#29701

The error is caused by IOException thrown in

java.awt.Font.createFont( Font.TRUETYPE_FONT, fontBytes )

call and usually caused by an attempt to load a custom font, which is not fully compatible with Java’s font subsystem. Usually the standard arial.ttf (for example from our sample fonts.jar) causes no problem.

Can you reproduce the problem when run PD4ML in command line mode?

See http://pd4ml.com/html-to-pdf-command-line-tool.htm

#29709

By a font configuration procedure PD4ML scans a given font directory for .TTF, .TTC and .OTF font files.

#27010

Hi,
in our HTML we are including dynamic params (http://pd4ml.com/cookbook/pdf_dynamic_values.htm)

We need to include a breakline inside this dynamic values.
We tried with
but HTML value are escaped …

Is-there any way to include a breakline ?

Thanks

#29715

No, it is not possible due to the current PD4ML architecture specifics.

Obviously the only solution is to define a number placeholders:
$[line1]
$[line2]
and substitute them correspondingly to a needed number of lines

#29710

Thanks.

#29702

I have two tomcat instances on my windows machine and both run the same code and point to the same font directory. Both of these instances use the same Java runtime environment. I get this error with only one of my instances. (I can confirm that the configurations for both are absolutely similar)

Again, as I mentioned already, I did try the Allfonts.jar provided by pd4ml @ http://pd4ml.com/i/allfonts.zip which also generates the same error.

-Regards,
Manooj

#29716

This is not an option for us because we don’t know the number of lines when building the HTML template

I tried another way:

My HTML template contains

$[line1]

I tried to include breakline usign the java string : n
But this fails also...

Does this means that you are also escaping "n" or did I miss something ?

#29717

A placeholder (i.e. $[line1]) substitution is performed after HTML is parsed: that means a DOM-like structure is already build.

$[line1]

code results a creation of a block element contains a single content element.

$[line1]
$[line2]

results a block element with two content elements.

A placeholder substitution is applied to the content elements.

When line1=”text 1
text 2″ dynamic data comes, in theory it should place “text 1” instead of $[line1], delete the trailing content (if any), create a new content element object, assign it to the same parent block object, put “text 2” to it, append the trailing content, deleted before. The functionality is not foreseen and not implemented.

If the situation is critical and there is no feasible workaround – we could urgently implement that on a paid basis, but I need to estimate efforts with the developers.

#29703

Hmm… As it works on one of the machines, that is obviously not a problem of PD4ML, but a server configuration issue or a JVM installation corruption.

Could you please send us 2 PDF samples produced on the healthy and the problematic machines?

#29718

Yes please estimate the effort required.

To get back to my question on

: if I have

$[line1]

with

line1 = text 1ntext 2

then the HTML output should be

text 1
text 2

which is a way to display a breakline

but actually it is:

text 1 text 2

Why ? Did I miss something ?

#29704

Well, actually this is how it is,,,

Tomcat instance 1 : I set the font directory as c:windowsfonts to PD4ML and PDF is generated fine. Even the fonts appear fine.

Tomcat instance 2 : I do the same as above, but I get the error in the subject of this discussion.

Now,

Tomcat instance 1 & 2 : I try to load the font files using the jar files PD4ML has provided instead of pointing them to a font directory. I still see the same error.

So, I have a problem with jar way of loading fonts. I can’t seem to get a fonts jar that would work for me.

-Thanks,
Manooj

#29705

Most probably the issue with JAR has a different nature: for example the font .JAR is packed to .WAR, and the web application is not configured to be unpacked to a work directory. In the case PD4ML may have (not in every configuration) fail to load resources from .JAR

But the issue with c:windowsfonts tells that the Tomcat instance #2 has a configuration (or inconsistence) problem. I would start with a resolving the issue.

Enable debug info on both instances and compare PD4ML’s diagnostic output. Is it identical by identical PDF conversion requests (except the failure messages)?

Also requested PDF samples (from even blank HTML) would give us some additional info.

#29719

According to our moderate estimations an implementation of the feature (+ reduced QA pass) would require 5 m/hours which would cost 400 EUR

But… I’ve just tried line1 = text 1ntext 2 – and it works (correctly breaks the line) on my workstation even without

. It is not quite clear why it works and under which conditions - I am going to investigate.


BTW: please publish a Java/JSP/etc code snippet of the way you define line1 = text 1ntext 2

#29720

dynamicParametersMap.put(“report.generation”, “ericntestnbreakline”);

#29721

It is strange, your sample also works on my side. Which PD4ML version do you currently use?

I did some more testing – even if it works on my workstation, it still renders not correctly: it does not inform its block parent element, that the new content requires more vertical space. As a result we have a broken layout. See attached.

So a proper feature implementation is still actual.

[Author]

Posts