Date   

Re: Ant and using xmlcatalog with xslt or xmltask #XSLT

Eric Sirois
 

Looks like this is not the first time folks have run into this issue.

Ant Trax cannot find DTD or read catalog in HTML plugin


I will try the solution that Ray used.  If anyone else has any suggestions, please feel free to reply.



On Thursday, April 23, 2020, 08:58:21 p.m. EDT, Eric Sirois <easirois@...> wrote:


Hi,

I want to run some processes in the depend.preprocess.pre stage of the OT.  So that means that the topics still have the doctypes in the documents.  The issue is that neither the <xmltask> nor the <xslt> process is able to properly find the catalog entry for the DTD (custom or OASIS). I am even trying to run things independently of the OT in oxygen, but just using the OT catalogs and it's still not working.

In oxygen I loaded the libraries saxon, xercesImpl xml-apis, xml-resolver, and ant-apache-resolver.  Am I missing some? Has anyone else been able to run an Ant task using the catalogs in a pre-process task. I have run things inside our publishing engine without issue, but it loads a ton of jars into memory but runs without any issues. I need it to run within an OT plugin.


Ant and using xmlcatalog with xslt or xmltask #XSLT

Eric Sirois
 

Hi,

I want to run some processes in the depend.preprocess.pre stage of the OT.  So that means that the topics still have the doctypes in the documents.  The issue is that neither the <xmltask> nor the <xslt> process is able to properly find the catalog entry for the DTD (custom or OASIS). I am even trying to run things independently of the OT in oxygen, but just using the OT catalogs and it's still not working.

In oxygen I loaded the libraries saxon, xercesImpl xml-apis, xml-resolver, and ant-apache-resolver.  Am I missing some? Has anyone else been able to run an Ant task using the catalogs in a pre-process task. I have run things inside our publishing engine without issue, but it loads a ton of jars into memory but runs without any issues. I need it to run within an OT plugin.


--args.output.base="%~ni"

scott ashmead
 

Hi All,

I'm using a DOS "for loop" to execute a DITA-OT command line.
The problem is that I need to change the output file name for each loop so that I don't overwrite the previous output file.

I'm trying to use --args.output.base="%~ni" but it has no effect. 
The output file is always named Global.xml because that's what the name of the ditamap is (Global.ditamap)

Is there another parameter I can/should use?

For more context, here's the complete command:

for %i in (c:\temp_test\*.xml) do (
dita -input C:\Global_Map\Global.ditamap -format resxml -output C:\curl_XML\ --args.filter="C:\ditaval\%~ni.ditaval" -DditaValFilePath="%~ni" --args.output.base=" %~ni")

Thank you,
Scott


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Chris Papademetrious
 

Hi Jang,

You're absolutely right, I need to be careful not to conflate the structured FrameMaker binary format (.fm) with the tool itself. Thanks for the reminder.

 - Chris


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Thomas Rathkamp
 

Chris, 

Jang makes an excellent point. My DITA experience started years ago while we were already using FrameMaker. I am unaware of Frame’s built-in support nowadays (I have since transitioned to Oxygen). But at the time, I discovered a wonderful Frame DITA plug-in called DITAFmx by Leximation. 

Not sure if I ever would have stuck with DITA at the time if I had not used DITA-FMx. 


Regards,

Tom


On Apr 19, 2020, at 10:02 AM, jang <jang@...> wrote:


Chris,

I want to make a small note on “transitioning from FrameMaker to DITA”.

You may be unaware of the DITA support in FrameMaker (which is hard to believe - FrameMaker has been supporting DITA for more than a decade and in fact Adobe is organising the biggest DITA conference on the planet since 3 years).. So it must be that you are not satisfied with the DITA support in FrameMaker. That is fine, as everyone is entitled to their own preferences.

But it would be nice if you could refrain from insinuating FrameMaker and DITA are two separate worlds. They are not. You can use DITA in FrameMaker at least as easily as in a number of other authoring tools. I have personally done at least 10 transitions from unstructured FrameMaker (or MS Word) to DITA without ever leaving FrameMaker behind, and all those customers are quite happy staying in (or moving to) an environment that does not require their authors to learn another style of working.

Kind regards

4everJang

Smart Information Design
Amsterdam, Netherlands
Cell: +31 646 854 996
On 19 Apr 2020, 16:21 +0200, Chris Papademetrious <chrispitude@...>, wrote:
Hi Jarno,

We are in the middle of transitioning from FrameMaker to DITA, but we are using both SVG and MathML and the capability you describe sounds like it could be useful.

Could this somehow enable profiling attributes (conditional graphical content) or key-based variables (e.g., product names) to be applied to external SVG files?

 - Chris


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

jang
 

Chris,

I want to make a small note on “transitioning from FrameMaker to DITA”.

You may be unaware of the DITA support in FrameMaker (which is hard to believe - FrameMaker has been supporting DITA for more than a decade and in fact Adobe is organising the biggest DITA conference on the planet since 3 years).. So it must be that you are not satisfied with the DITA support in FrameMaker. That is fine, as everyone is entitled to their own preferences.

But it would be nice if you could refrain from insinuating FrameMaker and DITA are two separate worlds. They are not. You can use DITA in FrameMaker at least as easily as in a number of other authoring tools. I have personally done at least 10 transitions from unstructured FrameMaker (or MS Word) to DITA without ever leaving FrameMaker behind, and all those customers are quite happy staying in (or moving to) an environment that does not require their authors to learn another style of working.

Kind regards

4everJang

Smart Information Design
Amsterdam, Netherlands
Cell: +31 646 854 996

On 19 Apr 2020, 16:21 +0200, Chris Papademetrious <chrispitude@...>, wrote:
Hi Jarno,

We are in the middle of transitioning from FrameMaker to DITA, but we are using both SVG and MathML and the capability you describe sounds like it could be useful.

Could this somehow enable profiling attributes (conditional graphical content) or key-based variables (e.g., product names) to be applied to external SVG files?

 - Chris


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Jarno Elovirta
 

Hi,

Not really, you could do profiling and indirect addressing without this. The problem with those is that they are not DITA content, thus those two features of DITA are not extended to SVG or MathML.

J

On 19. Apr 2020, at 17.21, Chris Papademetrious <chrispitude@...> wrote:

Hi Jarno,

We are in the middle of transitioning from FrameMaker to DITA, but we are using both SVG and MathML and the capability you describe sounds like it could be useful.

Could this somehow enable profiling attributes (conditional graphical content) or key-based variables (e.g., product names) to be applied to external SVG files?

 - Chris


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Chris Papademetrious
 

Hi Jarno,

We are in the middle of transitioning from FrameMaker to DITA, but we are using both SVG and MathML and the capability you describe sounds like it could be useful.

Could this somehow enable profiling attributes (conditional graphical content) or key-based variables (e.g., product names) to be applied to external SVG files?

 - Chris


Re: DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Radu Coravu
 

Hi Jarno,

One use case one of our customers had was that they had referenced external SVG images (and most people use external SVG images instead of internal ones) and those SVG images had Javascript code inside. In order for the HTML to run the Javascript the SVG images needed to be embedded inside the HTML, about a year ago I created a small plugin for that but probably it does not take into account references from SVGs to other images:

https://github.com/oxygenxml/dita-embed-html-mathml-svg

Regards,
Radu
Radu Coravu
Oxygen XML Editor
On 4/18/20 3:51 PM, Jarno Elovirta wrote:

Hi,

There's a DITA-OT feature I've been considering to implement for years now. What if there was a built-in preprocess step to extract inline SVG or MathML, or inline external SVG or MathML references? Would you as a DITA-OT plug-in developer find this useful?


Some of the use cases I've come up with include:
  • DITA source uses a mixture of inline and referenced MathML content. In transformation to Word OOXML, we need to inline all MathML when we convert it to OOXML equations. This would allow OOXML to not need to read MathML references during processing.
  • DITA source uses referenced SVG images. In processing, only inline SVG is read to discover bitmap images referenced by SVG. This would allow us to copy images referenced by SVG with default image handling.
  • DITA source uses referenced SVG images. In transforming to HTM5, we want to inline SVG into HTML5 so we can style it with CSS. If we handle inlining during preprocess, HTML5 plug-in doesn't have to deal with this.

So for DITA-OT users this will not be visible in any way, this is a question for plug-in developers. Would you find it a useful feature?

Jarno


  


DITA-fix-xbook-html-links - DITA cross-book links in HTML deliverables #HTML #linking

Chris Papademetrious
 

I have a basic publishing solution for DITA cross-book links in HTML-based outputs:

https://github.com/chrispy-snps/DITA-fix-xbook-html-links

The solution consists of a DITA plugin (com.synopsys.preserve.keyrefs) that preserves additional scoped-key reference and definition information in the HTML output, along with a perl post-processing script (fix_html_xbook_links.pl) that modifies the HTML files in-place with resolved references.

My approach is to convert cross-book links to pseudo-URL references during preprocess:

<xref href="keyref://book.keyname">...</xref>

This does two things:

  • It allows the information to survive publishing.
  • It ensures that publishing formats the reference as a link instead of an unresolved-link text span.

I tested the solution with the default DITA-OT html5 and xhtml transformations, and with Oxygen's WebHelp transformation. There are some limitations, described in the Git repo page above.

I'm also hoping to develop a solution for PDFs, but that will be a much harder problem.

Feedback is welcomed!

 - Chris


Re: Relationship tables in 3.4.1 #reltable #DITA-OT

Jarno Elovirta
 

See https://www.dita-ot.org/dev/parameters/parameters-base.html#base__args.rellinks

The args.rellinks parameter changes which related links are output. Where the related links come from, reltables or related-links, is not taken into consideration.

J


On Tue, Apr 14, 2020 at 10:02 AM, Briggs, John wrote:

for the 3.4.1 user guide "Note that links generated from a <reltable> in a DITA map will have the role attribute set to "friend". " but when I look at the files in the temp folder, not a single topic has role="friend".


DITA-OT feature to inline/extract SVG and MathML #DITA-OT #MathML #SVG

Jarno Elovirta
 

Hi,

There's a DITA-OT feature I've been considering to implement for years now. What if there was a built-in preprocess step to extract inline SVG or MathML, or inline external SVG or MathML references? Would you as a DITA-OT plug-in developer find this useful?


Some of the use cases I've come up with include:
  • DITA source uses a mixture of inline and referenced MathML content. In transformation to Word OOXML, we need to inline all MathML when we convert it to OOXML equations. This would allow OOXML to not need to read MathML references during processing.
  • DITA source uses referenced SVG images. In processing, only inline SVG is read to discover bitmap images referenced by SVG. This would allow us to copy images referenced by SVG with default image handling.
  • DITA source uses referenced SVG images. In transforming to HTM5, we want to inline SVG into HTML5 so we can style it with CSS. If we handle inlining during preprocess, HTML5 plug-in doesn't have to deal with this.

So for DITA-OT users this will not be visible in any way, this is a question for plug-in developers. Would you find it a useful feature?

Jarno


Re: Any Interest in a Hardware Domain for Non-computer Buttons/Controls? #DITAv2-0

David Yin
 

I am going to put a vote in that YES hardware controls need an element.  I would request a separate element to define a hardware control itself and possibly another mechanism (element or an attribute of an element) that identifies the control’s function/operation.  For the current landscape of our products, more than these two suggestions would be an overkill.

 

In the automotive industry, there are a proliferation of buttons/wheels being designed into the steering wheel.  And the current vocabulary doesn’t allow us to delineate those controls from the software UI controls.

 

David Yin

Technical Publication Manager  |  david.yin@...  |  940.591.4196

DRIVING UPTIME EVERY DAY logo w RED PB EMAIL SIG One Line SMALL 2

 


Re: Non file based content #DITA-OT

Jarno Elovirta
 

Hi

We don’t have an extension point to hook up code to do the retrieval, but if you can use the XML  catalog to rewrite your IDs into URLs, OT will handle the HTTPS traffic for you. Then, you need to add a custom parser to convert your REST output into DITA. See https://github.com/jelovirt/org.lwdita/blob/master/src/main/plugin.xml#L43 for example.

Cheers,

J

On 16. Apr 2020, at 20.57, Paul Danaher <danaher.paul@...> wrote:

We have maps and topics stored in a database which can be retrieved via a REST service we wrote. What I'd like to do is pass in a DitaMap id as an input to the DITA-OT and have a cusyom parse retrieve the map content and pass the content to the toolkit for parsing. Every topic in the map has an id also whose content can be retrieved via the REST service and the toolkit would need to call the parser to retrieve the topic content, etc. . Is there anyway of customising the toolkit to allow this type of behaviour ocustom plugin maybe)?  Note I have managed to get this working by editing the core code as a proof of concept which I know is a NO NO and this has the downside of preventing me to upgrade easily. Any thoughts. Thanks.

Paul


Re: Using SVG in DITA? #SVG

Dan Vint
 

I seem to be seeing that the height and width values are more important to getting the pdf size correct. If I enlarge those the image changes size. If I only change the values in the viewport or only have a viewport value than it shrinks



Sent from my Verizon, Samsung Galaxy smartphone


-------- Original message --------
From: dvint <dvint@...>
Date: 4/16/20 10:51 AM (GMT-08:00)
To: Mica Semrick <mica@...>, main@dita-users.groups.io
Cc: dvint@...
Subject: Re: [dita-users] Using SVG in DITA?

So the Lucidchart graphic had a height and width, the second file had a view box, but no height and width. These are generated files so I had hoped they would be doing the right thing.

Anyway I added the viewport to the Lucidchart image and removed the scalefit from the image tag. The image is too wide in the pdf. I reduced both the viewport and width with no change.



Sent from my Verizon, Samsung Galaxy smartphone


-------- Original message --------
From: Mica Semrick <mica@...>
Date: 4/16/20 10:07 AM (GMT-08:00)
To: main@dita-users.groups.io, Dan Vint <dvint@...>
Subject: Re: [dita-users] Using SVG in DITA?

Hey Dan,

Are you setting the viewbox or width and height on the root svg element? That'd be the proper way to size it:

<svg viewBox="0 0 120 120" width="120" height="120">...</svg>

On April 16, 2020 9:22:17 AM PDT, Dan Vint <dvint@...> wrote:
I'm trying to writeup guidelines for using graphics for our writers. I 
think I have a process identified for making sure that a PNG file will
reproduce properly in HTML and PDF output - trick is to make sure the
dpi is set that corresponds to the number of pixels in the image. Note
that I'm just using this with and <image> element and none of the SVG
specific markup.

I'm now trying to create the corresponding process for SVG images. Our
primary tool is Lucidchart for creating vector art. I've then been using
Inkscape to try and verify the results and loading directly into Chrome.

So in Lucid chart I have specified a 6x8in image and export that to SVG.
When I place that in a topic I observe the following:
- With no other attributes set, the HTML is ok but the PDF the image is
about 2x's too large and runs off the page
- If I use the scale or scalefit attribute the image is not changed in
the html, but it does work for the PDF

I have a different little icon, 1.5x1.5in, when I look at it in
Inkscape. I see the following:
- with no attributes set in HTML it fills out to be a full width object
instead of being small.

Other things I have tired include trying to work around limitations of
imagemaps and graphics. I have drawing a small flowchart and linked the
boxes to a variety of external webpages in Lucidchart. I observe:
- loading image in Chrome the links work perfectly
- Putting this same image in a topic and producing the HTML, the links
do not work. What really surprised me, was that same image in the PDF,
the links did work!

I've been googling around DITA and SVG and have not really found
anything addressing the results I have seen. Are people not using SVG in
DITA? In the past I avoided SVG just because I could never guarantee
that the image I exported was the image that would appear on the web. I
haven't tried to verify the stability of the format (our tools) this
time and went to testing more on how well it worked with DITA.

Has anyone been successful in making SVG graphics work consistently in
DITA? Do you have any guidelines you can share? In particular I'm
looking for:
- what elements in the svg actually control the page scaling such that
if those numbers are correct that the image appears properly without
setting attributes in the DITA markup
- do scale/scalefit work for you in HTML output? Either the OT is not
applying these values or something about the browser use is ignoring
those values in the HTML
- are you able to make links in SVG work? I did find some articles on
issues here. It seems that if you have links to other SVG pages that
those pages are not being copied to the output folder. Also saw
something that indicated there might be an issue with linking to topics
in the same publication. This is actually what I plan to do is link the
various process blocks to steps or sections in the same topic to connect
the workflow view to the specific steps required to accomplish the step.
- what tools are you using to create SVG graphics? I'm finding Inkscape
a little awkward to use, and doesn't seem to help. Lucidchart is ok but
I'm not sure if it is good at writing SVG results. I'm thinking about
trying Adobe illustrator to test the results - it seems like a
sledgehammer for our use. Anything else I should be looking at?

At this point I'm about to give up and just have the writers create a
PNG from Lucidchart.

..dan
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#45379): https://dita-users.groups.io/g/main/message/45379
Mute This Topic: https://groups.io/mt/73058001/2985651
Group Owner: main+owner@dita-users.groups.io
Unsubscribe: https://dita-users.groups.io/g/main/leave/defanged [mica@...]


Non file based content #DITA-OT

Paul Danaher
 

We have maps and topics stored in a database which can be retrieved via a REST service we wrote. What I'd like to do is pass in a DitaMap id as an input to the DITA-OT and have a cusyom parse retrieve the map content and pass the content to the toolkit for parsing. Every topic in the map has an id also whose content can be retrieved via the REST service and the toolkit would need to call the parser to retrieve the topic content, etc. . Is there anyway of customising the toolkit to allow this type of behaviour ocustom plugin maybe)?  Note I have managed to get this working by editing the core code as a proof of concept which I know is a NO NO and this has the downside of preventing me to upgrade easily. Any thoughts. Thanks.

Paul


Re: Using SVG in DITA? #SVG

Dan Vint
 

So the Lucidchart graphic had a height and width, the second file had a view box, but no height and width. These are generated files so I had hoped they would be doing the right thing.

Anyway I added the viewport to the Lucidchart image and removed the scalefit from the image tag. The image is too wide in the pdf. I reduced both the viewport and width with no change.



Sent from my Verizon, Samsung Galaxy smartphone


-------- Original message --------
From: Mica Semrick <mica@...>
Date: 4/16/20 10:07 AM (GMT-08:00)
To: main@dita-users.groups.io, Dan Vint <dvint@...>
Subject: Re: [dita-users] Using SVG in DITA?

Hey Dan,

Are you setting the viewbox or width and height on the root svg element? That'd be the proper way to size it:

<svg viewBox="0 0 120 120" width="120" height="120">...</svg>

On April 16, 2020 9:22:17 AM PDT, Dan Vint <dvint@...> wrote:
I'm trying to writeup guidelines for using graphics for our writers. I 
think I have a process identified for making sure that a PNG file will
reproduce properly in HTML and PDF output - trick is to make sure the
dpi is set that corresponds to the number of pixels in the image. Note
that I'm just using this with and <image> element and none of the SVG
specific markup.

I'm now trying to create the corresponding process for SVG images. Our
primary tool is Lucidchart for creating vector art. I've then been using
Inkscape to try and verify the results and loading directly into Chrome.

So in Lucid chart I have specified a 6x8in image and export that to SVG.
When I place that in a topic I observe the following:
- With no other attributes set, the HTML is ok but the PDF the image is
about 2x's too large and runs off the page
- If I use the scale or scalefit attribute the image is not changed in
the html, but it does work for the PDF

I have a different little icon, 1.5x1.5in, when I look at it in
Inkscape. I see the following:
- with no attributes set in HTML it fills out to be a full width object
instead of being small.

Other things I have tired include trying to work around limitations of
imagemaps and graphics. I have drawing a small flowchart and linked the
boxes to a variety of external webpages in Lucidchart. I observe:
- loading image in Chrome the links work perfectly
- Putting this same image in a topic and producing the HTML, the links
do not work. What really surprised me, was that same image in the PDF,
the links did work!

I've been googling around DITA and SVG and have not really found
anything addressing the results I have seen. Are people not using SVG in
DITA? In the past I avoided SVG just because I could never guarantee
that the image I exported was the image that would appear on the web. I
haven't tried to verify the stability of the format (our tools) this
time and went to testing more on how well it worked with DITA.

Has anyone been successful in making SVG graphics work consistently in
DITA? Do you have any guidelines you can share? In particular I'm
looking for:
- what elements in the svg actually control the page scaling such that
if those numbers are correct that the image appears properly without
setting attributes in the DITA markup
- do scale/scalefit work for you in HTML output? Either the OT is not
applying these values or something about the browser use is ignoring
those values in the HTML
- are you able to make links in SVG work? I did find some articles on
issues here. It seems that if you have links to other SVG pages that
those pages are not being copied to the output folder. Also saw
something that indicated there might be an issue with linking to topics
in the same publication. This is actually what I plan to do is link the
various process blocks to steps or sections in the same topic to connect
the workflow view to the specific steps required to accomplish the step.
- what tools are you using to create SVG graphics? I'm finding Inkscape
a little awkward to use, and doesn't seem to help. Lucidchart is ok but
I'm not sure if it is good at writing SVG results. I'm thinking about
trying Adobe illustrator to test the results - it seems like a
sledgehammer for our use. Anything else I should be looking at?

At this point I'm about to give up and just have the writers create a
PNG from Lucidchart.

..dan
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#45379): https://dita-users.groups.io/g/main/message/45379
Mute This Topic: https://groups.io/mt/73058001/2985651
Group Owner: main+owner@dita-users.groups.io
Unsubscribe: https://dita-users.groups.io/g/main/leave/defanged [mica@...]


Re: Any Interest in a Hardware Domain for Non-computer Buttons/Controls? #DITAv2-0

Tom Magliery
 

Here is an example usage that maybe can go into the next version of the spec.


Re: Using SVG in DITA? #SVG

Mica Semrick
 

Hey Dan,

Are you setting the viewbox or width and height on the root svg element? That'd be the proper way to size it:

<svg viewBox="0 0 120 120" width="120" height="120">...</svg>


On April 16, 2020 9:22:17 AM PDT, Dan Vint <dvint@...> wrote:
I'm trying to writeup guidelines for using graphics for our writers. I 
think I have a process identified for making sure that a PNG file will
reproduce properly in HTML and PDF output - trick is to make sure the
dpi is set that corresponds to the number of pixels in the image. Note
that I'm just using this with and <image> element and none of the SVG
specific markup.

I'm now trying to create the corresponding process for SVG images. Our
primary tool is Lucidchart for creating vector art. I've then been using
Inkscape to try and verify the results and loading directly into Chrome.

So in Lucid chart I have specified a 6x8in image and export that to SVG.
When I place that in a topic I observe the following:
- With no other attributes set, the HTML is ok but the PDF the image is
about 2x's too large and runs off the page
- If I use the scale or scalefit attribute the image is not changed in
the html, but it does work for the PDF

I have a different little icon, 1.5x1.5in, when I look at it in
Inkscape. I see the following:
- with no attributes set in HTML it fills out to be a full width object
instead of being small.

Other things I have tired include trying to work around limitations of
imagemaps and graphics. I have drawing a small flowchart and linked the
boxes to a variety of external webpages in Lucidchart. I observe:
- loading image in Chrome the links work perfectly
- Putting this same image in a topic and producing the HTML, the links
do not work. What really surprised me, was that same image in the PDF,
the links did work!

I've been googling around DITA and SVG and have not really found
anything addressing the results I have seen. Are people not using SVG in
DITA? In the past I avoided SVG just because I could never guarantee
that the image I exported was the image that would appear on the web. I
haven't tried to verify the stability of the format (our tools) this
time and went to testing more on how well it worked with DITA.

Has anyone been successful in making SVG graphics work consistently in
DITA? Do you have any guidelines you can share? In particular I'm
looking for:
- what elements in the svg actually control the page scaling such that
if those numbers are correct that the image appears properly without
setting attributes in the DITA markup
- do scale/scalefit work for you in HTML output? Either the OT is not
applying these values or something about the browser use is ignoring
those values in the HTML
- are you able to make links in SVG work? I did find some articles on
issues here. It seems that if you have links to other SVG pages that
those pages are not being copied to the output folder. Also saw
something that indicated there might be an issue with linking to topics
in the same publication. This is actually what I plan to do is link the
various process blocks to steps or sections in the same topic to connect
the workflow view to the specific steps required to accomplish the step.
- what tools are you using to create SVG graphics? I'm finding Inkscape
a little awkward to use, and doesn't seem to help. Lucidchart is ok but
I'm not sure if it is good at writing SVG results. I'm thinking about
trying Adobe illustrator to test the results - it seems like a
sledgehammer for our use. Anything else I should be looking at?

At this point I'm about to give up and just have the writers create a
PNG from Lucidchart.

..dan
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#45379): https://dita-users.groups.io/g/main/message/45379
Mute This Topic: https://groups.io/mt/73058001/2985651
Group Owner: main+owner@dita-users.groups.io
Unsubscribe: https://dita-users.groups.io/g/main/leave/defanged [mica@...]


Using SVG in DITA? #SVG

Dan Vint
 

I'm trying to writeup guidelines for using graphics for our writers. I think I have a process identified for making sure that a PNG file will reproduce properly in HTML and PDF output - trick is to make sure the dpi is set that corresponds to the number of pixels in the image. Note that I'm just using this with and <image> element and none of the SVG specific markup.

I'm now trying to create the corresponding process for SVG images. Our primary tool is Lucidchart for creating vector art. I've then been using Inkscape to try and verify the results and loading directly into Chrome.

So in Lucid chart I have specified a 6x8in image and export that to SVG. When I place that in a topic I observe the following:
- With no other attributes set, the HTML is ok but the PDF the image is about 2x's too large and runs off the page
- If I use the scale or scalefit attribute the image is not changed in the html, but it does work for the PDF

I have a different little icon, 1.5x1.5in, when I look at it in Inkscape. I see the following:
- with no attributes set in HTML it fills out to be a full width object instead of being small.

Other things I have tired include trying to work around limitations of imagemaps and graphics. I have drawing a small flowchart and linked the boxes to a variety of external webpages in Lucidchart. I observe:
- loading image in Chrome the links work perfectly
- Putting this same image in a topic and producing the HTML, the links do not work. What really surprised me, was that same image in the PDF, the links did work!

I've been googling around DITA and SVG and have not really found anything addressing the results I have seen. Are people not using SVG in DITA? In the past I avoided SVG just because I could never guarantee that the image I exported was the image that would appear on the web. I haven't tried to verify the stability of the format (our tools) this time and went to testing more on how well it worked with DITA.

Has anyone been successful in making SVG graphics work consistently in DITA? Do you have any guidelines you can share? In particular I'm looking for:
- what elements in the svg actually control the page scaling such that if those numbers are correct that the image appears properly without setting attributes in the DITA markup
- do scale/scalefit work for you in HTML output? Either the OT is not applying these values or something about the browser use is ignoring those values in the HTML
- are you able to make links in SVG work? I did find some articles on issues here. It seems that if you have links to other SVG pages that those pages are not being copied to the output folder. Also saw something that indicated there might be an issue with linking to topics in the same publication. This is actually what I plan to do is link the various process blocks to steps or sections in the same topic to connect the workflow view to the specific steps required to accomplish the step.
- what tools are you using to create SVG graphics? I'm finding Inkscape a little awkward to use, and doesn't seem to help. Lucidchart is ok but I'm not sure if it is good at writing SVG results. I'm thinking about trying Adobe illustrator to test the results - it seems like a sledgehammer for our use. Anything else I should be looking at?

At this point I'm about to give up and just have the writers create a PNG from Lucidchart.

..dan