Ejemplo n.º 1
0
    def get_metadata(self,
                     recordID,
                     sandbox=False,
                     exceptions=True,
                     timeout=15):
        if not sandbox:
            url = 'https://zenodo.org/api/records/'
        else:
            url = 'https://sandbox.zenodo.org/api/records/'

        try:
            r = requests.get(url + recordID, timeout=timeout)
        except requests.exceptions.ConnectTimeout:
            self.logger.critical('Connection timeout during metadata reading.')
            raise
        except Exception:
            self.logger.critical('Connection error during metadata reading.')
            raise

        js = {}
        if r.ok:
            js = json.loads(r.text)['files']
            for f in json.loads(r.text)['files']:
                path = 'zenodo' if not sandbox else 'sandbox'
                self.attr_cache[f'/{path}/{recordID}/{f["key"]}'] = SBox(
                    f, default_box=True)

            self.content[f"/{path}/{recordID}.json"] = (
                SBox(metadata=js).to_json() + "\n").encode()
            self.content[f"/{path}/{recordID}.yaml"] = SBox(
                metadata=js).to_yaml().encode()
        return js
Ejemplo n.º 2
0
 def __init__(self,
              recordIDs,
              sandbox_recordIDs,
              chunksize=64,
              largefile=1024):
     self.records = {
         'sandbox': [],
         'zenodo': [],
     }
     self.attr_cache = SBox(default_box=True)
     self.dir_cache = SBox(default_box=True)
     self.open_files = {}
     self.content = {}
     self.chunksize = chunksize
     self.largefile = largefile
     self.logger = logging.getLogger()
     for rid in recordIDs:
         self.get_metadata(rid, sandbox=False)
     for rid in sandbox_recordIDs:
         self.get_metadata(rid, sandbox=True)
Ejemplo n.º 3
0
    def test_property_box(self):
        td = test_dict.copy()
        td["inner"] = {"CamelCase": "Item"}

        pbox = SBox(td, camel_killer_box=True)
        assert isinstance(pbox.inner, SBox)
        assert pbox.inner.camel_case == "Item"
        assert json.loads(pbox.json)["inner"]["camel_case"] == "Item"
        test_item = yaml.load(pbox.yaml, Loader=yaml.SafeLoader)
        assert test_item["inner"]["camel_case"] == "Item"
        assert repr(pbox["inner"]).startswith("<ShorthandBox")
        assert not isinstance(pbox.dict, Box)
        assert pbox.dict["inner"]["camel_case"] == "Item"
        assert pbox.toml.startswith('key1 = "value1"')
Ejemplo n.º 4
0
    def test_property_box(self):
        td = test_dict.copy()
        td['inner'] = {'CamelCase': 'Item'}

        pbox = SBox(td, camel_killer_box=True)
        assert isinstance(pbox.inner, SBox)
        assert pbox.inner.camel_case == 'Item'
        assert json.loads(pbox.json)['inner']['camel_case'] == 'Item'
        test_item = yaml.load(pbox.yaml, Loader=yaml.SafeLoader)
        assert test_item['inner']['camel_case'] == 'Item'
        assert repr(pbox['inner']).startswith('<ShorthandBox')
        assert not isinstance(pbox.dict, Box)
        assert pbox.dict['inner']['camel_case'] == 'Item'
        assert pbox.toml.startswith('key1 = "value1"')
Ejemplo n.º 5
0
    'frozen': 'box',
}, frozen_box=True)
# frozen_box.frozen = 'Box'
# box.BoxError: Box is frozen

empty_box = Box(default_box=True)

print(empty_box.a.b.c)
# <Box: {}>

empty_box.a.b.c.d = "h"
print(empty_box)

camel_killer_box = Box(HelloWorld="HELLO, WORLD!", camel_killer_box=True)
print(camel_killer_box.hello_world)

box_of_order = Box(ordered_box=True)
box_of_order.c = 1
box_of_order.a = 2
box_of_order.d = 3
print(box_of_order.keys() == ['c', 'a', 'd'])

from box import BoxList
box_list = BoxList({'item': x} for x in range(10))

print(box_list[5].item)

from box import SBox
s_box = SBox(hello='world')
print(s_box.json)
Ejemplo n.º 6
0
class ZenodoFS(LoggingMixIn, Operations):
    def __init__(self,
                 recordIDs,
                 sandbox_recordIDs,
                 chunksize=64,
                 largefile=1024):
        self.records = {
            'sandbox': [],
            'zenodo': [],
        }
        self.attr_cache = SBox(default_box=True)
        self.dir_cache = SBox(default_box=True)
        self.open_files = {}
        self.content = {}
        self.chunksize = chunksize
        self.largefile = largefile
        self.logger = logging.getLogger()
        for rid in recordIDs:
            self.get_metadata(rid, sandbox=False)
        for rid in sandbox_recordIDs:
            self.get_metadata(rid, sandbox=True)

    @cache(maxsize=1024)
    def get_metadata(self,
                     recordID,
                     sandbox=False,
                     exceptions=True,
                     timeout=15):
        if not sandbox:
            url = 'https://zenodo.org/api/records/'
        else:
            url = 'https://sandbox.zenodo.org/api/records/'

        try:
            r = requests.get(url + recordID, timeout=timeout)
        except requests.exceptions.ConnectTimeout:
            self.logger.critical('Connection timeout during metadata reading.')
            raise
        except Exception:
            self.logger.critical('Connection error during metadata reading.')
            raise

        js = {}
        if r.ok:
            js = json.loads(r.text)['files']
            for f in json.loads(r.text)['files']:
                path = 'zenodo' if not sandbox else 'sandbox'
                self.attr_cache[f'/{path}/{recordID}/{f["key"]}'] = SBox(
                    f, default_box=True)

            self.content[f"/{path}/{recordID}.json"] = (
                SBox(metadata=js).to_json() + "\n").encode()
            self.content[f"/{path}/{recordID}.yaml"] = SBox(
                metadata=js).to_yaml().encode()
        return js

    def readdir(self, path, fh):
        level = len(path.split('/'))
        content = [
            name for name in self.attr_cache.keys() if name.startswith(path)
        ]
        if path == '/':
            return ['.', 'sandbox', 'zenodo']

        elif path in ('/sandbox', '/zenodo'):
            content = [
                name for name in self.attr_cache.keys()
                if name.startswith(path)
            ]
        else:
            parts = path.split('/')
            if len(parts) >= 3:
                recordID = parts[2]
                self.get_metadata(recordID)
                content = [
                    name for name in self.attr_cache.keys()
                    if name.startswith(path)
                ]

        N = len(path) + 1
        content = list({
            name[N:].split('/')[0]
            for name in content if len(name) > N and name[N - 1] == '/'
        })
        if level == 2:
            content = content + [
                f"{name}.yaml" for name in content if name.find('.') == -1
            ] + [f"{name}.json" for name in content if name.find('.') == -1]
        return list(set(content))

    def getattr(self, path, fh=None):
        parts = path.split('/')
        level = len(parts)
        st = {}
        if path in ['/', '/sandbox', '/zenodo']:
            st['st_mode'] = (S_IFDIR | 0o755)
            st['st_nlink'] = 2
        elif level == 3:
            if path.find('.') > -1:
                size = len(self.content[path])
                st = {'st_mode': (S_IFREG | 0o444), 'st_size': size}
            else:
                st['st_mode'] = (S_IFDIR | 0o755)
                st['st_nlink'] = 2
        else:
            size = 0
            st = {'st_mode': (S_IFREG | 0o444), 'st_size': size}
            if level >= 3:
                recordID = parts[2]
                self.get_metadata(recordID)

            if level == 4:
                fn = self.attr_cache[path]
                if 'size' in fn:
                    st['st_size'] = fn['size']
        st['st_ctime'] = st['st_mtime'] = st['st_atime'] = time()
        return st

    def open(self, path, mode):
        if path not in self.open_files:
            url = self.attr_cache[path]['links'].get('self')
            size = self.attr_cache[path].get('size', 0)
            self.open_files[path] = WebFile(url, size, self.chunksize,
                                            self.largefile)
        return 0

    def read(self, path, size, offset, fh):
        if path in self.content:
            return self.content[path][offset:offset + size]

        return self.open_files[path][offset:offset + size]

    def release(self, path, fh):
        if path in self.open_files:
            wf = self.open_files.pop(path)
            wf.close()