cutoff of implicit H atoms in JPEG

User 870ab5b546

08-06-2006 13:06:28

This molecule:

Code:

<?xml version="1.0" ?>

<MDocument>

<MChemicalStruct>

<molecule molID="m1">

<atomArray

atomID="a1 a2 a3"

elementType="N C R"

sgroupRef="0 0 sg1"

x2="-5.331760000000001 -3.7917600000000005 -2.25176"

y2="1.2250933333333336 1.2250933333333336 1.2250933333333336"

/>

<bondArray>

<bond atomRefs2="a2 a1" order="1" />

<bond atomRefs2="a2 a3" order="1" />

</bondArray>

<molecule id="sg1" role="SuperatomSgroup" title="CN" molID="m2">

<atomArray

atomID="a4 a5"

elementType="C N"

attachmentPoint="1 0"

x2="-2.25176 -1.4817600000000002"

y2="1.2250933333333336 -0.10864000000000001"

/>

<bondArray>

<bond atomRefs2="a4 a5" order="3" />

</bondArray>

</molecule>

</molecule>

</MChemicalStruct>

</MDocument>

produces the JPEG below with the molecule.toBinFormat() method. As you can see, the 2 is cut off.

User f359e526a1

09-06-2006 07:34:55

Thanks, I will test it in the new version - again. We had issues with indices in images hoping this had been solved since.

User f359e526a1

14-06-2006 08:41:15

Could you please send me the exact parameters of toBinFormat("jpeg") you are using? I hope I had fixed the thing but want to be certain it works with your exact version as well.

User 870ab5b546

14-06-2006 13:42:10

I was mistaken, we're using MolConverter for the image generation:

Code:

      if (mol.indexOf("Lewis ") > -1) { // LewisSketch

         String xml = MolToXML.getXML(mol);

         ip = new ByteArrayInputStream(xml.getBytes());

         convertParams = "jpeg:mono,H_off,wireframe,#ffffff";

      } else { // Marvin MOL

         ip = new ByteArrayInputStream(mol.getBytes());

         convertParams = "jpeg:mono,H_heteroterm,wireframe,#ffffff";

      }

      // compute a reasonable scaling factor

      MolImageSize molIimageSize = MolImporter.importMol(mol).

         getImageSize(convertParams + "scale" + initialScale);

      int myScale = (optWidth * initialScale) / molIimageSize.width;

      myScale = (myScale > maxScale) ? maxScale : myScale;

      DataOutputStream op =

         new DataOutputStream(new FileOutputStream(imgfile));

      MolConverter mcv = new MolConverter(ip, op, convertParams

         + "scale" + myScale, false);

User f359e526a1

28-06-2006 15:33:44

Seems it is OK by now - actually I found an other nasty bug with the help of your comment - thank you! The fix will appear in the shiny new release that is really not so far.

User 870ab5b546

30-06-2006 00:12:04

szilva wrote:

The fix will appear in the shiny new release that is really not so far.

Quite a number of fixes should appear in that shiny new release. I'm confident it will be bug-free! I eagerly await it.

User 870ab5b546

26-07-2006 01:39:27

I also find that when a jpeg is made of a document containing electron-flow arrows that are the nearest objects to the margin, the furthest loops of the electron-flow arrows are often cut off. See picture.

ChemAxon 7c2d26e5cf

26-07-2006 13:01:36

I have tested the development version of Marvin: I have drawn your structure, saved it then converted to jpeg. The generated image seems to be OK.