Merge pull request #998 from minrk/update-versions

minrk · web-flow · commit 91161d748515 · 2022-02-01T14:41:44.000+01:00
Update some versions
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -1,21 +1,22 @@
 repos:
 - repo: https://github.com/asottile/reorder_python_imports
-  rev: v1.3.5
+  rev: v2.6.0
   hooks:
   - id: reorder-python-imports
-    language_version: python3
 - repo: https://github.com/ambv/black
-  rev: 18.9b0
+  rev: "21.12b0"
   hooks:
   - id: black
+- repo: https://github.com/PyCQA/flake8
+  rev: "4.0.1"
+  hooks:
+  - id: flake8
 - repo: https://github.com/pre-commit/pre-commit-hooks
-  rev: v2.1.0
+  rev: v4.1.0
   hooks:
   - id: end-of-file-fixer
   - id: check-json
   - id: check-yaml
     exclude: ^helm-chart/nbviewer/templates/
   - id: check-case-conflict
   - id: check-executables-have-shebangs
-  - id: requirements-txt-fixer
-  - id: flake8
diff --git a/Dockerfile b/Dockerfile
@@ -1,30 +1,31 @@
 # Define a builder image
-FROM python:3.7-buster as builder
+FROM python:3.10-bullseye as builder
 
 ENV DEBIAN_FRONTEND=noninteractive
 ENV LANG=C.UTF-8
 RUN apt-get update \
  && apt-get install -yq --no-install-recommends \
     ca-certificates \
     libcurl4-gnutls-dev \
+    libgnutls28-dev \
+    libmemcached-dev \
     git \
     nodejs \
     npm
 
-# Python requirements
-COPY ./requirements-dev.txt /srv/nbviewer/
-COPY ./requirements.txt /srv/nbviewer/
-RUN python3 -mpip install -r /srv/nbviewer/requirements-dev.txt -r /srv/nbviewer/requirements.txt
+# Build requirements
+COPY ./requirements-dev.txt  /srv/nbviewer/
+RUN python3 -mpip install -r /srv/nbviewer/requirements-dev.txt
 
 WORKDIR /srv/nbviewer
 
 # Copy source tree in
 COPY . /srv/nbviewer
 RUN python3 setup.py build && \
-    python3 -mpip wheel -vv . -w /wheels
+    python3 -mpip wheel -vv -r requirements.txt . -w /wheels
 
 # Now define the runtime image
-FROM python:3.7-slim-buster
+FROM python:3.10-slim-bullseye
 LABEL maintainer="Jupyter Project <jupyter@googlegroups.com>"
 
 ENV DEBIAN_FRONTEND=noninteractive
@@ -34,6 +35,7 @@ RUN apt-get update \
  && apt-get install -yq --no-install-recommends \
     ca-certificates \
     libcurl4 \
+    libmemcached11 \
     git \
  && apt-get clean && rm -rf /var/lib/apt/lists/*
 
diff --git a/helm-chart/nbviewer/requirements.lock b/helm-chart/nbviewer/requirements.lock
@@ -1,6 +1,6 @@
 dependencies:
 - name: memcached
-  repository: https://kubernetes-charts.storage.googleapis.com
-  version: 3.2.2
-digest: sha256:b0f92f7e3f8bfeb286cf8566d86c9c795a2712d7c690bdf66eb037dbae7b9036
-generated: "2020-03-03T10:17:51.050357+01:00"
+  repository: https://charts.helm.sh/stable
+  version: 3.2.3
+digest: sha256:b2795a847595ae51ef43c7818c6ed17d0ae66265f4613393581238418c184fb4
+generated: "2022-01-28T11:34:43.737858+01:00"
diff --git a/helm-chart/nbviewer/values.yaml b/helm-chart/nbviewer/values.yaml
@@ -1,7 +1,6 @@
 image: "jupyter/nbviewer"
 imagePullPolicy: null
 
-service: {}
 nodeSelector: null
 
 pdbMinAvailable: 2
diff --git a/nbviewer/cache.py b/nbviewer/cache.py
@@ -114,7 +114,7 @@ async def incr(self, key):
 
 class AsyncMemcache(object):
     """Wrap pylibmc.Client to run in a background thread
-    
+
     via concurrent.futures.ThreadPoolExecutor
     """
 
@@ -157,7 +157,7 @@ async def incr(self, *args, **kwargs):
 
 class AsyncMultipartMemcache(AsyncMemcache):
     """subclass of AsyncMemcache that splits large files into multiple chunks
-    
+
     because memcached limits record size to 1MB
     """
 
diff --git a/nbviewer/client.py b/nbviewer/client.py
@@ -28,18 +28,18 @@
 
 class NBViewerAsyncHTTPClient(object):
     """Subclass of AsyncHTTPClient with bonus logging and caching!
-    
+
     If upstream servers support 304 cache replies with the following headers:
-    
+
     - ETag : If-None-Match
     - Last-Modified : If-Modified-Since
-    
+
     Upstream requests are still made every time,
     but resources and rate limits may be saved by 304 responses.
-    
+
     If upstream responds with 304 or an error and a cached response is available,
     use the cached response.
-    
+
     Responses are cached as long as possible.
     """
 
diff --git a/nbviewer/log.py b/nbviewer/log.py
@@ -12,7 +12,7 @@
 
 def log_request(handler):
     """log a bit more information about each request than tornado's default
-    
+
     - move static file get success to debug-level (reduces noise)
     - get proxied IP instead of proxy IP
     - log referer for redirect and failed requests
diff --git a/nbviewer/providers/__init__.py b/nbviewer/providers/__init__.py
@@ -17,16 +17,16 @@
 
 def provider_handlers(providers, **handler_kwargs):
     """Load tornado URL handlers from an ordered list of dotted-notation modules
-       which contain a `default_handlers` function
+     which contain a `default_handlers` function
 
-       `default_handlers` should accept a list of handlers and returns an
-       augmented list of handlers: this allows the addition of, for
-       example, custom URLs which should be intercepted before being
-       handed to the basic `url` handler
+     `default_handlers` should accept a list of handlers and returns an
+     augmented list of handlers: this allows the addition of, for
+     example, custom URLs which should be intercepted before being
+     handed to the basic `url` handler
 
-      `handler_kwargs` is a dict of dicts: first dict is `handler_names`, which
-      specifies the handler_classes to load for the providers, the second
-      is `handler_settings` (see comments in `format_handlers` in nbviewer/handlers.py)
+    `handler_kwargs` is a dict of dicts: first dict is `handler_names`, which
+    specifies the handler_classes to load for the providers, the second
+    is `handler_settings` (see comments in `format_handlers` in nbviewer/handlers.py)
     """
     handler_names = handler_kwargs["handler_names"]
     handler_settings = handler_kwargs["handler_settings"]
@@ -45,23 +45,23 @@ def provider_handlers(providers, **handler_kwargs):
 
 def provider_uri_rewrites(providers):
     """Load (regex, template) tuples from an ordered list of dotted-notation
-       modules which contain a `uri_rewrites` function
+    modules which contain a `uri_rewrites` function
 
-       `uri_rewrites` should accept a list of rewrites and returns an
-       augmented list of rewrites: this allows the addition of, for
-       example, the greedy behavior of the `gist` and `github` providers
+    `uri_rewrites` should accept a list of rewrites and returns an
+    augmented list of rewrites: this allows the addition of, for
+    example, the greedy behavior of the `gist` and `github` providers
     """
     return _load_provider_feature("uri_rewrites", providers)
 
 
 def _load_provider_feature(feature, providers, **handler_names):
     """Load the named feature from an ordered list of dotted-notation modules
-       which each implements the feature.
+     which each implements the feature.
 
-       The feature will be passed a list of feature implementations and must
-       return that list, suitably modified.
+     The feature will be passed a list of feature implementations and must
+     return that list, suitably modified.
 
-      `handler_names` is the same as the `handler_names` attribute of the NBViewer class
+    `handler_names` is the same as the `handler_names` attribute of the NBViewer class
     """
 
     # Ex: provider = 'nbviewer.providers.url'
diff --git a/nbviewer/providers/base.py b/nbviewer/providers/base.py
@@ -330,7 +330,7 @@ def get_page_links(self, response):
 
     def client_error_message(self, exc, url, body, msg=None):
         """Turn the tornado HTTP error into something useful
-        
+
         Returns error code
         """
         str_exc = str(exc)
@@ -466,8 +466,7 @@ def cache_headers(self):
 
     @property
     def cache_key(self):
-        """Use checksum for cache key because cache has size limit on keys
-        """
+        """Use checksum for cache key because cache has size limit on keys"""
 
         if self._cache_key is None:
             to_hash = utf8(getattr(self.request, self._cache_key_attr))
@@ -635,12 +634,12 @@ def filter_formats(self, nb, raw):
     # empty methods to be implemented by subclasses to make GET requests more modular
     def get_notebook_data(self, **kwargs):
         """
-        Pass as kwargs variables needed to define those variables which will be necessary for 
-        the provider to find the notebook. (E.g. path for LocalHandler, user and repo for GitHub.) 
+        Pass as kwargs variables needed to define those variables which will be necessary for
+        the provider to find the notebook. (E.g. path for LocalHandler, user and repo for GitHub.)
         Return variables the provider needs to find and load the notebook. Then run custom logic
         in GET or pass the output of get_notebook_data immediately to deliver_notebook.
 
-        First part of any provider's GET method. 
+        First part of any provider's GET method.
 
         Custom logic, if applicable, is middle part of any provider's GET method, and usually
         is implemented or overwritten in subclasses, while get_notebook_data and deliver_notebook
diff --git a/nbviewer/providers/gist/handlers.py b/nbviewer/providers/gist/handlers.py
@@ -255,8 +255,8 @@ async def get_notebook_data(self, gist_id, filename, many_files_gist, file):
     # Only called by file_get
     async def deliver_notebook(self, user, gist_id, filename, gist, file, content):
         """
-        user, gist_id, filename, gist, file, are the same values as those 
-        passed into file_get, whereas content is returned from 
+        user, gist_id, filename, gist, file, are the same values as those
+        passed into file_get, whereas content is returned from
         get_notebook_data using user, gist_id, filename, gist, and file.
         """
         # Enable a binder navbar icon if a binder base URL is configured
diff --git a/nbviewer/providers/github/client.py b/nbviewer/providers/github/client.py
@@ -105,7 +105,7 @@ def _log_rate_limit(self, future):
 
     def github_api_request(self, path, **kwargs):
         """Make a GitHub API request to URL
-        
+
         URL is constructed from url and params, if specified.
         **kwargs are passed to client.fetch unmodified.
         """
@@ -165,7 +165,7 @@ def get_tags(self, user, repo, **kwargs):
     def extract_tree_entry(self, path, tree_response):
         """extract a single tree entry from
         a tree response using for a path
-        
+
         raises 404 if not found
 
         Useful for finding the blob url for a given path.
diff --git a/nbviewer/providers/github/handlers.py b/nbviewer/providers/github/handlers.py
@@ -334,8 +334,10 @@ class GitHubBlobHandler(GithubClientMixin, RenderingHandler):
 
     async def get_notebook_data(self, user, repo, ref, path):
         if os.environ.get("GITHUB_API_URL", "") == "":
-            raw_url = u"https://raw.githubusercontent.com/{user}/{repo}/{ref}/{path}".format(
-                user=user, repo=repo, ref=ref, path=quote(path)
+            raw_url = (
+                u"https://raw.githubusercontent.com/{user}/{repo}/{ref}/{path}".format(
+                    user=user, repo=repo, ref=ref, path=quote(path)
+                )
             )
         else:  # Github Enterprise has a different URL pattern for accessing raw files
             raw_url = url_path_join(
diff --git a/nbviewer/providers/github/tests/test_client.py b/nbviewer/providers/github/tests/test_client.py
@@ -28,7 +28,7 @@ def _get_url(self):
 
     def assertStartsWith(self, string, beginning):
         """Assert that a url has the correct beginning.
-        
+
         Github API requests involve non-trivial query strings.  This is useful
         when you want to compare URLs, but don't care about the querystring.
         """
diff --git a/nbviewer/providers/url/tests/test_content.py b/nbviewer/providers/url/tests/test_content.py
@@ -6,8 +6,7 @@
 
 class ForceUTF8TestCase(NBViewerTestCase):
     def test_utf8(self):
-        """ #507, bitbucket returns no content headers, but _is_ serving utf-8
-        """
+        """#507, bitbucket returns no content headers, but _is_ serving utf-8"""
         response = requests.get(
             self.url("/urls/bitbucket.org/sandiego206/asdasd/raw/master/Untitled.ipynb")
         )
diff --git a/nbviewer/ratelimit.py b/nbviewer/ratelimit.py
@@ -17,7 +17,7 @@ def __init__(self, limit, interval, cache):
 
     def key_for_handler(self, handler):
         """Identify a visitor.
-        
+
         Currently combine ip + user-agent.
         We don't need to be perfect.
         """
@@ -29,9 +29,9 @@ def key_for_handler(self, handler):
 
     async def check(self, handler):
         """Check the rate limit for a handler.
-        
+
         Identifies the source by ip and user-agent.
-        
+
         If the rate limit is exceeded, raise HTTPError(429)
         """
         if not self.limit:
diff --git a/nbviewer/utils.py b/nbviewer/utils.py
@@ -58,8 +58,8 @@ def quote(s):
 
 
 def clean_filename(fn):
-    """ Github url sanitizes gist filenames to produce their permalink. This is
-    not provided over API, so we recreate it here. """
+    """Github url sanitizes gist filenames to produce their permalink. This is
+    not provided over API, so we recreate it here."""
     return re.sub("[^0-9a-zA-Z]+", "-", fn)
 
 
@@ -231,7 +231,7 @@ def time_block(message, logger, debug_limit=1):
     """context manager for timing a block
 
     logs millisecond timings of the block
-    
+
     If the time is longer than debug_limit,
     then log level will be INFO,
     otherwise it will be DEBUG.
diff --git a/requirements.in b/requirements.in
@@ -0,0 +1,12 @@
+elasticsearch
+ipython
+jupyter_client
+jupyter_server>=0.2.0
+markdown>=3.0,==3.1.1 # pin until we workaround #909, which is a regression in 3.2
+nbconvert>=5.4,<6.0.0 # pin to fix #958
+nbformat>=4.2
+newrelic!=2.80.0.60
+pycurl
+pylibmc
+statsd
+tornado>=6.0
diff --git a/requirements.txt b/requirements.txt
diff --git a/setup.py b/setup.py
diff --git a/tasks.py b/tasks.py
