potential issue with syntax in GDAL

[smmorey]

I'm currently trying to get some GeoClaw results up and running in Google Earth but am running into what I think are some Python 3.7 incompatibilities. When following the documented workflow (here: http://www.clawpack.org/googleearth_plotting.html), I get this error when testing the instalation of GDAL:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/home/smmorey/anaconda3/share/gdal/gdal_test.py", line 79, in <module>
    docfile.write(etree.tostring(etree.ElementTree(doc),pretty_print=True))
TypeError: write() argument must be str, not bytes

This is referencing the gdal_test.py file I downloaded from the above link. I'm pretty sure this is just a change in syntax python updates, but I'm can't figure out how to update this line appropriately.

This doesn't seem to be the biggest issue, as GDAL seems optional, but I thought I'd bring it up anyway.

[mandli]

I have seen this come up in other contexts as well, mostly plotting though. I think it may actually be a change in matplotlib though. Definitely deserves some attention to what is really going on though. @ donnacalhoun have you run into this yet?

[donnaaboise]

Thanks for catching this. I am updating the test now, but here is a quick fix you can try :

Change line 79 to add a “decode()” statement.

docfile.write(etree.tostring(etree.ElementTree(doc),pretty_print=True).decode())

Can you let me know whether this works or not?

Thanks,

Donna

---

Donna Calhoun Associate Professor Department of Mathematics, Mathematics Bldg 241A Boise State University 1910 University Drive Boise, ID 83725-1555 (208) 426-3386 http://math.boisestate.edu/~calhoun

On Oct 16, 2019, at 3:35 PM, susannah [email protected] wrote:

I'm currently trying to get some GeoClaw results up and running in Google Earth but am running into what I think are some Python 3.7 incompatibilities. When following the documented workflow (here: http://www.clawpack.org/googleearth_plotting.html http://www.clawpack.org/googleearth_plotting.html), I get this error when testing the instalation of GDAL:

Traceback (most recent call last): File "", line 1, in File "/home/smmorey/anaconda3/share/gdal/gdal_test.py", line 79, in docfile.write(etree.tostring(etree.ElementTree(doc),pretty_print=True)) TypeError: write() argument must be str, not bytes This is referencing the gdal_test.py file I downloaded from the above link. I'm pretty sure this is just a change in syntax python updates, but I'm can't figure out how to update this line appropriately.

This doesn't seem to be the biggest issue, as GDAL seems optional, but I thought I'd bring it up anyway.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/408?email_source=notifications&email_token=AAKXMQNLUG4UBWYUKTWVIXDQO6CI3A5CNFSM4JBRK5RKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HSJFRLA, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXMQN46XWQ4D6DBRO26ADQO6CI3ANCNFSM4JBRK5RA.

[smmorey]

That worked! Thanks, Donna!

[donnaaboise]

Thanks for confirming that.

You probably had to make other changes to the gdal_test.py example (e.g. the print statement).

Did you have to make any more substantial changes?

Thanks again,

Donna

---

Donna Calhoun Associate Professor Department of Mathematics, Mathematics Bldg 241A Boise State University 1910 University Drive Boise, ID 83725-1555 (208) 426-3386 http://math.boisestate.edu/~calhoun

On Oct 16, 2019, at 3:57 PM, susannah [email protected] wrote:

That worked! Thanks, Donna!

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/408?email_source=notifications&email_token=AAKXMQO7PLZDUNBAO2SMI7DQO6E3HA5CNFSM4JBRK5RKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBOCASI#issuecomment-542908489, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXMQLJOQYOO6VTXAYMSDLQO6E3HANCNFSM4JBRK5RA.

[mandli]

Does this work across versions of Python?

[donnaaboise]

Not sure about that, but I know that it was required when converting to Python 3.x

I can try it in Python 2.x, once I get my version of gdal working.

---

Donna Calhoun Associate Professor Department of Mathematics, Mathematics Bldg 241A Boise State University 1910 University Drive Boise, ID 83725-1555 (208) 426-3386 http://math.boisestate.edu/~calhoun

On Oct 16, 2019, at 4:02 PM, Kyle Mandli [email protected] wrote:

Does this work across versions of Python?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/408?email_source=notifications&email_token=AAKXMQPCP3MKHVMJCQVKRPDQO6FQHA5CNFSM4JBRK5RKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBOCO6Q#issuecomment-542910330, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXMQP7GCF6VNY7J3NMWZLQO6FQHANCNFSM4JBRK5RA.

[smmorey]

I had to change the print statements in gdal_test.py and also a couple of print statements in factory.py within pykml, but nothing major besides that.

[smmorey]

It ran fine for me without the updated line in Python 2.7.

[donnaaboise]

Right - the updated line wasn’t needed in Python 2.7. But does the updated line work in 2.7? (I think it should; the .decode() statement was probably just assumed in 2.7)

---

Donna Calhoun Associate Professor Department of Mathematics, Mathematics Bldg 241A Boise State University 1910 University Drive Boise, ID 83725-1555 (208) 426-3386 http://math.boisestate.edu/~calhoun

On Oct 16, 2019, at 4:06 PM, susannah [email protected] wrote:

It ran fine for me without the updated line in Python 2.7.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/clawpack/geoclaw/issues/408?email_source=notifications&email_token=AAKXMQPLHFKIIA2KFTWAWGTQO6F6DA5CNFSM4JBRK5RKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBOCXWA#issuecomment-542911448, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKXMQMUOC2QBNTGHZ34SCTQO6F6DANCNFSM4JBRK5RA.

[mandli]

I am not sure but I think your assumption is probably correct. At the very least we should test this and if there is a difference then fence it off with a check on the version of Python if need be.