pylint

.pylintrc

@@ -1,6 +1,6 @@
 [FORMAT]
 max-line-length=88
-good-names=i,x1,x2,y1,y2
+good-names=i,x1,x2,y1,y2,id
 
 [MESSAGES CONTROL]
 disable=missing-docstring,too-many-ancestors,too-few-public-methods,no-else-return,no-member,attribute-defined-outside-init,similarities,import-outside-toplevel,cyclic-import,no-member,no-else-raise,too-many-instance-attributes,too-many-arguments,ungrouped-imports,useless-object-inheritance,no-else-continue

documentcloud/addon.py

@@ -24,7 +24,7 @@ class BaseAddOn:
 
     def __init__(self):
         args = self._parse_arguments()
-        client = self._create_client(args)
+        self._create_client(args)
 
         # a unique identifier for this run
         self.id = args.pop("id", None)
@@ -65,7 +65,7 @@ def _create_client(self, args):
             self.client.refresh_token = args["refresh_token"]
         if args["token"] is not None:
             self.client.session.headers.update(
-                {"Authorization": "Bearer {}".format(args["token"])}
+                {"Authorization": f"Bearer {args['token']}"}
             )
 
         # custom user agent for AddOns
@@ -119,7 +119,7 @@ def _parse_arguments(self):
 
         # validate parameter data
         try:
-            with open("config.yaml") as config:
+            with open("config.yaml", encoding="utf-8") as config:
                 schema = yaml.safe_load(config)
                 args["data"] = fastjsonschema.validate(schema, args["data"])
                 # add title in case the add-on wants to reference its own title
@@ -207,6 +207,8 @@ def get_document_count(self):
             documents = self.client.documents.search(self.query)
             return documents.count
 
+        return 0
+
     def get_documents(self):
         """Get documents from either selected or queried documents"""
         if self.documents:

documentcloud/base.py

@@ -98,9 +98,7 @@ def get(self, id_, expand=None):
             params = {"expand": ",".join(expand)}
         else:
             params = {}
-        response = self.client.get(
-            "{}/{}/".format(self.api_path, get_id(id_)), params=params
-        )
+        response = self.client.get(f"{self.api_path}/{get_id(id_)}/", params=params)
         # pylint: disable=not-callable
         return self.resource(self.client, response.json())
 
@@ -120,7 +118,7 @@ class ChildAPIClient(BaseAPIClient):
     """Base client for sub resources"""
 
     def __init__(self, client, parent):
-        super(ChildAPIClient, self).__init__(client)
+        super().__init__(client)
         self.parent = parent
 
     def list(self, **params):
@@ -169,7 +167,7 @@ def delete(self):
 
 class APISet(list):
     def __init__(self, iterable, resource):
-        super(APISet, self).__init__(iterable)
+        super().__init__(iterable)
         self.resource = resource
         if not all(isinstance(obj, self.resource) for obj in self):
             raise TypeError(
@@ -191,7 +189,7 @@ def append(self, obj):
             raise DuplicateObjectError(
                 f"Object with ID {obj.id} appears in the list more than once"
             )
-        super(APISet, self).append(copy(obj))
+        super().append(copy(obj))
 
     def add(self, obj):
         if not isinstance(obj, self.resource):
@@ -200,7 +198,7 @@ def add(self, obj):
             )
         # skip duplicates silently
         if obj.id not in [i.id for i in self]:
-            super(APISet, self).append(copy(obj))
+            super().append(copy(obj))
 
     def extend(self, list_):
         if not all(isinstance(obj, self.resource) for obj in list_):
@@ -213,4 +211,4 @@ def extend(self, list_):
                 raise DuplicateObjectError(
                     f"Object with ID {id_} appears in the list more than once"
                 )
-        super(APISet, self).extend(copy(obj) for obj in list_)
+        super().extend(copy(obj) for obj in list_)

documentcloud/client.py

@@ -169,6 +169,6 @@ def raise_for_status(self, response):
             response.raise_for_status()
         except requests.exceptions.RequestException as exc:
             if exc.response.status_code == 404:
-                raise DoesNotExistError(response=exc.response)
+                raise DoesNotExistError(response=exc.response) from exc
             else:
-                raise APIError(response=exc.response)
+                raise APIError(response=exc.response) from exc

documentcloud/documents.py

@@ -3,11 +3,11 @@
 """
 
 # Standard Library
+import datetime
 import logging
 import os
 import re
 import warnings
-import datetime
 from functools import partial
 
 # Third Party
@@ -62,7 +62,7 @@ def __init__(self, client, dict_):
                 dict_[f"_{name}"] = None
                 dict_[f"{name}_id"] = value
 
-        super(Document, self).__init__(client, dict_)
+        super().__init__(client, dict_)
 
         self.sections = SectionClient(client, self)
         self.annotations = AnnotationClient(client, self)
@@ -187,7 +187,10 @@ def get_page_text_url(self, page=1):
         return f"{self.asset_url}documents/{self.id}/pages/{self.slug}-p{page}.txt"
 
     def get_page_position_json_url(self, page=1):
-        return f"{self.asset_url}documents/{self.id}/pages/{self.slug}-p{page}.position.json"
+        return (
+            f"{self.asset_url}documents/{self.id}/pages/"
+            f"{self.slug}-p{page}.position.json"
+        )
 
     def get_json_text_url(self):
         return f"{self.asset_url}documents/{self.id}/{self.slug}.txt.json"
@@ -265,6 +268,16 @@ def list(self, **params):
 
     def upload(self, pdf, **kwargs):
         """Upload a document"""
+
+        def check_size(size):
+            # DocumentCloud's size limit is set to 501MB to give people a little leeway
+            # for OS rounding
+            if size >= 501 * 1024 * 1024:
+                raise ValueError(
+                    "The pdf you have submitted is over the DocumentCloud API's 500MB "
+                    "file size limit. Split it into smaller pieces and try again."
+                )
+
         # if they pass in a URL, use the URL upload flow
         if is_url(pdf):
             return self._upload_url(pdf, **kwargs)
@@ -275,19 +288,13 @@ def upload(self, pdf, **kwargs):
                 size = os.fstat(pdf.fileno()).st_size
             except (AttributeError, OSError):  # pragma: no cover
                 size = 0
+            check_size(size)
+            return self._upload_file(pdf, **kwargs)
         else:
             size = os.path.getsize(pdf)
-            pdf = open(pdf, "rb")
-
-        # DocumentCloud's size limit is set to 501MB to give people a little leeway
-        # for OS rounding
-        if size >= 501 * 1024 * 1024:
-            raise ValueError(
-                "The pdf you have submitted is over the DocumentCloud API's 500MB "
-                "file size limit. Split it into smaller pieces and try again."
-            )
-
-        return self._upload_file(pdf, **kwargs)
+            check_size(size)
+            with open(pdf, "rb") as pdf_file:
+                return self._upload_file(pdf_file, **kwargs)
 
     def _format_upload_parameters(self, name, **kwargs):
         """Prepare upload parameters from kwargs"""
@@ -371,11 +378,13 @@ def _collect_files(self, path, extensions):
 
     def upload_directory(self, path, handle_errors=False, extensions=".pdf", **kwargs):
         """Upload files with specified extensions in a directory"""
+        # pylint: disable=too-many-locals, too-many-branches
 
         # Do not set the same title for all documents
         kwargs.pop("title", None)
 
-        # If extensions are specified as None, it will check for all supported filetypes.
+        # If extensions are specified as None, it will check for all supported
+        # filetypes.
         if extensions is None:
             extensions = SUPPORTED_EXTENSIONS
 
@@ -444,9 +453,8 @@ def upload_directory(self, path, handle_errors=False, extensions=".pdf", **kwarg
             for url, file_path in zip(presigned_urls, file_paths):
                 logger.info(f"Uploading {file_path} to S3...")
                 try:
-                    response = requests_retry_session().put(
-                        url, data=open(file_path, "rb").read()
-                    )
+                    with open(file_path, "rb") as file:
+                        response = requests_retry_session().put(url, data=file.read())
                     self.client.raise_for_status(response)
                 except (APIError, RequestException) as exc:
                     if handle_errors:

documentcloud/exceptions.py

@@ -16,7 +16,7 @@ def __init__(self, *args, **kwargs):
         else:
             self.error = None
             self.status_code = None
-        super(DocumentCloudError, self).__init__(*args, **kwargs)
+        super().__init__(*args, **kwargs)
 
 
 class DuplicateObjectError(DocumentCloudError):

documentcloud/projects.py

@@ -14,7 +14,7 @@ class Project(BaseAPIObject):
 
     def __init__(self, *args, **kwargs):
         per_page = kwargs.pop("per_page", PER_PAGE_MAX)
-        super(Project, self).__init__(*args, **kwargs)
+        super().__init__(*args, **kwargs)
         self._document_list = None
         self._per_page = per_page
 
@@ -23,7 +23,7 @@ def __str__(self):
 
     def save(self):
         """Add the documents to the project as well"""
-        super(Project, self).save()
+        super().save()
         if self._document_list:
             self.clear_documents()
             self.add_documents(self._document_list)
@@ -100,7 +100,7 @@ def all(self, **params):
         return self.list(user=self.client.user_id, **params)
 
     def get(self, id=None, title=None):
-        # pylint:disable=redefined-builtin, arguments-differ
+        # pylint:disable=redefined-builtin, arguments-renamed
         # pylint disables are necessary for backward compatibility
         if id is not None and title is not None:
             raise ValueError(
@@ -115,7 +115,7 @@ def get(self, id=None, title=None):
             return self.get_by_title(title)
 
     def get_by_id(self, id_):
-        return super(ProjectClient, self).get(id_)
+        return super().get(id_)
 
     def get_by_title(self, title):
         response = self.client.get(

documentcloud/toolbox.py

@@ -2,12 +2,14 @@
 A few toys the API will use.
 """
 
+# Standard Library
+from itertools import zip_longest
+from urllib.parse import urlparse
+
 # Third Party
 import requests
 from requests.adapters import HTTPAdapter
 from urllib3.util.retry import Retry
-from urllib.parse import urlparse
-from itertools import zip_longest
 
 
 def requests_retry_session(