Repository navigation
Conversation
|
Have tried this(https://github.com/pfernique/python-docx/tree/feature/add_svg_picture) with word 2013. |
|
Unfortunatly SVG works only on word >= 2016 (more details here) |
Thank you for the link. I think that is not very important which vector format to use. def _dimensions_from_stream(cls, stream):
#https://docs.microsoft.com/en-us/openspecs/windows_protocols/ms-wmf/828e1864-7fe7-42d8-ab0a-1de161b32f27
stream.seek(0)
META_PLACEABLE = stream.read()
print("META_PLACEABLE")
mkey = META_PLACEABLE[0:4] # Key (4 bytes): Identification value that indicates the presence of a placeable metafile header. This value MUST be 0x9AC6CDD7.
mhmw = META_PLACEABLE[4:6] # HWmf (2 bytes): The resource handle to the metafile, when the metafile is in memory. When the metafile is on disk, this field MUST contain 0x0000. This attribute of the metafile is specified in the Type field of the META_HEADER Record.
mbbox = META_PLACEABLE[6:14] # BoundingBox (8 bytes): The rectangle in the playback context (or simply the destination rectangle), measured in logical units, for displaying the metafile. The size of a logical unit is specified by the Inch field.
minch = META_PLACEABLE[14:16] # Inch (2 bytes): The number of logical units per inch used to represent the image. This value can be used to scale an image.
|
|
Merge at some point? |
|
Would like to see this merged. |
| """ | ||
| pic_id = 0 # Word doesn't seem to use this, but does not omit it | ||
| pic = CT_Picture.new(pic_id, filename, rId, cx, cy) | ||
| if filename.endswith('.svg'): |
There was a problem hiding this comment.
Unless I'm misreading this - this would only capture SVG is actually provided as a file (that has a filename go with it). There are many use cases where pictures are generated on the fly in in-memory buffers and therefore not provided as files. Any way to account for that?
(For example the PNG equivalent for matplotlib + docx would be something like:
import base64
import io
import docx
from docx.shared import Inches
import matplotlib.figure as figure
fig = figure.Figure()
ax = fig.subplots()
ax.plot([1, 2])
buf = io.BytesIO()
fig.savefig(buf, format='png')
document = docx.Document()
document.add_heading('Document Title', 0)
document.add_picture(buf, width=Inches(1.25))
document.save('demo.docx')
There was a problem hiding this comment.
That's true that this seems strange but when dispatch is done, a filename image.svg is given for a stream (must be the same for png). So even if you provide in-memory buffers, it's good.
There was a problem hiding this comment.
Didn't realize that - thank you for calling it out. I looked a bit further (which I should have from the beginning):filename is generated here in image.py.
|
Definitely looking forward to this. Word now supports SVG natively, i.e. images are imported as editable vector objects. It's converted to DrawingML in the docx file with a reference to the embedded SVG file: <asvg:svgBlip xmlns:asvg="http://schemas.microsoft.com/office/drawing/2016/SVG/main" r:embed="rId5"/>Then, in Word, it can be converted into an actual shape, and the SVG file is removed from the docx media to be replaced with the usual |
4762810 to
0a8e9c4
Compare
shape.pyandpkgwriter.pyare particularly ugly but it works with Word for Office 365 (16.0)