Selected Text offset

[jykim]

Hi,

Is it possible to get the offset of currently selected text?
I see that there's way to get the selected text using -

Code: Select all

DoVerb("Documents[0].Pages[0].Text", "GetSelected", DataIn(1), DataOut, 0);

but, it only gives me the text.
Please help.

[Vasyl - PDF-XChange]

Hi,
No way for this in current realization.
But new method "GetSelectedRanges" will be added into the next release (2.42.5 build, coming soon) such as:

Code: Select all

DoVerb("Documents[#<docID>].Pages[0].Text", "GetSelectedRanges", dataIn, dataOut, 0);
// or
DoDocumentVerb(docID, "Pages[0].Text", "GetSelectedRanges", dataIn, dataOut, 0);
// we can obtain one(usually), two(and more) segment(s):
int firstChar1 = dataOut[0];
int countChars1 = dataOut[1];
int firstChar2 = dataOut[2];
int countChars2 = dataOut[3];
...

Best
Regards

[Vasyl - PDF-XChange]

Hi,
Additionally, new method will be added for obtain pages with selection, such as:

Code: Select all

DoDocumentVerb(docID, "Pages", "GetSelectedRanges", dataIn, dataOut, 0);
// we can obtain one(usually), two(and more) page-ranges(s):
int firstPageWithSel1 = dataOut[0];
int countPagesWithSel1 = dataOut[1];
int firstPageWithSel2 = dataOut[2];
int countPagesWithSel2 = dataOut[3];
...

Best
Regards

[jykim]

Vasyl,

Thanks very much for your reply. I have some follow-up questions -

1. In the sample code for the new method "GetSelectedRanges", what's the meaning of "countChars" (int countChars1 = dataOut[1]? Is it number of characters in the selection? or, is it number of characters within the page upto the selection?
2. How would the new method "GetSelectedRanges" be useful for my requirement? (to get the offset - character count upto the selection within the page)
3. With the viewer product, you can select some text and you can highlight it using a toolbar button. So, I was guessing, there should be a way to detect current text selection (as an object or some sort) in your internal viewer implementation. If that's true, would it be possible to make that API available in SDK?

Thanks for your help.

[jykim]

I had more questions for Vasyl,

1. In your sample code for the new method "GetSelectedRanges", what's the meaning of "countChars" (int countChars1 = dataOut[1]? Is it number of characters in the selection? or, is it number of characters within the page upto the selection?
2. How would the new method "GetSelectedRanges" be useful for my requirement? (to get the offset - character count upto the selection within the page)
3. With the viewer product, you can select some text and you can highlight it using a toolbar button. So, I was guessing, there should be a way to detect current text selection (as an object or some sort) in your internal viewer implementation. If that's true, would it be possible to make that API available in SDK?

I was thinking of something like these -

Code: Select all

DoVerb("Documents[0].Pages[0].Text.Selection.Offset", "get", DataIn, DataOut, 0);

or

Code: Select all

DoVerb("Documents[0].Pages[0].SelectedText.Offset", "get", DataIn, DataOut, 0);

[Vasyl - PDF-XChange]

Hi,

1. In the sample code for the new method "GetSelectedRanges", what's the meaning of "countChars" (int countChars1 = dataOut[1]? Is it number of characters in the selection? or, is it number of characters within the page upto the selection?
2. How would the new method "GetSelectedRanges" be useful for my requirement? (to get the offset - character count upto the selection within the page)

The text-selection can contains some(typically one) ranges, for example, you can select some text by mouse with pressed 'Alt' key (alternate text-selection mode).
Each text-selection range contains:
1. index of start-character(begin of range)
2. count of characters in this range.
the index of last character in each range can be calculated as: <index of start-character> + <count of characters> - 1.
So, to get selection offset on the page you can use fist number from obtained array of numbers from the dataOut.
For example some ranges: 1-1, 5-10, 100-1.

Code: Select all

3. With the viewer product, you can select some text and you can highlight it using a toolbar button. So, I was guessing, there should be a way to detect current text selection (as an object or some sort) in your internal viewer implementation. If that's true, would it be possible to make that API available in SDK?

1. In the new build you will be able to receive special events about selection changes (look to Reference->Named Items->Objects->Notifications->Selection in the new SDK Help).
2. For detect pages which contains selected text you will be able to use:

Code: Select all

DoDocumentVerb(docID, "Pages", "GetSelectedRanges", dataIn, dataOut, 0);

3. For obtain text-selection ranges for each page, call:

Code: Select all

DoDocumentVerb(docID, "Pages[<Index>].Text", "GetSelectedRanges", dataIn, dataOut, 0);

4. You may obtain text from each text-selection range by:

Code: Select all

DoDocumentVerb(docID, "Pages[<Index>].Text", "Get", dataIn(FirstCharIndex, CountOfChars), dataOut, 0);

5. You may highlight/underline/strikeout each text-selection range by:

Code: Select all

DoDocumentVerb(docID, "Pages[<Index>].Text", "Highlight", dataIn(FirstCharIndex, CountOfChars), dataOut, 0);
DoDocumentVerb(docID, "Pages[<Index>].Text", "Underline", dataIn(FirstCharIndex, CountOfChars), dataOut, 0);
DoDocumentVerb(docID, "Pages[<Index>].Text", "StrikeOut", dataIn(FirstCharIndex, CountOfChars), dataOut, 0);

And about:

Code: Select all

DoVerb("Documents[0].Pages[0].Text.Selection.Offset", "get", DataIn, DataOut, 0);
DoVerb("Documents[0].Pages[0].SelectedText.Offset", "get", DataIn, DataOut, 0);

-in the future we planning to realize special object named as 'Selection' with more features like as:

Code: Select all

DoDocumentVerb(docID, "Selection.Pages", "get", dataIn, dataOut, 0);
DoDocumentVerb(docID, "Selection.Annots", "get", dataIn, dataOut, 0);
DoDocumentVerb(docID, "Selection.Bookmarks", "get", dataIn, dataOut, 0);
DoDocumentVerb(docID, "Selection", "Delete", dataIn, dataOut, 0);
DoDocumentVerb(docID, "Selection", "Copy", dataIn, dataOut, 0);
DoDocumentVerb(docID, "Selection", "Clear", dataIn, dataOut, 0);
...

Thanks.

[jykim]

Thanks Vasyl. You're the best!
I'm looking forward to trying out the new version with new API.
Cheers!

[John - Tracker Supp]

[jykim]

Hi Vasyl,

I've downloaded the new version (v.2.042.6), but I can't find "GetSelectedRanges" in the manual document.
Is it included in the new version, and just missing from the manual document?
Or, is it not included in the new version?

[jykim]

Hi Vasyl,

Regarding the sample code you've provided above -

Code: Select all

DoVerb("Documents[#<docID>].Pages[0].Text", "GetSelectedRanges", dataIn, dataOut, 0);
// or
DoDocumentVerb(docID, "Pages[0].Text", "GetSelectedRanges", dataIn, dataOut, 0);
// we can obtain one(usually), two(and more) segment(s):
int firstChar1 = dataOut[0];
int countChars1 = dataOut[1];
int firstChar2 = dataOut[2];
int countChars2 = dataOut[3];
...

I guess that "dataOut" is a Variant that contains an array. If that's the case, I won't be able to implement such a thing in my Java/SWT ActiveX environment. Please see this thread - http://www.docu-track.com/forum3/viewto ... =36&t=7079.
So, I'm wondering if you could add new interfaces to support that. I'm thinking of something like this -

Code: Select all

Variant dataOut = new Variant(OS.GlobalAlloc(OS.GMEM_FIXED | OS.GMEM_ZEROINIT, Variant.sizeof), (short)(OLE.VT_BYREF | OLE.VT_DISPATCH));
DoDocumentVerb(docID, "Pages", "GetSelectedRanges", dataIn, dataOut, 0);

OleAutomation outputs = dataOut.getAutomation();

Variant firstPageWithSel1  = outputs.invoke(getDispId(outputs,"Get"), new Variant[]{new Variant(0)});
Variant countPagesWithSel1 = outputs.invoke(getDispId(outputs,"Get"), new Variant[]{new Variant(1)});
Variant firstPageWithSel1  = outputs.invoke(getDispId(outputs,"Get"), new Variant[]{new Variant(2)});
Variant countPagesWithSel2 = outputs.invoke(getDispId(outputs,"Get"), new Variant[]{new Variant(3)});

[Corwin - Tracker Sup]

In new build we will try to add possibility to get output values thru IPDFXCargs (our own implementation of SafeArray).

Thanks for patience.

[jykim]

Thanks Corwin. I'm looking forward to it.