def ocropus_find_file(fname, gz=True):
"""Search for `fname` in one of the OCRopus data directories, as well as
the current directory). If `gz` is True, search also for gzipped files.
Result of searching $fname is the first existing in:
* $base/$fname
* $base/$fname.gz # if gz
* $base/model/$fname
* $base/model/$fname.gz # if gz
* $base/data/$fname
* $base/data/$fname.gz # if gz
* $base/gui/$fname
* $base/gui/$fname.gz # if gz
$base can be four base paths:
* `$OCROPUS_DATA` environment variable
* current working directory
* ../../../../share/ocropus from this file's install location
* `/usr/local/share/ocropus`
* `$PREFIX/share/ocropus` ($PREFIX being the Python installation
prefix, usually `/usr`)
"""
possible_prefixes = []
if os.getenv("OCROPUS_DATA"):
possible_prefixes.append(os.getenv("OCROPUS_DATA"))
possible_prefixes.append(os.curdir)
possible_prefixes.append(os.path.normpath(os.path.join(
os.path.dirname(inspect.getfile(inspect.currentframe())),
os.pardir, os.pardir, os.pardir, os.pardir, "share", "ocropus")))
possible_prefixes.append("/usr/local/share/ocropus")
# datarootdir is None in windows so don't add it to search list
if sysconfig.get_config_var("datarootdir") is not None:
possible_prefixes.append(os.path.join(
sysconfig.get_config_var("datarootdir"), "ocropus"))
# Unique entries with preserved order in possible_prefixes
# http://stackoverflow.com/a/15637398/201318
possible_prefixes = [possible_prefixes[i] for i in
sorted(numpy.unique(possible_prefixes, return_index=True)[1])]
for prefix in possible_prefixes:
if not os.path.isdir(prefix):
continue
for basename in [".", "models", "data", "gui"]:
if not os.path.isdir(os.path.join(prefix, basename)):
continue
full = os.path.join(prefix, basename, fname)
if os.path.exists(full):
return full
if gz and os.path.exists(full + ".gz"):
return full + ".gz"
raise FileNotFound(fname)
def glob_all(args):
"""Given a list of command line arguments, expand all of them with glob."""
result = []
for arg in args:
if arg[0] == "@":
with open(arg[1:], "r") as stream:
expanded = stream.read().split("\n")
expanded = [s for s in expanded if s != ""]
else:
expanded = sorted(glob.glob(arg))
if len(expanded) < 1:
raise FileNotFound("%s: expansion did not yield any files" % arg)
result += expanded
return result
def finddir(name):
"""Find some OCRopus-related resource by looking in a bunch off standard places.
(This needs to be integrated better with setup.py and the build system.)"""
local = getlocal()
path = name
if os.path.exists(path) and os.path.isdir(path): return path
path = local+name
if os.path.exists(path) and os.path.isdir(path): return path
_,tail = os.path.split(name)
path = tail
if os.path.exists(path) and os.path.isdir(path): return path
path = local+tail
if os.path.exists(path) and os.path.isdir(path): return path
raise FileNotFound("file '"+path+"' not found in . or /usr/local/share/ocropus/")