Applied patches from Vaclav Dvorak

xlhtml/ChangeLog

@@ -1,19 +1,31 @@
 							xlHtml ChangeLog
 
+HEAD
+	* Temporary fix for non-Unicode 8-bit characters in utf-8 output until
+	  real charset conversion is in place
+	* Fixed some typos and formatting
+	* Fixed closing of <FONT> tag
+	* Fixed raw Unicode (A.B.C.D., where . is character 0x00) in output
+	  when string in XLS begins as 8-bit but continues as 16-bit
+	* Fixed invalid XML output - missing quotes aroud size attribute
+	  of <FONT>
+	* Made rowspan and colspan attributes instead of content of <cell>
+	  in XML output
+
 0.5 04/13/02
-	*Fixed some memory leaks
-	*Increased file name limit to 256 characters
+	* Fixed some memory leaks
+	* Increased file name limit to 256 characters
 
 0.4.9.3 03/19/02
-	*Began code modularization
-	*Changed pointers to xlhtml web site
+	* Began code modularization
+	* Changed pointers to xlhtml web site
 
 0.4.9.2 03/13/02
-	*Turned over xlhtml to Charles Wyble
-	*Added a patch to support bold formatting 
+	* Turned over xlhtml to Charles Wyble
+	* Added a patch to support bold formatting 
 
 0.4.9.1 03/10/02
-    * Fixed bug in cole.c which causes errors reading ppthtml files - Takahiro Kambe
+	* Fixed bug in cole.c which causes errors reading ppthtml files - Takahiro Kambe
 	* Fixed bug in page name output for non-html output
 	* xlhtml now uses Microsoft CodePages - Victor Wagner
 
@@ -31,10 +43,10 @@
 	* Added doxygen doco generation - Rafael Kitover
 	* Added patch to support String Formulas - Rafael Kitover
 	* Changed unsupported numeric formats to output using floating point
-	technique with the decimal suppressed. Was truncating to MAXINT.
+	  technique with the decimal suppressed. Was truncating to MAXINT.
 	* Fixed bug where CSV was not escaping currency fields - Rafael Kitover
 	* Fixed bug where unknown field types were not escaped in CSV 
-	- Rafael Kitover
+	  - Rafael Kitover
 	* Fixed bug in cell merging where the whole row was selected
 	* Fixed bug in mailto hyperlink
 	* Now supports custom colors
@@ -45,7 +57,7 @@
 	* Optimized html further so it suppresses alignment of blank rows
 	* Improved portability to DOS & WIN32
 	* Moved everything over to U8, U16, U32, F32, & F64 data types. This is in
-	effort to improve support on 64 bit platforms.
+	  effort to improve support on 64 bit platforms.
 	* Added the xlv script from Asher Blum
 
 0.3 10/19/01
@@ -53,7 +65,7 @@
 	* Fixed situation where ellipses were not being rendered correctly
 	* Fixed Unicode problem in Label Cells
 	* Fixed problem where cells created by row data were not being updated 
-	when the cell was officially created.
+	  when the cell was officially created.
 	* Program renamed to xlhtml...all lowercase.
 	* Moved xlHtml over to a new makefile system. It does a static build now.
 	* Finished moving cole library to mkstemp from tempnam()
@@ -82,7 +94,7 @@
 	* Fixed extended string + rich string
 	* Fixed a misplaced array initialization for ws_array
 	* Fixed problem where the html closing tags were being output during error
- 	  messages when -asc option was given
+	  messages when -asc option was given
 	* Changed all the comments over to standard C comments for increased
 	  portability
 
@@ -100,9 +112,9 @@
 0.2.4 12/04/1999
 	* Added Cell Merging capabilities
 	* Added -te command line option to "trim the edges" (empty rows or
-      columns at the edges of a worksheet).
+	  columns at the edges of a worksheet).
 	* Added the contributed excel2text script to let mutt clients see xls 
-      attachments
+	  attachments
 	* Added a -v command line switch for version information
 	* Added code to optimize font size statements similar to font face
 	* Fixed Unicode bugs in font name and sheet name - Thanks Hironori !
@@ -123,14 +135,14 @@
 	* Added -tc command line optine to override the text color
 	* Added -bi command line option to add a background image
 	* Added better support for Excel 95 & 5.0. Had to disable
-      colors since it looked crazy. Probably has a different palette.
+	  colors since it looked crazy. Probably has a different palette.
 	* Added -a command line to use aggressive optimization. This doesn't
-      generate closing tags for TD & TR tags. This could be a problem
-      for older browsers.
+	  generate closing tags for TD & TR tags. This could be a problem
+	  for older browsers.
 	* Further reduced generated html output.
 	* Program now displays the last saved value of a formula.
 	* Added command line option to suppress warning that formula data
-      may be stale or inaccurate.
+	  may be stale or inaccurate.
 	* Updated uploader.pl documentation
 	* Added nsopen *bash* script in contrib directory
 
@@ -142,16 +154,16 @@
 
 0.1.6 9/25/1999
 	* Re-designed the Test.xls file to really workout xlHtml
-	* Added support for most built-in cell formats :date, time, 
-      number, and scientific formats
+	* Added support for most built-in cell formats: date, time, 
+	  number, and scientific formats
 	* Added support for general formatting
 	* Fixed MS induced glitch in LabelSST - only very large files
-      would have this problem.
+	  would have this problem.
 	* Fixed bug in default Row XF's - only seen in colored blank cells
 	* Optimized html tags. Avg 40% less html!
 	* Added more documentation in README-xlHtml
 	* Cleaned out document folder. It had some files that were created 
-      from the ./configure script. Reduced total tar size significantly.
+	  from the ./configure script. Reduced total tar size significantly.
 	* Fixed bug in RK number conversion
 
 0.1.4 9/15/1999
@@ -197,5 +209,4 @@
 	* Labels are now supported.
 
 0.0.1 08/8/1999 Initial Release. 
-	*So far it supports most strings, empty cells, and integers. 
-
+	* So far it supports most strings, empty cells, and integers. 

xlhtml/html.c

@@ -145,14 +145,14 @@ void OutputTableHTML(void)
 	/* Print the author's name in itallics... */
 	if (author.str)
 	{
-		printf("<FONT SIZE=-1><I>Spreadsheet's Author:&nbsp;");
+		printf("<FONT SIZE=\"-1\"><I>Spreadsheet's Author:&nbsp;");
 		OutputString(&author);
 		printf("</I></FONT><br>");
 		do_cr();
 	}
 
 	/* Print when & how the file was last updated. */
-	printf("<FONT SIZE=-1><I>Last Updated ");
+	printf("<FONT SIZE=\"-1\"><I>Last Updated ");
 	if (lastUpdated)
 		printf("%s&nbsp; ", lastUpdated);
 	switch (file_version)
@@ -244,7 +244,7 @@ void OutputTableHTML(void)
 	do_cr();
 
 	/* Output Credit */
-	printf("<hr><FONT SIZE=-1>Created with <a href=\"http://chicago.sf.net/xlhtml\">xlhtml %s</a></FONT><br>", VERSION);
+	printf("<hr><FONT SIZE=\"-1\">Created with <a href=\"http://chicago.sf.net/xlhtml\">xlhtml %s</a></FONT><br>", VERSION);
 	do_cr();
 
 	/* Output Tail */
@@ -389,4 +389,9 @@ void output_end_html_attr(html_attr *h)
 			printf("</A>");
 		h->uflag = 0;
 	}
+	if (h->fflag)
+	{
+		printf("</FONT>");
+		h->fflag = 0;
+	}
 }

xlhtml/xlhtml.c

@@ -980,6 +980,17 @@ void main_line_processor(U16 opcode, U16 version, U32 count, U16 last, U8 data)
 /*					printf("I:%04X BL:%04X\n", bufidx, buflast); */
 					cont_str_array = 1;
 					cont_grbit = data;
+					if ((cont_grbit & 0x01) && !(grbit & 0x01))
+					{	/* previous chunk was not Unicode but this one is */
+						int i;
+						grbit |= 0x01;
+						for (i = bufidx-1; i >= 0; i--) {
+							working_buffer[2*i] = working_buffer[i];
+							working_buffer[2*i+1] = 0;
+						}
+						bufidx = 2 * bufidx;
+						buflast = 2 * buflast;
+					}
 					return;
 				}
 
@@ -2516,16 +2527,20 @@ void output_cell(cell *c, int xml)
 	html_attr h;
 
 	if (c == NULL)
-		printf( xml ? "" : "<TD>&nbsp;");
+		printf( xml ? ">" : "<TD>&nbsp;");
 	else if (c->spanned != 0)
+	{
+		if (xml)
+			printf(">");
 		return;
+	}
 	else
 	{	/* Determine whether or not it is of numeric origin.. */
 		int numeric = IsCellNumeric(c);	/* 0=Text 1=Numeric */
 		html_flag_init(&h);
 		if (c->xfmt == 0)
 		{	/* Unknown format... */
-			printf( xml ? "" : "<TD>");		/* This section doesn't use Unicode */
+			printf( xml ? ">" : "<TD>");		/* This section doesn't use Unicode */
 			if (c->ustr.str)
 				OutputString(&(c->ustr));
 			else
@@ -2551,9 +2566,9 @@ void output_cell(cell *c, int xml)
 			if ((c->rowspan != 0)||(c->colspan != 0))
 			{
 				if (c->colspan)
-					printf( xml ? "<colspan>%d</colspan>" : " COLSPAN=\"%d\"", c->colspan);
+					printf( xml ? " colspan=\"%d\"" : " COLSPAN=\"%d\"", c->colspan);
 				if (c->rowspan)
-					printf( xml ? "<rowspan>%d</rowspan>" : " ROWSPAN=\"%d\"", c->rowspan);
+					printf( xml ? " rowspan=\"%d\"" : " ROWSPAN=\"%d\"", c->rowspan);
 			}
 			if ((safe > 0)&&(!nullString))
 			{
@@ -2667,8 +2682,8 @@ void output_cell(cell *c, int xml)
 				}
 			}
 
-			/* Close up the <TD>... */
-			printf(xml ? "" : ">");
+			/* Close up the <TD> or <cell>... */
+			printf(">");
 
 			/* Next set font properties */
 			if (safe > 1 && !xml )
@@ -2719,8 +2734,6 @@ void output_cell(cell *c, int xml)
 
 		/* Now close the tags... */
 		output_end_html_attr(&h);
-		if (h.fflag)
-			printf("</FONT>");
 	}
 
 	if (!aggressive)
@@ -3584,8 +3597,6 @@ void OutputString(uni_string *u)
 						if (format_changed)
 						{	/* if old attributs, close */
 							output_end_html_attr(&h_flags);
-							if (h_flags.fflag)
-								printf("</FONT>");
 						}
 						else
 						{	/* FIXME: Also need to consider that a font may already be set for
@@ -3609,8 +3620,6 @@ void OutputString(uni_string *u)
 				if (format_changed)
 				{
 					output_end_html_attr(&h_flags);
-					if (h_flags.fflag)
-						printf("</FONT>");
 				}
 			}
 			else
@@ -3906,11 +3915,11 @@ void output_start_font_attribute(html_attr *h, U16 fnt_idx)
 	if (font_array[fnt_idx]->super & 0x0003)
 	{
 		if (h->fflag)
-			printf(" SIZE=2");		/* Sub & Superscript */
+			printf(" SIZE=\"2\"");			/* Sub & Superscript */
 		else
 		{
 			h->fflag = 1;
-			printf("<FONT SIZE=2");		/* Sub & Superscript */
+			printf("<FONT SIZE=\"2\"");		/* Sub & Superscript */
 		}
 	}
 	else
@@ -3918,14 +3927,14 @@ void output_start_font_attribute(html_attr *h, U16 fnt_idx)
 		if (h->fflag)
 		{
 			if (font_array[fnt_idx]->size != default_fontsize)
-				printf(" SIZE=%d", font_array[fnt_idx]->size);
+				printf(" SIZE=\"%d\"", font_array[fnt_idx]->size);
 		}
 		else
 		{
 			if (font_array[fnt_idx]->size != default_fontsize)
 			{
 				h->fflag = 1;
-				printf("<FONT SIZE=%d", font_array[fnt_idx]->size);
+				printf("<FONT SIZE=\"%d\"", font_array[fnt_idx]->size);
 			}
 		}
 	}

xlhtml/xml.c

@@ -98,7 +98,7 @@ void OutputTableXML(void)
 			printf("\t\t\t\t<row>\n");
 			for (k=ws_array[i]->first_col; k<=ws_array[i]->biggest_col; k++)
 			{
-				printf("\t\t\t\t\t<cell row=\"%d\" col=\"%d\">", j, k );
+				printf("\t\t\t\t\t<cell row=\"%d\" col=\"%d\"", j, k );
 				output_cell(ws_array[i]->c_array[(j*ws_array[i]->max_cols)+k], 1); /* This stuff happens for each cell... */
 				printf("</cell>\n" );
 				if (ws_array[i]->c_array[(j*ws_array[i]->max_cols)+k])
@@ -109,7 +109,7 @@ void OutputTableXML(void)
                                 
 			}
 
-			printf("</row>\n");
+			printf("\t\t\t\t</row>\n");
 		}
 		printf( "\t\t\t</rows>\n" );
 		printf("\t\t</sheet>\n");