Changeset 1279 for branches/New_Config_Format-branch/gsdl/perllib/multiread.pm

Timestamp:

2000-07-13T10:21:53+12:00 (24 years ago)

Author:

sjboddie

Message:

merged changes to trunk into New_Config_Format branch

File:

• branches/New_Config_Format-branch/gsdl/perllib/multiread.pm (modified) (5 diffs)

branches/New_Config_Format-branch/gsdl/perllib/multiread.pm

@@ -26 +26 @@
 # encodings currently supported are
 #
-# utf8     - either utf8 or unicode (automatically detected)
-# unicode  - just unicode (doesn't currently do endian detection)
-# gb       - GB
-# extended - extended ascii
-
+# utf8         - either utf8 or unicode (automatically detected)
+# unicode      - just unicode (doesn't currently do endian detection)
+# gb           - GB
+# iso_8859_1   - extended ascii (iso-8859-1)
+# iso_8859_6   - 8 bit arabic (iso-8859-6)
+# windows_1256 - Windows codepage 1256 (Arabic)
 
 package multiread;
@@ -169 +170 @@
     }
 
-    if ($self->{'encoding'} eq "extended") {
-    # extended ascii
+    if ($self->{'encoding'} eq "iso_8859_1") {
+    # Latin 1 extended ascii (ISO-8859-1)
     return undef if (eof ($handle));
     return &unicode::ascii2utf8 (getc ($handle));
+    }
+
+    if ($self->{'encoding'} eq "iso_8859_6") {
+    # 8 bit Arabic (IOS-8859-6)
+    return undef if (eof ($handle));
+    return &unicode::unicode2utf8(&unicode::arabic2unicode (getc ($handle)));
+    }
+
+    if ($self->{'encoding'} eq "windows_1256") {
+    # Windows 1256 (Arabic)
+    return undef if (eof ($handle));
+    return &unicode::unicode2utf8(&unicode::windows2unicode ("1256", getc ($handle)));
     }
 
@@ -236 +249 @@
     }
     
-    if ($self->{'encoding'} eq "extended") {
-    # extended ascii
+    if ($self->{'encoding'} eq "iso_8859_1") {
+    # extended ascii (ISO-8859-1)
     my $line = "";
     if (defined ($line = <$handle>)) {
@@ -244 +257 @@
     return undef;
     }
+    
+    if ($self->{'encoding'} eq "iso_8859_6") {
+    # 8 bit arabic (ISO-8859-6)
+    my $line = "";
+    if (defined ($line = <$handle>)) {
+        return &unicode::unicode2utf8(&unicode::arabic2unicode ($line));
+    }
+    return undef;
+    }
+    
+    if ($self->{'encoding'} eq "windows_1256") {
+    # Windows 1256 (Arabic)
+    my $line = "";
+    if (defined ($line = <$handle>)) {
+        return &unicode::unicode2utf8(&unicode::windows2unicode ("1256", $line));
+    }
+    return undef;
+    }
 
     # unknown encoding
@@ -250 +281 @@
 
 
+# will convert entire contents of file to utf8 and append result to $outputref
+# this may be a slightly faster way to get the contents of a file than by 
+# recursively calling read_line()
+sub read_file {
+    my $self = shift (@_);
+    my ($outputref) = @_;
+
+    # make sure we have a file handle
+    return if ($self->{'handle'} eq "");
+
+    my $handle = $self->{'handle'};
+
+    if ($self->{'first'} && $self->{'encoding'} eq "utf8") {
+    # special case for the first line of utf8 text to detect whether
+    # the file is in utf8 or unicode
+    $$text .= $self->read_line ();
+    }
+
+    if ($self->{'encoding'} eq "utf8") {
+    undef $/;
+    $$outputref .=  <$handle>;
+    $/ = "\n";
+    return;
+    }
+
+    if ($self->{'encoding'} eq "unicode") {
+    my $line = "";
+    while (defined ($line = $self->read_line())) {
+        $$outputref .= $line;
+    }
+    return;
+    }
+
+    if ($self->{'encoding'} eq "gb") {
+    undef $/;
+    my $text = <$handle>;
+    $/ = "\n";
+    $$outputref .= &unicode::unicode2utf8 (&gb::gb2unicode ($text));
+    return;
+    }
+    
+    if ($self->{'encoding'} eq "iso_8859_1") {
+    undef $/;
+    my $text = <$handle>;
+    $/ = "\n";
+    $$outputref .= &unicode::ascii2utf8 ($text);
+    return;
+    }
+    
+    if ($self->{'encoding'} eq "iso_8859_6") {
+    my $text = <$handle>;
+    undef $/;
+    $/ = "\n";
+    $$outputref .= &unicode::unicode2utf8(&unicode::arabic2unicode ($text));
+    return;
+    }
+
+    if ($self->{'encoding'} eq "windows_1256") {
+    undef $/;
+    my $text = <$handle>;
+    $/ = "\n";
+    $$outputref .= &unicode::unicode2utf8(&unicode::windows2unicode ("1256", $text));
+    return;
+    }
+}
+
+
 1;