From 382606fb71a6b0192c8b40d362b12d6b017ac144 Mon Sep 17 00:00:00 2001
From: "Pierre Habouzit (MadCoder" <pierre.habouzit@m4x.org>
Date: Tue, 24 May 2005 07:13:23 +0000
Subject: [PATCH] =?utf8?q?=A4=20thing=20+=20#294?=
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

git-archimport-id: opensource@polytechnique.org--2005/banana--mainline--1.0--patch-18
---
 Changelog            |  8 +++++
 banana/misc.inc.php  |  5 +++
 banana/post.inc.php  |  3 +-
 banana/spool.inc.php |  1 +
 banana/utf8.php      | 88 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 104 insertions(+), 1 deletion(-)
 create mode 100644 banana/utf8.php

diff --git a/Changelog b/Changelog
index f06f42f..b950355 100644
--- a/Changelog
+++ b/Changelog
@@ -1,4 +1,12 @@
 ================================================================================
+VERSION 1.1
+
+Tue, 24 May 2005					Pierre Habouzit <pierre.habouzit@m4x.org>
+
+	* Fix one problem with $this->ids beeing empty in Spool.
+	* Fix encoding issues since iconv truncates strings sometimes.
+
+================================================================================
 VERSION 1.0
 
 Fri, 07 Jan 2005					Pierre Habouzit <pierre.habouzit@m4x.org>
diff --git a/banana/misc.inc.php b/banana/misc.inc.php
index a72e6cf..ea88ccd 100644
--- a/banana/misc.inc.php
+++ b/banana/misc.inc.php
@@ -13,6 +13,11 @@
 
 function _b_($str) { return utf8_decode(dgettext('banana', utf8_encode($str))); }
 
+function to_html($str, $charset) {
+    require_once 'banana/utf8.php';
+    return utf8entities(htmlentities(iconv($charset, 'utf8', $str), ENT_NOQUOTES, 'UTF-8'));
+}
+
 /********************************************************************************
  *  HEADER STUFF
  */
diff --git a/banana/post.inc.php b/banana/post.inc.php
index 340d218..60f2986 100644
--- a/banana/post.inc.php
+++ b/banana/post.inc.php
@@ -44,7 +44,8 @@ class BananaPost
         }
 
         if (preg_match('!charset=([^;]*)\s*(;|$)!', $this->headers['content-type'], $matches)) {
-            $this->body = iconv($matches[1], 'iso-8859-15', $this->body);
+            require_once 'banana/misc.inc.php';
+            $this->body = to_html($this->body, $matches[1]);
         }
     }
 
diff --git a/banana/spool.inc.php b/banana/spool.inc.php
index 1f9777d..f199342 100644
--- a/banana/spool.inc.php
+++ b/banana/spool.inc.php
@@ -202,6 +202,7 @@ class BananaSpool
         if (empty($since)) { return; }
 
         if (is_array($newpostsids = $banana->nntp->newnews($since, $this->group))) {
+            if (!is_array($this->ids)) { $this->ids = array(); }
             $newpostsids = array_intersect($newpostsids, array_keys($this->ids));
             foreach ($newpostsids as $mid) {
                 $this->overview[$this->ids[$mid]]->isread     = false;
diff --git a/banana/utf8.php b/banana/utf8.php
new file mode 100644
index 0000000..6a9fb51
--- /dev/null
+++ b/banana/utf8.php
@@ -0,0 +1,88 @@
+<?php
+
+/**
+ * Transforme une chaine encodée en UTF-8, et la convertit
+ * en entitiées unicode &#xxx; pour que ça s'affiche correctement
+ * dans les navigateurs, sans forcément tenir compte du meta
+ * content-type charset...
+ * @param String $source la chaine en UTF-8
+ * @return String les entitées
+ * @access public
+ * @see http://www.php.net/utf8_decode
+ */
+function utf8entities($source)
+{
+   // array used to figure what number to decrement from character order value 
+   // according to number of characters used to map unicode to ascii by utf-8
+   $decrement[4] = 240;
+   $decrement[3] = 224;
+   $decrement[2] = 192;
+   $decrement[1] = 0;
+   
+   // the number of bits to shift each charNum by
+   $shift[1][0] = 0;
+   $shift[2][0] = 6;
+   $shift[2][1] = 0;
+   $shift[3][0] = 12;
+   $shift[3][1] = 6;
+   $shift[3][2] = 0;
+   $shift[4][0] = 18;
+   $shift[4][1] = 12;
+   $shift[4][2] = 6;
+   $shift[4][3] = 0;
+   
+   $pos = 0;
+   $len = strlen($source);
+   $encodedString = '';
+   while ($pos < $len)
+   {
+      $charPos = $source{$pos};
+      $asciiPos = ord($charPos);
+      if ($asciiPos < 128)
+      {
+         $encodedString .= $charPos;
+         $pos++;
+         continue;
+      }
+      
+      $i=1;
+      if (($asciiPos >= 240) && ($asciiPos <= 255)) // 4 chars representing one unicode character
+         $i=4;
+      else if (($asciiPos >= 224) && ($asciiPos <= 239)) // 3 chars representing one unicode character
+         $i=3;
+      else if (($asciiPos >= 192) && ($asciiPos <= 223)) // 2 chars representing one unicode character
+         $i=2;
+      else // 1 char (lower ascii)
+         $i=1;
+      $thisLetter = substr($source, $pos, $i);
+      $pos += $i;
+      
+      // process the string representing the letter to a unicode entity
+      $thisLen = strlen($thisLetter);
+      $thisPos = 0;
+      $decimalCode = 0;
+      while ($thisPos < $thisLen)
+      {
+         $thisCharOrd = ord(substr($thisLetter, $thisPos, 1));
+         if ($thisPos == 0)
+         {
+            $charNum = intval($thisCharOrd - $decrement[$thisLen]);
+            $decimalCode += ($charNum << $shift[$thisLen][$thisPos]);
+         }
+         else
+         {
+            $charNum = intval($thisCharOrd - 128);
+            $decimalCode += ($charNum << $shift[$thisLen][$thisPos]);
+         }
+         
+         $thisPos++;
+      }
+      
+      $encodedLetter = '&#'. str_pad($decimalCode, ($thisLen==1)?3:5, '0', STR_PAD_LEFT).';';
+      $encodedString .= $encodedLetter;
+   }
+   
+   return $encodedString;
+}
+
+?>
-- 
2.1.4