let's tell people to use bioconda or pip instead of easy_install.

Deprecation notice: https://setuptools.readthedocs.io/en/latest/easy_install.html

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

It's not possible to run a template modifying the value of an editable constraint.
If a template has an editable constraint with code A, using the method template.add_constraint (same constraint field, same constraint operator, different constraint value, code="A") will add a new constraint with code B, and when you execute template.results it will return an error (which is correct because it's not possible to add a new constraint in a template)

The InterMine python client allows users to create lists of things (e.g. lists of genes). If the list size is zero, the list should not be created.

See https://mybinder.org/v2/gh/intermine/intermine-ws-python-docs/master?filepath=09-tutorial.ipynb for more info on the process to create a list, or look at the full set of tutorials here: https://github.com/intermine/intermine-ws-python-docs.

If someone uses intermine-ws-python to try to create a list with 0 items, we should prevent this and give them a sensible clear error, e.g. "Lists must have one or more elements - the current list has 0".

At present we have the docs hosted via pydocs, which will be in html files in a separate folder named 'docs' after the merge of #15.

Originally, the documentation was hosted here but it hasn't been updated with the current registry update. This is because the documentation aspect of pypi is undergoing some changes and doesn't allow updation of documentation at present. Refer to pypi/warehouse#582 .

As soon as this gets fixed, pypi can again be used as a way to keep the documentation.

The registry methods should allow us to get a list of all intermines without requiring an organism name. Can we add this method? It would have been useful when we were working on a Galaxy tool for the registry.

:)

Our travis ci build is failing on the two pypy settings because the default version of pypy for trusty is too old - it's 2.8 and needs to be at least 2.9.x in order to install pandas correctly. See https://travis-ci.org/intermine/intermine-ws-python/jobs/580760283 for an example of the failure.

This is a link of different python and pypy versions: travis-ci/docs-travis-ci-com#1249 (comment)

To test this, you could enable travis ci for your fork of this repo, or create a WIP (work in progress) PR - either is fine.

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

From @rachellyne

I created  two lists in humanmine using  lm.create_list(file, list_type="Gene", name=listname, description=description) where file is the path to the file.   It created two empty lists that I can't delete - it says "The query is well formatted but you do not have access to the following mentioned lists: PL_GenomicsEngland_GenePanel:Stickler_syndrome".  

It seems like when applying multiple options to one constraint, they need not be all enclosed by quotes (i.e. outside the sq. brackets), as when I removed that the error stopped showing.

--------------------------------------------------------------------------- NameError Traceback (most recent call last) <ipython-input-4-ea8f991734d6> in <module> ----> 1 b.query_to_barchart_log('<query model="genomic" view="Gene.name Gene.symbol Gene.length" sortOrder="Gene.name ASC" ><constraint path="Gene.length" op="&lt;" value="450074" /><constraint path="Gene.name" op="=" value="translation initiation factor IF-2-like" /></query>','true') ~/virtualenv/python3.5.6/lib/python3.5/site-packages/intermine/bar_chart.py in query_to_barchart_log(xml, resp) 237 ax = y.plot(kind='bar') 238 ax.set_title(list[0][0]) --> 239 ax.set_xlabel(l[1]) 240 if resp == 'true': 241 ax.set_ylabel('log(' + l[2] + ')') NameError: name 'l' is not defined NameError: name 'l' is not defined

This would have helped us spot #40 earlier.

Right now adding python 3.7 is non-trivial - I tried a few different things and couldn't get a successful build. See the commits attached to #42 (which I closed for now as it wasn't playing nicely...).

Love the package! Noticed Intermine isn't compatible with Python 3.10 or 3.11 (same issue as in #96). Any chance of compatibility for both 3.11 and 3.10?
Thanks in advance.
Best Lucas

I changed my mind! See intermine/intermine#1676

To remind you, we have a parameter called xml that is for the InterMine query. However, we now accept JSON queries! I have changed the name of the parameter to be query instead.

However, this messes up your code as you are using the xml parameter. Instead, can you do this:

if version >= 27, then parameter = query
else parameter = xml

This will use xml for current mines, and switch to query for newer mines.

I know it's not very elegant!

We have the travis config already, you just have to enable Travis for this repo. (I think?)

People want to view the list contents. I don't see an easy way to do that.

The new docs look great, so thanks @nupurgunwant :) http://intermine.org/intermine-ws-python

.....buuuuttttt it would be nice to have some more descriptions for the registry methods - see http://intermine.org/intermine-ws-python/intermine.html#module-intermine.registry - they don't have in-code descriptions of what they do in a format that sphinx can read.

HTML documentation for our library can be generated using the pydoc -w command.

It would be nice to use travis or github actions to automatically generate docs and deploy them to the gh-pages branch of this repo.

This should only happen for the master branch.

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

we removed it in response to persistent build failures that were non-trivial to fix - see #53

Non-urgent task - I believe @asherpasha may already be looking at this :) 🚀

Right now when we return query results, they're returned as in iterator. If you wish to work with query results in a dataframe, you have to manually build the dataframe step by step in a for loop, which can be a pain.

We should add a backwards-compatible way of getting query results in a dataframe - so our old tutorials and code should still work, but perhaps add another argument or use different methods to fetch the results as a dataframe.

Presumably this will make us require pandas by default.

We have a tutorial here that shows how to handle intermine query data: https://mybinder.org/v2/gh/intermine/intermine-ws-python-docs/master?filepath=02-tutorial.ipynb and generally there is more info in out other tutorials as well: (scroll down and look at the list of tutorials in the readme https://github.com/intermine/intermine-ws-python-docs/ )

I don't see any specific guidance on using a specific python version but, using Python 3.10.8, things don't seem to work out of the box. Using a query I made from the MouseMine website (which seem to provide python 2-style print...), it has:

from intermine.webservice import Service

Which fails in its import block, I had to manually edit webservice.py to use "collections.abc" vs. "collections".

Not sure if developer intent is to use a specific python version, or if this should be fixed, or handled for different python versions.

Since we have the pep8speaks bot commenting on our issues, we ought to make it easy for people to format things in a way that makes pep8 happy. This project looks like it would do it: https://pypi.org/project/autopep8/

task

• add autopep8 to requirements.txt
• add instructions to the readme file telling people how to run it before making PRs.

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

Hi,

I would like to download the transcripts sequences from a list of genes on the PhytoMine?

would anyone give me some suggestions on how to fetch the sequence from server by python API?

My best,

Ming

We want to update the README to have instructions on how to install the client.

1. Link to bioconda docs
2. Show how to install via PyPi sudo easy_install intermine

Don't make people download directly. If they are at github, they'll know they can make a clone of the repo!

We have a method, getInfo, that fetches the version of a given InterMine from the registry here: https://github.com/intermine/intermine-ws-python/blob/dev/intermine/registry.py#L11

getInfo method only prints results to the screen. I would be nice to have a similar method called getVersion that returns the result (rather than printing) so it can be used programmatically, and without the extra bits printed to the screen.

Labelled as easy since much of this method can be copied from getInfo.

Ideally the return response should contain keys for the following three version options:

API Version: 25
Release Version: 45.1 2017 August
InterMine Version: 1.8.5

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

My code is like this:

from intermine.webservice import Service
service = Service("https://phytozome.jgi.doe.gov/phytomine/service")
query=service.new_query()
query.select("*")
query.add_constraint("Gene","LOOKUP","AT5G67180")

this code works fine. but if i wrote using wildcard,like,

query.add_constraint("Gene","LOOKUP","*5G67180")

the error comes up,

Traceback (most recent call last):
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/site-packages/intermine/results.py", line 640, in open
return urlopen(req)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 222, in urlopen
return opener.open(url, data, timeout)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 531, in open
response = meth(req, response)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 641, in http_response
'http', request, response, code, msg, hdrs)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 569, in error
return self._call_chain(*args)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 503, in _call_chain
result = func(*args)
File "/Users/myang/Ananconda/anaconda3/lib/python3.7/urllib/request.py", line 649, in http_error_default
raise HTTPError(req.full_url, code, msg, hdrs, fp)
urllib.error.HTTPError: HTTP Error 500: Internal Server Error

Could anyone help to fix it? thanks a lot.

There is no way to create a list using synonyms
The method create_list(content=genes,list_type="Gene",name="a random name") in the list_manager, only adds exact matches.
We should add an additional group 'SYNONYM' to the inut param add

Using the Python API with Python 3.10 leads to the following error

from intermine.webservice import Service
Traceback (most recent call last):
  File "/Users/jmarie/opt/anaconda3/envs/intermine-test/lib/python3.10/site-packages/intermine/webservice.py", line 9, in <module>
    from urlparse import urlparse
ModuleNotFoundError: No module named 'urlparse'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/jmarie/opt/anaconda3/envs/intermine-test/lib/python3.10/site-packages/intermine/webservice.py", line 16, in <module>
    from collections import MutableMapping as DictMixin
ImportError: cannot import name 'MutableMapping' from 'collections' (/Users/jmarie/opt/anaconda3/envs/intermine-test/lib/python3.10/collections/__init__.py)

No problem with Python version <3.10

Pep8speaks is a GitHub app that automatically review Python code style over Pull Requests.
The configuration is quite simple, and it can be used to review all the Python files.
Here is the repository: https://github.com/OrkoHunter/pep8speaks , for more information.

We have an intermine-bio package. Why?

https://github.com/intermine/intermine-ws-bio-python

I don't see a utility in keeping these separate. Just making another dependency. (@justinccdev may disagree?).

Let's merge the functionality into the main intermine package and delete this repo.

What does the bio package do? Here are the docs:

# Get all sequences for proteins on "h", "r", "eve", "bib" and "zen":

    from intermine.webservice import Service
    from interminebio import SequenceQuery

    s = Service("www.flymine.org/query")
    q = SequenceQuery(s, "Gene")

    syms = ["h", "r", "eve", "bib", "zen"] 

    print q.select_sequence("proteins").where(s.model.Gene.symbol == syms).fasta()

# Process the locations of these genes one at a time:

    for line in q.select_sequence("Gene").where(s.model.Gene.symbol == syms).bed():
        process(line)

Here are the end points it uses:

query/fasta
query/gff
query/ned

• region search (already available in main client?)

    LIST_PATH = "/regions/list"
    BED_PATH = "/regions/bed"
    FASTA_PATH = "/regions/fasta"
    GFF3_PATH = "/regions/gff3"

TODO

1. copy over init and iterators files
2. rename them something bio specific (I think we want a bio directory?)
3. write tutorial
4. test!

I was messing about with pandas and plotly for fun at a study group and I thought I'd pull in some familiar data into a data frame. I wanted to get the results in a raw-ish format, so I delved into the massive tutorial-in-comments in query.py and thought I'd found what I wanted: a nice way to get the results as a simple string rather than an iterator, based on this comment here:

         for row in query.result("string")
             print(row)

I think it was a trick, though! query.result("whatever") isn't a method and query.results("whatever") (note the additional s) IS a method but string isn't a valid arg.

Maybe we just need to remove that line ;)

We have pep8 speaks set up to check files that are edited in PRs - but we still have some files that aren't checked because they haven't been edited in a while.

Task: Apply pep8 formatting to all files in this repo. You probably want to try a command something like autopep8 . -r --in-place.

Note that this command might not fix all long lines - some might need to be done manually.

if you want to pick up this task

0. Take a look at the InterMine contributing guidelines
1. Comment on this issue stating that you intend to work on the task
2. When you're ready, add your work to the repo and create a pull request.

What to do if you need help

Mention @yochannah, tweet @yoyehudi, pop by to say hi on chat or if needed email [email protected]. Don't forget we're usually only available during uk office hours and will not be able to respond at other times :)

See comments here: #36

The testmodel has an idea of an extraValue. I forget what it is, company department maybe?

Executing service.get_template() in Python 3.7 results in RuntimeError: generator raised StopIteration. The example is below is when accessing FlyMine; however, I have seen the same thing happens for YeastMine.
Demonstrating the issue:
Go to here and scroll fown to the 'Tutorial 7: Templates' section. Under there click the launch binder badge and try running the first cell after the session spins up. The error will come up. as currently the Binder instances are defaulting to Python 3.7.

Example output from the demonstration:

I suspect it is related to the second item under 'Changes in Python Behavior' here as currently the Binder instances are defaulting to Python 3.7.. A possible fix to a related issue is shown here. It looks like incorporating that approach to line 477 of intermine/query.py may fix it.

There is a link on intermine.readthedocs.io to https://pythonhosted.org/intermine/ but this presents just several pages with empty frames.

Does the Python client have any way to create anonymous session tokens? It could be useful for demo code where I didn't want to expose the token associated with my account and I didn't need the list operations I was performing to persist, long-term.

I couldn't see one, and ended up writing something like this (it's probably awful, given my python skills... ;)

service_url = "http://www.humanmine.org/humanmine/"
token = requests.get(url=service_url + "/service/session")
token = token.json()["token"]
service = Service(service_url, token=token)

I had an issue created against my repo since to the naive outsider it appears to be the main one. Would you like to take ownership of it (after merging back your progress of course)?