def readInfoFile(self,
informationsdata,
filesrc,
srclanguage):
"""
TextDBDict.readInfoFile
Read the informations from the informations'file <src>.
informationsdata : InformationsData object
filesrc : (file descriptor)
Return (bool)success
"""
textdbdictinfofile = TextDBDictInfoFile(logotherasdata = self.logotherasdata,
errors = self.errors)
success = textdbdictinfofile.readFromData(informationsdata = informationsdata,
reading_position = self.reading_position,
filesrc = filesrc,
srclanguage = srclanguage)
return success
def writeTextDBDictInto(self,
destpath,
srclanguage):
"""
TextDBDict.writeTextDBDictInto
destpath : (str)
srclanguage : Language object
"""
self.errors.debug("TextDBDict.writeTextDBDictInto; destpath=" + str(destpath) )
# maximal length for the name of the files :
max_length = \
logotheras.options.OPTIONS["textdbdict::maximal number of characters in a filename"]
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# Is <destpath> valid ?
if not os.path.exists(destpath):
msg = "(ERR021) (TextDBDict.writeTextDBDictInto) Missing directory : '{0}'"
self.errors.error(msg.format(destpath))
return
if not os.path.isdir(destpath):
msg = "(ERR020) (TextDBDict.writeTextDBDictInto) '{0}' is not a directory"
self.errors.error(msg.format(destpath))
return
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# creation of the source directory :
source_dir = os.path.join(destpath,
SOURCEDIRECTORY_DEFAULTNAME)
if os.path.exists(source_dir):
shutil.rmtree(source_dir)
os.makedirs( source_dir )
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# license :
# adding "#COMMENT#" before each line :
sep_license = NEWLINE + "#COMMENT# "
_license = sep_license.join( logotheras.options.LICENSE_TEXT.split("\n") )
# we add "#COMMENT#" before the first line since .join() doesn't.
_license = "#COMMENT# " + _license
# information file :
infofile = TextDBDictInfoFile(logotherasdata = self.logotherasdata,
errors = self.errors)
infofile.writeFromData(destpath = destpath)
# data file :
if logotheras.options.OPTIONS["textdbdict::share out TextDBDict files"] is None:
#~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
# everything is packed into one file :
datafile_name = "unique_file.textdbdict"
datafile_name = datafile_name[:max_length]
with open( os.path.join(source_dir,
datafile_name),
"w" ) as datafile:
for line in _license:
datafile.write(line)
for articledata in self.logotherasdata.getArticles():
article = Article( errors = self.errors,
logotherasdata = self.logotherasdata )
article.setData( articledata )
self.errors.info("writing data into the unique file : " + datafile.name)
datafile.write( article.getTextDBDictRepr() )
elif logotheras.options.OPTIONS["textdbdict::share out TextDBDict files"].startswith('a'):
#~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
# several outputfiles : all the words begin with the same letters
length = len(logotheras.options.OPTIONS["textdbdict::share out TextDBDict files"])
for (datafile_number,
(articles, initial)) in enumerate(self.logotherasdata.getArticlesGroupedBy(
method="initial",
length=length,
srclanguage=srclanguage)):
# name of this new file ?
if not logotheras.options.OPTIONS["textdbdict::" \
"write TextDBDict files with detailed names"]:
# short name :
datafile_name = "{0:0>6}.textdbdict".format(datafile_number)
else:
# long name :
first_article = articles[0]
first_article = first_article.strip()
last_article = articles[-1]
last_article = last_article.strip()
datafile_name = \
logotheras.options.OPTIONS["textdbdict::detailed filename"].format(
datafile_number,
first_article,
last_article)
datafile_name = os.path.join( source_dir, datafile_name )
datafile_name = datafile_name[:max_length]
with open(datafile_name, 'w') as datafile:
self.errors.info("writing data into a new file : " + datafile.name)
for line in _license:
datafile.write(line)
for article_name in articles:
article = Article( errors = self.errors,
logotherasdata = self.logotherasdata )
article.setData( self.logotherasdata[article_name] )
datafile.write( article.getTextDBDictRepr() )
# let's add the articles marked with NOTALEXICAL_ARTINAME in a
# special file.
self.writeAddendum(destpath, _license)
else :
#~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
# several outputfiles : each outputfile contain the same number of articles
length = int(logotheras.options.OPTIONS["textdbdict::share out TextDBDict files"])
for (datafile_number,
articles) in enumerate(self.logotherasdata.getArticlesGroupedBy(
method="articles' number",
length=length,
srclanguage=srclanguage)):
# name of this new file ?
if not logotheras.options.OPTIONS["textdbdict::" \
"write TextDBDict files with detailed names"]:
# short name :
datafile_name = "{0:0>6}.textdbdict".format(datafile_number)
else:
# long name :
first_article = articles[0]
first_article = first_article.strip()
last_article = articles[-1]
last_article = last_article.strip()
datafile_name = \
logotheras.options.OPTIONS["textdbdict::" \
"detailed filename"].format(datafile_number,
first_article,
last_article)
datafile_name = os.path.join( source_dir, datafile_name )
datafile_name = datafile_name[:max_length]
with open(datafile_name, 'w') as datafile:
self.errors.info("writing data into a new file : " + datafile.name)
for line in _license:
datafile.write(line)
for article_name in articles:
article = Article( errors = self.errors,
logotherasdata = self.logotherasdata )
article.setData( self.logotherasdata[article_name] )
datafile.write( article.getTextDBDictRepr() )
# let's add the articles marked with NOTALEXICAL_ARTINAME in a
# special file.
self.writeAddendum(destpath, _license)
