1 Commits

Author SHA1 Message Date
Alok Saldanha
a6c13a7606 Applied "Introduction of ItemSource interface" patch 2020-10-08 08:37:31 -04:00
23 changed files with 858 additions and 509 deletions

View File

@@ -59,7 +59,12 @@ cellxgene-gateway
Here's what the environment variables mean:
* `CELLXGENE_LOCATION` - the location of the cellxgene executable, e.g. `~/anaconda2/envs/cellxgene/bin/cellxgene`
At least one of the following is required:
* `CELLXGENE_DATA` - a directory that can contain subdirectories with `.h5ad` data files, *without* trailing slash, e.g. `/mnt/cellxgene_data`
* `CELLXGENE_BUCKET` - an s3 bucket that can contain keys with `.h5ad` data files, e.g. `my-cellxgene-data-bucket`
Cellxgene Gateway is designed to make it easy to add additional data sources, please see the source code for gateway.py and the ItemSource interface in items/item_source.py
Optional environment variables:
* `CELLXGENE_ARGS` - catch-all variable that can be used to pass additional command line args to cellxgene server
* `EXTERNAL_HOST` - the hostname and port from the perspective of the web browser, typically `localhost:5005` if running locally. Defaults to "localhost:{GATEWAY_PORT}"
@@ -67,7 +72,6 @@ Optional environment variables:
* `GATEWAY_IP` - ip addess of instance gateway is running on, mostly used to display SSH instructions. Defaults to `socket.gethostbyname(socket.gethostname())`
* `GATEWAY_PORT` - local port that the gateway should bind to, defaults to 5005
* `GATEWAY_EXTRA_SCRIPTS` - JSON array of script paths, will be embedded into each page and forwarded with `--scripts` to cellxgene server
* `GATEWAY_ENABLE_UPLOAD` - Set to `true` or `1` to enable HTTP uploads. This is not recommended for a public server.
* `GATEWAY_ENABLE_ANNOTATIONS` - Set to `true` or to `1` to enable cellxgene annotations.
* `GATEWAY_ENABLE_BACKED_MODE` - Set to `true` or to `1` to load AnnData in file-backed mode. This saves memory and speeds up launch time but may reduce overall performance.
@@ -126,9 +130,8 @@ For convenience, the code repo includes a `run.sh.example` shell script to run t
pip install isort flake8 black
```bash
isort -rc .
flake8 .
black -l 79 .
isort -rc . # rc means recursive, and was deprecated in dev version of isort
black .
```
# Getting Help

View File

@@ -14,8 +14,10 @@ from flask_api import status
from cellxgene_gateway import env
from cellxgene_gateway.cache_entry import CacheEntry, CacheEntryStatus
from cellxgene_gateway.cache_key import CacheKey
from cellxgene_gateway.cellxgene_exception import CellxgeneException
from cellxgene_gateway.subprocess_backend import SubprocessBackend
from typing import List
process_backend = SubprocessBackend()
@@ -35,13 +37,13 @@ class BackendCache:
contents = self.entry_list
return [c.port for c in contents]
def check_entry(self, key):
def check_path(self, source, path):
contents = self.entry_list
matches = [
c
for c in contents
if c.key.dataset == key.dataset
and c.key.annotation_file == key.annotation_file
if c.key.source.name == source.name
and path.startswith(c.key.descriptor)
and c.status != CacheEntryStatus.terminated
]
@@ -52,10 +54,28 @@ class BackendCache:
else:
raise CellxgeneException(
status.HTTP_500_INTERNAL_SERVER_ERROR,
"Found " + str(len(matches)) + " for " + dataset,
"Found " + str(len(matches)) + " for " + path,
)
def create_entry(self, key, scripts):
def check_entry(self, key):
contents = self.entry_list
matches = [
c
for c in contents
if c.key.equals(key) and c.status != CacheEntryStatus.terminated
]
if len(matches) == 0:
return None
elif len(matches) == 1:
return matches[0]
else:
raise CellxgeneException(
status.HTTP_500_INTERNAL_SERVER_ERROR,
"Found " + str(len(matches)) + " for " + key.dataset,
)
def create_entry(self, key: CacheKey, scripts: List[str]):
port = 8000
existing_ports = self.get_ports()

View File

@@ -8,9 +8,10 @@
# the specific language governing permissions and limitations under the License.
import datetime
import logging
import urllib.parse
from enum import Enum
import psutil
from enum import Enum
from flask import make_response, render_template, request
from requests import get, post, put
import re
@@ -69,6 +70,10 @@ class CacheEntry:
None,
)
@property
def source_name(self):
return self.key.source_name
def set_loaded(self, pid):
self.pid = pid
self.status = CacheEntryStatus.loaded
@@ -98,9 +103,13 @@ class CacheEntry:
for child in children:
child.terminate()
psutil.wait_procs(children, callback=on_terminate)
terminated.append(p.pid)
p.terminate()
psutil.wait_procs([p], callback=on_terminate)
# the parent process may automatically die once its children have --
try:
p.terminate()
psutil.wait_procs([p], callback=on_terminate)
except psutil.NoSuchProcess:
pass
logging.getLogger("cellxgene_gateway").info(
f"terminated {terminated}"
)
@@ -120,15 +129,19 @@ class CacheEntry:
return gateway_content
def gateway_basepath(self):
return f"{env.external_protocol}://{env.external_host}/view/{self.key.pathpart}/"
source_path = (
f"/source/{urllib.parse.quote_plus(self.source_name)}"
if self.source_name
else ""
)
return f"{env.external_protocol}://{env.external_host}{source_path}/view/{self.key.descriptor}/"
def cellxgene_basepath(self):
return f"http://127.0.0.1:{self.port}"
def serve_content(self, path):
gateway_basepath = self.gateway_basepath()
subpath = path[len(self.key.pathpart) :] # noqa: E203
subpath = path[len(self.key.descriptor) :] # noqa: E203
if len(subpath) == 0:
r = make_response(f"Redirect to {gateway_basepath}\n", 301)
r.headers["location"] = gateway_basepath + querystring()
@@ -201,5 +214,4 @@ class CacheEntry:
cellxgene_response.status_code,
resp_headers,
)
return gateway_response

View File

@@ -9,15 +9,62 @@
# There are three kinds of CacheKey:
# 1) somedir/dataset.h5ad: a dataset
# in this case, pathpart == dataset == 'somedir/dataset.h5ad'
# 2) somedir/dataset_annotations/my_annotations.csv : an actual annotaitons file.
# in this case, pathpart == 'dataset_annotations/my_annotations.csv', dataset == 'somedir/dataset.h5ad'
# in this case, descriptor == dataset == 'somedir/dataset.h5ad'
# 2) somedir/dataset_annotations/my_annotations.csv : an actual annotations file.
# in this case, descriptor == 'somedir/dataset_annotations/my_annotations.csv', dataset == 'somedir/dataset.h5ad'
# 3) somedir/dataset_annotations: an annotation directory. The corresponding h5ad must exist, but the directory may not.
# in this case, pathpart == 'dataset_annotations', dataset == 'somedir/dataset.h5ad'
# in this case, descriptor == 'somedir/dataset_annotations', dataset == 'somedir/dataset.h5ad'
from cellxgene_gateway.items.item import Item
from cellxgene_gateway.items.item_source import ItemSource, LookupResult
class CacheKey:
def __init__(self, pathpart, dataset, annotation_file):
self.pathpart = pathpart
self.dataset = dataset
self.annotation_file = annotation_file
@property
def descriptor(self):
if self.annotation_item is None:
return self.h5ad_item.descriptor
else:
return self.annotation_item.descriptor
@property
def file_path(self):
return self.source.get_local_path(self.h5ad_item)
@property
def annotation_file_path(self):
if self.annotation_item is None:
return None
else:
return self.source.get_local_path(self.annotation_item)
@property
def source_name(self):
return self.source.name
@property
def annotation_descriptor(self):
if self.annotation_item is None:
return None
else:
return self.annotation_item.descriptor
def equals(self, other):
return (
(self.source.name == other.source.name)
and (self.h5ad_item.descriptor == other.h5ad_item.descriptor)
and (self.annotation_descriptor == other.annotation_descriptor)
)
def __init__(
self, h5ad_item: Item, source: ItemSource, annotation_item: Item = None
):
assert h5ad_item is not None
assert source is not None
self.h5ad_item = h5ad_item
self.annotation_item = annotation_item
self.source = source
@classmethod
def for_lookup(cls, source: ItemSource, lookup: LookupResult):
return CacheKey(lookup.h5ad_item, source, lookup.annotation_item)

View File

@@ -53,10 +53,11 @@ def create_dir(parent_path, dir_name):
annotations_suffix = "_annotations"
h5ad_suffix = ".h5ad"
def make_h5ad(el):
return el[: -len(annotations_suffix)] + ".h5ad"
return el[: -len(annotations_suffix)] + h5ad_suffix
def make_annotations(el):

View File

@@ -22,13 +22,9 @@ external_host = os.environ.get(
external_protocol = os.environ.get(
"EXTERNAL_PROTOCOL", os.environ.get("GATEWAY_PROTOCOL", "http")
)
ip = os.environ.get("GATEWAY_IP", "127.0.0.1")
ip = os.environ.get("GATEWAY_IP")
extra_scripts = os.environ.get("GATEWAY_EXTRA_SCRIPTS")
ttl = os.environ.get("GATEWAY_TTL")
enable_upload = os.environ.get("GATEWAY_ENABLE_UPLOAD", "").lower() in [
"true",
"1",
]
enable_annotations = os.environ.get(
"GATEWAY_ENABLE_ANNOTATIONS", ""
).lower() in [
@@ -44,7 +40,6 @@ enable_backed_mode = os.environ.get(
env_vars = {
"CELLXGENE_LOCATION": cellxgene_location,
"CELLXGENE_DATA": cellxgene_data,
}
optional_env_vars = {
@@ -54,10 +49,10 @@ optional_env_vars = {
"GATEWAY_PORT": gateway_port,
"GATEWAY_EXTRA_SCRIPTS": extra_scripts,
"GATEWAY_TTL": ttl,
"GATEWAY_ENABLE_UPLOAD": enable_upload,
"GATEWAY_ENABLE_ANNOTATIONS": enable_annotations,
"GATEWAY_ENABLE_BACKED_MODE": enable_backed_mode,
"CELLXGENE_ARGS": cellxgene_args,
"CELLXGENE_DATA": cellxgene_data,
}

View File

@@ -8,113 +8,61 @@
# the specific language governing permissions and limitations under the License.
import os
from cellxgene_gateway import env
from cellxgene_gateway.dir_util import (
make_h5ad,
make_annotations,
annotations_suffix,
)
import urllib.parse
def recurse_dir(path):
if not os.path.exists(path):
raise CellxgeneException(
"The given path does not exist.", status.HTTP_400_BAD_REQUEST
)
all_entries = sorted(os.listdir(path))
def is_h5ad(el):
return el.endswith(".h5ad") and os.path.isfile(os.path.join(path, el))
h5ad_entries = [x for x in all_entries if is_h5ad(x)]
annotation_dir_entries = [
x
for x in all_entries
if x.endswith(annotations_suffix) and make_h5ad(x) in h5ad_entries
]
def list_annotations(el):
full_path = os.path.join(path, el)
if not os.path.isdir(full_path):
entries = []
else:
entries = [
{
"name": x[:-13]
if (len(x) > 13 and x[-13] in ["-", "_"])
else (x[:-4] if x.endswith(".csv") else x),
"path": os.path.join(full_path, x).replace(
env.cellxgene_data, ""
),
}
for x in sorted(os.listdir(full_path))
if x.endswith(".csv")
and os.path.isfile(os.path.join(full_path, x))
]
return [
{
"name": "new",
"class": "new",
"path": full_path.replace(env.cellxgene_data, ""),
}
] + entries
def make_entry(el):
full_path = os.path.join(path, el)
if el in h5ad_entries:
return {
"path": full_path.replace(env.cellxgene_data, ""),
"name": el,
"type": "file",
"annotations": list_annotations(make_annotations(el)),
}
elif os.path.isdir(full_path) and el not in annotation_dir_entries:
return {
"path": full_path.replace(env.cellxgene_data, ""),
"name": el,
"type": "directory",
"children": recurse_dir(full_path),
}
else:
return {
"path": full_path,
"name": el,
"type": "neither",
}
return [make_entry(x) for x in all_entries]
def render_entries(entries):
return "<ul>" + "\n".join([render_entry(e) for e in entries]) + "</ul>"
def get_url(entry):
return f"/view/{ entry['path'].lstrip('/') }"
def get_class(entry):
return f" class='{entry['class']}'" if "class" in entry else ""
def render_annotations(entry):
if len(entry["annotations"]) > 0:
return " | annotations: " + ", ".join(
def render_annotations(item, item_source):
subpath = f"/source/{urllib.parse.quote_plus(item_source.name)}/view/"
new_annotation = f"<a class='new' href='{subpath}{item_source.get_annotations_subpath(item)}'>new</a>"
annotations = (
", ".join(
[
f"<a href='{get_url(a)}'{get_class(a)}>{a['name']}</a>"
for a in entry["annotations"]
f"<a href='{subpath}{a.descriptor}/'>{a.name}</a>"
for a in item.annotations
]
)
else:
return ""
+ ", "
if item.annotations
else ""
)
return " | annotations: " + annotations + new_annotation
def render_entry(entry):
if entry["type"] == "file":
return f"<li> <a href='{ get_url(entry) }'>{entry['name']}</a> {render_annotations(entry)}</li>"
elif entry["type"] == "directory":
url = f"/filecrawl/{entry['path'].lstrip('/')}"
return f"<li><a href='{url}'>{entry['name']}</a>{render_entries(entry['children'])}</li>"
def render_item(item, item_source):
url = f"/source/{urllib.parse.quote_plus(item_source.name)}/view/{item.descriptor}/"
item_string = f"<li> <a href='{ url }'>{item.name}</a> {render_annotations(item, item_source)}</li>"
return item_string
def render_item_tree(item_tree, item_source):
items = (
"\n".join([render_item(i, item_source) for i in item_tree.items])
if item_tree.items
else ""
)
branches = (
"\n".join(
[render_item_tree(b, item_source) for b in item_tree.branches]
)
if item_tree.branches
else ""
)
html = "<ul>" + items + branches + "</ul>"
if item_tree.descriptor:
descriptor = item_tree.descriptor.lstrip("/")
url = f"/filecrawl/{descriptor}?source={item_source.name}"
name = (
descriptor.rsplit("/")[1]
if descriptor.find("/") >= 0
else descriptor
)
return f"<li><a href='{url}'>{name}</a>{html}</li>"
else:
return ""
return html
def render_item_source(item_source, filter=None):
item_tree = item_source.list_items(filter)
heading = f"<h6><a href='/filecrawl.html?source={urllib.parse.quote_plus(item_source.name)}'>{item_source.name}</a></h6>"
return heading + render_item_tree(item_tree, item_source)

View File

@@ -6,12 +6,11 @@
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
# import BaseHTTPServer
import json
import logging
# import BaseHTTPServer
import os
import urllib.parse
from threading import Lock, Thread
from flask import (
@@ -32,14 +31,17 @@ from cellxgene_gateway.cache_entry import CacheEntryStatus
from cellxgene_gateway.cellxgene_exception import CellxgeneException
from cellxgene_gateway.dir_util import create_dir, is_subdir
from cellxgene_gateway.extra_scripts import get_extra_scripts
from cellxgene_gateway.filecrawl import recurse_dir, render_entries
from cellxgene_gateway.path_util import get_key
from cellxgene_gateway.filecrawl import render_item_source
from cellxgene_gateway.process_exception import ProcessException
from cellxgene_gateway.prune_process_cache import PruneProcessCache
from cellxgene_gateway.util import current_time_stamp
from cellxgene_gateway.cache_key import CacheKey
app = Flask(__name__)
item_sources = []
default_item_source = None
def _force_https(app):
def wrapper(environ, start_response):
@@ -88,8 +90,8 @@ def handle_invalid_process(error):
http_status=error.http_status,
stdout=error.stdout,
stderr=error.stderr,
dataset=error.key.dataset,
annotation_file=error.key.annotation_file,
dataset=error.key.h5ad_item.descriptor,
annotation_file=error.key.annotation_descriptor,
),
error.http_status,
)
@@ -106,84 +108,40 @@ def favicon():
@app.route("/")
def index():
users = [
name
for name in os.listdir(env.cellxgene_data)
if os.path.isdir(os.path.join(env.cellxgene_data, name))
]
return render_template(
"index.html",
ip=env.ip,
cellxgene_data=env.cellxgene_data,
extra_scripts=get_extra_scripts(),
users=users,
enable_upload=env.enable_upload,
)
def make_user():
dir_name = request.form["directory"]
create_dir(env.cellxgene_data, dir_name)
return redirect(location, code=302)
def make_subdir():
parent_path = os.path.join(env.cellxgene_data, request.form["usernames"])
dir_name = request.form["directory"]
create_dir(parent_path, dir_name)
return redirect(location, code=302)
def upload_file():
upload_dir = request.form["path"]
full_upload_path = os.path.join(env.cellxgene_data, upload_dir)
if is_subdir(full_upload_path, env.cellxgene_data) and os.path.isdir(
full_upload_path
):
if request.method == "POST":
if "file" in request.files:
f = request.files["file"]
if f and f.filename.endswith(".h5ad"):
f.save(
os.path.join(
full_upload_path, secure_filename(f.filename)
)
)
return redirect("/filecrawl.html", code=302)
else:
raise CellxgeneException(
"Uploaded file must be in anndata (.h5ad) format.",
status.HTTP_400_BAD_REQUEST,
)
else:
raise CellxgeneException(
"A file must be chosen to upload.",
status.HTTP_400_BAD_REQUEST,
)
else:
raise CellxgeneException(
"Invalid directory.", status.HTTP_400_BAD_REQUEST
@app.route("/filecrawl.html")
@app.route("/filecrawl/<path:path>")
def filecrawl(path=None):
source_name = request.args.get("source")
sources = (
filter(
lambda x: x.name == urllib.parse.unquote_plus(source_name),
item_sources,
)
return redirect(location, code=302)
if env.enable_upload:
app.add_url_rule("/make_user", "make_user", make_user, methods=["POST"])
app.add_url_rule(
"/make_subdir", "make_subdir", make_subdir, methods=["POST"]
if source_name
else item_sources
)
app.add_url_rule(
"/upload_file", "upload_file", upload_file, methods=["POST"]
# loop all data sources --
rendered_sources = [
render_item_source(item_source, path) for item_source in sources
] # will we need to make this async in the page???
rendered_html = "\n".join(rendered_sources)
resp = make_response(
render_template(
"filecrawl.html",
extra_scripts=get_extra_scripts(),
rendered_html=rendered_html,
path=path,
)
)
def set_no_cache(resp):
resp.headers["Cache-Control"] = "no-cache, no-store, must-revalidate"
resp.headers["Pragma"] = "no-cache"
resp.headers["Expires"] = "0"
@@ -191,52 +149,44 @@ def set_no_cache(resp):
return resp
@app.route("/filecrawl.html")
def filecrawl():
entries = recurse_dir(env.cellxgene_data)
rendered_html = render_entries(entries)
resp = make_response(
render_template(
"filecrawl.html",
extra_scripts=get_extra_scripts(),
rendered_html=rendered_html,
)
)
return set_no_cache(resp)
@app.route("/filecrawl/<path:path>")
def do_filecrawl(path):
filecrawl_path = os.path.join(env.cellxgene_data, path)
if not os.path.isdir(filecrawl_path):
raise CellxgeneException(
"Path is not directory: " + filecrawl_path,
status.HTTP_400_BAD_REQUEST,
)
entries = recurse_dir(filecrawl_path)
rendered_html = render_entries(entries)
return render_template(
"filecrawl.html",
extra_scripts=get_extra_scripts(),
rendered_html=rendered_html,
path=path,
)
entry_lock = Lock()
def matching_source(source_name):
if source_name is None:
source_name = default_item_source.name
matching = [i for i in item_sources if i.name == source_name]
if len(matching) != 1:
raise Exception(f"Could not find matching item source {source_name}")
source = matching[0]
return source
@app.route(
"/source/<path:source_name>/view/<path:path>",
methods=["GET", "PUT", "POST"],
)
@app.route("/view/<path:path>", methods=["GET", "PUT", "POST"])
def do_view(path):
key = get_key(path)
print(
f"view path={path}, dataset={key.dataset}, annotation_file= {key.annotation_file}, key={key.pathpart}"
)
with entry_lock:
match = cache.check_entry(key)
if match is None:
uascripts = get_extra_scripts()
match = cache.create_entry(key, uascripts)
def do_view(path, source_name=None):
source = matching_source(source_name)
match = cache.check_path(source, path)
if match is None:
lookup = source.lookup(path)
if lookup is None:
raise CellxgeneException(
f"Could not find item for path {path} in source {source.name}",
404,
)
key = CacheKey.for_lookup(source, lookup)
print(
f"view path={path}, source_name={source_name}, dataset={key.file_path}, annotation_file= {key.annotation_file_path}, key={key.descriptor}, source={key.source_name}"
)
with entry_lock:
match = cache.check_entry(key)
if match is None:
uascripts = get_extra_scripts()
match = cache.create_entry(key, uascripts)
match.timestamp = current_time_stamp()
@@ -275,7 +225,9 @@ def do_GET_status_json():
@app.route("/relaunch/<path:path>", methods=["GET"])
def do_relaunch(path):
key = get_key(path)
source_name = request.args.get("source") or default_item_source.name
source = matching_source(source_name)
key = CacheKey.for_lookup(source, source.lookup(path))
match = cache.check_entry(key)
if not match is None:
match.terminate()
@@ -288,25 +240,24 @@ def do_relaunch(path):
@app.route("/terminate/<path:path>", methods=["GET"])
def do_terminate(path):
key = get_key(path)
source_name = request.args.get("source_name") or default_item_source.name
source = matching_source(source_name)
key = CacheKey.for_lookup(source, source.lookup(path))
match = cache.check_entry(key)
if not match is None:
match.terminate()
return redirect(url_for("do_GET_status"), code=302)
@app.route("/metadata/ip_address", methods=["GET"])
def ip_address():
resp = make_response(env.ip)
return set_no_cache(resp)
def main():
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s:%(name)s:%(levelname)s:%(message)s",
)
def launch():
env.validate()
if not item_sources or not len(item_sources):
raise Exception("No data sources specified for Cellxgene Gateway")
global default_item_source
if default_item_source is None:
default_item_source = item_sources[0]
pruner = PruneProcessCache(cache)
background_thread = Thread(target=pruner)
@@ -316,5 +267,29 @@ def main():
app.run(host="0.0.0.0", port=env.gateway_port, debug=False)
def main():
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s:%(name)s:%(levelname)s:%(message)s",
)
cellxgene_data = os.environ.get("CELLXGENE_DATA", None)
cellxgene_bucket = os.environ.get("CELLXGENE_BUCKET", None)
if cellxgene_bucket is not None:
from cellxgene_gateway.items.s3.s3item_source import S3ItemSource
item_sources.append(S3ItemSource(cellxgene_bucket, name="s3"))
default_item_source = "s3"
if cellxgene_data is not None:
from cellxgene_gateway.items.file.fileitem_source import FileItemSource
item_sources.append(FileItemSource(cellxgene_data, name="local"))
default_item_source = "local"
if len(item_sources) == 0:
raise Exception("Please specify CELLXGENE_DATA or CELLXGENE_BUCKET")
launch()
if __name__ == "__main__":
main()

View File

@@ -0,0 +1,27 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
import os
from cellxgene_gateway.items.item import Item
class FileItem(Item):
"""e.g. FileItem(subpath = subpath, name = filename, type = ItemType.h5ad)
The Item superclass expects a 'name' and 'type'.
"""
def __init__(self, subpath: str, *args, **kwargs):
super().__init__(*args, **kwargs)
self.subpath = subpath
@property
def descriptor(self) -> str:
return os.path.join(self.subpath, self.name).strip("/")

View File

@@ -0,0 +1,185 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
import os
from typing import List
from cellxgene_gateway import dir_util
from cellxgene_gateway.items.file.fileitem import FileItem
from cellxgene_gateway.items.item import ItemTree, ItemType
from cellxgene_gateway.items.item_source import ItemSource, LookupResult
class FileItemSource(ItemSource):
def __init__(
self,
base_path,
name=None,
h5ad_suffix=dir_util.h5ad_suffix,
annotation_dir_suffix=dir_util.annotations_suffix,
annotation_file_suffix=".csv",
):
self._name = name
self.base_path = base_path
self.h5ad_suffix = h5ad_suffix
self.annotation_dir_suffix = annotation_dir_suffix
self.annotation_file_suffix = annotation_file_suffix
@property
def name(self):
return self._name or f"Files:{self.base_path}"
def is_h5ad_file(self, path: str) -> bool:
return path.endswith(self.h5ad_suffix) and os.path.isfile(path)
def convert_annotation_path_to_h5ad(self, path):
return path[: -len(self.annotation_dir_suffix)] + self.h5ad_suffix
def convert_h5ad_path_to_annotation(self, path):
return path[: -len(self.h5ad_suffix)] + self.annotation_dir_suffix
def get_local_path(self, item: FileItem) -> str:
return os.path.join(self.base_path, item.descriptor)
def get_annotations_subpath(self, item) -> str:
return self.convert_h5ad_path_to_annotation(item.descriptor)
def list_items(self, filter: str = None) -> ItemTree:
item_tree = self.scan_directory()
"""def get_items(dir):
if dir.branches:
return [*dir.items, *[item for subdir in dir.branches for item in get_items(subdir)]]
else:
return dir.items
return get_items(self.item_tree)"""
return item_tree
def scan_directory(self, subpath="") -> dict:
base_path = os.path.join(self.base_path, subpath)
if not os.path.exists(base_path):
raise Exception(
f"Path for local files '{base_path}' does not exist."
)
filepath_map = dict(
(filepath, os.path.join(base_path, filepath))
for filepath in sorted(os.listdir(base_path))
)
def is_annotation_dir(dir):
return (
dir.endswith(self.annotation_dir_suffix)
and self.convert_annotation_path_to_h5ad(dir) in h5ad_paths
)
h5ad_paths = [
filepath
for filepath, full_path in filepath_map.items()
if self.is_h5ad_file(full_path)
]
subdirs = [
filepath
for filepath, full_path in filepath_map.items()
if os.path.isdir(full_path) and not is_annotation_dir(filepath)
]
items = [
self.make_fileitem_from_path(filename, subpath)
for filename in h5ad_paths
]
branches = None
if len(subdirs) > 0:
branches = [
self.scan_directory(os.path.join(subpath, subdir))
for subdir in subdirs
]
return ItemTree(subpath, items, branches)
def create_annotation(self, item: FileItem, name: str) -> FileItem:
annotation = self.make_fileitem_from_path(
name, self.get_annotations_subpath(item), is_annotation=True
)
item.annotations = (item.annotations or []).append(annotation)
return annotation
def update(self, item: FileItem) -> None:
pass
def full_path(self, p):
return os.path.join(self.base_path, p)
def lookup_item(self, descriptor):
full_path = self.full_path(descriptor)
if self.is_h5ad_file(full_path):
return self.shallowitem_from_descriptor(descriptor)
def lookup(self, indescriptor: str) -> LookupResult:
descriptor = indescriptor.strip("/")
if descriptor.endswith(self.annotation_file_suffix):
annotation_item = self.shallowitem_from_descriptor(
descriptor, True
)
h5ad_descriptor = self.convert_annotation_path_to_h5ad(
annotation_item.subpath
)
item = self.lookup_item(h5ad_descriptor)
if item is not None:
return LookupResult(item, annotation_item)
else:
item = self.lookup_item(descriptor)
if item is not None:
return LookupResult(item)
def shallowitem_from_descriptor(self, descriptor, is_annotation=False):
filename = os.path.basename(descriptor)
subpath = os.path.dirname(descriptor)
return self.make_fileitem_from_path(
filename,
subpath,
is_annotation,
True,
)
def make_fileitem_from_path(
self, filename, subpath, is_annotation=False, is_shallow=False
) -> FileItem:
item = FileItem(
subpath=subpath,
name=filename,
type=ItemType.annotation if is_annotation else ItemType.h5ad,
)
if not is_annotation and not is_shallow:
annotations = self.make_annotations_for_fileitem(item)
item.annotations = annotations
return item
def make_annotations_for_fileitem(self, item: FileItem) -> List[FileItem]:
annotations_subpath = self.get_annotations_subpath(item)
annotations_fullpath = self.full_path(annotations_subpath)
if os.path.isdir(annotations_fullpath):
return [
self.make_fileitem_from_path(
annotation, annotations_subpath, True
)
for annotation in sorted(os.listdir(annotations_fullpath))
if annotation.endswith(self.annotation_file_suffix)
and os.path.isfile(
os.path.join(annotations_fullpath, annotation)
)
]
else:
return None

View File

@@ -0,0 +1,43 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
from abc import ABC, abstractmethod
from enum import Enum
from typing import List
class ItemType(Enum):
annotation = "annotation"
h5ad = "h5ad"
class Item(ABC):
def __init__(
self, name: str, type: ItemType, annotations: List["Item"] = None
):
self.name = name
self.type = type
self.annotations = annotations
@property
@abstractmethod
def descriptor(self):
raise Exception('"descriptor" not implemented')
class ItemTree:
def __init__(
self,
descriptor: str,
items: List[Item] = None,
branches: List["ItemTree"] = None,
):
self.descriptor = descriptor
self.items = items
self.branches = branches

View File

@@ -0,0 +1,50 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
from abc import ABC, abstractmethod
from typing import List
from cellxgene_gateway.items.item import Item
class LookupResult:
def __init__(self, h5ad_item: Item, annotation_item: Item = None):
self.h5ad_item = h5ad_item
self.annotation_item = annotation_item
class ItemSource(ABC):
@abstractmethod
def list_items(self, filter: str = None) -> List[Item]:
raise Exception('"list_items" unimplemented')
@abstractmethod
def get_local_path(self, item: Item) -> str:
raise Exception('"local_path" unimplemented')
@abstractmethod
def get_annotations_subpath(self, item) -> str:
raise Exception('"annotations_path" unimplemented')
@abstractmethod
def create_annotation(self, item: Item, name: str) -> Item:
raise Exception('"annotation" unimplemented')
@abstractmethod
def update(self, item: Item) -> None:
raise Exception('"update" unimplemented')
@abstractmethod
def lookup(self, descriptor: str) -> LookupResult:
raise Exception('"lookup" unimplemented')
@property
@abstractmethod
def name(self):
pass

View File

@@ -0,0 +1,27 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
import os
from cellxgene_gateway.items.item import Item
class S3Item(Item):
"""e.g. FileItem(subpath = subpath, name = filename, type = ItemType.h5ad)
The Item superclass expects a 'name' and 'type'.
"""
def __init__(self, s3key: str, *args, **kwargs):
super().__init__(*args, **kwargs)
self.s3key = s3key
@property
def descriptor(self) -> str:
return self.s3key

View File

@@ -0,0 +1,171 @@
# Copyright 2019 Novartis Institutes for BioMedical Research Inc. Licensed
# under the Apache License, Version 2.0 (the "License"); you may not use
# this file except in compliance with the License. You may obtain a copy
# of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless
# required by applicable law or agreed to in writing, software distributed
# under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES
# OR CONDITIONS OF ANY KIND, either express or implied. See the License for
# the specific language governing permissions and limitations under the License.
from typing import List
from os.path import join, dirname, basename
from cellxgene_gateway import dir_util
import s3fs
from cellxgene_gateway.items.s3.s3item import S3Item
from cellxgene_gateway.items.item import ItemTree, ItemType
from cellxgene_gateway.items.item_source import ItemSource, LookupResult
class S3ItemSource(ItemSource):
def __init__(
self,
bucket,
name=None,
h5ad_suffix=dir_util.h5ad_suffix,
annotation_dir_suffix=dir_util.annotations_suffix,
annotation_file_suffix=".csv",
):
self._name = name
self.s3 = s3fs.S3FileSystem()
self.bucket = bucket
self.h5ad_suffix = h5ad_suffix
self.annotation_dir_suffix = annotation_dir_suffix
self.annotation_file_suffix = annotation_file_suffix
def url(self, path):
return "s3://" + join(self.bucket, path)
@property
def name(self):
return self._name or f"Items:{self.url('')}"
def is_h5ad_url(self, s3url: str) -> bool:
return s3url.endswith(self.h5ad_suffix) and self.s3.exists(s3url)
def convert_annotation_key_to_h5ad(self, s3key):
return s3key[: -len(self.annotation_dir_suffix)] + self.h5ad_suffix
def convert_h5ad_key_to_annotation(self, s3key):
return s3key[: -len(self.h5ad_suffix)] + self.annotation_dir_suffix
def get_local_path(self, item: S3Item) -> str:
return self.url(item.descriptor)
def get_annotations_subpath(self, item) -> str:
return self.convert_h5ad_key_to_annotation(item.descriptor)
def list_items(self, filter: str = None) -> ItemTree:
item_tree = self.scan_directory()
return item_tree
def scan_directory(self, subpath="") -> dict:
url = self.url(subpath)
if not self.s3.exists(url):
raise Exception(f"S3 url '{url}' does not exist.")
s3key_map = dict(
(filepath[len(self.bucket) :], "s3://" + filepath)
for filepath in sorted(self.s3.ls(url))
)
def is_annotation_dir(dir_s3key):
return (
dir_s3key.endswith(self.annotation_dir_suffix)
and self.convert_annotation_key_to_h5ad(dir_s3key)
in h5ad_paths
)
h5ad_paths = [
filepath
for filepath, item_url in s3key_map.items()
if self.is_h5ad_url(item_url)
]
subdirs = [
filepath
for filepath, item_url in s3key_map.items()
if self.s3.isdir(item_url) and not is_annotation_dir(filepath)
]
items = [
self.make_s3item_from_key(filename, join(subpath, filename))
for filename in h5ad_paths
]
branches = None
if len(subdirs) > 0:
branches = [
self.scan_directory(join(subpath, subdir))
for subdir in subdirs
]
return ItemTree(subpath, items, branches)
def create_annotation(self, item: S3Item, name: str) -> S3Item:
annotation = self.make_s3item_from_key(
name, self.get_annotations_subpath(item), is_annotation=True
)
item.annotations = (item.annotations or []).append(annotation)
return annotation
def update(self, item: S3Item) -> None:
pass
def lookup_item(self, descriptor):
full_path = self.url(descriptor)
if self.is_h5ad_url(full_path):
return self.shallowitem_from_descriptor(descriptor)
def lookup(self, indescriptor: str) -> LookupResult:
descriptor = indescriptor.strip("/")
if descriptor.endswith(self.annotation_file_suffix):
annotation_item = self.shallowitem_from_descriptor(
descriptor, True
)
if not self.s3.exists(self.url(annotation_item.s3key)):
with self.s3.open(self.url(annotation_item.s3key), "w") as f:
f.write("")
h5ad_descriptor = self.convert_annotation_key_to_h5ad(
dirname(annotation_item.s3key)
)
item = self.shallowitem_from_descriptor(h5ad_descriptor)
return LookupResult(item, annotation_item)
else:
item = self.lookup_item(descriptor)
if item is not None:
return LookupResult(item)
def shallowitem_from_descriptor(self, descriptor, is_annotation=False):
return self.make_s3item_from_key(
basename(descriptor), descriptor, is_annotation, True
)
def make_s3item_from_key(
self, name, s3key, is_annotation=False, is_shallow=False
) -> S3Item:
item = S3Item(
s3key=s3key,
name=name,
type=ItemType.annotation if is_annotation else ItemType.h5ad,
)
if not is_annotation and not is_shallow:
annotations = self.make_annotations_for_fileitem(item)
item.annotations = annotations
return item
def make_annotations_for_fileitem(self, item: S3Item) -> List[S3Item]:
annotations_subpath = self.get_annotations_subpath(item)
annotations_fullpath = self.url(annotations_subpath)
if self.s3.isdir(annotations_fullpath):
return [
self.make_s3item_from_key(
annotation, join(annotations_subpath, annotation), True
)
for annotation in sorted(self.s3.ls(annotations_fullpath))
if annotation.endswith(self.annotation_file_suffix)
and self.s3.isfile(join(annotations_fullpath, annotation))
]
else:
return None

View File

@@ -11,43 +11,11 @@ import os
from flask_api import status
from cellxgene_gateway import env
from cellxgene_gateway.cache_key import CacheKey
from cellxgene_gateway.cellxgene_exception import CellxgeneException
from cellxgene_gateway.dir_util import make_h5ad
def get_key(path):
if path == "/" or path == "":
raise CellxgeneException(
"No matching dataset found.", status.HTTP_404_NOT_FOUND
)
trimmed = path[:-1] if path[-1] == "/" else path
try:
# valid paths come in three forms:
if trimmed.endswith(".h5ad") and data_file_exists(trimmed):
# 1) somedir/dataset.h5ad: a dataset
return CacheKey(trimmed, trimmed, None)
elif trimmed.endswith(".csv"):
# 2) somedir/dataset_annotations/my_annotations.csv : an actual annotations file.
annotations_dir = os.path.split(trimmed)[0]
dataset = make_h5ad(annotations_dir)
if data_file_exists(dataset):
data_dir_ensure(annotations_dir)
return CacheKey(trimmed, dataset, trimmed)
elif trimmed.endswith("_annotations") and data_dir_exists(trimmed):
# 3) somedir/dataset_annotations: an annotation directory. The corresponding h5ad must exist, but the directory may not.
dataset = make_h5ad(trimmed)
if data_file_exists(dataset):
return CacheKey(trimmed, dataset, "")
except CellxgeneException:
pass
split = os.path.split(trimmed)
return get_key(split[0])
def validate_exists(file_path):
if not os.path.exists(file_path):
raise CellxgeneException(
@@ -71,37 +39,3 @@ def validate_is_dir(file_path):
"Path is not dir: " + file_path, status.HTTP_400_BAD_REQUEST
)
return
def data_file_exists(dataset):
file_path = os.path.join(env.cellxgene_data, dataset)
validate_is_file(file_path)
return True
def data_dir_exists(dataset):
file_path = os.path.join(env.cellxgene_data, dataset)
validate_is_dir(file_path)
return True
def data_dir_ensure(dataset):
file_path = os.path.join(env.cellxgene_data, dataset)
if not os.path.exists(file_path):
os.makedirs(file_path)
def get_file_path(key):
dataset = key.dataset
file_path = os.path.join(env.cellxgene_data, dataset)
validate_is_file(file_path)
return file_path
def get_annotation_file_path(key):
if key.annotation_file is None:
return None
if key.annotation_file == "":
return ""
file_path = os.path.join(env.cellxgene_data, key.annotation_file)
return file_path

View File

@@ -1,4 +1,5 @@
// neandertal javascript
// TODO: rewrite this --
const new_annotation_callback = (() =>{
const suffix = `.csv`;
return (e) => {

View File

@@ -11,9 +11,9 @@ import logging
import subprocess
from flask_api import status
from cellxgene_gateway.cache_entry import CacheEntryStatus
from cellxgene_gateway.dir_util import make_annotations
from cellxgene_gateway.path_util import get_annotation_file_path, get_file_path
from cellxgene_gateway.env import (
enable_annotations,
enable_backed_mode,
@@ -45,8 +45,7 @@ class SubprocessBackend:
cmd = (
f"yes | {cellxgene_loc} launch {file_path}"
+ " --port "
+ str(port)
+ f" --port {port}"
+ " --host 127.0.0.1"
+ extra_args
)
@@ -57,13 +56,12 @@ class SubprocessBackend:
return cmd
def launch(self, cellxgene_loc, scripts, cache_entry):
cmd = self.create_cmd(
cellxgene_loc,
get_file_path(cache_entry.key),
cache_entry.key.file_path,
cache_entry.port,
scripts,
get_annotation_file_path(cache_entry.key),
cache_entry.key.annotation_file_path,
)
logging.getLogger("cellxgene_gateway").info(f"launching {cmd}")
process = subprocess.Popen(

View File

@@ -10,59 +10,68 @@
-->
<html>
<head>
<title>Cellxgene Gateway - FILE CRAWLER</title>
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.3.1/jquery.min.js"></script>
<link rel="icon" type="image/png" href="{{ url_for('static', filename='nibr.ico') }}">
{% for script in extra_scripts %}
<script src="{{ script }}"></script>
{% endfor %}
<link rel="stylesheet" href="https://stackpath.bootstrapcdn.com/bootstrap/4.1.3/css/bootstrap.min.css" integrity="sha384-MCw98/SFnGE8fJT3GXwEOngsV7Zt27NXFoaoApmYm81iuXoPkFOJwJ8ERdknLPMO" crossorigin="anonymous">
{% for script in extra_scripts %}
<script src="{{ script }}"></script>
{% endfor %}
<link rel="stylesheet" href="https://stackpath.bootstrapcdn.com/bootstrap/4.1.3/css/bootstrap.min.css"
integrity="sha384-MCw98/SFnGE8fJT3GXwEOngsV7Zt27NXFoaoApmYm81iuXoPkFOJwJ8ERdknLPMO" crossorigin="anonymous">
</head>
<body>
<header class="navbar navbar-expand navbar-dark flex-column flex-md-row bd-navbar">
<h3>Cellxgene Gateway - Cache Status</h3>
<h3>Cellxgene Gateway - Cache Status</h3>
</header>
<br>
<table class="table">
<thead>
<tr>
<th>PID</th>
<th>dataset</th>
<th>annotation_file</th>
<th>port</th>
<th>launchtime</th>
<th>last access</th>
<th>status</th>
<th>message</th>
<th>http_status</th>
<th>actions</th>
</tr>
</thead>
<tbody>
{% for entry in entry_list %}
<tr>
<td>{{ entry.pid }}</td>
<td><a href="{{ url_for('do_view', path=entry.key.pathpart) }}">{{ entry.key.dataset }}</a></td>
<td>{{ entry.key.annotation_file }}</td>
<td>{{ entry.port }}</td>
<td class="timestamp">{{ entry.launchtime }}</td>
<td class="timestamp">{{ entry.timestamp }}</td>
<td>{{ entry.status.name }}</td>
<td>{{ entry.message }}</td>
<td>{{ entry.http_status }}</td>
<td>
{% if entry.status.name == 'loaded' %}
<a href="{{ url_for('do_terminate', path=entry.key.pathpart) }}"> terminate </a>
{% endif %}
</td>
</tr>
{% endfor %}
</tbody>
<tr>
<th>PID</th>
<th>dataset</th>
<th>annotation_file</th>
<th>source</th>
<th>port</th>
<th>launchtime</th>
<th>last access</th>
<th>status</th>
<th>message</th>
<th>http_status</th>
<th>actions</th>
</tr>
</thead>
<tbody>
{% for entry in entry_list %}
<tr>
<td>{{ entry.pid }}</td>
<td><a
href="{{ url_for('do_view', path=entry.key.descriptor, source_name=entry.key.source_name) }}">{{ entry.key.h5ad_item.descriptor }}</a>
</td>
<td>{{ entry.key.annotation_descriptor }}</td>
<td>{{ entry.source_name }}</td>
<td>{{ entry.port }}</td>
<td class="timestamp">{{ entry.launchtime }}</td>
<td class="timestamp">{{ entry.timestamp }}</td>
<td>{{ entry.status.name }}</td>
<td>{{ entry.message }}</td>
<td>{{ entry.http_status }}</td>
<td>
{% if entry.status.name == 'loaded' %}
<a
href="{{ url_for('do_terminate', path=entry.key.descriptor, source_name=entry.key.source_name) }}">
terminate </a>
{% endif %}
</td>
</tr>
{% endfor %}
</tbody>
</table>
<script>
$(() => {
$(".timestamp").each(function(){
$(".timestamp").each(function () {
const el = $(this);
const ts = el.text();
const dt = new Date(parseInt(ts * 1000));
@@ -71,4 +80,5 @@
})
</script>
</body>
</html>
</html>

View File

@@ -44,47 +44,6 @@
<u>Cache Status: view status of launched cellxgene servers.</u></a>
</div>
{% if enable_upload %}
<br>
<h1 style="padding-left:35px">
How To Upload Data via HTTP:
</h1>
<ol style="padding-left:85px;">
<li>
Create a folder for your Username:
</li>
<br>
<form action="{{ url_for('make_user') }}" method="post">
Username <input type="text" name="directory">
<input type="submit" value="Create">
</form>
<li>
Create a subdirectory under the selected Folder:
</li>
<br>
<form action="{{ url_for('make_subdir') }}" method="post">
<select name="usernames" id="usernames">
{% for user in users %}
<option value="{{ user }}">{{ user }}</option>
{% endfor %}
</select>
<br>
Subdirectory Name <input type="text" name="directory">
<input type="submit" value="Create">
</form>
<li>Choose a folder to copy your data to, then upload your data file (must be in .h5ad format).</li>
<br>
<form action="{{ url_for('upload_file') }}" method="post" enctype="multipart/form-data">
Type in the name of the directory and subdirectory you wish to upload to, i.e. "USER/cells". <input type="text" name="path">
<br>
File: <input type="file" name="file"><br>
<input style="position:relative; top:10px;" type="submit" value="Upload">
</form>
<br>
<li>Take a look at your data using the file crawler link above</li>
</ol>
{% endif %}
<br>
<h1 style="padding-left:35px">

View File

@@ -1,8 +1,9 @@
import os
import codecs
from setuptools import find_packages, setup
import os
import sys
from setuptools import find_packages, setup
if sys.version_info < (3, 6):
sys.exit("Sorry, Python < 3.6 is not supported")
@@ -25,8 +26,8 @@ def get_version(rel_path):
def parse_requirements():
reqs = []
with open("requirements.txt", "r") as f:
for l in f.readlines():
reqs.append(l.strip("\n"))
for line in f.readlines():
reqs.append(line.strip("\n"))
return reqs
@@ -49,7 +50,7 @@ setup(
license="MIT",
keywords="visualization, genomics",
url="http://github.com/Novartis/cellxgene-gateway",
packages=["cellxgene_gateway"],
packages=find_packages(),
package_data={
"cellxgene_gateway": [
"static/css/homepagestyle.css",

View File

@@ -1,8 +1,14 @@
import unittest
from cellxgene_gateway.cache_entry import CacheEntry, CacheEntryStatus
from cellxgene_gateway.cache_key import CacheKey
from cellxgene_gateway.items.item import ItemType
from cellxgene_gateway.items.file.fileitem import FileItem
from cellxgene_gateway.items.file.fileitem_source import FileItemSource
key = CacheKey("czi/pbmc3k.h5ad", "pbmc3k.h5ad", "tmp.csv")
key = CacheKey(
FileItem("/czi/", "pbmc3k.h5ad", ItemType.h5ad),
FileItemSource("/tmp", "local"),
)
class TestRenderEntry(unittest.TestCase):
@@ -14,16 +20,14 @@ class TestRenderEntry(unittest.TestCase):
actual = CacheEntry.for_key(key, 8000).rewrite_text_content(
"src:url(/static/assets/"
)
expected = (
"src:url(http://localhost:5005/view/czi/pbmc3k.h5ad/static/assets/"
)
expected = "src:url(http://localhost:5005/source/local/view/czi/pbmc3k.h5ad/static/assets/"
self.assertEqual(actual, expected)
def test_GIVEN_absolute_src_THEN_include_path(self):
actual = CacheEntry.for_key(key, 8000).rewrite_text_content(
'<link rel="shortcut icon" href="/static/assets/favicon.ico">'
)
expected = '<link rel="shortcut icon" href="http://localhost:5005/view/czi/pbmc3k.h5ad/static/assets/favicon.ico">'
expected = '<link rel="shortcut icon" href="http://localhost:5005/source/local/view/czi/pbmc3k.h5ad/static/assets/favicon.ico">'
self.assertEqual(actual, expected)

View File

@@ -1,49 +0,0 @@
import unittest
from unittest.mock import MagicMock, patch
from cellxgene_gateway.filecrawl import render_entry
class TestRenderEntry(unittest.TestCase):
def test_GIVEN_path_both_slash_THEN_view_has_single_slash(self):
entry = {
"path": "/somepath/",
"name": "entry",
"type": "file",
"annotations": [],
"children": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
def test_GIVEN_path_starts_slash_THEN_view_has_single_slash(self):
entry = {
"path": "/somepath",
"name": "entry",
"type": "file",
"annotations": [],
"children": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
def test_GIVEN_path_ends_slash_THEN_view_has_single_slash(self):
entry = {
"path": "somepath/",
"name": "entry",
"type": "file",
"annotations": [],
"children": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
def test_GIVEN_path_no_slash_THEN_view_has_single_slash(self):
entry = {
"path": "somepath",
"name": "entry",
"type": "file",
"annotations": [],
"children": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)

View File

@@ -1,46 +1,33 @@
import unittest
from unittest.mock import MagicMock, patch
from cellxgene_gateway.filecrawl import render_entry
from cellxgene_gateway.filecrawl import render_item
from cellxgene_gateway.items.item import ItemType
from cellxgene_gateway.items.file.fileitem import FileItem
from cellxgene_gateway.items.file.fileitem_source import FileItemSource
source = FileItemSource("/tmp")
class TestRenderEntry(unittest.TestCase):
def test_GIVEN_path_both_slash_THEN_view_has_single_slash(self):
entry = {
"path": "/somepath/",
"name": "entry",
"type": "file",
"annotations": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
entry = FileItem(
subpath="/somepath/", name="entry", type=ItemType.h5ad
)
rendered = render_item(entry, source)
self.assertIn("view/somepath/entry/'", rendered)
def test_GIVEN_path_starts_slash_THEN_view_has_single_slash(self):
entry = {
"path": "/somepath",
"name": "entry",
"type": "file",
"annotations": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
entry = FileItem(subpath="/somepath", name="entry", type=ItemType.h5ad)
rendered = render_item(entry, source)
self.assertIn("view/somepath/entry/'", rendered)
def test_GIVEN_path_ends_slash_THEN_view_has_single_slash(self):
entry = {
"path": "somepath/",
"name": "entry",
"type": "file",
"annotations": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
entry = FileItem(subpath="somepath/", name="entry", type=ItemType.h5ad)
rendered = render_item(entry, source)
self.assertIn("view/somepath/entry/'", rendered)
def test_GIVEN_path_no_slash_THEN_view_has_single_slash(self):
entry = {
"path": "somepath",
"name": "entry",
"type": "file",
"annotations": [],
}
rendered = render_entry(entry)
self.assertIn("view/somepath", rendered)
entry = FileItem(subpath="somepath", name="entry", type=ItemType.h5ad)
rendered = render_item(entry, source)
self.assertIn("view/somepath/entry/'", rendered)