#28237

Here you are faced with restrictions of PD4ML.

1. PD4ML offers only a lightweight support of

. It understands only text in it and ignores nested tag structures if any. You may apply text style and color to the legend text (addressing

tag), but background colors and borders are going to be ignored.

2. PD4ML allows to apply styles to . But

tags are not a part of internal document structure tree, so they are not visible to CSS selector runtime. As a workaround you would need to remove .form from the CSS selectors.

[language=css:lhug3mi0].main_body .form P, .main_body .form DIV.customControl
{
font-weight : bold;
color: #004677;
line-height: normal;
margin: 0px !IMPORTANT;
padding: 1px 3px 1px 3px !IMPORTANT;
vertical-align: top;
}[/language:lhug3mi0]

#28243

Any news on this or anymore information needed to solve?

#28244

Layout box is referring to this post by the way;

pdf-generation-troubleshooting-f4/is-it-possible-to-toggle-multicolumn-from-page-to-page-t255.html

#28240

I am afraid it is not going to be supported in the nearest future.

1. PD4ML uses a third-party CSS parser ss_css2.jar. We did minor patches of it, but a support of Chinese font names would require a significant refactoring.

2. PD4ML font subsystem currently works only with latin font names. We have plans to totally refactor it, but it will take some time until we begin.

#26469

After generating a pDF file (size 470 kB) the size of a document sent to a printer is greater than 70 MB. It cause capacity problems with network printer. Is there any way how to optimize the PDF for printing?

#28241

@PD4ML wrote:

I am afraid it is not going to be supported in the nearest future.

1. PD4ML uses a third-party CSS parser ss_css2.jar. We did minor patches of it, but a support of Chinese font names would require a significant refactoring.

2. PD4ML font subsystem currently works only with latin font names. We have plans to totally refactor it, but it will take some time until we begin.

FROM THE HOME PAGE:
“PD4ML Pro is Unicode© compatible to support most of Asian, Middle-East, European and Latin character sets; True Type and Open Type fonts. ”

you say that you just support Latin, but the home pages says different, its a little confusing, what do you support for fonts. arabic and japanese?
or is what you are saying onthe home page incorrect?

#28242

PD4ML does support Chinese, but it does not support chinese font names. For example it will work if you specify , but if you give chinese glyphs instead of “SimSun”, it will fail. The same is with Russian: – ok, – not.

#26470

I have been testing Pd4ml demo version on Linux with Java 1.6 installed. I am using PHP so the php wrapper is used.

Under the pd4ml_php_wrapper2probe folder, I have replaced the HTML content from xdemo.php just for testing. However, when I put a image tag with a very short source name (i.e. test.jpg), the JVM uses 100% of CPU and never finishes the PDF document (the size of the PDF is just keep increasing). At the end I have to kill the running process to get the server up on running again.

Could you explain why this is happening as we don’t have control on what our clients are going to put into the HTML?

#28238

The issue is not reproducible with actual buiulds of PD4ML. Please download v360fx1 or v361b2 and replace pd4ml(_demo).jar and ss_css2.jar.

#28253

The issue is not reproducible with actual builds of PD4ML. Please download v360fx1 or v361b2 and replace pd4ml(_demo).jar and ss_css2.jar.

We’ll update PHP wrapper distribution with the new JARs asap.

#28216

PD4ML API includes enableSmartTableBreaks() (formerly enableTableBreaks()) method. The method forces PD4ML to split tables protecting table rows content and to replicate table headers (if any) from page to page. Unfortunately there is no way to trigger the mode from HTML code yet.

Table header ia a row (or multiple rows) consists of

tags only.

In v361b2 the table break logic was significantly refactored. So it is recommended to use the build or a newer one.

#28196

The version implements the feature (v361b2) is already available for download.

#27970

Since v361b2 there 2 new API calls: getCache() and setCache(). The methods allow to save and reuse PD4ML cache instance for a case if its static hash tables are aggressively garbage collected by JVM.

Please avoid to use a single cache wrapper instance from a multiple threads simultaneously.

#28184

PD4ML table border rendering algorithms are significantly refactored by v361b2. The issue is solved.

#28044

“Warning: Invalid syntax in CSS selector:” may cause the entire CSS stylesheet interpretation to fail. Your CSS sample should not cause the exception, obviously it comas from another stylesheet used in your documents.

CSS parser, supplied with the most recent v361b2, addresses the issue.

PD4ML does not support paddings, given in percents. But for width it should work. Please test it with the most recent version.

[Author]

Posts