +# Inspired by code from Jesse Glick <jglick@sig.bsh.com> and +# Peter Seibel <peter@weblogic.com> + +package Error::subs; + +use Exporter (); +use vars qw(@EXPORT_OK @ISA %EXPORT_TAGS); + +@EXPORT_OK = qw(try with finally except otherwise); +%EXPORT_TAGS = (try => \@EXPORT_OK); + +@ISA = qw(Exporter); + + +sub blessed { + my $item = shift; + local $@; # don't kill an outer $@ + ref $item and eval { $item->can('can') }; +} + + +sub run_clauses ($$$\@) { + my($clauses,$err,$wantarray,$result) = @_; + my $code = undef; + + $err = $Error::ObjectifyCallback->({'text' =>$err}) unless ref($err); + + CATCH: { + + # catch + my $catch; + if(defined($catch = $clauses->{'catch'})) { + my $i = 0; + + CATCHLOOP: + for( ; $i < @$catch ; $i += 2) { + my $pkg = $catch->[$i]; + unless(defined $pkg) { + #except + splice(@$catch,$i,2,$catch->[$i+1]->()); + $i -= 2; + next CATCHLOOP; + } + elsif(blessed($err) && $err->isa($pkg)) { + $code = $catch->[$i+1]; + while(1) { + my $more = 0; + local($Error::THROWN); + my $ok = eval { + if($wantarray) { + @{$result} = $code->($err,\$more); + } + elsif(defined($wantarray)) { + @{$result} = (); + $result->[0] = $code->($err,\$more); + } + else { + $code->($err,\$more); + } + 1; + }; + if( $ok ) { + next CATCHLOOP if $more; + undef $err; + } + else { + $err = defined($Error::THROWN) + ? $Error::THROWN : $@; + $err = $Error::ObjectifyCallback->({'text' =>$err}) + unless ref($err); + } + last CATCH; + }; + } + } + } + + # otherwise + my $owise; + if(defined($owise = $clauses->{'otherwise'})) { + my $code = $clauses->{'otherwise'}; + my $more = 0; + my $ok = eval { + if($wantarray) { + @{$result} = $code->($err,\$more); + } + elsif(defined($wantarray)) { + @{$result} = (); + $result->[0] = $code->($err,\$more); + } + else { + $code->($err,\$more); + } + 1; + }; + if( $ok ) { + undef $err; + } + else { + $err = defined($Error::THROWN) + ? $Error::THROWN : $@; + + $err = $Error::ObjectifyCallback->({'text' =>$err}) + unless ref($err); + } + } + } + $err; +} + +sub try (&;$) { + my $try = shift; + my $clauses = @_ ? shift : {}; + my $ok = 0; + my $err = undef; + my @result = (); + + unshift @Error::STACK, $clauses; + + my $wantarray = wantarray(); + + do { + local $Error::THROWN = undef; + local $@ = undef; + + $ok = eval { + if($wantarray) { + @result = $try->(); + } + elsif(defined $wantarray) { + $result[0] = $try->(); + } + else { + $try->(); + } + 1; + }; + + $err = defined($Error::THROWN) ? $Error::THROWN : $@ + unless $ok; + }; + + shift @Error::STACK; + + $err = run_clauses($clauses,$err,wantarray,@result) + unless($ok); + + $clauses->{'finally'}->() + if(defined($clauses->{'finally'})); + + if (defined($err)) + { + if (blessed($err) && $err->can('throw')) + { + throw $err; + } + else + { + die $err; + } + } + + wantarray ? @result : $result[0]; +} + +# Each clause adds a sub to the list of clauses. The finally clause is +# always the last, and the otherwise clause is always added just before +# the finally clause. +# +# All clauses, except the finally clause, add a sub which takes one argument +# this argument will be the error being thrown. The sub will return a code ref +# if that clause can handle that error, otherwise undef is returned. +# +# The otherwise clause adds a sub which unconditionally returns the users +# code reference, this is why it is forced to be last. +# +# The catch clause is defined in Error.pm, as the syntax causes it to +# be called as a method + +sub with (&;$) { + @_ +} + +sub finally (&) { + my $code = shift; + my $clauses = { 'finally' => $code }; + $clauses; +} + +# The except clause is a block which returns a hashref or a list of +# key-value pairs, where the keys are the classes and the values are subs. + +sub except (&;$) { + my $code = shift; + my $clauses = shift || {}; + my $catch = $clauses->{'catch'} ||= []; + + my $sub = sub { + my $ref; + my(@array) = $code->($_[0]); + if(@array == 1 && ref($array[0])) { + $ref = $array[0]; + $ref = [ %$ref ] + if(UNIVERSAL::isa($ref,'HASH')); + } + else { + $ref = \@array; + } + @$ref + }; + + unshift @{$catch}, undef, $sub; + + $clauses; +} + +sub otherwise (&;$) { + my $code = shift; + my $clauses = shift || {}; + + if(exists $clauses->{'otherwise'}) { + require Carp; + Carp::croak("Multiple otherwise clauses"); + } + + $clauses->{'otherwise'} = $code; + + $clauses; +} + +1; +__END__ + +=head1 NAME + +Error - Error/exception handling in an OO-ish way + +=head1 SYNOPSIS + + use Error qw(:try); + + throw Error::Simple( "A simple error"); + + sub xyz { + ... + record Error::Simple("A simple error") + and return; + } + + unlink($file) or throw Error::Simple("$file: $!",$!); + + try { + do_some_stuff(); + die "error!" if $condition; + throw Error::Simple -text => "Oops!" if $other_condition; + } + catch Error::IO with { + my $E = shift; + print STDERR "File ", $E->{'-file'}, " had a problem\n"; + } + except { + my $E = shift; + my $general_handler=sub {send_message $E->{-description}}; + return { + UserException1 => $general_handler, + UserException2 => $general_handler + }; + } + otherwise { + print STDERR "Well I don't know what to say\n"; + } + finally { + close_the_garage_door_already(); # Should be reliable + }; # Don't forget the trailing ; or you might be surprised + +=head1 DESCRIPTION + +The C<Error> package provides two interfaces. Firstly C<Error> provides +a procedural interface to exception handling. Secondly C<Error> is a +base class for errors/exceptions that can either be thrown, for +subsequent catch, or can simply be recorded. + +Errors in the class C<Error> should not be thrown directly, but the +user should throw errors from a sub-class of C<Error>. + +=head1 PROCEDURAL INTERFACE + +C<Error> exports subroutines to perform exception handling. These will +be exported if the C<:try> tag is used in the C<use> line. + +=over 4 + +=item try BLOCK CLAUSES + +C<try> is the main subroutine called by the user. All other subroutines +exported are clauses to the try subroutine. + +The BLOCK will be evaluated and, if no error is throw, try will return +the result of the block. + +C<CLAUSES> are the subroutines below, which describe what to do in the +event of an error being thrown within BLOCK. + +=item catch CLASS with BLOCK + +This clauses will cause all errors that satisfy C<$err-E<gt>isa(CLASS)> +to be caught and handled by evaluating C<BLOCK>. + +C<BLOCK> will be passed two arguments. The first will be the error +being thrown. The second is a reference to a scalar variable. If this +variable is set by the catch block then, on return from the catch +block, try will continue processing as if the catch block was never +found. + +To propagate the error the catch block may call C<$err-E<gt>throw> + +If the scalar reference by the second argument is not set, and the +error is not thrown. Then the current try block will return with the +result from the catch block. + +=item except BLOCK + +When C<try> is looking for a handler, if an except clause is found +C<BLOCK> is evaluated. The return value from this block should be a +HASHREF or a list of key-value pairs, where the keys are class names +and the values are CODE references for the handler of errors of that +type. + +=item otherwise BLOCK + +Catch any error by executing the code in C<BLOCK> + +When evaluated C<BLOCK> will be passed one argument, which will be the +error being processed. + +Only one otherwise block may be specified per try block + +=item finally BLOCK + +Execute the code in C<BLOCK> either after the code in the try block has +successfully completed, or if the try block throws an error then +C<BLOCK> will be executed after the handler has completed. + +If the handler throws an error then the error will be caught, the +finally block will be executed and the error will be re-thrown. + +Only one finally block may be specified per try block + +=back + +=head1 CLASS INTERFACE + +=head2 CONSTRUCTORS + +The C<Error> object is implemented as a HASH. This HASH is initialized +with the arguments that are passed to it's constructor. The elements +that are used by, or are retrievable by the C<Error> class are listed +below, other classes may add to these. + + -file + -line + -text + -value + -object + +If C<-file> or C<-line> are not specified in the constructor arguments +then these will be initialized with the file name and line number where +the constructor was called from. + +If the error is associated with an object then the object should be +passed as the C<-object> argument. This will allow the C<Error> package +to associate the error with the object. + +The C<Error> package remembers the last error created, and also the +last error associated with a package. This could either be the last +error created by a sub in that package, or the last error which passed +an object blessed into that package as the C<-object> argument. + +=over 4 + +=item throw ( [ ARGS ] ) + +Create a new C<Error> object and throw an error, which will be caught +by a surrounding C<try> block, if there is one. Otherwise it will cause +the program to exit. + +C<throw> may also be called on an existing error to re-throw it. + +=item with ( [ ARGS ] ) + +Create a new C<Error> object and returns it. This is defined for +syntactic sugar, eg + + die with Some::Error ( ... ); + +=item record ( [ ARGS ] ) + +Create a new C<Error> object and returns it. This is defined for +syntactic sugar, eg + + record Some::Error ( ... ) + and return; + +=back + +=head2 STATIC METHODS + +=over 4 + +=item prior ( [ PACKAGE ] ) + +Return the last error created, or the last error associated with +C<PACKAGE> + +=item flush ( [ PACKAGE ] ) + +Flush the last error created, or the last error associated with +C<PACKAGE>.It is necessary to clear the error stack before exiting the +package or uncaught errors generated using C<record> will be reported. + + $Error->flush; + +=cut + +=back + +=head2 OBJECT METHODS + +=over 4 + +=item stacktrace + +If the variable C<$Error::Debug> was non-zero when the error was +created, then C<stacktrace> returns a string created by calling +C<Carp::longmess>. If the variable was zero the C<stacktrace> returns +the text of the error appended with the filename and line number of +where the error was created, providing the text does not end with a +newline. + +=item object + +The object this error was associated with + +=item file + +The file where the constructor of this error was called from + +=item line + +The line where the constructor of this error was called from + +=item text + +The text of the error + +=back + +=head2 OVERLOAD METHODS + +=over 4 + +=item stringify + +A method that converts the object into a string. This method may simply +return the same as the C<text> method, or it may append more +information. For example the file name and line number. + +By default this method returns the C<-text> argument that was passed to +the constructor, or the string C<"Died"> if none was given. + +=item value + +A method that will return a value that can be associated with the +error. For example if an error was created due to the failure of a +system call, then this may return the numeric value of C<$!> at the +time. + +By default this method returns the C<-value> argument that was passed +to the constructor. + +=back + +=head1 PRE-DEFINED ERROR CLASSES + +=over 4 + +=item Error::Simple + +This class can be used to hold simple error strings and values. It's +constructor takes two arguments. The first is a text value, the second +is a numeric value. These values are what will be returned by the +overload methods. + +If the text value ends with C<at file line 1> as $@ strings do, then +this infomation will be used to set the C<-file> and C<-line> arguments +of the error object. + +This class is used internally if an eval'd block die's with an error +that is a plain string. (Unless C<$Error::ObjectifyCallback> is modified) + +=back + +=head1 $Error::ObjectifyCallback + +This variable holds a reference to a subroutine that converts errors that +are plain strings to objects. It is used by Error.pm to convert textual +errors to objects, and can be overridden by the user. + +It accepts a single argument which is a hash reference to named parameters. +Currently the only named parameter passed is C<'text'> which is the text +of the error, but others may be available in the future. + +For example the following code will cause Error.pm to throw objects of the +class MyError::Bar by default: + + sub throw_MyError_Bar + { + my $args = shift; + my $err = MyError::Bar->new(); + $err->{'MyBarText'} = $args->{'text'}; + return $err; + } + + { + local $Error::ObjectifyCallback = \&throw_MyError_Bar; + + # Error handling here. + } + +=head1 KNOWN BUGS + +None, but that does not mean there are not any. + +=head1 AUTHORS + +Graham Barr <gbarr@pobox.com> + +The code that inspired me to write this was originally written by +Peter Seibel <peter@weblogic.com> and adapted by Jesse Glick +<jglick@sig.bsh.com>. + +=head1 MAINTAINER + +Shlomi Fish <shlomif@iglu.org.il> + +=head1 PAST MAINTAINERS + +Arun Kumar U <u_arunkumar@yahoo.com> + +=cut |