def indexFile(cls, writer, path, baseDir):
input = file(path)
props = {}
while True:
line = input.readline().strip()
if not line:
break
name, value = line.split("=", 1)
props[name] = value.decode("unicode-escape")
input.close()
doc = Document()
# category comes from relative path below the base directory
category = os.path.dirname(path)[len(baseDir) :]
if os.path.sep != "/":
category = category.replace(os.path.sep, "/")
isbn = props["isbn"]
title = props["title"]
author = props["author"]
url = props["url"]
subject = props["subject"]
pubmonth = props["pubmonth"]
print title.encode("utf8")
print author.encode("utf-8")
print subject.encode("utf-8")
print category.encode("utf-8")
print "---------"
doc.add(Field("isbn", isbn, Field.Store.YES, Field.Index.NOT_ANALYZED))
doc.add(Field("category", category, Field.Store.YES, Field.Index.NOT_ANALYZED))
doc.add(Field("title", title, Field.Store.YES, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS))
doc.add(
Field(
"title2",
title.lower(),
Field.Store.YES,
Field.Index.NOT_ANALYZED_NO_NORMS,
Field.TermVector.WITH_POSITIONS_OFFSETS,
)
)
# split multiple authors into unique field instances
authors = author.split(",")
for a in authors:
doc.add(
Field("author", a, Field.Store.YES, Field.Index.NOT_ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS)
)
doc.add(Field("url", url, Field.Store.YES, Field.Index.NOT_ANALYZED_NO_NORMS))
doc.add(
Field("subject", subject, Field.Store.NO, Field.Index.ANALYZED, Field.TermVector.WITH_POSITIONS_OFFSETS)
)
doc.add(NumericField("pubmonth", Field.Store.YES, True).setIntValue(int(pubmonth)))
d = DateTools.stringToDate(pubmonth)
d = int(d.getTime() / (1000 * 3600 * 24.0))
doc.add(NumericField("pubmonthAsDay").setIntValue(d))
doc.add(
Field(
"contents",
" ".join([title, subject, author, category]),
Field.Store.NO,
Field.Index.ANALYZED,
Field.TermVector.WITH_POSITIONS_OFFSETS,
)
)
doc.add(Field("path", path, Field.Store.YES, Field.Index.NOT_ANALYZED))
doc.add(Field("modified", DateField.dateToString(samplesModified), Field.Store.YES, Field.Index.NOT_ANALYZED))
writer.addDocument(doc)
def indexFile(cls, writer, path, baseDir):
input = file(path)
props = {}
while True:
line = input.readline().strip()
if not line:
break
name, value = line.split('=', 1)
props[name] = value.decode('unicode-escape')
input.close()
doc = Document()
# category comes from relative path below the base directory
category = os.path.dirname(path)[len(baseDir):]
if os.path.sep != '/':
category = category.replace(os.path.sep, '/')
isbn = props['isbn']
title = props['title']
author = props['author']
url = props['url']
subject = props['subject']
pubmonth = props['pubmonth']
print title.encode('utf8')
print author.encode('utf-8')
print subject.encode('utf-8')
print category.encode('utf-8')
print "---------"
doc.add(Field("isbn", isbn, Field.Store.YES, Field.Index.NOT_ANALYZED))
doc.add(
Field("category", category, Field.Store.YES,
Field.Index.NOT_ANALYZED))
doc.add(
Field("title", title, Field.Store.YES, Field.Index.ANALYZED,
Field.TermVector.WITH_POSITIONS_OFFSETS))
doc.add(
Field("title2", title.lower(), Field.Store.YES,
Field.Index.NOT_ANALYZED_NO_NORMS,
Field.TermVector.WITH_POSITIONS_OFFSETS))
# split multiple authors into unique field instances
authors = author.split(',')
for a in authors:
doc.add(
Field("author", a, Field.Store.YES, Field.Index.NOT_ANALYZED,
Field.TermVector.WITH_POSITIONS_OFFSETS))
doc.add(
Field("url", url, Field.Store.YES,
Field.Index.NOT_ANALYZED_NO_NORMS))
doc.add(
Field("subject", subject, Field.Store.NO, Field.Index.ANALYZED,
Field.TermVector.WITH_POSITIONS_OFFSETS))
doc.add(
NumericField("pubmonth", Field.Store.YES,
True).setIntValue(int(pubmonth)))
d = DateTools.stringToDate(pubmonth)
d = int(d.getTime() / (1000 * 3600 * 24.0))
doc.add(NumericField("pubmonthAsDay").setIntValue(d))
doc.add(
Field("contents", ' '.join([title, subject, author, category]),
Field.Store.NO, Field.Index.ANALYZED,
Field.TermVector.WITH_POSITIONS_OFFSETS))
doc.add(Field("path", path, Field.Store.YES, Field.Index.NOT_ANALYZED))
doc.add(
Field("modified", DateField.dateToString(samplesModified),
Field.Store.YES, Field.Index.NOT_ANALYZED))
writer.addDocument(doc)
def indexFile(cls, writer, path, baseDir):
input = file(path)
props = {}
while True:
line = input.readline().strip()
if not line:
break
name, value = line.split('=', 1)
props[name] = value.decode('unicode-escape')
input.close()
doc = Document()
# category comes from relative path below the base directory
category = os.path.dirname(path)[len(baseDir):]
if os.path.sep != '/':
category = category.replace(os.path.sep, '/')
isbn = props['isbn']
title = props['title']
author = props['author']
url = props['url']
subject = props['subject']
pubmonth = props['pubmonth']
print title.encode('utf8')
print author.encode('utf-8')
print subject.encode('utf-8')
print category.encode('utf-8')
print "---------"
doc.add(Field("isbn", isbn, StringField.TYPE_STORED))
doc.add(Field("category", category, StringField.TYPE_STORED))
# note: ft should be initialized once and re-used
ft = FieldType()
ft.setIndexed(True)
ft.setTokenized(True)
ft.setStored(True)
ft.setStoreTermVectorPositions(True)
ft.setStoreTermVectorOffsets(True)
ft.freeze()
doc.add(Field("title", title, ft))
ft = FieldType(StringField.TYPE_STORED)
ft.setIndexed(True)
ft.setTokenized(False)
ft.setOmitNorms(True)
ft.setStoreTermVectorPositions(True)
ft.setStoreTermVectorOffsets(True)
doc.add(Field("title2", title.lower(), ft))
# split multiple authors into unique field instances
authors = author.split(',')
ft = FieldType()
ft.setIndexed(True)
ft.setTokenized(False)
ft.setStored(True)
ft.setStoreTermVectorPositions(True)
ft.setStoreTermVectorOffsets(True)
for a in authors:
doc.add(Field("author", a, ft))
ft = FieldType()
ft.setIndexed(True)
ft.setTokenized(False)
ft.setStored(True)
ft.setOmitNorms(True)
doc.add(Field("url", url, ft))
ft = FieldType()
ft.setIndexed(True)
ft.setTokenized(True)
ft.setStored(False)
ft.setStoreTermVectorPositions(True)
ft.setStoreTermVectorOffsets(True)
doc.add(Field("subject", subject, ft))
doc.add(IntField("pubmonth", int(pubmonth), Field.Store.YES))
d = DateTools.stringToDate(pubmonth)
d = int(d.getTime() / (1000 * 3600 * 24.0))
doc.add(IntField("pubmonthAsDay", d, IntField.TYPE_NOT_STORED))
ft = FieldType()
ft.setIndexed(True)
ft.setTokenized(True)
ft.setStored(False)
ft.setStoreTermVectorPositions(True)
ft.setStoreTermVectorOffsets(True)
doc.add(Field("contents", ' '.join([title, subject, author, category]),
ft))
doc.add(Field("path", path,
StringField.TYPE_STORED))
doc.add(Field("modified", DateTools.dateToString(samplesModified, DateTools.Resolution.MILLISECOND),
StringField.TYPE_STORED))
writer.addDocument(doc)