view release on metacpan or  search on metacpan

lib/Perl/Tidy.pm  view on Meta::CPAN

                    Warn(
"file: $display_name: bad guess to decode source as $encoding_in\n"
                    );
                    $encoding_in = EMPTY_STRING;
                }
                else {
                    $encoding_log_message .= <<EOM;
Guessed encoding '$encoding_in' successfully decoded
EOM
                    $decoded_input_as = $encoding_in;
                    $rinput_string    = \$buf;
                }
            }
        }
        else {
            $encoding_log_message .= <<EOM;
Does not look like utf8 encoded text so processing as raw bytes
EOM
        }
    }

    # Case 4. Decode with a specific encoding
    else {
        $encoding_in = $rOpts_character_encoding;
        my $buf;
        if (
            !eval {
                $buf = Encode::decode( $encoding_in, ${$rinput_string},
                    Encode::FB_CROAK | Encode::LEAVE_SRC );
                1;
            }
          )
        {

            # Quit if we cannot decode by the requested encoding;
            # Something is not right.
            Warn(
"skipping file: $display_name: Unable to decode source as $encoding_in\n"
            );

            # return nothing on error
            return;
        }
        else {
            $encoding_log_message .= <<EOM;
Specified encoding '$encoding_in' successfully decoded
EOM
            $decoded_input_as = $encoding_in;
            $rinput_string    = \$buf;
        }
    }

    # Set the encoding to be used for all further i/o: If we have
    # decoded the data with any format, then we must continue to
    # read and write it as encoded data, and we will normalize these
    # operations with utf8.  If we have not decoded the data, then
    # we must not treat it as encoded data.
    my $is_encoded_data = $encoding_in ? 'utf8' : EMPTY_STRING;
    $self->[_is_encoded_data_] = $is_encoded_data;

    # Delete any Byte Order Mark (BOM), which can cause trouble
    if ($is_encoded_data) {
        ${$rinput_string} =~ s/^\x{FEFF}//;
    }

    $rstatus->{'input_name'}       = $display_name;
    $rstatus->{'opt_encoding'}     = $rOpts_character_encoding;
    $rstatus->{'char_mode_used'}   = $encoding_in ? 1 : 0;
    $rstatus->{'input_decoded_as'} = $decoded_input_as;

    # Define the function to determine the display width of character
    # strings
    my $length_function;
    if ($is_encoded_data) {

        # Try to load Unicode::GCString for defining text display width, if
        # requested, when the first encoded file is encountered
        if ( !defined($loaded_unicode_gcstring) ) {
            if ( eval { require Unicode::GCString; 1 } ) {
                $loaded_unicode_gcstring = 1;
            }
            else {
                $loaded_unicode_gcstring = 0;
                if ( $rOpts->{'use-unicode-gcstring'} ) {
                    Warn(<<EOM);
----------------------
Unable to load Unicode::GCString: $EVAL_ERROR
Processing continues but some vertical alignment may be poor
To prevent this warning message, you can either:
- install module Unicode::GCString, or
- remove '--use-unicode-gcstring' or '-gcs' from your perltidyrc or command line
----------------------
EOM
                }
            }
        }
        if ($loaded_unicode_gcstring) {
            $length_function = sub {
                return Unicode::GCString->new( $_[0] )->columns;
            };
            $encoding_log_message .= <<EOM;
Using 'Unicode::GCString' to measure horizontal character widths
EOM
            $rstatus->{'gcs_used'} = 1;
        }
    }
    return (
        $rinput_string,
        $is_encoded_data,
        $decoded_input_as,
        $encoding_log_message,
        $length_function,

    );
} ## end sub get_decoded_string_buffer

{ #<<<

my $LF;
my $CR;
my $CRLF;