Changeset 17058 for gsdl/trunk/perllib/ghtml.pm

Timestamp:

2008-08-28T13:44:31+12:00 (16 years ago)

Author:

ak19

Message:

1. Moved the mime_type hashmap out of the guess_mime_type subroutine since this hashmap might get recreated everytime that subroutine is called. 2. Converted the filename from which the mimetype is guessed into lowercase first since the mime_type hashmap only contains lowercase file extension.

File:

• gsdl/trunk/perllib/ghtml.pm (modified) (2 diffs)

gsdl/trunk/perllib/ghtml.pm

@@ -161 +161 @@
 
 
-# returns the character as a raw utf-8 character. It assumes that the
-# & and ; have been stripped off the string.
-sub getcharequiv {
-    my ($entity, $convertsymbols) = @_;
-
-    # a numeric entity
-    if ($entity =~ /^\#0*(\d+)/) {
-    my $code=$1;
-    # non-standard Microsoft breakage, as usual
-    if ($code < 0x9f) { # code page 1252 uses reserved bytes
-        if ($code == 0x91) {$code=0x2018} # 145 = single left quote
-        elsif ($code == 0x92) {$code=0x2019} # 146 = single right quote
-        elsif ($code == 0x93) {$code=0x201c} # 147 = double left quote
-        elsif ($code == 0x94) {$code=0x201d} # 148 = double right quote
-        # ...
-    }
-    return &unicode::unicode2utf8([$code]);
-    }
-    
-    # a named character entity
-    if (defined $charnetosf{$entity}) {
-    return &unicode::unicode2utf8([$charnetosf{$entity}]);
-    }
-
-    # a named symbol entity
-    if ($convertsymbols && defined $symnetosf{$entity}) {
-    return &unicode::unicode2utf8([$symnetosf{$entity}]);
-    }
-
-    return "&$entity;"; # unknown character
-}
-
-# convert character entities from named equivalents to html font
-sub convertcharentities {
-    # args: the text that you want to convert
-
-    $_[0] =~ s/&([^;]+);/&getcharequiv($1,0)/gse;
-}
-
-# convert any entities from named equivalents to html font
-sub convertallentities {
-    # args: the text that you want to convert
-
-    $_[0] =~ s/&([^;]+);/&getcharequiv($1,1)/gse;
-}
-
-sub html2txt {
-    # args: the text that you want converted to ascii, 
-    # and whether to strip out sgml tags
-
-    # strip out sgml tags if needed
-    $_[0] =~ s/<[^>]*>//g if $_[1];
-
-    # convert the char entities to the standard html font
-    &convertcharentities($_[0]); 
-    
-    # convert the html character set to a plain ascii character set
-    my $pos = 0;
-    while ($pos < length($_[0])) {
-    my $charnum = ord(substr($_[0], $pos, 1));
-    if ($charnum >= 32) { # only convert characters above #32
-        my $replacechars = " ";
-        $replacechars = $sftotxt{$charnum} if defined $sftotxt{$charnum};
-        substr($_[0], $pos, 1) = $replacechars;
-        $pos += length ($replacechars);
-
-    } else {
-        $pos ++;
-    }
-    }
-}  
-
-
-# look for mime.types (eg in /etc, or apache/conf directories), or have a look
-# at <ftp://ftp.iana.org/in-notes/iana/assignments/media-types/> for defaults.
-sub guess_mime_type {
-    my ($filename) = @_;
-
-    my ($fileext) = $filename =~ /\.(\w+)$/;
-    return "unknown" unless defined $fileext;
-
-    my %mime_type = ("ai"=>"application/postscript", "aif"=>"audio/x-aiff", 
+my %mime_type = ("ai"=>"application/postscript", "aif"=>"audio/x-aiff", 
              "aifc"=>"audio/x-aiff", "aiff"=>"audio/x-aiff", 
              "au"=>"audio/basic", "avi"=>"video/x-msvideo",
@@ -298 +217 @@
              "xyz"=>"chemical/x-pdb", "zip"=>"application/zip");
 
-    return $mime_type{$fileext} if (defined $mime_type{$fileext});
+
+# returns the character as a raw utf-8 character. It assumes that the
+# & and ; have been stripped off the string.
+sub getcharequiv {
+    my ($entity, $convertsymbols) = @_;
+
+    # a numeric entity
+    if ($entity =~ /^\#0*(\d+)/) {
+    my $code=$1;
+    # non-standard Microsoft breakage, as usual
+    if ($code < 0x9f) { # code page 1252 uses reserved bytes
+        if ($code == 0x91) {$code=0x2018} # 145 = single left quote
+        elsif ($code == 0x92) {$code=0x2019} # 146 = single right quote
+        elsif ($code == 0x93) {$code=0x201c} # 147 = double left quote
+        elsif ($code == 0x94) {$code=0x201d} # 148 = double right quote
+        # ...
+    }
+    return &unicode::unicode2utf8([$code]);
+    }
+    
+    # a named character entity
+    if (defined $charnetosf{$entity}) {
+    return &unicode::unicode2utf8([$charnetosf{$entity}]);
+    }
+
+    # a named symbol entity
+    if ($convertsymbols && defined $symnetosf{$entity}) {
+    return &unicode::unicode2utf8([$symnetosf{$entity}]);
+    }
+
+    return "&$entity;"; # unknown character
+}
+
+# convert character entities from named equivalents to html font
+sub convertcharentities {
+    # args: the text that you want to convert
+
+    $_[0] =~ s/&([^;]+);/&getcharequiv($1,0)/gse;
+}
+
+# convert any entities from named equivalents to html font
+sub convertallentities {
+    # args: the text that you want to convert
+
+    $_[0] =~ s/&([^;]+);/&getcharequiv($1,1)/gse;
+}
+
+sub html2txt {
+    # args: the text that you want converted to ascii, 
+    # and whether to strip out sgml tags
+
+    # strip out sgml tags if needed
+    $_[0] =~ s/<[^>]*>//g if $_[1];
+
+    # convert the char entities to the standard html font
+    &convertcharentities($_[0]); 
+    
+    # convert the html character set to a plain ascii character set
+    my $pos = 0;
+    while ($pos < length($_[0])) {
+    my $charnum = ord(substr($_[0], $pos, 1));
+    if ($charnum >= 32) { # only convert characters above #32
+        my $replacechars = " ";
+        $replacechars = $sftotxt{$charnum} if defined $sftotxt{$charnum};
+        substr($_[0], $pos, 1) = $replacechars;
+        $pos += length ($replacechars);
+
+    } else {
+        $pos ++;
+    }
+    }
+}  
+
+
+# look for mime.types (eg in /etc, or apache/conf directories), or have a look
+# at <ftp://ftp.iana.org/in-notes/iana/assignments/media-types/> for defaults.
+sub guess_mime_type {
+    my ($filename) = @_;
+    # make the filename lowercase, since the mimetypes hashmap looks for lowercase
+    $filename = lc($filename);
+
+    my ($fileext) = $filename =~ /\.(\w+)$/;
+    return "unknown" unless defined $fileext;
+
+    # else
+    my $mimetype =  $mime_type{$fileext};
+    return $mimetype if (defined $mimetype);
 
     return "unknown";