Trying to install PyXB on a production server and the installation procedure halts on compiling "...pyxb/bundles/common/raw/xhtml1.py to xhtml1.pyc". Tried installing with pip and with setup.py install. Here's the full output:
$ python setup.py installFound bundle in pyxb/bundles/common
Found bundle in pyxb/bundles/dc
Found bundle in pyxb/bundles/saml20
Found bundle in pyxb/bundles/wssplat
running install
running build
running build_py
running build_scripts
running install_lib
byte-compiling ...-venv/lib/python2.7/site-packages/pyxb/bundles/common/raw/xhtml1.py to xhtml1.pyc
Killed
I have PyXB on my desktop computer (Ubuntu 12.10) installed without any issue. Server runs Ubuntu 12.04.
I have installed all xml/html libs I could think off, but it's still the same.
Please help!
Last edit: Milan 2013-04-15
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
running python setup.py test shows like xml2 library is missing:
......................................WARNING:pyxb.binding.content:Multiple accepting paths for tData
WARNING:pyxb.binding.content:Multiple accepting paths for tData
..WARNING:pyxb.binding.basis:Unable to convert DOM node {http://www.w3.org/1999/xhtml}ul at <unknown>[6:49] to binding
..............WARNING:pyxb.binding.content:Incomplete content, expect kid per <unknown>[8:13]
..........................WARNING:pyxb.binding.basis:Unable to convert DOM node {http://www.w3.org/2001/10/xml-exc-c14n#}InclusiveNamespaces at <unknown>[10:10] to binding
.......WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testASCII_libxml2_file bypassed since libxml2 not present
.WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testASCII_libxml2_str bypassed since libxml2 not present
....WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testNihongo_libxml2_file bypassed since libxml2 not present
.WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testNihongo_libxml2_str bypassed since libxml2 not present
..............WARNING:pyxb.binding.content:Multiple accepting paths for tElt
...........WARNING:pyxb.binding.content:Incomplete content, expect {URN:test-trac-0057}assignedPriority per <unknown>[7:8]
WARNING:pyxb.binding.content:Incomplete content, expect {URN:test-trac-0057}assignedPriority per <unknown>[7:8]
...............................
----------------------------------------------------------------------
Ran 632 tests in 1.281s
</unknown></unknown></unknown></unknown></unknown>
I have installed libxml2, libxml2-dev, python-libxml2, libxslt-dev and probably a few more. I have also tried to copy the libxml2.so.2* files to /usr/lib, but still the same error.
Thanks a lot,
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I was able to solve this, but not in a way I wanted: I have copied the relevant files/folders from site-packages on my local machine to server and then reran the setup.py install and it worked. Pyxb is accessible now from the console and it works.
Still not sure though why regular install wasn't working.
Is there anything else I should make sure with this approach?
Last edit: Milan 2013-04-15
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It may be related to Python versions on Ubuntu 12.10, which I have not tried, or something about your local machine. Normally when a process is "killed" it has run out of memory or there's some other fatal error at a system level. Does your shell have a ulimit in effect? xhtml1 is probably one of the larger schemas in the base distribution, and it can take a lot of memory to generate bindings. dmesg may also be more informative.
libxml2 is not required for Python; the messages about it are merely warning you that certain tests won't work in its absence.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
@Peter, thanks a lot for the reply. Memory might be the issue. I have tried installing it on the Amazon EC2 Micro instance, with only 613MB of memory so that might have caused it to halt.
Wouldn't have thought of it.
If compiling wasn't working, does that mean that same can happen while during "normal" conversions?
Thanks a lot.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
613MB is pretty small these days. Whether you're likely to run into problems using the bindings depends on the size and complexity of the documents you need to manipulate. I suggest running a few experiments to see how it goes.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi everyone,
Trying to install PyXB on a production server and the installation procedure halts on compiling "...pyxb/bundles/common/raw/xhtml1.py to xhtml1.pyc". Tried installing with pip and with setup.py install. Here's the full output:
I have PyXB on my desktop computer (Ubuntu 12.10) installed without any issue. Server runs Ubuntu 12.04.
I have installed all xml/html libs I could think off, but it's still the same.
Please help!
Last edit: Milan 2013-04-15
Update:
running python setup.py test shows like xml2 library is missing:
......................................WARNING:pyxb.binding.content:Multiple accepting paths for tData WARNING:pyxb.binding.content:Multiple accepting paths for tData ..WARNING:pyxb.binding.basis:Unable to convert DOM node {http://www.w3.org/1999/xhtml}ul at <unknown>[6:49] to binding ..............WARNING:pyxb.binding.content:Incomplete content, expect kid per <unknown>[8:13] ..........................WARNING:pyxb.binding.basis:Unable to convert DOM node {http://www.w3.org/2001/10/xml-exc-c14n#}InclusiveNamespaces at <unknown>[10:10] to binding .......WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testASCII_libxml2_file bypassed since libxml2 not present .WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testASCII_libxml2_str bypassed since libxml2 not present ....WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testNihongo_libxml2_file bypassed since libxml2 not present .WARNING:test_trac_0139:tests/trac/test-trac-0139.py: testNihongo_libxml2_str bypassed since libxml2 not present ..............WARNING:pyxb.binding.content:Multiple accepting paths for tElt ...........WARNING:pyxb.binding.content:Incomplete content, expect {URN:test-trac-0057}assignedPriority per <unknown>[7:8] WARNING:pyxb.binding.content:Incomplete content, expect {URN:test-trac-0057}assignedPriority per <unknown>[7:8] ............................... ---------------------------------------------------------------------- Ran 632 tests in 1.281s </unknown></unknown></unknown></unknown></unknown>I have installed libxml2, libxml2-dev, python-libxml2, libxslt-dev and probably a few more. I have also tried to copy the libxml2.so.2* files to /usr/lib, but still the same error.
Thanks a lot,
I was able to solve this, but not in a way I wanted: I have copied the relevant files/folders from site-packages on my local machine to server and then reran the setup.py install and it worked. Pyxb is accessible now from the console and it works.
Still not sure though why regular install wasn't working.
Is there anything else I should make sure with this approach?
Last edit: Milan 2013-04-15
It may be related to Python versions on Ubuntu 12.10, which I have not tried, or something about your local machine. Normally when a process is "killed" it has run out of memory or there's some other fatal error at a system level. Does your shell have a ulimit in effect? xhtml1 is probably one of the larger schemas in the base distribution, and it can take a lot of memory to generate bindings. dmesg may also be more informative.
libxml2 is not required for Python; the messages about it are merely warning you that certain tests won't work in its absence.
@Peter, thanks a lot for the reply. Memory might be the issue. I have tried installing it on the Amazon EC2 Micro instance, with only 613MB of memory so that might have caused it to halt.
Wouldn't have thought of it.
If compiling wasn't working, does that mean that same can happen while during "normal" conversions?
Thanks a lot.
613MB is pretty small these days. Whether you're likely to run into problems using the bindings depends on the size and complexity of the documents you need to manipulate. I suggest running a few experiments to see how it goes.