def test_get_marker(self):
# Make sure, sorted dicts get the same marker
result1 = get_marker()
result2 = get_marker(options={})
result3 = get_marker(options={'b': '0', 'a': '1'})
result4 = get_marker(options={'a': '1', 'b': '0'})
assert result1 == 'W10'
assert result2 == 'W10'
assert result3 == result4
assert result2 != result3
def test_get_marker(self):
# Make sure, sorted dicts get the same marker
result1 = get_marker()
result2 = get_marker(options={})
result3 = get_marker(options={'b': '0', 'a': '1'})
result4 = get_marker(options={'a': '1', 'b': '0'})
assert result1 == 'W10'
assert result2 == 'W10'
assert result3 == result4
assert result2 != result3
def get_cached_by_source(self, src_doc_path, options={}):
"""Get the document from cache by source doc and options.
Find a cached document, which was created from the given
`src_doc_path` and `options`.
Returns the path to the document and a cache key you are
encouraged to use for future access.
Please note that this method is much more expensive than
:meth:`get_cached`. Use it only if the `cache_key` returned
upon registering a doc is absolutely not available any more.
Returns ``(None, None)`` if no such file can be found or no
cache dir was set at all.
.. warning:: The returned path (if any) is part of cache! Do
not remove or change the file. Copy it to another
location instead.
.. versionadded:: 1.1
"""
repr_key = get_marker(options)
if self.cache_manager is not None:
return self.cache_manager.get_cached_file_by_source(
src_doc_path, repr_key)
return None, None
def get_cached_by_source(self, src_doc_path, options={}):
"""Get the document from cache by source doc and options.
Find a cached document, which was created from the given
`src_doc_path` and `options`.
Returns the path to the document and a cache key you are
encouraged to use for future access.
Please note that this method is much more expensive than
:meth:`get_cached`. Use it only if the `cache_key` returned
upon registering a doc is absolutely not available any more.
Returns ``(None, None)`` if no such file can be found or no
cache dir was set at all.
.. warning:: The returned path (if any) is part of cache! Do
not remove or change the file. Copy it to another
location instead.
.. versionadded:: 1.1
"""
repr_key = get_marker(options)
if self.cache_manager is not None:
return self.cache_manager.get_cached_file_by_source(
src_doc_path, repr_key)
return None, None
def register_fakedoc_in_cache(self, src, options):
# register a fake doc in cache. Result cache_key is based on
# path to src document and options given.
cm = CacheManager(self.cachedir)
fake_result_path = os.path.join(self.workdir, 'result.html')
open(fake_result_path, 'w').write('A fake result.')
marker = get_marker(options)
cache_key = cm.register_doc(src, fake_result_path, repr_key=marker)
return cache_key
def convert_doc(src_doc, options, cache_dir):
"""Convert `src_doc` according to the other parameters.
`src_doc` is the path to the source document. `options` is a dict
of options for processing, passed to the processors.
`cache_dir` may be ``None`` in which no caching is requested
during processing.
Generates a converted representation of `src_doc` by calling
:class:`ulif.openoffice.processor.MetaProcessor` with `options` as
parameters.
Afterwards the conversion result is stored in cache (if
allowed/possible) for speedup of upcoming requests.
Returns a triple:
``(<PATH>, <CACHE_KEY>, <METADATA>)``
where ``<PATH>`` is the path to the resulting document,
``<CACHE_KEY>`` an identifier (string) to retrieve a generated doc
from cache on future requests, and ``<METADATA>`` is a dict of values
returned during request (and set by the document processors,
notably setting the `error` keyword).
If errors happen or caching is disabled, ``<CACHE_KEY>`` is
``None``.
"""
result_path = None
cache_key = None
repr_key = get_marker(options) # Create unique marker out of options
metadata = dict(error=False)
# Generate result
input_copy_dir = tempfile.mkdtemp()
input_copy = os.path.join(input_copy_dir, os.path.basename(src_doc))
shutil.copy2(src_doc, input_copy)
try:
proc = MetaProcessor(options=options) # Removes original doc
result_path, metadata = proc.process(input_copy)
except Exception as exc:
shutil.rmtree(input_copy_dir)
raise exc
error_state = metadata.get('error', False)
if cache_dir and not error_state and result_path is not None:
# Cache away generated doc
cache_key = CacheManager(cache_dir).register_doc(
src_doc, result_path, repr_key)
return result_path, cache_key, metadata
def convert_doc(src_doc, options, cache_dir):
"""Convert `src_doc` according to the other parameters.
`src_doc` is the path to the source document. `options` is a dict
of options for processing, passed to the processors.
`cache_dir` may be ``None`` in which no caching is requested
during processing.
Generates a converted representation of `src_doc` by calling
:class:`ulif.openoffice.processor.MetaProcessor` with `options` as
parameters.
Afterwards the conversion result is stored in cache (if
allowed/possible) for speedup of upcoming requests.
Returns a triple:
``(<PATH>, <CACHE_KEY>, <METADATA>)``
where ``<PATH>`` is the path to the resulting document,
``<CACHE_KEY>`` an identifier (string) to retrieve a generated doc
from cache on future requests, and ``<METADATA>`` is a dict of values
returned during request (and set by the document processors,
notably setting the `error` keyword).
If errors happen or caching is disabled, ``<CACHE_KEY>`` is
``None``.
"""
result_path = None
cache_key = None
repr_key = get_marker(options) # Create unique marker out of options
metadata = dict(error=False)
# Generate result
input_copy_dir = tempfile.mkdtemp()
input_copy = os.path.join(input_copy_dir, os.path.basename(src_doc))
shutil.copy2(src_doc, input_copy)
try:
proc = MetaProcessor(options=options) # Removes original doc
result_path, metadata = proc.process(input_copy)
except Exception as exc:
shutil.rmtree(input_copy_dir)
raise exc
error_state = metadata.get('error', False)
if cache_dir and not error_state and result_path is not None:
# Cache away generated doc
cache_key = CacheManager(cache_dir).register_doc(
src_doc, result_path, repr_key)
return result_path, cache_key, metadata
def test_show_with_cache(self, conv_env):
# we can retrieve cached files
app = RESTfulDocConverter(cache_dir=str(conv_env / "cache"))
conv_env.join("sample_in.txt").write("Fake source.")
conv_env.join("sample_out.pdf").write("Fake result.")
marker = get_marker(dict(foo='bar', bar='baz'))
doc_id = app.cache_manager.register_doc(
source_path=str(conv_env.join("sample_in.txt")),
to_cache=str(conv_env.join("sample_out.pdf")),
repr_key=marker)
assert doc_id == '3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1'
url = 'http://localhost/docs/3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1'
req = Request.blank(url)
resp = app(req)
assert resp.status == "200 OK"
assert resp.content_type == "application/pdf"
def test_show_with_cache(self, conv_env):
# we can retrieve cached files
app = RESTfulDocConverter(cache_dir=str(conv_env / "cache"))
conv_env.join("sample_in.txt").write("Fake source.")
conv_env.join("sample_out.pdf").write("Fake result.")
marker = get_marker(dict(foo='bar', bar='baz'))
doc_id = app.cache_manager.register_doc(
source_path=str(conv_env.join("sample_in.txt")),
to_cache=str(conv_env.join("sample_out.pdf")),
repr_key=marker)
assert doc_id == '3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1'
url = 'http://localhost/docs/3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1'
req = Request.blank(url)
resp = app(req)
assert resp.status == "200 OK"
assert resp.content_type == "application/pdf"
def test_show_with_cache(self):
# we can retrieve cached files
app = RESTfulDocConverter(cache_dir=self.cachedir)
fake_src = os.path.join(self.workdir, 'sample_in.txt')
fake_result = os.path.join(self.workdir, 'sample_out.pdf')
open(fake_src, 'w').write('Fake source.')
open(fake_result, 'w').write('Fake result.')
marker = get_marker(dict(foo='bar', bar='baz'))
doc_id = app.cache_manager.register_doc(
source_path=fake_src, to_cache=fake_result, repr_key=marker)
self.assertEqual('3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1', doc_id)
doc_id = '%s_%s' % (doc_id, marker)
url = 'http://localhost/docs/3fe6f0d4c5e62ff9a1deca0a8a65fe8d_1_1'
req = Request.blank(url)
resp = app(req)
self.assertEqual(resp.status, '200 OK')
self.assertEqual(resp.content_type, 'application/pdf')