XSL Formatter V2 Q&A

To make a second line in the xsl:param dialog, press down arrow key after inputting a name or a value of xsl:param. This feature should be modified to be easier.

Take the following steps if you want to turn off the error message on screen.

1. On the Formatter Menu, click Formatter Option.
2. In the Formatter Options dialog box, click the File Output, etc. tab.
3. Check Formatting Error Logging.
4. Click OK.

In V2.2, the function of starting auto formatting is supported. By registering constantly used stylesheet, it eliminates the need for specifying the stylesheet on each time. Furthermore, by checking the auto-formatting check box, and specifying the document, it starts formatting automatically.

We are now investing the functionality so that the selected path becomes valid. And it will be possible in the next version V2.3.

Yes. User Interface can register a separate editor into a document and a stylesheet for each from V2.4.

This problem was fixed with XSL Formatter V2.5 MR3.

Yes. You may use an interface program of command line to the engine (AXFO.dll) of XSL Formatter that enables you to pass file names as the arguments from command line.

The option.ini file is not a necessary condition for PDF file generating. XSL Formatter can generate PDF file without specifying option ini file. If you don't specify option.ini file, XSL Formatter uses default value in the formatter option.

e.g. AXFOCMD
 -d "c:\My Documents\xml\sample.xml"
 -s "c:\My Documents\xml\sample.xsl"
 -p "Acrobat PDFWriter"
 -o "c:\My Documents\xml\sample.pdf" 

If you want to specify option.ini file, please make option.ini file before formatting. Click here to get sample option.ini file.

Yes. We support passing parameters on the commandline. You can specify parameter name and value of xsl:param. When value includes white space, specify "name=value" using double-quortation mark. The following three patterns are available.

e.g.
    -param name=value
    -param "name"=value
    -param name="value"

You can use '='(equal mark) within " " or ' '. Specify "&guot;" as a string double quortation. Specify "'" as a string single quortation.

I you want to use PDF Option by commandline utility, specify @PDF as printername.

AXFOCMD -p "@PDF" -d document.fo -o document.pdf

Yes. Please specify a file name or a printer name in case sensitive.

Only when the COM interface is used on the ASP application, it is possible.

1. Specify "@PDF" to IPrinterName property.
2. Don't specify OutputFilePath property.
3. Perform "Response.End" after calling "Execute" method.

And, in current XSL Formatter V2, the temporary file is used.

From V2.2, security function is supported. It is possible to set user password, master password and some access privileges. Refer to Online Manual attached to PDF Output Option for details.

No. It's no supported yet.

No. It's no supported yet.

Yes. It will have the same number of pages.

Support team has checked the FO instance. The reason it takes time to create PDF is background-images. PDF Output Option supports JPEG-based compression and ZLIB-based compression when outputting raster graphics. By specifying 'Auto' selection of compression method, this version PDF Output Option program calculates the size of compressed image for both compression methods for each page, selects smaller size of the two methods. To shorten the time to create PDF, you are recommended to specify JPEG or ZLIB compression method depending on your type of graphics, by taking following steps:

1. On the XSL Formatter Tools menu, click PDF Option .
2. In the PDF Option dialog, click Compression to display the Compression dialog box and select JPEG as Compression Setting.

Antenna House will improve PDF Output Option in the future.

In the current PDF Option version, the original image is not output to PDF as is. Once the image is transformed into BMP or WMF/EMF, then according to the specified compression or quality, it is transformed into JPEG-based compression and ZLIB-based compression of BMP image again.

That's why the size of the image becomes bigger or smaller.

Not yet in XSL Formatter. You told us that it was possible in V2.2 but not work in V2.3. However, both V2.2 and V2.3 doesn't support this functionality. In order to achieve it Antenna House extension functionality is required, which is not implemented yet.

It's possible to specify these document properties from V2.5 by using Antenna House XSL Extensions. Please refer to the XSL Formatter Online Manual for more details.

XSL Formatter is based on C++, while FOP is based on Java. We assume this makes the difference in the amount of memory use between them.

It often happens when the compression method in the PDF Option dialog is set as Automatic.

It's because PDF Output Option supports JPEG-based compression and ZLIB-based compression when outputting raster graphics. By specifying 'Automatic' selection of compression method, this version PDF Output Option program calculates the size of compressed image for both compression methods for each page, selects smaller size of the two methods.

That is the images are created twice, the processing speed becomes lower.

To shorten the processing time to create PDF, you are recommended to specify JPEG or ZLIB compression method depending on your type of graphics. If your image is like a picture, select "JPEG", if it is like a business graphic, select "ZLIB".

1. In the case of the image type that is treatable by XSL Formatter and can be treated in PDF file without change, the original image file will be inserted in PDF.Therefore, the speed of processing has been improved.
2. In the case of the image type that is treatable by XSL Formatter but can not be treated in PDF file, it is converted to the type which can be treated in PDF. In this case, the processing speed depends on resolution setting/compression setting as well as V2.
3. On the Windows environment, the image type that is not treatable by XSL Formatter will be converted to BMP format by using the function of Windows or LeadTools. In this case, the processing speed depends on resolution setting/compression setting as well as V2.

When a PDF file is created as a PDF output option, it turns out that file size becomes large according to the following conditions.

1. Problem of Graphics.

   According to your file, the same images are embedded repeatedly. PDF Output Option embeds all the images even if they are the same images. For this reason, the size of PDF becomes bigger. This problem will be removed with the next version of XSL Formatter.

2. Problem of Link.

   In case the link is used, the id property is supposed to be added to fo:block. PDF output option divides the stream in PDF per unit of the block with the id property. For this reason, the compression ratio becomes inefficient and the size becomes bigger.

These two problems were fixed with XSL Formatter V2.5. However, there might be a possibility that the size becomes bigger when the word-spacing,letter-spacing,justify property is used for justifying letter spacing.

In addition, This problem does not occur in XSL Formatter V3.

If the font file of 0 Byte exists in the "font" folder of the Windows system currently used, it turns out that this error occurs. This problem will be generated if XSL Formatter V2.4 or earlier are used. This problem was corrected in XSL Formatter V2.4 Maintenance Release 1.

This is a bug of PDF Output Option. When fo:basic-link external-destination is used and the link address ends with ')' , inaccurate PDF is outputted. This problem is due to be corrected in the next version. Please understand the situation.

PDF file does not store the paper size as "mm" but as "point". 1 point is equal to 1/72 inches.

• Using PDF Option

  The value of mm is transferred to the value of point as follows and the size 595 × 841 is stored in PDF.

  210 × 72 / 25.4 = 595
  297 × 72 / 25.4 = 841
  

  Next, when the PDF is displayed in Acrobat , the value of point is transferred to the value of mm on the contrary.

  595 / 72 × 25.4 = 209.9
  841 / 72 × 25.4 = 296.69
  

  Therefore the paper size is shown as 209.9 × 296.69.

• Using Acrobat Distiller

  In case of Acrobat Distiller, the calculation is a little complicated. When an output destination is a printer driver (Distiller), XSL Formatter passes not the value of "210 × 297" but the information of "A4" to the printer driver (Distiller).

  Since the size of A4 is "210.2241 × 297.3018", conversion from mm to point is performed with this value, and the size of "595 × 842" point is stored in PDF.

  210.224 × 72 / 25.4 = 595
  297.3018 × 72 / 25.4 = 842
  

  And when it is displayed, since conversion from point to mm is conversely performed. the paper size is shown as 209.9 × 297.04.

  595 / 72 × 25.4 = 209.9
  842 / 72 × 25.4 = 297.04
  

XSL Formatter V2 and PDF output option are not supporting CMYK. Therefore, the PDF file of PDF/X standard (ISO15930) cannot be made, using PDF output option or Acrobat Distiller unfortunately. However it's possible to convert to CMYK by post-processing the file with a third party application.

PDF/X standard will be examined from now on.

This problem is a limitation of XSL Formatter V2.

When XSL Formatter V2 processes JPEG, the image displayed on a screen is processed. Therefore, if a screen setup of client PC differs from that of server PC, such phenomenon will occur.

Please set the same value to the "color" in the "setting" item of "Display Property" for each PC.

When we examined it by using the sample data, the space with the width of 0.01mm was generated between cells of fo:table.

For instance, there seems to be a thin line in zoomed PDF when background-color is specified for fo:table-cell.

This problem was fixed with XSL Formatter V2.5 MR3.

When PDF is generated by XSL Formatter, the color space used in PDF is DeviceRGB.

The test data and the line art data in PDF file generated by PDF Option are always set as compression format. This setting is not changeable.

It is possible for the image files to select the way of compression by ZLIB/JPEG/Auto.

('Auto' can select the smaller size of the file automatically. Please see also the online manual of XSL Formatter V2.5 PDF output option.)

The bold character might not be displayed with Acrobat 5.0. It is displayed with Acrobat 5.0.5 or the later version. Please update the version of Acrobat.

There are some differences as follows:

1. Difference in the way of creating bookmarks.

   In XSL Formatter, bookmarks are created automatically by writing 'Section 1' in the text.

   <fo:flow flow-name="xsl-region-body">
       ...
       <fo:block axf:outline-level="1">Section 1</fo:block>
       ...
   </fo:flow>

   In FOP, first you write the following.

   <fo:root>
       <fox:outline internal-destination="sec1">
           <fox:label>What is FOP?</fox:label>
       </fox:outline>
       ...

   Then, add the following description.

   <fo:flow flow-name="xsl-region-body">
       ...
       <fo:block id="sec1" >A   What is FOP?</fo:block>
       ...
   </fo:flow>

2. Difference in viewing the tree structure

   Suppose you are making the following tree structured document.

   Section 1
     Section 1-1
      Section 1-1-1
      Section 1-1-2

   In XSL Formatter, you can write the tree structure simply by using the outline-level attribute.

   <fo:block axf:outline-level="1">Section 1</fo:block>
   <fo:block axf:outline-level="2">Section 1-1</fo:block>
   <fo:block axf:outline-level="3">Section 1-1-1</fo:block>
   <fo:block axf:outline-level="3">Section 1-1-2</fo:block>

   In FOP, you need to write the tree structure as it is constructed.

   <fox:outline internal-destination="sec1">
     <fox:label>Section 1</fox:label>
     <fox:outline internal-destination="sec1-1">
       <fox:label>Section 1-1</fox:label>
       <fox:outline internal-destination="sec1-1-1">
         <fox:label>Section 1-1-1</fox:label>
       </fox:outline>
       <fox:outline internal-destination="sec1-1-2">
         <fox:label>Section 1-1-2</fox:label>
       </fox:outline>
     </fox:outline>
   </fox:outline>

The function of creating bookmarks in XSL Formatter is much more efficient, easy to handle and saving your time compared to that of FOP. Please refer to XSL Formatter Online Manual for creating bookmarks.

The difference between PDF Output Option and Postscript in creating PDF is:

XSL-FO -->(FORMATTER)-->AreaTree-->(PDF Output Option) --> PDF
               |
               |
          (Windows GDI)
               |
               +--> (Postscript Printer Driver) -> postscript

The functionality "PDF Mark" can be embedded in Postscript. When outputting Postscript by Postscript Printer Diver, it will be possible to set the function of Bookmark and Link by using PDF Mark.

Then, Bookmark and Link can be set to PDF generated by Distiller.

This functionality is available with the version of V2.3.

Acrobat and Acrobat Reader can not show Thai characters in bookmark. They become always dots in bookmark. Acrobat and Acrobat Reader do not support Thai language for bookmark.

No, it's impossible.

This was a bug of XSL Formatter until V2.3. When axf:outline-level="2" is specified to the nested block, axf:outline is ignored.

<fo:block axf:outline-level="1" axf:outline-title="Level1" >
     Level1
     <fo:block axf:outline-level="2" axf:outline-title="Level2" >
          Level2
          <fo:block axf:outline-level="3" axf:outline-title="Level3" >
               Level3
          </fo:block>
     </fo:block>
</fo:block>

In the FO sample above, the bookmarks of axf:outline-level="2", axf:outline-title="Level2", axf:outline-level="3" or axf:outline-title="Level3" cannot be created. This problem will be fixed with the next version of XSl Formatter.

BTW, if you remove the nested block, bookmarks are created correctly.

<fo:block axf:outline-level="1" axf:outline-title="Level1" >
     Level1
</fo:block>
<fo:block axf:outline-level="2" axf:outline-title="Level2" >
     Level2
</fo:block>
<fo:block axf:outline-level="3" axf:outline-title="Level3" >
     Level3
</fo:block>

This problem occurs when the paper orientation is landscape. It's because there is a problem in transforming coordinate with some PS printer driver (Distiller) when the paper orientation changes from portrait to landscape. As the result the link spot is not placed correctly.

Please use the newest version of your PS printer Driver (Distiller) in such case.

We fixed this problem in the Maintenance Relase 2 version of XSL Formatter.

Automatic division of PDF, the crossing reference for the link or the bookmark cannot be achieved.

We will consider for this issue to be improved by adding an extension function in the future.

This is impossible. Only a static content is made for Bookmark of PDF by the Antenna House Extensions. Bookmark that uses a dynamic content such as fo:page-number and fo:marker cannot be generated. This is a limitation of XSL Formatter.

In your FO, you specified as follows:

external-destination="url("www.antennahouse.com")"

When you want to specify absolute URI as external destination, please be sure to write 'http://' as follows:

external-destination="url("http://www.antennahouse.com")"

Without specifying 'http://', it doesn't work correctly because it is considered to be relative path. (BTW It is accidental that it seemed that the external link of PDF operated correctly by V2.2.)

This is considered to be the bug of Acrobat Distiller. When "PDF" is created by "Distiller" from "PSfile" created by "Distiller", we found that the correct "PDF" was output.

Antenna House avoids this problem of Distiller from XSL Formatter V2.3MR1.

When creating PDF using Distiller, up to 127 characters are acceptable. XSL Formatter makes the title length of a bookmark valid up to 100 characters at the maximum when using Distiller. The characters beyond 100 are omitted, then "..." is used instead of the omitted characters.

It's because Distiller rotates the page direction automatically by judging the text direction so that text can be read.

If all the direction of the pages to be output to PDF is 'portrait' or 'landscape', you can avoid this problem by selecting 'Collectively by File' in 'Auto-Rotated Pages'.

If the page direction is mixed with portrait and landscape, select 'Individually' in 'Auto-Rotated Pages'. In this case, I suppose Distiller decides the page direction by judging the text direction in a page. Then selects the direction of larger number of text as the direction of the whole text.

Therefore if you put larger number of the text to meet the page direction, you can generate page direction as you like.

BTW, The problem and the way around are only suppositions by Antenna House investigation. The other information (images etc.) may affect this problem. Please understand that this is a part of Distiller's engineering

The reason why the ruler width are not consistent:

In this case of creating PDF by Acrobat Distiller. PDF was output through Windows GDI. The width of the rule depends on the output resolution. Even though you specify a small value like 0.1pt, if the resolution is 72dpi, the minimum width becomes 1pt. Furthermore, there is a possibility that the width becomes 2 pixel-width of a rule if the rule is placed expanding to the next pixel.

In order to avoid this behavior,

• Use PDF Option instead of Distiller.
• If you use Distiller, raise the resolution setting. (When outputting, in the Paper/Quality tab in the Acrobat Distiller Document Properties dialog, click Advanced... button. If you see lower value such as 71dpi, change the value into 600dpi or more.

Not only using Command line Interface, even if Com Interface is used, a PDF file is not created.

XSL Formatter cannot distinguish the output destination as Acrobat Distiller, but outputs a formatted result like using an usual printer. And the file name specified to "output file"parameter turns into the file name specifed to "Print to file" in a print dialog.

Since "Acrobat Distiller" is a PostScript printer driver, PS file is outputted to the specified destination.

Yes. You can do it by using the interface program of command line

Yes. It is possible to select paper trays without using printer-setting dialog. XSL Formatter can initialize printer setting by making a data file that records the printer settings and specifying it to XSL Formatter.

In Windows, DEVMODE structure is used to specify the information regarding printer setting. The printer driver runs according to the information of the DEVMODE structure.

XSL Formatter package includes AXFODEV.exe as an application for making DEVMODE structure and saving it to a file. Please refer to XSL Formatter Online Manual for more details.

When the printer name includes a space, you need to double-quote the printer name you want to use like the following:

AXFOCMD -p "EPSON LP-LP8700"

Also, double-quote a file name of XML, XSL, PDF, if a file name includes a space.

XSL Formatter cannot generate Postscript by itself. If you install Postscript Pinter driver on Windows and select the Postscript Printer driver from print dialog of XSL Formatter, then you can get Postscript file via Pinter driver.

No, it's not supported when creating Postscript file. Linking is supported only when you create PDF files. Therefore when creating a postscript file with XSL Formatter, bookmarks and links are sometimes not placed correctly.

If you really need to create a correct Postscript file, specify Acrobat Distiller as a Postscript printer name.

This issue is impossible.

Arial does not have the glyph for U+2003. Therefore U+2003 shows up as a box in XSL Formatter. The fonts that have these space code glyph are limited to Tahoma, Lucida Sans Unicode, Arial Unicode MS and etc. If you specify font-family="Tahoma", the U+2003 will be indicated by XSL Formatter.

While, the Zero-width space is processed by XSL Formatter (for justification). Therefore U+200B does not show as a box.

Next code shows up as a box:

• U+2000 (EN QUAD)
• U+2001 (EM QUAD)
• U+2002 (EN SPACE)
• U+2003 (EM SPACE)
• U+2004 (THREE-PER-EM SPACE)
• U+2005 (FOUR-PER-EM SPACE)
• U+2006 (SIX-PER-EM SPACE)
• U+2007 (FIGURE SPACE)
• U+2008 (PUNCTUATION SPACE)
• U+2009 (THIN SPACE)
• U+200A (HAIR SPACE)

As the "Mincho" font does not exist in the system folder, XSL Formatter retrieves the font of "Lucida Sans Unicode", "Arial Unicode MS", "Tahoma, Microsoft SansSerif" and use the fist retrieved font as a fallback font. The reason why these fonts are retrieved is that they are Unicode and they include many Unicode characters, that is easy to alternate.

When "LucidaSansUnicode" is used as a fallback font family, it's impossible to use Japanese character because Japanese character (glyph) does not exist in "LucidaSansUnicode". Then XSL Formatter cannot embed the non-existing Japanese character in PDF and it becomes a white space.

The reason why Japanese character can be displayed in XSL Formatter is that when XSL Formatter gives a command to display to the system (Windows), Windows automatically use a font such as MS PGothic as a fallback font family and display it.

Then it happens that Chinese character becomes white space in PDF, while it can be displayed correctly in XSL Formatter. To avoid this:

1. Specify the attribute to the font-family property that currently exists in the system.
2. Specify the attribute such as serif to the font-family and use the font that is specified on the Language-Fonts i18n tab in the Formatter Option dialog.

Yes. Put the fonts in reverse order like this:

font-family="Arial , MS Mincho"

The first placed font takes priority. If MS Mincho is placed first, It is applicable in Western font, so Arial is not used. If Arial is placed first, Western font uses Arial and Japanese font uses MS Mincho because Arial is not applicable in Japanese.

XSL Formatter uses the fonts installed on your system.

Plantin ATM font is Type1 font. When Type1 font is specified, text in Antenna GUI may look differently compared to PDF. It's because XSL Formatter V2.3 of PDF Option does not support Type1 font except Adobe Standard 14 fonts.

If you output PDF by Acrobat PDF Distiller, there is no problem creating PDF.

This problem was fixed with XSL Formatter V2.4.

Acrobat and Acrobat Reader don't support specifying "Non-breaking spaces" character and "trademark" character to bookmarks. Therefore, PDF file views garbage characters where the "Non-breaking spaces" character and "trademark" character are written.

This is an issue of line braking by The Specification of Unicode Line Breaking Properties: http://www.unicode.org/unicode/reports/tr14/ According to this specification, it's acceptable to break lines after the colon of 2001/02:Sk322.

IS - Numeric Separator (Infix) (XB)

Characters that usually occur inside a numerical expression may not be
separated from following numeric characters, unless space character
intervenes. Since they are otherwise sentence ending punctuation, they
prevent breaks before.

There is no break in “100.00” or “10,000”, nor in “12:59”

002C  COMMA
002E  FULL STOP
003A  COLON
003B  SEMICOLON
0589  ARMENIAN FULL STOP

It means that if numbers follow after the COLON, there should be no break, however, if non-numbers follow after the COLON, the line break is acceptable. In order to avoid breaking after the COLON, please put ⁠ after the COLON as follows: 2001/02:⁠Sk322 See also Unicode 3.2 Specification to get more information about ⁠ (U+2060 = WORD JOINER): http://www.unicode.org/unicode/reports/tr28/

It sometimes happens that it looks as if the space between words is gone depending on the font size or the font type you specify. First using the zoom in function of XSL Formatter, confirm how it looks like. Then adjust the font size or font type if necessary.

This was a bug of XSL Formatter until V2.4. For example, when the fonts for different languages are set out as follows:

 font-family="MS Mincho, SimSun"

According to the current XSL Formatter, if the text is recognized as Chinese, font for Chinese character is supposed to be selected, even though it can be shown as Japanese font.

In the following case,

 font-family="MS Mincho,serif"

the result is the same as above if serif is specified as Chinese font in the setting on the Language-Fonts i18n tab.

The reason why the script is shown as Chinese font in FO with language="en" specified is that when the script is recognized as Chinese character and XSL Formatter cannot know exactly which language's font it is, Chinese character takes priority.

From V2.5, in this case, the font is selected by the locale of the system (OS) where XSL Formatter runs.

Font name should be exactly the same as the standard font name.

For example, in the name "Times New Roman", the white space between Time and New should exist, number of white space should be one.

In order to specify the font name without fail, I recommend to view a font name in a font dialog box in any applications, and copy and paste it.

If the specified font does not have glyphs for some characters, XSL Formatter GUI displays these characters by alternative font which has the glyphs for the characters. This is automatically done by Windows Operating System. Acrobat Reader ties to use the specified font for the characters, but the font does not have glyphs for the characters. It causes error messages of Acrobat Reader. The reason of this problem is that you may have specified the incorrect fonts to CJK. To resolve this phenomenon, you must specify the correct fonts for the CJK.

Yes, it's possible to embed TrueType font matrix information within PDF files. (However，as an exception, the fonts having limitation on embedding are not included.)

• If you are using XSL Formatter PDF option, please select 'Font Embedding' tab and check 'Embed All Embeddable Fonts' checkbox, in the 'PDF option' dialog on the 'Tools' menu.
• If you are using Adobe acrobat Distiller, Please select 'Font' tab and check 'Embed All Fonts' checkbox, in the 'Job Option' dialog.

It's also possible to specify fonts to be embedded individually.

As the width of the space is calculated by the specified font, it is supposed to be the same if the glyph is the same. However, it looks as if the width of the space is different because of the part of space existing in the glyph of font. Please format this file in XSL Formatter GUI and display the border.

This problem is a bug of PDF Output Option V2.4. We will fix this problem in the next version.

These two fonts are the Type1 font. Information of the font is as mentioned below.

'Avenir-Heavy' Font
avenh___.pfm
    dfWeight=400(0x190)
    WindowsName='Avenir'
    mPostScriptName='Avenir-Heavy'

'Avenir-Roman' Font
avenr___.pfm
    dfWeight=400(0x190)
    WindowsName='Avenir'
    mPostScriptName='Avenir-Roman'

From the above-mentioned information, with Windows, the indicatory name of both fonts becomes Avenir. ( That's the reason why two fonts are not visible with MS-Word.)

Usually, with Windows a font distinguishes the font name (WindowsName) or dfWeight value (Bold or Italic, for example). However, as these two font have the same value of WindowsName and dfWeight, they cannot be distinguished.

Usually, Windows distinguishes the font name by the font name (WindowsName) or by the dfWeight value (Bold or Italic, for example). However, as these two font have the same value of WindowsName and dfWeight, they cannot be distinguished.

Because XSL Formatter uses GDI to select the fonts, it cannot distinguish two fonts as well as MS-Word, only one font can be embedded in PDF.

When two fonts are installed on the Windows system, the Avenir-Roman font is supposed to be embedded in PDF. If you probably would like to use the Avenir-Heavy font, by deleting the font of Avenir-Roman from your Windows system, the Avenir-Heavy font will be embedded in PDF.

In order to use external characters, please specify the font where external character is stored, and describe the character code of the external character which you would like to use.

In XML and XSL, it's possible to embed the Unicode of external characters into XSL-FO by describing as follows:

In XML

    .
    .
    .
<FONT FACE="Font name of external character">
 &amp;#x260E;
</FONT>
    .
    .
    .

In XSL

    .
    .
    .
<xsl:template match="FONT">
 <fo:block font-family="{@FACE}">
  <xsl:value-of disable-output-escaping="yes" select="." />
 </fo:block>
</xsl:template>
    .
    .
    .

This was a bug of XSL Formatter until V2.4. It happens when the proportional font is used.

Therefore, if fonts other than the proportional font are specified, it is possible to evade.

This problem was corrected in XSL Formatter V2.5.

XSL Formatter V2 can not display the character of some Type 1 font such as ZapfDingbats and Symbol on GUI when they are specified by Unicode. Please use the Postscript encoding for them instead of Uniocde.

The PostScript encoding can be confirmed by the text (Zapfdingbats.txt) attached to the font.

For example, please use the code as follows:

<fo:block font-size="12pt" font-family="ZapfDingbats">
    &#xE4;
</fo:block>

You have to put a unit of measure when specifying the size in XSL-FO. It is an error if there are no units.

In addition, you have to use the unit px when specifying the width, height of the img tag in HTML. The following is the example.

<xsl:if test="@width">
  <xsl:attribute name="content-width">
    <xsl:value-of select="@width" />px
  </xsl:attribute>
</xsl:if>

It is possible by installing InternetExploler 5.0 or later, installing SVG Viewer provided from the Adobe homepage, although XSL Formatter cannot display and print SGV by itself.

XSL Formatter treats relative path as XML document based. Therefore when there is no path or when you specify a image file as a relative path, XSL Formatter search the image file using the pass as XML document based.

In this case, you need to specify the absolute path of the image file, or need to save the image file in the folder where XML file exists.

We add powerful functions of graphics with the release of V2.5 in comparison with V2.0. Click here for more details.

Yes. You can use GIF with XSL Formatter. UNISYS requires the direct contract between customer and UNISYS if the customer uses GIF function on the Internet or Intranet servers. Especially for Web server, I think each customer is required to contract with UNISYS.

For standalone purpose or in-house use, Antenna House can supply GIF function to users depending of the contract between Antenna House and UNISYS. It is only a matter of amount of fee.

By using PDF Output Option, it's impossible to embed vector graphics without change in PDF.

By using Distiller or PDF Writer, it's possible to embed other than SVG vector graphics as vector format without transforming to rasterized format in PDF.

In case of PDF Option:

• You can change the value of file compression when transforming vector graphics into rasterized format. Set Compression rate by clicking PDF Option on the Tool menu of XSL Formatter. You can adjust the graphics quality. See Online Manual attached to the PDF Option.
• In case of EPS graphics, preview image is embedded, if there is no preview image, it's possible to embed EPS by reflecting a dummy image on window.

In case of Acrobat Distiller or Acrobat PDF Writer:

• By using Distiller or PDF Writer, it's possible to embed other than SVG vector graphics as vector format without transforming to rasterized format in PDF.
• In the case of SVG, the quality of image can be upgraded by the following steps
  1. In the Print dialog, select Acrobat Distiller, and click Property. The Acrobat Distiller Document Property dialog is displayed on the screen.
  2. On the Adobe PDF Setting tab, Change CJKScreen or Screen to Print or Press in the Job Option. You may see how the value of file compression changes. The graphics are downsampled to 72dpi compression rate in Screen or in CJKScreen. Also in case of PDFWriner, you can change the setting of graphics downsampling in the Property setting.
• EPS can be output to vector format only by using Acrobat Distiller in Windows2000/XP, in other case preview image of EPS is embedded in PDF.

XSL Formatter V3 includes only SVG image in PDF with a vector image. Therefore, an image does not deteriorate.

XSL Formatter supports CGM. It's possible to display CGM using Corel ActiveCGM, ITEDO ISO-View as a plug-in for XSL Formatter. Both are downloadable CGM Viewers free of charge.

In V2.2, Setting of compression rate is supported. You can specify the value of resolution, specify the quality of the Raster Graphics when stored in JPEG format in the range of 1-100, or low, medium, high, max. The quality becomes higher in proportion to the increase of number, however the size of the output file becomes bigger.

Compression Setting for PDF graphics output is available in the PDF Option dialog. On the XSL Formatter Tools menu, click PDF Option and select Compression.

No. It's our future plan. When creating a PDF file using PDF output option, the image of CMYK is converted to RGB and embedded in PDF. Concerning this issue, we would like to keep examining in the future.

Not yet. We are now investigating if it's possible to add this functionality to our future version.

From V2.3 it's available to display and print MathML by installing MathPlayer plug-in by Design Science, Inc. (MS IE6.0 or later required.)

Preview Image and Postscript Code are included in EPS.

From V2.2, EPS image can be displayed on screen, output to printer, inserted to PDF by embedding Postscript code in EPS into Postscript output file via Postscript printer driver. (Available only in Windows2000/XP)

In case of generating PDF by Acrobat Distiller, the printer driver within Acrobat Distiller is distinguished as Postscript Printer Driver. Then the Postscript code in EPS is output, converted and embedded into PDF, it means Acrobat Distiller can generate PDF with high quality EPS graphics on Windows2000/XP.

From V2.3, even if EPS has no preview image, it's possible to embed EPS by reflecting a dummy image on window. When printing by PS Printer, outputting PDF by Adobe Acrobat Distiller, Postscript code in EPS is output and EPS image can be embedded.

It was revised in the version V2.2 and it's possible to specify external-graphic using http. However, in case you use XSL Formatter on a server, the target file is downloaded and stored in the local temporary folder. Therefore if the image cannot be downloaded for authorization reason, it's not shown.

If the EPS does not contain preview image, the special graphics that means "preview image does not exist" will be inserted on XSL Formatter GDI window. When printing to "Distiller" or other "Postscript Print Driver", Postscript code within EPS file will be output in Postscript.

In your case, you changed the original data into GIF. However, it doesn't seem to make sense because XSL Formatter converts it to BMP.

I look over the fo in your file and actually converted it to PDF using the smallest data of which buffy.fo(13,249Byte) is attached.

Please see how I converted it to PDF as shown below:

Image: ReallySmall.gif 4,039Byte

Compression method: Auto (JPEG compression rate selected)

 V2.2: 15,112 Byte

 V2.3: 14,837 Byte

       13,394 Byte (CJKSCreen setting in Distiller)

In this data, image became 4,588Byte (a little bit bigger than original), text became 3,001Byte, and the rest of the main part is character width data.

Times New Roman is used as default formatting, however four types of font type, Regular, Bold, Italic and BoldItalic are used in this document. As the character width data is different for each font type, PDF has four different character width data. When I modified fo and leave only Regular type in the data, it became 10,652 Byte. 4KByte is reduced and it is the part of 3 types of font types concerning character width data.

I assume there is not so much room for reducing the volume in this document so far.

There are two ways of embedding SVG. One is to link to the external destination by using 'external-graphic'. The other is to describe the code of svg or abc:svg xmlns:abc="http://www.w3.org/2000/svg" directly to the document by using 'instream-foreign-object'.

.
.
.
<fo:instream-foreign-object content-width="215.9mm" content-height="279.4mm" >
<svg xmlns="http://www.w3.org/2000/svg">
.
（Code of SVG)
.
</svg>
</fo:instream-foreign-object>
.
.
.

In this case please be sure to make the first element as svg or abc:svg xmlns:abc="http://www.w3.org/2000/svg". (abc as namespace prefix)

When embedding SVG in the document, the size of SVG becomes the absolute value of content-width/content-height properties specified to 'instream-foreign-object' or 'external-graphic'. If these properties don't exist, the absolute value of width/height specified to the svg tag becomes the SVG size.

If there aren't these properties or are invalid, the height of SVG becomes the same as that of line-area (normally font-size).

Please specify content-width and content-height or specify the absolute value of width/height to the svg tag as follows:

.
.
.
<fo:instream-foreign-object content-width="215.9mm" content-height="279.4mm" >
<svg xmlns="http://www.w3.org/2000/svg">
.
（Code of SVG）
.
</svg>
</fo:instream-foreign-object>
.
.
.

XSL Formatter does not have such a function. It's our future plan.

In XSL Formatter V2.4 or later and XSL Formatter V3, the resolution value of the raster images can be specified. Moreover, in XSL Formatter V3, SVG, EMF and WMF are drawn as vector images in the PDF without being converted into the raster images. Please refer to an Online Manual for details.

The U.S. LZW patent expired on June 20, 2003, and the Japanese counterpart patents expired June 20, 2004. GIF/TIFF-LZW License/Fee is not required to display or output in PDF the LZW compression-based GIF/TIFF files any more. The cansellation of the above restriction is available from XSL Formatter V3.1 MR3.

Please confirm your file path.

1. When specifying with a relative path, XSL Formatter searches a file on the basis of the path with which an XML document exists. Therefore, when an XML document and an XSL style sheet exist in another path, you need to be careful.
2. Even if an absolute path is specified, when the reference place is specified by URL, such as a file server and HTTP, you need to check the upper case/lower case of a file name. If the upper case/lower case is different when a WWW server or a file server are UNIX systems, it will be regarded as another file and there will be no file.
3. Be sure if the specified image needs Plug-in. XSL Formatter requires Plug-in for some images. Please refer to the following URL to comfirm the type of images. http://www.antennahouse.com/product/axfo20/graphics.htm

Only background-color is affected by the image with transparent. background-image is not affected (not transparent). It's a limitation of the current XSL Formatter.

In XSL Fromatter V3.1, function of setting transparent color PNG/GIF is added.

Not apply only to MathML, there are some image types that can be described in XSL-FO. When describing them directly, you need to use the fo:instream-foreign-object method before writing the image data.

Following shows the example of MathML:

<fo:block>
  <fo:instream-foreign-object content-type="application/mathml+xml">
    <m:math xmlns:m="http://www.w3.org/1998/Math/MathML">
      .
      . (MathML data)
      .
    </m:math>
  </fo:instream-foreign-object>
</fo:block>

In case of embedding a SVGZ image, it sometimes happens that the size of the original image displayed is not obtained.Please specify the size of a displayed image to content-width, content-height explicitly.

XSL Formatter V2.4 or earlier encoded all the images even though there are the same images. This was the cause of making the file size big. From V2.5, XSL Formatter encodes the images one time and reduces the size of PDF output file.

Preview Image and Postscript Code are included in EPS. Preview image of EPS is displayed on GUI. It's a specification of XSL Formatter. And in the PDF Output Option, preview image is embedded also.

By using Adobe Acrobat Distiller, EPS is outputted to PDF by embedding Postscript code in EPS into Postscript output file via Postscript printer driver. (Available only in Windows2000/XP)

As Raster Imaging Pro library is not included in Developer License (SDK), it's not available to display TIFF, that is supported with evaluation version.

It has to do with the licensing from Lead Technologies that prevents us from redistributing the imaging software/modules as part of the SDK. This is covered in the description of the SDK on our website.

Since XSL Formatter V3 processes only baseline-TIFF in the library of the XSL Formatter itself, it can be displayed also by SDK. TIFF other than baseline-TIFF will also be supported in the future.

No, it's not necessary to display BMP image.

RasterImagingPro is not necessary for:
BMP, JPEG, GIF, WMF, EMF, SVG, CGM, MathML

RasterImagingPro is necessary for:
TIFF, PNG, JPEG in CMYK

PS:
RasterImageingPro is sometimes used for identifying the file formats. If the file extension and content-type are specified explicitly, it's not necessary to identify the file format by using RasterImagingPro.

XSL Formatter needs ActiveCGM plug-in for CGM output. When 2byte character exists in the path name, ActiveCGM does not display the image.

This is a limitation of current XSL Formatter. Please change to no inclusion of 2byte character in CGM path.

This problem is a bug of XSL Formatter V2.5. This problem was fixed with XSL Formatter V2.5 MR3.

When Windows standard CJK fonts are available, XSL Formatter uses these standard CJK fonts as default for each language.

In the Formatter Options dialog box, click the Language-Fonts, i18n tab.

Following is the normal setting:

language: ja - Japanese

Generic font family setting for the language:
serif: MS Mincho
sans-serif: MS Gothic
cursive: (default)
fantasy: (default)
monospace: MS Mincho

If your setting is different, initialize all the settings of generic font family to default.

Click the Default Fonts tab in the formatter Options dialog box.

Check Reset generic font family settings to set as default.

Then you can see Asian fonts without specifying font name to the font-family property. e.g.,

<fo:block font-family="serif" xml:lang="ja">
   (Japanese text here)
</fo:block>

If you want to specify a font name explicitly, please confirm that the font name is correct.( for example, MS Mincho should have a space, MSMincho is incorrect.) e.g.,

<fo:block font-family="'Times New Roman', 'MS Mincho', Batang, serif">
   (Latin and Japanese and Korean mixed text)
</fo:block>

Note: CJK-Unicode is slightly different in character glyph by locales, even if you use the same CJK Unicode. Therefore specify Japanese fonts for Japanese documents, specify Chinese fonts for Chinese documents.

XSL Formatter initializes the setting for standard languages and fonts when XSL Formatter is installed and first executed. As this setting is stored in the registry, the default setting is done only when XSL Formatter is first executed after the installation and there are no changes in the registry.

If you install Chinese fonts and Korean fonts after the default setting, these fonts are not stored as specified fonts in the registry.

To make additional fonts effective:

1. On the Formatting menu, select the Formatter Options tab.
2. In the Formatter Options dialog box, click the Default Fonts tab.
3. Check Reset generic font family settings and click Yes.

Take the following steps:

1. It's necessary to install Arabic language.
   • In Windows 2000, select 'language settings for the system' in the 'Regional Options' in the Control Panel and check Arabic.
   • In Windows XP, select Supplemental language support in language in the 'Regional and Language Options' in Control Panel and check 'Install files for complex script and right-to-left language'
2. In order to make the text in right-to-left mode, you need to specify the writing mode="rl-tb" property to fo block.

It's because the font not listed in Online Manual is used. In your file, the font might be sans-serif specified in Formatter option. In order to create Thai PDF, specify Microsoft Sans Serif instead of Arial that is initial. Then it is suppose to work.

Arial is not available because Arial doesn't have the point code glyph of Thai, as well as Times New Roman. GDI displays Thai using other font, but the creator cannot take this process.

Default Font Family specified in the Formatter Option is used when there is no font-family specified in the stylesheet and it is not used when font-family is specified in the stylesheet.

When the font that is not installed on the system or the font not found is specified in the font-family setting, defaultFontFamily is not accepted but fallbackFontFamily which exists insiade AXFO.dll is used instead.

fallbackFontFamily = "Lucida Sans Unicode, Arial Unicode MS, Tahoma, Microsoft Sans Serif"

Then, the font becomes Lucida Sans Unicode. Please change the setting as font-family="Arial Unicode MS, serif" in your stylesheet.

Yes. XSL Formatter supports HAT. Because XSL Formatter is a UNICODE application on Windows, the languages can be formatted as far as there are suitable fonts installed on your PC. Details are as follows:

Thai language:

• Automatic line breaking (word breaking by dictionary) is not supported from V2.5 effective only with WindowsNT/2000/XP.
• Supports the function of glyph substitution, such as the adjustment of tone mark position, depending on its character sequence.(Effective only on Windows 2000/xp).

Arabic and Hebrew:

• From V2.3 Kashida is supported and justification of Arabic text is available. The glyph named Kashida is inserted between the characters of a word and extends the width of the word.

UnicodeBIDI processing is supported only with Windows2000/XP.

xml:lang is a shorthand of the country property and the language property. You can specify xml:lang where the country property and the language property can be specified. And the language can be changed.

<fo:block xml:lang="en">
    English flow（language code en）
</fo:block>

<fo:block xml:lang="zh-CN">
    Simplified Chinese flow（language code zh, country code CN）
</fo:block>

Arial Unicode MS supports almost all of Unicode characters. However, Arial Unicode MS cannot display all the languages correctly.

As CJK (Chinese, Japanese and Korean) have common area in Unicode, it's impossible to distinguish which of CJK character is from the character code. Concerning CJK, you have to specify the specific font name for CJK code point characters. In other words, it is necessary to specify the specific font of the language in order to display the appropriate language as follows.

<fo:block font-family="MingLiU">
    沈下大海的島嶼(Traditional Chinese text)
</fo:block>

<fo:block font-family="SimSun">
    〓下大海的〓〓(Simplified Chinese text)
</fo:block>

<fo:block font-family="MS Gothic">
    海に沈む島(Japanese text)
</fo:block>

It's not a bug. It follows the Unicode spec.

The directionality of Unicode text is decided by bidi algorithm defined in the Unicode specification. Latin Alphabet goes from left to right, Arabic Alphabet and Hebrew Alphabet go from right to left.

The enclosing parentheses are neutral characters that are available for both writing mode in left-to-right and right-to-left. The glyph direction changes reversibly when writing directions changes from 'right-to-left' to 'left-to-right'. It's so called 'Mirroring'.

In other words, as the enclosing parentheses are neutral characters and also characters acceptable for 'Mirroring', left parenthesis '(' becomes right parenthesis ')' and right parenthesis ')' becomes left parenthesis '('. Strictly saying, this phenomenon occurs by reversing the glyph, so the right parenthesis doesn't become the exact left parenthesis.

In the following example, suppose 'Arabic' stands for an Arabic word and 'English1', 'English2' stand for English words, and English2 is enclosed by parentheses, appearing order is as follows:

<fo:block-container writing-mode="rl-tb" language="ar">
<fo:block>
Arabic English1 (English2)
</fo:block>
</fo:block-container>

When displaying in writing mode 'right-to left', it's shown as follows:

(English1 (English2 cibarA

As the Unicode Spec says that ' ... Generally, neutrals take on the direction of the surrounding text.', the direction of the neutral between the left-to-right text and the left-to-right text becomes left-to-right. So, the left parenthesis between 'English1' and 'English2' becomes left-to-right mode.

However, as the Unicode Spec. says that 'In case of a conflict, they take on the embedding direction.', the right parenthesis placed after 'English2' becomes right-to-left mode because writing-mode="rl-tb" is specified to fo:block-container. Then, the right parenthesis becomes left parenthesis by the process of mirroring and placed at the last of line (left side) by writing-mode="rl-tb".

There are two ways considerable to avoid this. First, please use the Unicode control character U+200E (LRM, Left-to-Right MARK). LRM is a zero-width character with directionality. If you add a character with left-to-right mode on the right side of '(English2)', the right parenthesis also becomes left-to-right mode.

<fo:block-container writing-mode="rl-tb" language="ar">
<fo:block>
Arabic English1 (English2) &#x200E;
</fo:block>
</fo:block-container>

This character is also effective with IE. In Notepad installed on Windows 2000/XP, LRM can be entered in [Insert Unicode control character] in right click menu.

Second, please use fo:bidi-override. Enclose '(English2)' with <fo:bidi-override unicode-bidi="embed" direction="ltr">

<fo:block-container writing-mode="rl-tb" language="ar">
<fo:block>
Arabic English1
<fo:bidi-override unicode-bidi="embed" direction="ltr">
(English2)
</fo:bidi-override>
</fo:block>
</fo:block-container>

FYI, as half-width space is also a neutral character, its position can be changed. Please use either way of the above mentioned.

Get more info from 'Unicode Standard Annex #9 The Bidirectional Algorithm': http://www.unicode.org/unicode/reports/tr9/

Yes. The level of the numbers are level 2. Therefore the value (EN) is reversed and after that the Hebrew text whose level is 1 is reversed.

Please refer to Unicode Standard Annex #9 to see more BIDI Algorithm. Also you can test it by Reference Code from the following URL.

• http://www.unicode.org/unicode/reports/tr9/BidiReferenceCpp/
• http://www.unicode.org/unicode/reports/tr9/BidiReferenceJava/

The followings shows the sample of BidiReferenceCpp. Pease find the answer here.

E:\BidiReferenceCpp\Debug>BidiReferenceCpp -verbose "abc 12 - GHI 12 - 34def"
Input     1: abc 12 - GHI 12 - 34def
Input Types: >>>_11_ _<<<_11_ _11>>>
Base Level : 0
Levels (A) : 00000000000000000000000
Types (A)  : >>> >>   <<< 11   11>>>
Types (B)  : >>>>>>>>><<<<11<<<11>>>
Levels (B) : 00000000011112211122000
Output    1: abc 12 - 34 - 12 IHGdef

BIDI Algorithm of XSL Formatter V2.3 uses Windows Uniscribe API. Because of this reason, The algorithm is influenced by Uniscribe.

Uniscribe API probably implements the BIDI Algorithm of Unicode 2.x. The BIDI Algorithm of Unicode 2.x doesn't have the rule that EN (numbers) coming after L(left-to-right characters) become L. http://www.unicode.org/unicode/reports/tr9/tr9-2.html

For this reason , BIDI level of numbers raises even though the numbers come after Western characters.

The latest Unicode BIDI Algorithm has the following rules:

Form the URL http://www.unicode.org/unicode/reports/tr9/tr9-10.html

W7. Search backwards from each instance of a European number until the first
strong type (R, L, or sor) is found.
If an L is found, then change the type of the European number to L.

        L  N EN => L  N  L

BIDI Algorithm of IE Looks more obedient to the above latest Unicode BIDI than that of Uniscrige API. That is, IE doesn't seem to use Uniscribe API. Therefore there may be differences in IE and XSL Formatter in BIDI Algorithm.

BTW, regarding this matter, we were not investigating accurately, it's only an Antenna House's prospect. And Windows Edit control shows the same result as XSL Formatter. I suppose it's because Edit control uses Uniscribe API.

Last of all, this test was confirmed by Windows XP Professional (SPI).

It's impossible in XSL-FO to change the flow of a part of hole pages in one document. （For example, the pages running from the left to the right are embedded in the pages running from the right to the left)

This problem was fixed with XSL Formatter V2.5 MR3.

When a Thai text is written directly not in fo:inline but in fo:block, the problem might be solved.

This error occurs when MSXML3 or MSXML4 is not installed on your PC. Please confirm if you have installed MSXML3 or MSXML4.

If you have already installed MSXML3 or MSXML4, please Re-install it or install IE6 because in some case installation has not completed. And try again.

FYI: The error message saying "MSXML is not installed" will be shown from V2.5.

There is an "Export INI file" function in XSL Formatter GUI. This function outputs each option set up in GUI with INI file format. This INI file can be handed over with a command line interface or COM interface without change.