I am trying to use the python-epo-ops-client library to perform a search; however, for some searches (not all), I am getting a 404 response status.

Here is a minimal example of a situation where this error occurs:

import epo_ops

key = '****'
consumer_secret = '*******'

client = epo_ops.Client(key=key, secret=consumer_secret)  # Instantiate client

query = '(txt all "water" AND "derailleur" AND "reuse") AND (cl all "E03B")'

response = client.published_data_search(
    cql=query, 
    range_begin=1,
    range_end=100
)

I am using Python 3.9 and python_epo_ops_client version 4.0.0.

Does anyone know how to fix this or indicate what I am doing wrong?

Travis restricts PR access to secure environment variables. Let's make sure tests can be run using anonymous clients.

See https://docs.travis-ci.com/user/pull-requests#Security-Restrictions-when-testing-Pull-Requests

Dear George,

in the backlog file, I've picked up that item which raised my interest.

• Testing: Replace Apiary tests with monkeypatch? Or https://pypi.python.org/pypi/pytest-localserver.

Maybe you can quickly educate me what this Apiary API is used for?

• What does this API do?
• Is the test suite running against the Apiary API only? Why do I need an OPS account then?
• Where is the source code for the Apiary API, and how would one make changes to it?

Following up on that, I will be more than happy to help replacing that testing subsystem with something else more self-contained, if that makes sense and would actually be possible.

With kind regards,
Andreas.

Hi,

First, thank you for creating this package. I am a beginner when it comes to using API. I am trying to extract data but not receiving any text.

I ran this code:

mport epo_ops

anonymous_client = epo_ops.Client() # Instantiate a default client
response = anonymous_client.published_data( # Retrieve bibliography data
reference_type = 'publication', # publication, application, priority
input = epo_ops.models.Docdb('1000000', 'EP', 'A1'), # original, docdb, epodoc
endpoint = 'biblio', # optional, defaults to biblio in case of published_data
constituents = [] # optional, list of constituents
)

registered_client = epo_ops.RegisteredClient(key='abc', secret='xyz')
registered_client.access_token # To see the current token
response = registered_client.published_data(…)

Replaced the last line:

response = registered_client.published_data('publication', epo_ops.models.Docdb('1000000', 'EP', 'A1'), endpoint='biblio', constituents=None)

Output I got : <Response [200]>

How should I modify the code to retrieve text? I would appreciate your help.

Need to make sure package is compatible with Python 3.6

Dear @gsong,

first things first: Thanks again for your exceptional work on maintaining this library.

We just tried to integrate the most recent release python-epo-ops-client 3.1.1 into PatZilla as we are looking forward to the updates coming from #33 and #34.

However, we hit a speed bump we would like to share with you. When requesting family information, python_epo_ops_client-3.1.1 croaked like

  File "/Users/amo/dev/elmyra/sources/ip-navigator/patzilla/access/epo/ops/api.py", line 718, in ops_family_inpadoc
    response = ops.family(reference_type, ops_id, constituents=to_list(constituents))
  File "/Users/amo/dev/elmyra/sources/ip-navigator/.venv2/lib/python2.7/site-packages/python_epo_ops_client-3.1.1-py2.7.egg/epo_ops/api.py", line 50, in family
    return self._make_request(url, None, params=input.as_api_input(), use_get=True)
TypeError: _make_request() got an unexpected keyword argument 'use_get'

While we haven't investigated this more thoroughly, it looks like the problem is coming from within the module.

Thanks already for looking into this and with kind regards,
Andreas.

See https://packaging.python.org/guides/migrating-to-pypi-org/#uploading for more details.

I ran the following basic code for OPS authentication using my API keys. Code looks like below

import epo_ops

client = epo_ops.Client(key='Key1', secret='Seckey1') # Instantiate client
response = client.published_data( # Retrieve bibliography data
reference_type = 'publication', # publication, application, priority
input = epo_ops.models.Docdb('1000000', 'EP', 'A1'), # original, docdb, epodoc
endpoint = 'biblio', # optional, defaults to biblio in case of published_data
constituents = [] # optional, list of constituents
)

When I ran, it showed ImportError: No module named epo_ops. Fixes I have tried so far

1. pip install epo_ops
2. Downloaded zip file from GitHub and ran the setup file but still it did not recognized epo_ops.

First of all, great work you've done! Thanks a lot!

This is what I found in EPO FAQ:

By default, OPS displays only the first 25 results but if you want to see more, you can change the interval at the end of the URL ("&Range=xx-xx"). You will then be able to download batches of up to 100 results at a time:

"&Range=1-100"

However, OPS can only display a total of 2,000 results. If your list of results exceeds this number, you will need to limit your query to reduce the number of search results.

Also, I saw here that you explicitly put the range. Is it possible to change the range from 1 to 1000 as it's said in the FAQ?

Thank you.

Document the Client interface more clearly in the README.

From announcement [1]:

The planned schedule is as follows:

• Current: version 3.1 production environment – version 3.2 public beta
• January 2017: version 3.2 in full production – version 3.1 still available
• As from Q2 2017: discontinuation of version 3.1

@gsong, do you have any availability to coordinate this effort?

[1] http://forums.epo.org/open-patent-services-and-publication-server-web-service/topic5378.html

Hi,

Since last night, I'm having problems sending requests to EPO OPS.

For example, with python-epo-ops-client v2.3.0:

import epo_ops
import epo_ops.models
client = epo_ops.Client(key='X', secret='X')
inp = epo_ops.models.Epodoc('CA826085')
client.published_data(reference_type='publication', input=inp, endpoint='biblio')

And I receive a 500 error with the body:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
  <fault xmlns="http://ops.epo.org">
    <code>SERVER.NotSupported</code>
    <message>An unknown error occurred. Please contact Open Patent Services Team (http://forums.epo.org/open-patent-services-and-publication-server-web-service)</message>
   </fault>

It seems that yesterday a patch was released for OPS: https://forums.epo.org/maintenance-release-of-open-patent-services-ops-v-3-2-expected-soon-7224

I've noticed that this works fine:

curl -X POST --header "Authorization: Bearer XXX" --header "Content-Type: text/plain" -d "(CA826085)" "http://ops.epo.org/3.2/rest-services/published-data/publication/epodoc/biblio"

But it generates a 500 error if the Content-Type header is missing:

curl -X POST --header "Authorization: Bearer XXX" -d "(CA826085)" "http://ops.epo.org/3.2/rest-services/published-data/publication/epodoc/biblio"

In epo_ops/api.py:64 _post method, a default 'Content-type' headers needs to be provided:

    def _post(self, url, data, extra_headers=None, params=None):
        headers = {
            'Accept': self.accept_type,
            'Content-Type': 'text/plain'
        }
        headers.update(extra_headers or {})
        return self.request.post(
            url, data=data, headers=headers, params=params
        )

Hello,

I would like to know if any of you has encountered the same issue as me.
Actually I get the feeling that the result is different from what we get via Espacenet platform and from the API.
I don't have the same number of documents and the same relevant documents.

Thanks,

About

At #63 (review), we reported that the test case test_mock_quota_exceeded currently fails. The reason is:

requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://opsv31.docs.apiary.io//individual-per-hour-exceeded/publication/docdb/biblio

References

• Maybe related: GH-65

Looking here, I see the Published Claims Retrieval Service (POST)

with the prototype http://ops.epo.org/3.2/rest-services/published-data/{type}/{format}/claims

When I call:

    # Retrieve bibliography data
    response = client.published_data(
        reference_type="publication",
        input=epo_ops.models.Docdb('2024054343', 'WO', 'A2'),
        endpoint="claims",
    )

I get an exception:

404 Client Error: Not Found for url: https://ops.epo.org/3.2/rest-services/published-data/publication/docdb/claims
  File "/home/dan/Build/python-epo-ops-client/epo_ops/api.py", line 177, in _make_request
    response.raise_for_status()
  File "/home/dan/Build/python-epo-ops-client/epo_ops/api.py", line 219, in _service_request
    return self._make_request(url, data)
  File "/home/dan/Build/python-epo-ops-client/epo_ops/api.py", line 78, in published_data
    return self._service_request(
  File "/home/dan/Geromics/PatentBot2000/EPO/getting_started_with_epo_ops_and_pydantic.py", line 97, in test_published_data
    response = client.published_data(
  File "/home/dan/Geromics/PatentBot2000/EPO/getting_started_with_epo_ops_and_pydantic.py", line 164, in <module>
    data = test_published_data(client, test_docdb)
requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://ops.epo.org/3.2/rest-services/published-data/publication/docdb/claims
20220105

The url parameter seems to match the prototype... Not sure what I'm doing wrong

Dear George,

some users and the lovely people from @Patent2net reported that family responses from OPS stopped yielding bibliographic information for some weeks already. This broke some features of Patent2Net [2] as well as PatZilla [3].

As outlined on [1], it looks like only POST requests would fail while GET requests would still succeed.

With kind regards,
Andreas.

[1] https://meta.ip-tools.org/t/bibliographic-data-missing-from-ops-family-response/161
[2] https://github.com/Patent2net/P2N-v3
[3] https://github.com/ip-tools/ip-navigator

Hi I want to do a basic published data search request.

On my home PC it's working fine, but on my Laptop I get the

requests.exceptions.SSLError: HTTPSConnectionPool(host='ops.epo.org', port=443): Max retries exceeded with url: /3.2/auth/accesstoken (Caused by SSLError(SSLCertVerificationError(1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self signed certificate in certificate chain (_ssl.c:997)')))

error.

Code:

import epo_ops

client = epo_ops.Client(key='...', secret='...')  # Instantiate client
response = client.published_data_search('ti all "plastic"',1,25)

Is this a device configuration problem or some other library version problem?

Any ideas?

See #81 (comment) for more info.

We need to flesh out design details for this.

Use dotenv for setting local environments while testing. Make sure to gracefully fail for environments like Travis, where the envvar is injected by the testing runtime.

I ran the following basic code for OPS authentication using my API keys. Code looks like below

import epo_ops

client = epo_ops.Client(key='Key1', secret='Seckey1') # Instantiate client
response = client.published_data( # Retrieve bibliography data
reference_type = 'publication', # publication, application, priority
input = epo_ops.models.Docdb('1000000', 'EP', 'A1'), # original, docdb, epodoc
endpoint = 'biblio', # optional, defaults to biblio in case of published_data
constituents = [] # optional, list of constituents
)

When I ran, it showed ImportError: No module named epo_ops. Fixes I have tried so far

1. pip install epo_ops
2. Downloaded zip file from GitHub and ran the setup file but still it did not recognized epo_ops.

when running:
middlewares = [
epo_ops.middlewares.Dogpile(),
epo_ops.middlewares.Throttler(),
]
I get:
'epo_ops.middlewares' has no attribute 'Dogpile'

I checked the dir of epo_ops.middlewares
print(dir(epo_ops.middlewares))

result:
['Middleware', 'Throttler', 'builtins', 'cached', 'doc', 'file', 'loader', 'name', 'package', 'path', 'spec', 'middleware', 'throttle']

'Throttler' is there but not 'Dogpile'.

Is everything all right?

Issue:
Using client.image to fetch pages from EPO's API always returns the first page, regardless of the range parameter provided.

Steps to Reproduce:

1. Used client.image with range=page to fetch the nth page of a document.
2. Only the first page was returned in every attempt.

Workaround:
Manually constructing the URL and headers to specify the X-OPS-Range and making a direct request with client._make_request retrieves the correct page.

Expected:
client.image should interpret the range parameter to fetch the specified page number.

Actual:
Always retrieves the first page.

Details:

• Failing call: client.image("http://ops.epo.org/rest-services/published-data/images/EP/1000000/PA/fullimage", range=page, document_format="application/pdf")
• Successful workaround:

base_url = "http://ops.epo.org/rest-services/published-data/images"
publication_authority = "EP"
publication_number = "1000000"
publication_kind = "A1"
url = f"{base_url}/{publication_authority}/{publication_number}/{publication_kind}/fullimage"

# Make the request using client._make_request, specifying the Range through parameters and headers
image_response = client._make_request(
    url, 
    data="", 
    params={"Range": page}, 
    extra_headers={"Accept": "application/pdf", "X-OPS-Range": str(page)}, 
    use_get=True
)

(I found the above through trial and error, because I couldn't understand what the data parameter does here)

Hi there,

coming here from #35 (comment), @gsong was so kind to offer he would think about an option on how to toggle the behavior of one of the core methods within this library.

We are talking about the _make_request method here, which invokes response.raise_for_status() in order to raise exceptions on all HTTP requests to the OPS API which respond with HTTP status >= 400.

For reasons I am currently not able to remember exactly, we wanted to turn that behavior off within PatZilla.

Most probably it was because the exception bubbling up from this method lacked important information from the original request object we wanted to present to the user, like details from OAuth failures or rate limiting errors we are extracting from HTTP response headers or similar things.

Please advise if we are getting this wrong and the same things could be achieved with this call in place. Thanks!

With kind regards,
Andreas.

When using the library for an extended time we eventually got the following error:

• Your operating system name and version.
  • We got the error in Docker (Linux d65ebfed45f6 4.19.121-linuxkit x86_64 GNU/Linux) hosted on macOS Bug Sur (x84_64)
• Any details about your local setup that might be helpful in troubleshooting.
  • It is a very long running program, so it seems like a bug that only happens in rare cases
• Detailed steps to reproduce the bug.
  • Keep sending queries for ~20 hrs. It happens reliably for us but very rarely

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
...
~/.heroku/python/lib/python3.9/site-packages/epo_ops/api.py in published_data_search(self, cql, range_begin, range_end, constituents)
     90     ):
     91         range = dict(key="X-OPS-Range", begin=range_begin, end=range_end)
---> 92         return self._search_request(
     93             dict(
     94                 service=self.__published_data_search_path__, constituents=constituents
~/.heroku/python/lib/python3.9/site-packages/epo_ops/api.py in _search_request(self, info, cql, range)
    220     def _search_request(self, info, cql, range):
    221         url = self._make_request_url(info)
--> 222         return self._make_request(
    223             url, {"q": cql}, {range["key"]: "{begin}-{end}".format(**range)}
    224         )
~/.heroku/python/lib/python3.9/site-packages/epo_ops/api.py in _make_request(self, url, data, extra_headers, params, use_get)
    170             request_method = self.request.get
    171 
--> 172         response = request_method(url, data=data, headers=headers, params=params)
    173         response = self._check_for_expired_token(response)
    174         response = self._check_for_exceeded_quota(response)
~/.heroku/python/lib/python3.9/site-packages/epo_ops/models.py in post(self, url, data, **kwargs)
     84 
     85     def post(self, url, data=None, **kwargs):
---> 86         return self._request(_post_callback, url, data, **kwargs)
     87 
     88     def get(self, url, data=None, **kwargs):
~/.heroku/python/lib/python3.9/site-packages/epo_ops/models.py in _request(self, callback, url, data, **kwargs)
     93 
     94         for mw in self.middlewares:
---> 95             url, data, kwargs = mw.process_request(self.env, url, data, **kwargs)
     96 
     97         # Either get response from cache environment or request from upstream
~/.heroku/python/lib/python3.9/site-packages/epo_ops/middlewares/throttle/throttler.py in process_request(self, env, url, data, **kwargs)
     18         if not env["from-cache"]:
     19             service = service_for_url(url)
---> 20             time.sleep(self.history.delay_for(service))
     21         return url, data, kwargs
     22 
ValueError: sleep length must be non-negative

When I dig into the code it seems that it is in the calculation here epo_ops/middlewares/throttle/storages/sqlite.py#L123

I think it happens if the next run is in the past (aka. next_run < _now). It could probably be fixed by adding a check for that and returning 0

I have made a PR that should fix it, but maybe I'm missing something/not doing it the desired way: #49

Dear George,

first things first: Thanks for conceiving and maintaining this great library for accessing the EPO OPS service!

When working on the Patent2net project, we might have found a minor anomaly when using your library.

Anomaly

We found that the requests for register information might not get cached properly, while it works like a charm for published data search and family information requests.

How to reproduce?

For reproducing this, using the Patent2Net project is just three commands away (you might want to run this inside a virtualenv).

Prepare

Install Patent2Net:
pip install 'https://github.com/Patent2net/P2N/archive/3.0.0-dev5.tar.gz'

Configure (use your OPS credentials here ;]):
p2n ops init --key=ScirfedyifJiashwOckNoupNecpainLo --secret=degTefyekDevgew1

Run

p2n adhoc worldmap --expression='TA=lentille' --country-field='designated_states' --with-family --with-register

Observations

When running this a second time with a warm cache, you will recognize from the log output that all requests to published data search and family information are speedy as they don't hit the OPS API at all.

However, all requests for register information feel like they hit the OPS API each time again, which can be recognized by the significant visible delay between each request.

Conclusion

I tried to peek around the code a bit, but couldn't find anything reasonable what could cause this behavior. Maybe you can find some time to look into this issue?

Thank you very much in advance!

With kind regards,
Andreas.

See https://dustingram.com/articles/2018/03/16/markdown-descriptions-on-pypi for more info.

This will get rid of having to convert to rst!

It seems forks are not able to use the encrypted environment variables for testing.

Hello,
with the supplied example I always get the following errors:

Traceback (most recent call last):
  File "C:\Users\Philippe\Develop\Python\Divers\Patents\patents.py", line 17, in <module>
    constituents = []  # optional, list of constituents
  File "C:\Python27\lib\site-packages\epo_ops\api.py", line 92, in published_data
    constituents
  File "C:\Python27\lib\site-packages\epo_ops\api.py", line 85, in _service_request
    return self.make_request(url, input.as_api_input())
  File "C:\Python27\lib\site-packages\epo_ops\api.py", line 167, in make_request
    response.raise_for_status()
  File "C:\Python27\lib\site-packages\requests\models.py", line 795, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 404 Client Error: Not Found

I have Python 2.7 32 bits on Windows7 64 bits. Tried at work behind a proxy (looked ok : response was a valid 404...) and at home (no proxy). Tried as anonymous or registered user...
I traced in the code but couldn't find what's wrong...
Any idea ? Thanks !

Multiple document search is supported by EPO 3.2. But I don't see an interface for the same in the code. A constructor overloading in Docdb would help achieve the same in Python3. Here is an edited version of models.py that I am using to achieve the same.

import inspect
import types

class MultiMethod:
    '''
    Represents a single multimethod.
    '''
    def __init__(self, name):
        self._methods = {}
        self.__name__ = name

    def register(self, meth):
        '''
        Register a new method as a multimethod
        '''
        sig = inspect.signature(meth)

        # Build a type-signature from the method's annotations
        types = []
        for name, parm in sig.parameters.items():
            if name == 'self': 
                continue
            if parm.annotation is inspect.Parameter.empty:
                raise TypeError(
                    'Argument {} must be annotated with a type'.format(name)
                    )
            if not isinstance(parm.annotation, type):
                raise TypeError(
                    'Argument {} annotation must be a type'.format(name)
                    )
            if parm.default is not inspect.Parameter.empty:
                self._methods[tuple(types)] = meth
            types.append(parm.annotation)

        self._methods[tuple(types)] = meth

    def __call__(self, *args):
        '''
        Call a method based on type signature of the arguments
        '''
        types = tuple(type(arg) for arg in args[1:])
        meth = self._methods.get(types, None)
        if meth:
            return meth(*args)
        else:
            raise TypeError('No matching method for types {}'.format(types))
        
    def __get__(self, instance, cls):
        '''
        Descriptor method needed to make calls work in a class
        '''
        if instance is not None:
            return types.MethodType(self, instance)
        else:
            return self
    
class MultiDict(dict):
    '''
    Special dictionary to build multimethods in a metaclass
    '''
    def __setitem__(self, key, value):
        if key in self:
            # If key already exists, it must be a multimethod or callable
            current_value = self[key]
            if isinstance(current_value, MultiMethod):
                current_value.register(value)
            else:
                mvalue = MultiMethod(key)
                mvalue.register(current_value)
                mvalue.register(value)
                super().__setitem__(key, mvalue)
        else:
            super().__setitem__(key, value)

class MultipleMeta(type):
    '''
    Metaclass that allows multiple dispatch of methods
    '''
    def __new__(cls, clsname, bases, clsdict):
        return type.__new__(cls, clsname, bases, dict(clsdict))

    @classmethod
    def __prepare__(cls, clsname, bases):
        return MultiDict()

## Actual Docdb change
class Docdb(BaseInput, metaclass=MultipleMeta):
    def __init__(self, number:str, country_code:str, kind_code:str, date:str=None):
        if not all([country_code, kind_code]):
            raise MissingRequiredValue(
                'number, country_code, and kind_code must be present'
            )
        super(Docdb, self).__init__(number, country_code, kind_code, date)

    def __init__(self, data_points:list):
        self.data_points = []
        for dp in data_points:
            d = Docdb(dp[0], dp[1], dp[2])
            self.data_points.append(d)

    def as_api_input(self):
        print(self.__dict__.keys(), )
        if hasattr(self, 'data_points'):
            print(self.data_points)
            print("#" * 20)
            return '\n'.join([x.as_api_input() for x in self.data_points])
        return super(Docdb, self).as_api_input()

Dear George,

thank you for your trust in handing over the maintenance of the repository on behalf of GH-60. In order to closer approach the project and code base, I would like to spend some iterations on submitting a few maintenance patches, both modernizing the sandbox infrastructure, and the CI configuration for this repository.

While doing it, I would like to gradually work towards a modern Python project setup, ultimately using Ruff as a main linter, a pyproject.toml file for describing project metadata, poethepoet as task runner, versioningit for versioning, and GitHub Actions (GHA) for CI purposes, as I am doing on all our other newly conceived projects as well (see, for example, [1,2]), so I can re-use many parts from there already without much effort.

Alongside the same iterations of modernizing the repository, I will also add support for Python 3.10 and 3.11 to the CI configuration. As I do not expect any changes, I believe the functional side of the code base will not be touched at all. Along the way, the requirements-based dependency management will be dissolved, and, if you agree, I will also dissolve the usage of tox. I don't know much about it, and I would like to get rid of the maintenance burden.

The modernization process will happen gradually, and the Make-based sandbox will still be around for a while, so you can get gradually accustomed with the modernization process. It is important for me that you do not have the feeling to get lost in this process, still I would like to make reasonably progress on that, because regular work will also re-start soon on my end.

In this spirit, I am humbly asking about any concerns or objections you may have about this topic, how much you would like to still be in the boat to be concerned with all the details, or whether you trust me enough to welcome all the small decisions I will make on those next steps, and/or if you are giving a general chèque en blanc on this.

I hope you are appreciating the plan outlined in this post in general, otherwise please let me know so I can back off correspondingly.

With kind regards,
Andreas.

[1] https://github.com/panodata/aika
[2] https://github.com/daq-tools/lorrystream

I did not find "description" or "claims" under the table "Currently the Client knows how to issue request for the following services:".
Is it supported by this client yet? (I am not familiar with python...) thanks a lot.
what I have tried:

response = client.published_data(
    reference_type = 'publication', # publication, application, priority
    input = epo_ops.models.Epodoc('EP1000000'), # original, docdb, epodoc
    endpoint = 'description', # optional, defaults to biblio in case of published_data
    constituents = [] # optional, list of constituents
)

Please, add typehints, it will make the library way easier to use!

Tests are failing for Python 2.6 on Travis. Drop support or figure out what's wrong.

Hello!
I using the EPO OPS client but have some troubles with my request.
When I use the example request of the EPO OPS client everything seems to work fine, but when using my code i constantly get a 404 error.
HTTPError: 404 Client Error: Not Found for url: https://ops.epo.org/3.2/rest-services/register/application/epodoc/biblio
Seems like between /epodoc and /biblio the application number seems to be missing right?
This is my code:

#example
	client = epo_ops.Client(key='key', secret='secret')  # Instantiate client
        response = client.register(  # Retrieve bibliography data
		  		reference_type = 'application',  # publication, application, priority
		  		input = epo_ops.models.Epodoc('12345'),  # original, docdb, epodoc
		  		constituents = ['biblio']  # optional, list of constituents
				)

I tried to do i similar like in the example
input = epo_ops.models.Docdb('1000000', 'EP', 'A1'),
but using
input = epo_ops.models.Epodoc(appnum), instead, as i want to make a register request.

https://www.pydoc.io/pypi/python-epo-ops-client-2.3.0/autoapi/models/index.html says that models.Epodoc just needs a number as input.
Does anyone know what i misunderstood or what is going wrong?

Dear George,

first of all, I would like to thank you tremendously for conceiving and maintaining this package. I think it has a high value for open data and open access applications aiming to connect to EPO's OPS service to gather information.

At #59 (comment), you mentioned that you are not actively working on this project anymore, and that you would welcome anyone to take it over.

In this spirit, I am offering my support. A few references about our past work in this area are attached below. Let me know if this resonates with you. 🌻

With kind regards,
Andreas.

References

• https://github.com/marians/dwd-weather/issues/3
• xmpppy/xmpppy#39
• m0nhawk/grafana_api#88
• mqtt-tools/mqttwarn#389 (comment)
• wireviz/WireViz#316

Hello,

I encountered this new error.
Do you have any idea to resolve it?
FYI, I don't use any multithreading process neither sqlite in my code.

"Traceback (most recent call last):
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/middlewares/throttle/storages/sqlite.py", line 76, in prune
    self.db.execute(sql)
sqlite3.ProgrammingError: SQLite objects created in a thread can only be used in that same thread. The object was created in thread id 69052472574080 and this is thread id 69052152153856.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/api.py", line 97, in published_data_search
    range,
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/api.py", line 223, in _search_request
    url, {"q": cql}, {range["key"]: "{begin}-{end}".format(**range)}
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/api.py", line 172, in _make_request
    response = request_method(url, data=data, headers=headers, params=params)
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/models.py", line 86, in post
    return self._request(_post_callback, url, data, **kwargs)
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/models.py", line 95, in _request
    url, data, kwargs = mw.process_request(self.env, url, data, **kwargs)
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/middlewares/throttle/throttler.py", line 20, in process_request
    time.sleep(self.history.delay_for(service))
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/middlewares/throttle/storages/sqlite.py", line 110, in delay_for
    self.prune()
  File "/layers/google.python.pip/pip/lib/python3.7/site-packages/epo_ops/middlewares/throttle/storages/sqlite.py", line 76, in prune
    self.db.execute(sql)
sqlite3.ProgrammingError: SQLite objects created in a thread can only be used in that same thread. The object was created in thread id 69052472574080 and this is thread id 69052152153856."

When I try to query using Epodoc, I always get the same issue:

requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://ops.epo.org/3.2/rest-services/register/publication/epodoc/biblio

Is there something I am missing in the code?

data = 'EP07102956'
client = Client(key=USER, secret=PASS)  # Instantiate client
response = client.register(
        reference_type='publication',
        input=models.Epodoc(data),
  )
 print(response.text)