Combining two unrelated topics in one message:
1) I am seeing a bug under macOS in the export snapshot (of a player
object) command. This is under LC 9.6.7 (and I think 9.6.8). We have a
routine that takes a list of time codes in a movie, loads the movie into
a player object, set the currentTime to each time code, uses export
snapshot of the player to grab the displayed frame, and returns an array
of the images.
The possible bug is that occasionally one of the images returned will be
rotated 90 degrees. We have NO code that rotates the image and the
movie, from start to finish, does not not change the camera orientation.
I have not had the time/bandwidth to develop a reproducible recipe or
test stack (yet). My question is: Has anyone seen anything like this? Is
it a known bug?
2) Unrelated to the above, we also have a routine that takes a list of
PDF image coordinates (5 number, the page number, and a rect on the page
in PDF coordinates), loads the PDF using the XPDF EXTERNAL (XPDF-1.0.45,
NOT the widget), goes to the page, and grabs the image, and returns an
array of the images.
The confirmed bug, is that the first time the XPDF Viewer is opened (via
XPDFViewer_Open and then XPDFViewer_Set tName,"FILENAME",tFile to load
the PDF file), the XPDF calls to grab the image:
put empty into tImageData
XPDFViewer_SelectImage tName,(item 1 of tRef),(item 2 of
tRef),(item 3 of tRef),(item 4 of tRef),(item 5 of tRef)
XPDFViewer_GetSelectionImage tName, "tImageData"
put the result into tSize
set the width of img "imageConvert" of stack "libHrAPI"
to abs(item 1 of tSize)
set the height of img "imageConvert" of stack "libHrAPI"
to abs(item 2 of tSize)
set the imageData of img "imageConvert" of stack "libHrAPI"
to tImageData
set the topLeft of img "imageConvert" of stack "libHrAPI"
to (0,0)
return garbage in tImageData. The tize (tSize) is correct, but the image
is not.
This has a WORK-AROUND as this only occurs the first time the XPDF
viewer is opened. So, if I am asking for images, I have a conditional
that closes the XPDF External viewer (via XPDFViewer_Close tName) and
then opens the PDF again behind the scenes. This causes all the returned
PDF images to be correct for the same code.
Again, I have not had the time/bandwidth to develop a reproducible
recipe or test stack (yet). My question is: Has anyone seen anything
like this one? Is it also a known bug?
Thank you in advance for any info of whether you have seen these
possible bugs or know if the are related to an existing issue
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
