Attachment 576293 Details for Bug 772362 – diff between sigil and upstream libtidy

diff between sigil and upstream libtidy

libtidy-sigil.diff (text/plain), 16.08 KB, created by Dan Horák on 2012-04-09 19:47:51 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Dan Horák

Created: 2012-04-09 19:47:51 UTC

Size: 16.08 KB

patch

obsolete

>Index: include/tidyenum.h
>===================================================================
>RCS file: /cvsroot/tidy/tidy/include/tidyenum.h,v
>retrieving revision 1.18
>diff -u -r1.18 tidyenum.h
>--- include/tidyenum.h	18 Jun 2008 20:18:54 -0000	1.18
>+++ include/tidyenum.h	9 Apr 2012 19:43:25 -0000
>@@ -204,6 +204,7 @@
>   TidySortAttributes,      /**< Sort attributes */
>   TidyMergeSpans,       /**< Merge multiple SPANs */
>   TidyAnchorAsName,    /**< Define anchors as name attributes */
>+  TidyClassStartID,    /**< Added by Strahinja Markovic: initial value for Tidy CSS class suffix */
>   N_TIDY_OPTIONS       /**< Must be last */
> } TidyOptionId;
> 
>Index: src/attrs.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/attrs.c,v
>retrieving revision 1.132
>diff -u -r1.132 attrs.c
>--- src/attrs.c	26 Mar 2009 13:05:22 -0000	1.132
>+++ src/attrs.c	9 Apr 2012 19:43:25 -0000
>@@ -599,17 +599,21 @@
>     if (!atnam)
>         return NULL;
> 
>+    /* Changed by Strahinja Markovic: switched all calls to
>+       tmbstrcmp into calls to tmbstrcasecmp. This way Tidy 
>+       compares attribute names in a case-insensitive manner. */
>+
> #if ATTRIBUTE_HASH_LOOKUP
>     for (p = attribs->hashtab[attrsHash(atnam)]; p && p->attr; p = p->next)
>-        if (TY_(tmbstrcmp)(atnam, p->attr->name) == 0)
>+        if (TY_(tmbstrcasecmp)(atnam, p->attr->name) == 0)
>             return p->attr;
> 
>     for (np = attribute_defs; np && np->name; ++np)
>-        if (TY_(tmbstrcmp)(atnam, np->name) == 0)
>+        if (TY_(tmbstrcasecmp)(atnam, np->name) == 0)
>             return attrsInstall(doc, attribs, np);
> #else
>     for (np = attribute_defs; np && np->name; ++np)
>-        if (TY_(tmbstrcmp)(atnam, np->name) == 0)
>+        if (TY_(tmbstrcasecmp)(atnam, np->name) == 0)
>             return np;
> #endif
> 
>Index: src/clean.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/clean.c,v
>retrieving revision 1.111
>diff -u -r1.111 clean.c
>--- src/clean.c	14 Oct 2008 12:18:10 -0000	1.111
>+++ src/clean.c	9 Apr 2012 19:43:26 -0000
>@@ -1854,7 +1854,7 @@
> 
>                 p = TY_(PutUTF8)(p, c);
>             }
>-            node->end = p - lexer->lexbuf;
>+            node->end = (uint) (p - lexer->lexbuf);
>         }
> 
>         node = node->next;
>@@ -2454,7 +2454,7 @@
>                 p = TY_(PutUTF8)(p, c);
>             }
> 
>-            node->end = p - lexer->lexbuf;
>+            node->end = (uint) (p - lexer->lexbuf);
>         }
> 
>         if (node->content)
>Index: src/config.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/config.c,v
>retrieving revision 1.111
>diff -u -r1.111 config.c
>--- src/config.c	18 Jun 2008 20:18:54 -0000	1.111
>+++ src/config.c	9 Apr 2012 19:43:26 -0000
>@@ -315,6 +315,10 @@
>   { TidySortAttributes,          PP, "sort-attributes",             IN, TidySortAttrNone,ParseSorter,       sorterPicks     },
>   { TidyMergeSpans,              MU, "merge-spans",                 IN, TidyAutoState,   ParseAutoBool,     autoBoolPicks   },
>   { TidyAnchorAsName,            MU, "anchor-as-name",              BL, yes,             ParseBool,         boolPicks       },
>+
>+  /* Added by Strahinja Markovic */
>+  { TidyClassStartID,            MS, "class-start-id",              IN, 0,               ParseInt,          NULL            },
>+
>   { N_TIDY_OPTIONS,              XX, NULL,                          XY, 0,               NULL,              NULL            }
> };
> 
>Index: src/lexer.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/lexer.c,v
>retrieving revision 1.194
>diff -u -r1.194 lexer.c
>--- src/lexer.c	22 Mar 2008 21:06:55 -0000	1.194
>+++ src/lexer.c	9 Apr 2012 19:43:26 -0000
>@@ -839,6 +839,13 @@
> 
>     while ( (c = TY_(ReadChar)(doc->docIn)) != EndOfStream )
>     {
>+        /* Added by Strahinja Markovic: using a colon instead of a semicolon
>+           is a frequent typing error. Let's try to handle it. */
>+        if ( c == ':' )
>+        {
>+            c = ';';       
>+        }
>+
>         if ( c == ';' )
>         {
>             semicolon = yes;
>@@ -945,15 +952,55 @@
>                
>             }
>             else
>-                TY_(ReportEntityError)( doc, UNKNOWN_ENTITY,
>-                                        lexer->lexbuf+start, ch );
>+            {
>+                /* Changed by Strahinja Markovic: 
>+                   When the user has something like "&co." in his
>+                   source that should really be "&amp;co.", this
>+                   code makes it happen.
>+                */
>+
>+                /* Points to the part of the "entity" after
>+                   the ampersand. */
>+                tmbstr points_after_amp = lexer->lexbuf + start + 1;
>+                uint size = TY_(tmbstrlen)( points_after_amp );                
>+
>+                /* 50 bytes should certainly be big enough, 
>+                   but let's try not to introduce buffer overflows anyway. */
>+                if ( size < 50 )
>+                {
>+                    tmbchar string_after_amp[ 50 + 1 ] = { 0 };
>+
>+                    TY_(tmbstrncpy)( string_after_amp, points_after_amp, size + 1 );
>+
>+                    /* We move the lexer back to the start of the entity;
>+                       we will be writing over the old entity value. */
>+                    lexer->lexsize = start;
>+
>+                    AddStringToLexer( lexer, "&amp;" );
>+                    AddStringToLexer( lexer, string_after_amp );
>+                }
>+
>+                else
>+                {                   
>+                    TY_(ReportEntityError)( doc, UNKNOWN_ENTITY,
>+                                            lexer->lexbuf+start, ch );
>+                }
>+            }
> 
>             if (semicolon)
>                 TY_(AddCharToLexer)( lexer, ';' );
>         }
>         else /* naked & */
>-            TY_(ReportEntityError)( doc, UNESCAPED_AMPERSAND,
>-                                    lexer->lexbuf+start, ch );
>+        {
>+            /* Changed by Strahinja Markovic: instead of reporting
>+               the error, how about fixing it? */
>+
>+            /*TY_(ReportEntityError)( doc, UNESCAPED_AMPERSAND,
>+                                    lexer->lexbuf+start, ch );*/
>+
>+            if ( cfgBool(doc, TidyQuoteAmpersand) )
>+                AddStringToLexer( lexer, "amp;" );
>+        }
>     }
>     else
>     {
>@@ -1567,9 +1614,15 @@
>     {
>     case TidyDoctypeStrict:
>         /* XHTML 1.0 Strict */
>-        TY_(RepairAttrValue)(doc, doctype, pub, GetFPIFromVers(X10S));
>+        /*TY_(RepairAttrValue)(doc, doctype, pub, GetFPIFromVers(X10S));
>         TY_(RepairAttrValue)(doc, doctype, sys, GetSIFromVers(X10S));
>-        lexer->versionEmitted = X10S;
>+        lexer->versionEmitted = X10S;*/
>+
>+        /* Changed by Strahinja Markovic: We need XHTML 1.1 for OPS docs */
>+        /* XHTML 1.1 */
>+        TY_(RepairAttrValue)(doc, doctype, pub, GetFPIFromVers(XH11));
>+        TY_(RepairAttrValue)(doc, doctype, sys, GetSIFromVers(XH11));
>+        lexer->versionEmitted = XH11;
>         break;
>     case TidyDoctypeLoose:
>         /* XHTML 1.0 Transitional */
>@@ -2430,10 +2483,15 @@
>                 if ((mode != Preformatted && ExpectsContent(lexer->token))
>                     || nodeIsBR(lexer->token) || nodeIsHR(lexer->token))
>                 {
>-                    c = TY_(ReadChar)(doc->docIn);
>+                    /* Changed by Strahinja Markovic:
>+                       DON'T swallow the newline following the start tag!
>+                       Why would you ever want to do that for Christ's sake,
>+                       a newline counts as whitespace too.*/
>+
>+                    /*c = TY_(ReadChar)(doc->docIn);
> 
>                     if (c != '\n' && c != '\f')
>-                        TY_(UngetChar)(c, doc->docIn);
>+                        TY_(UngetChar)(c, doc->docIn);*/                                         
> 
>                     lexer->waswhite = yes;  /* to swallow leading whitespace */
>                 }
>@@ -2984,16 +3042,23 @@
>     tmbstr attr = NULL;
>     uint c, lastc;
> 
>+    /* This WILL explode if someone uses an attribute
>+    that has more than 50 characters. Fortunately, such
>+    attributes don't exist in XHTML, SVG or anything similar.*/
>+    uint attribute_buffer[ 50 ] = { 0 };
>+    int attribute_buffer_index = 0;
>+    int index = 0;
>+    Bool all_chars_uppercase = yes;
>+
>     *asp = NULL;  /* clear asp pointer */
>     *php = NULL;  /* clear php pointer */
>-
>- /* skip white space before the attribute */
>-
>+ 
>+    /* skip white space before the attribute */
>     for (;;)
>     {
>         c = TY_(ReadChar)( doc->docIn );
> 
>-
>+        // Tag is ending.
>         if (c == '/')
>         {
>             c = TY_(ReadChar)( doc->docIn );
>@@ -3060,6 +3125,7 @@
>     start = lexer->lexsize;
>     lastc = c;
> 
>+    // Read the attribute name
>     for (;;)
>     {
>      /* but push back '=' for parseValue() */
>@@ -3089,14 +3155,43 @@
>         /* what should be done about non-namechar characters? */
>         /* currently these are incorporated into the attr name */
> 
>-        if ( !cfgBool(doc, TidyXmlTags) && TY_(IsUpper)(c) )
>-            c = TY_(ToLower)(c);
>+        attribute_buffer[ attribute_buffer_index ] = c;
>+        ++attribute_buffer_index;
> 
>-        TY_(AddCharToLexer)( lexer, c );
>         lastc = c;
>         c = TY_(ReadChar)(doc->docIn);
>     }
> 
>+    /* Are all chars in buffer uppercase? 
>+       We leave mixed-case names alone. */
>+    for ( index = 0; index < attribute_buffer_index; ++index )
>+    {
>+        uint character = attribute_buffer[ index ];
>+
>+        if ( TY_(IsLetter)( character ) && !TY_(IsUpper)( character ) )
>+        {
>+            all_chars_uppercase = no;
>+            break;
>+        }
>+    }    
>+
>+    /* If they are, make them all lowercase */
>+    if ( all_chars_uppercase )
>+    {
>+        for ( index = 0; index < attribute_buffer_index; ++index )
>+        {
>+            if ( TY_(IsLetter)( attribute_buffer[ index ] ) )
>+
>+                attribute_buffer[ index ] = TY_(ToLower)( attribute_buffer[ index ] );
>+        }
>+    }
>+
>+    /* Add all the chars from the buffer to the lexer */
>+    for ( index = 0; index < attribute_buffer_index; ++index )
>+    {
>+        TY_(AddCharToLexer)( lexer, attribute_buffer[ index ] );
>+    }
>+
>     /* handle attribute names with multibyte chars */
>     len = lexer->lexsize - start;
>     attr = (len > 0 ? TY_(tmbstrndup)(doc->allocator,
>Index: src/localize.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/localize.c,v
>retrieving revision 1.178
>diff -u -r1.178 localize.c
>--- src/localize.c	18 Jun 2008 20:18:54 -0000	1.178
>+++ src/localize.c	9 Apr 2012 19:43:27 -0000
>@@ -882,25 +882,25 @@
>   switch ( level )
>   {
>   case TidyInfo:
>-    TY_(tmbstrncpy)( buf, "Info: ", count );
>+    TY_(tmbstrncpy)( buf, "Info: ", (uint) count );
>     break;
>   case TidyWarning:
>-    TY_(tmbstrncpy)( buf, "Warning: ", count );
>+    TY_(tmbstrncpy)( buf, "Warning: ", (uint) count );
>     break;
>   case TidyConfig:
>-    TY_(tmbstrncpy)( buf, "Config: ", count );
>+    TY_(tmbstrncpy)( buf, "Config: ", (uint) count );
>     break;
>   case TidyAccess:
>-    TY_(tmbstrncpy)( buf, "Access: ", count );
>+    TY_(tmbstrncpy)( buf, "Access: ", (uint) count );
>     break;
>   case TidyError:
>-    TY_(tmbstrncpy)( buf, "Error: ", count );
>+    TY_(tmbstrncpy)( buf, "Error: ", (uint) count );
>     break;
>   case TidyBadDocument:
>-    TY_(tmbstrncpy)( buf, "Document: ", count );
>+    TY_(tmbstrncpy)( buf, "Document: ", (uint) count );
>     break;
>   case TidyFatal:
>-    TY_(tmbstrncpy)( buf, "panic: ", count );
>+    TY_(tmbstrncpy)( buf, "panic: ", (uint) count );
>     break;
>   }
>   return buf + TY_(tmbstrlen)( buf );
>@@ -1373,14 +1373,14 @@
> {
>     ctmbstr fmt = GetFormatFromCode(code);
>     doc->badAccess |= BA_WAI;
>-    messageNode( doc, TidyAccess, node, fmt );
>+    messageNode( doc, TidyAccess, node, "%s", fmt );
> }
> 
> void TY_(ReportAccessError)( TidyDocImpl* doc, Node* node, uint code )
> {
>     ctmbstr fmt = GetFormatFromCode(code);
>     doc->badAccess |= BA_WAI;
>-    messageNode( doc, TidyAccess, node, fmt );
>+    messageNode( doc, TidyAccess, node, "%s", fmt );
> }
> 
> #endif /* SUPPORT_ACCESSIBILITY_CHECKS */
>@@ -1399,7 +1399,7 @@
>     switch (code)
>     {
>     case NESTED_QUOTATION:
>-        messageNode(doc, TidyWarning, rpt, fmt);
>+        messageNode(doc, TidyWarning, rpt, "%s", fmt);
>         break;
> 
>     case OBSOLETE_ELEMENT:
>@@ -1480,7 +1480,7 @@
>     case INCONSISTENT_NAMESPACE:
>     case DOCTYPE_AFTER_TAGS:
>     case DTYPE_NOT_UPPER_CASE:
>-        messageNode(doc, TidyWarning, rpt, fmt);
>+        messageNode(doc, TidyWarning, rpt, "%s", fmt);
>         break;
> 
>     case COERCE_TO_ENDTAG:
>@@ -1499,7 +1499,7 @@
>     case ENCODING_IO_CONFLICT:
>     case MISSING_DOCTYPE:
>     case SPACE_PRECEDING_XMLDECL:
>-        messageNode(doc, TidyWarning, node, fmt);
>+        messageNode(doc, TidyWarning, node, "%s", fmt);
>         break;
> 
>     case TRIM_EMPTY_ELEMENT:
>@@ -1548,7 +1548,7 @@
>     {
>     case SUSPECTED_MISSING_QUOTE:
>     case DUPLICATE_FRAMESET:
>-        messageNode(doc, TidyError, rpt, fmt);
>+        messageNode(doc, TidyError, rpt, "%s", fmt);
>         break;
> 
>     case UNKNOWN_ELEMENT:
>Index: src/parser.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/parser.c,v
>retrieving revision 1.188
>diff -u -r1.188 parser.c
>--- src/parser.c	27 Oct 2009 19:27:49 -0000	1.188
>+++ src/parser.c	9 Apr 2012 19:43:27 -0000
>@@ -5,9 +5,9 @@
>   
>   CVS Info :
> 
>-    $Author: krusch $ 
>-    $Date: 2009/10/27 19:27:49 $ 
>-    $Revision: 1.188 $ 
>+    $Author: arnaud02 $ 
>+    $Date: 2008/03/22 20:02:42 $ 
>+    $Revision: 1.187 $ 
> 
> */
> 
>@@ -3468,11 +3468,8 @@
>             }
>             else if (node->tag->model & (CM_TABLE | CM_ROWGRP | CM_ROW))
>             {
>-                /* http://tidy.sf.net/issue/2855621 */
>-                if (node->type != EndTag) {
>-                    TY_(UngetToken)( doc );
>-                    node = TY_(InferredTag)(doc, TidyTag_TABLE);
>-                }
>+                TY_(UngetToken)( doc );
>+                node = TY_(InferredTag)(doc, TidyTag_TABLE);
>                 lexer->excludeBlocks = yes;
>             }
>             else if ( nodeIsINPUT(node) )
>Index: src/pprint.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/pprint.c,v
>retrieving revision 1.119
>diff -u -r1.119 pprint.c
>--- src/pprint.c	22 Mar 2008 20:23:37 -0000	1.119
>+++ src/pprint.c	9 Apr 2012 19:43:28 -0000
>@@ -2023,14 +2023,19 @@
>             PPrintTag( doc, mode, indent, node );
> 
>             indent = 0;
>-            TY_(PFlushLine)( doc, indent );
>+
>+            if ( !nodeIsPRE(node) )
>+                TY_(PFlushLine)( doc, indent );
> 
>             for ( content = node->content; content; content = content->next )
>             {
>                 TY_(PPrintTree)( doc, (mode | PREFORMATTED | NOWRAP),
>                                  indent, content );
>             }
>-            PCondFlushLine( doc, indent );
>+
>+            if ( !nodeIsPRE(node) )
>+                PCondFlushLine( doc, indent );
>+            
>             indent = indprev;
>             PPrintEndTag( doc, mode, indent, node );
> 
>Index: src/tidylib.c
>===================================================================
>RCS file: /cvsroot/tidy/tidy/src/tidylib.c,v
>retrieving revision 1.75
>diff -u -r1.75 tidylib.c
>--- src/tidylib.c	18 Jun 2008 20:18:54 -0000	1.75
>+++ src/tidylib.c	9 Apr 2012 19:43:28 -0000
>@@ -915,6 +915,9 @@
>     TidyBuffer inbuf;
>     StreamIn* in = NULL;
> 
>+    /* Ugly hack added by Strahinja Markovic */
>+    doc->nClassId = cfg(doc, TidyClassStartID);
>+
>     if ( content )
>     {
>         tidyBufInitWithAllocator( &inbuf, doc->allocator );
>@@ -1324,7 +1327,10 @@
>             TY_(SetXHTMLDocType)(doc);
>             TY_(FixAnchors)(doc, &doc->root, wantNameAttr, yes);
>             TY_(FixXhtmlNamespace)(doc, yes);
>-            TY_(FixLanguageInformation)(doc, &doc->root, yes, yes);
>+
>+            /* Change by Strahinja Markovic - changed "yes, yes" to "yes, no";
>+               the 'lang' is attribute not allowed for XHTML 1.1.*/
>+            TY_(FixLanguageInformation)(doc, &doc->root, yes, no);
>         }
>         else
>         {

Actions: View

Attachments on bug 772362: 576293