def _query_parser_help(self):
field_docs = dict(
classifiers="""
The <a href="https://pypi.org/pypi?%3Aaction=list_classifiers" target="_blank">trove classifiers</a> of a package.
Use single quotes to specify a classifier, as they contain spaces:
<code>classifiers:'Programming Language :: Python :: 3'</code>""",
index="The name of the index. This is only the name part, without the user. For example: <code>index:pypi</code>",
keywords="The keywords of a package.",
name="The package name. For example: <code>name:devpi-client</code>",
path="The path of the package in the form '/{user}/{index}/{name}'. For example: <code>path:/root/pypi/devpi-server</code>",
text=None,
type="""
The type of text.
One of <code>project</code> for the project name,
<code>title</code> for the title of a documentation page,
<code>page</code> for a documentation page,
or one of the following project metadata fields:
<code>author</code>, <code>author_email</code>,
<code>description</code>, <code>keywords</code>,
<code>summary</code>. For example: <code>type:page</code>
""",
user="******")
schema = self.project_schema
fields = []
for name in schema.names():
field = schema[name]
if not field.indexed:
continue
if name not in field_docs:
fields.append((name, "Undocumented"))
continue
field_doc = field_docs[name]
if field_doc is None:
continue
fields.append((name, field_doc))
fields_doc = "<dl>%s</dl>" % ''.join("<dt><code>%s</code></dt><dd>%s</dd>" % x for x in fields)
return {
plugins.WhitespacePlugin:
None,
plugins.SingleQuotePlugin: """
To specify a term which contains spaces, use single quotes like this:
<code>'term with spaces'</code>""",
plugins.FieldsPlugin: """
By using a search like <code>fieldname:term</code>,
you can search in the following fields:<br />%s""" % fields_doc,
plugins.PrefixPlugin: """
End a term with an asterisk to search by prefix like this: <code>path:/fschulze/*</code>""",
plugins.GroupPlugin: """
Group query clauses with parentheses.""",
plugins.OperatorsPlugin: """
Use the <code>AND</code>, <code>OR</code>,
<code>ANDNOT</code>, <code>ANDMAYBE</code>, and <code>NOT</code><br />
operators to further refine your search.<br />
Write them in all capital letters, otherwise they will be interpreted as search terms.<br />
An example search would be: <code>devpi ANDNOT client</code>""",
plugins.BoostPlugin: """
Boost a term by adding a circumflex followed by the boost value like this:
<code>term^2</code>"""}
def instant_search():
results = list()
search_term_original = request.form["search"]
search_term_original = search_term_original.strip().strip("*")
search_term = search_term_original.lower()
if " " in search_term:
for word in search_term.split():
search_term = ("*%s*" % word).join(search_term.split(word))
else:
search_term = "*%s*" % search_term_original
for model in get_model_classes():
fields = list()
for field in model()._get_indexable_columns():
if field != "id":
fields.append(field)
fields.append("model_type")
mparser = MultifieldParser(fields, schema=model()._get_schema())
query = mparser.parse(search_term)
with model()._get_index().searcher() as searcher:
for h in searcher.search(query, terms=True):
model_id, model_type, ci_name, score = h["model_id"], h[
"model_name"], h["name"], h.score
matched_terms = list()
for field, text in h.matched_terms():
field = "<code>%s</code>" % pretty_print(field)
for search_term_word in search_term_original.split():
text = ("<mark class='bg-danger'>%s</mark>" %
search_term_word).join(
text.split(search_term_word))
matched_terms.append(": ".join([field, text]))
model_name = pretty_print(model_type)
results.append((score, model_id, model_name, model_type,
ci_name, matched_terms))
results = sorted(results, key=lambda attr: attr[0], reverse=True)
html_results = ""
for score, model_id, model_name, model_type, ci_name, matched_terms in results:
if model_id is None or model_id == u"None":
continue
html_results += "<tr>"
html_results += '<td><a href="/assets/edit/%s/%s">%s</a></td>' % (
model_type.lower(), model_id, model_name)
html_results += '<td><a href="/assets/edit/%s/%s">%s</a></td>' % (
model_type.lower(), model_id, ci_name)
html_results += '<td><a href="/assets/edit/%s/%s">' % (
model_type.lower(), model_id)
for matched_term in matched_terms:
html_results += "<p>" + matched_term + "</p>"
html_results += "</a></td></tr>"
return html_results
def search_index(self, search_term):
model_index = self._get_index()
schema = self._get_schema()
fields = list()
for field in self._get_indexable_columns():
if field == "id":
field = "model_id"
value = getattr(self, field)
# Do not search the primary key
if not value.primary_key:
fields.append(field)
parser = qparser.MultifieldParser(fields, schema)
query = parser.parse(search_term)
with model_index.searcher() as searcher:
results = searcher.search(query)
for result in results:
yield result
def instant_search():
results = list()
search_term_original = request.form["search"]
search_term_original = search_term_original.strip().strip("*")
search_term = search_term_original.lower()
if " " in search_term:
for word in search_term.split():
search_term = ("*%s*" % word).join(search_term.split(word))
else:
search_term = "*%s*" % search_term_original
for model in get_model_classes():
fields = list()
for field in model()._get_indexable_columns():
if field != "id":
fields.append(field)
fields.append("model_type")
mparser = MultifieldParser(fields, schema=model()._get_schema())
query = mparser.parse(search_term)
with model()._get_index().searcher() as searcher:
for h in searcher.search(query, terms=True):
model_id, model_type, ci_name, score = h["model_id"], h["model_name"], h["name"], h.score
matched_terms = list()
for field, text in h.matched_terms():
field = "<code>%s</code>" % pretty_print(field)
for search_term_word in search_term_original.split():
text = ("<mark class='bg-danger'>%s</mark>" % search_term_word).join(text.split(search_term_word))
matched_terms.append(": ".join([field, text]))
model_name = pretty_print(model_type)
results.append((score, model_id, model_name, model_type, ci_name, matched_terms))
results = sorted(results, key=lambda attr: attr[0], reverse=True)
html_results = ""
for score, model_id, model_name, model_type, ci_name, matched_terms in results:
if model_id is None or model_id == u"None":
continue
html_results += "<tr>"
html_results += '<td><a href="/assets/edit/%s/%s">%s</a></td>' % (model_type.lower(), model_id, model_name)
html_results += '<td><a href="/assets/edit/%s/%s">%s</a></td>' % (model_type.lower(), model_id, ci_name)
html_results += '<td><a href="/assets/edit/%s/%s">' % (model_type.lower(), model_id)
for matched_term in matched_terms:
html_results += "<p>" + matched_term + "</p>"
html_results += "</a></td></tr>"
return html_results
def Map(Qinput, concept):
# Check if not correct syntax
# IF THERE ARE NOT FIELDS
if Qinput.find('[') == -1 or Qinput.find(']') == -1:
return ('top', ['top'])
# ==========
# Replace 'and','or'
Qinput = Qinput.replace(' and', ' ')
Qinput = Qinput.replace(' or', ' ')
# Replace 'AND', 'OR'
if Qinput.strip().split()[-1] == 'AND' or Qinput.strip().split(
)[-1] == 'OR':
Qinput = ' '.join(Qinput.strip().split()[0:-1])
#print Qinput
map1 = Qinput
map1 = map1.replace('[', '|')
map1 = map1.replace(']', '|')
map2 = map1.split('|')
#print map2
#print len(map2)
terms = []
fields = []
conj = []
for index1 in range(0, len(map2), 2):
if len(map2[index1].strip()) > 0:
text = map2[index1]
if text.find('AND') >= 0:
conj.append('AND')
text_str = text.replace('AND', '')
elif text.find('OR') >= 0:
conj.append('OR')
text_str = text.replace('OR', '')
else:
text_str = text
terms.append(text_str.strip())
if index1 + 1 < len(map2):
fields.append(map2[index1 + 1])
#print terms
#print fields
#print conj
# Fix parenthesis, first remove all parenthesis
for i in range(0, len(terms)):
terms[i] = terms[i].replace("(", '').strip()
terms[i] = terms[i].replace(")", '').strip()
if terms[-1] == '':
terms = terms[0:-1]
#print "=========="
#print terms
#print fields
#print conj
# ==============================================
# Convert to Whoosh query
# ==============================================
schema = Schema(IDName=TEXT(stored=True),
path=ID(stored=True),
title=TEXT(stored=True),
desc=TEXT(stored=True),
Type=TEXT(stored=True),
cohort=NUMERIC(stored=True),
inexclude=TEXT(stored=True),
platform=TEXT(stored=True),
MESHterm=TEXT(stored=True),
history=TEXT(stored=True),
attributes=TEXT(stored=True),
topic=TEXT(stored=True),
disease=TEXT(stored=True),
measurement=TEXT(stored=True),
demographics=TEXT(stored=True),
geography=TEXT(stored=True),
age=TEXT(stored=True),
gender=TEXT(stored=True),
category=TEXT(stored=True),
IRB=TEXT(stored=True),
ConsentType=TEXT(stored=True),
phen=TEXT(stored=True),
phenID=TEXT(stored=True),
phenDesc=TEXT(stored=True),
phenName=TEXT(stored=True),
phenCUI=TEXT(stored=True),
phenMap=TEXT(stored=True),
AgeMin=NUMERIC(stored=True),
AgeMax=NUMERIC(stored=True),
MaleNum=NUMERIC(stored=True),
FemaleNum=NUMERIC(stored=True),
OtherGenderNum=NUMERIC(stored=True),
UnknownGenderNum=NUMERIC(stored=True),
Demographics=TEXT(stored=True),
phenType=TEXT(stored=True))
## Convert into Advanced Search
MainQuery = ''
count = 0
for iQuery in terms:
# Default MParser
mparser = MultifieldParser([
"IDName", "path", "title", "desc", "Type", "cohort", "platform",
"topic", "disease", "measurement", "demographics", "geography",
"age", "gender", "category", "phenID", "phenName", "phenDesc",
"phenCUI", "phenMap", "Demographics", "phenType"
], schema)
# Work with SampleSize
if fields[count].find('SampleSize') >= 0:
MinSize = terms[count].split(',')[0].strip('(')
MaxSize = terms[count].split(',')[1].strip(')')
if MinSize.find('*') >= 0:
MinSizeN = 0
else:
MinSizeN = int(MinSize)
if MaxSize.find('*') >= 0:
MaxSizeN = 100000
else:
MaxSizeN = int(MaxSize)
if count < len(terms) - 1:
MainQuery += 'cohort:' + '[' + str(MinSizeN) + ' to ' + str(
MaxSizeN) + '] ' + conj[count] + ' '
else:
MainQuery += 'cohort:' + '[' + str(MinSizeN) + ' to ' + str(
MaxSizeN) + ']'
# Work with Age
if fields[count] == 'Age':
#print terms
MinAge = terms[count].strip().split(',')[0].strip('(')
MaxAge = terms[count].strip().split(',')[1].strip(')')
if MinAge.find('*') >= 0:
MinAgeN = 0
else:
MinAgeN = int(MinAge)
if MaxAge.find('*') >= 0:
MaxAgeN = 150
else:
MaxAgeN = int(MaxAge)
if count < len(terms) - 1:
MainQuery += 'AgeMin:' + '[' + str(
MinAgeN) + ' to 150] ' + ' AND AgeMax:' + '[0 to ' + str(
MaxAgeN) + '] ' + conj[count] + ' '
else:
MainQuery += 'AgeMin:' + '[' + str(
MinAgeN) + ' to 150] ' + ' AND AgeMax:' + '[0 to ' + str(
MaxAgeN) + ']'
# Work with StudySubject
if fields[count].find('StudySubject') >= 0:
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (Type:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Type:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Type:' + terms[count].strip(
) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' Type:' + terms[count].strip() + ') '
else:
MainQuery += ' Type:' + terms[count].strip() + ' '
# Work with Ethnicity
if fields[count].find('Ethnicity') >= 0:
#print terms[count]
if terms[count].find('Hispanic') == 0:
terms[count] = 'Hispanic'
elif terms[count].find('Not Hispanic') == 0:
terms[count] = 'NonLatino'
else:
terms[count] = 'Hispanic OR Demographics:NonLatino'
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (Demographics:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Demographics:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Demographics:' + terms[count].strip(
) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' Demographics:' + terms[count].strip() + ') '
else:
MainQuery += ' Demographics:' + terms[count].strip() + ' '
# Work with Platform
if fields[count].find('Platform') >= 0:
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '") ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' platform:"' + terms[count].strip() + '") '
else:
MainQuery += ' platform:"' + terms[count].strip() + '" '
# Work with DataAnalysisMethod
if fields[count].find('DataAnalysisMethod') >= 0:
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '") ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' platform:"' + terms[count].strip() + '") '
else:
MainQuery += ' platform:"' + terms[count].strip() + '" '
# Work with Machine
if fields[count].find('Machine') >= 0:
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '") ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' platform:"' + terms[count].strip() + '") '
else:
MainQuery += ' platform:"' + terms[count].strip() + '" '
# Work with SenquencingTechnique
if fields[count].find('SequencingTechnique') >= 0:
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '" ' + conj[count]
else:
MainQuery += ' platform:"' + terms[count].strip(
) + '") ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' platform:"' + terms[count].strip() + '") '
else:
MainQuery += ' platform:"' + terms[count].strip() + '" '
# Work with IRB
if fields[count].find('IRB') >= 0:
if terms[count] == 'Not Required':
terms[count] = 'No'
if count < len(terms) - 1:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += '(IRB:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' IRB:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' IRB:' + terms[count].strip(
) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' IRB:' + terms[count].strip() + ') '
else:
MainQuery += ' IRB:' + terms[count].strip() + ' '
# Work with Consent Type
if fields[count].find('Consent') >= 0:
#print terms
if terms[count] == 'Unrestricted':
terms[count] = 'No'
if terms[count] == 'Restricted':
terms[count] = 'Restricted'
if terms[count] == 'Unspecified':
terms[count] = 'Unspecified'
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (ConsentType:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' ConsentType:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' ConsentType:' + terms[count].strip(
) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' ConsentType:' + terms[count].strip() + ') '
else:
MainQuery += ' ConsentType:' + terms[count].strip() + ' '
#if count<len(terms)-1:
# MainQuery+= ' ConsentType:' + terms[count].strip() + ' ' +conj[count]
#else:
# MainQuery+= ' ConsentType:' + terms[count].strip()
# Work with Nationality
if fields[count].find('Nationality') >= 0:
mparser = MultifieldParser(["Demographics", "demographics"],
schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
#if count<=len(terms)-2:
# MainQuery+=str(query) + ' ' +conj[count] + ' '
#else:
# MainQuery+=str(query)
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' ( ' + str(query) + ' ' + conj[count]
else:
MainQuery += ' ' + str(query) + ' ' + conj[count]
else:
MainQuery += ' ' + str(query) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' ' + str(query) + ') '
else:
MainQuery += ' ' + str(query) + ' '
# Work with Gender
if fields[count].find('Sex') >= 0:
if terms[count] == 'Male':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] '
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (MaleNum:[1 to] ' + conj[count]
else:
MainQuery += ' MaleNum:[1 to] ' + conj[count]
else:
MainQuery += ' MaleNum:[1 to]) ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' MaleNum:[1 to]) '
else:
MainQuery += ' MaleNum:[1 to] '
if terms[count] == 'Female':
#if count<len(terms)-1:
# MainQuery+= ' FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' FemaleNum:[1 to] '
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (FemaleNum:[1 to] ' + conj[count]
else:
MainQuery += ' FemaleNum:[1 to] ' + conj[count]
else:
MainQuery += ' FemaleNum:[1 to]) ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' FemaleNum:[1 to]) '
else:
MainQuery += ' FemaleNum:[1 to] '
if terms[count] == 'Both':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] AND FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] AND FemaleNum:[1 to] '
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (MaleNum:[1 to] AND FemaleNum:[1 to] ' + conj[
count]
else:
MainQuery += ' MaleNum:[1 to] AND FemaleNum:[1 to] ' + conj[
count]
else:
MainQuery += ' MaleNum:[1 to] AND FemaleNum:[1 to]) ' + conj[
count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' MaleNum:[1 to] AND FemaleNum:[1 to]) '
else:
MainQuery += ' MaleNum:[1 to] AND FemaleNum:[1 to] '
if terms[count] == 'Either':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] OR FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] OR FemaleNum:[1 to] '
if count <= len(terms) - 2:
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (MaleNum:[1 to] OR FemaleNum:[1 to] ' + conj[
count]
else:
MainQuery += ' MaleNum:[1 to] OR FemaleNum:[1 to] ' + conj[
count]
else:
MainQuery += ' MaleNum:[1 to] OR FemaleNum:[1 to]) ' + conj[
count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' MaleNum:[1 to] OR FemaleNum:[1 to]) '
else:
MainQuery += ' MaleNum:[1 to] OR FemaleNum:[1 to] '
# Work with Study Design
if fields[count].find('StudyDesign') >= 0:
if terms[count].find('Genome Wide Association Study') >= 0:
terms[count] = 'gwas'
if terms[count] == 'Case-Control Study':
terms[count] = 'Case-Control'
if terms[count] == 'Cross Sectional Study':
terms[count] = 'Cross Sectional'
if terms[count] == 'Double Blind Study':
terms[count] = 'Double Blind'
if terms[count] == 'Interventional Studies':
terms[count] = 'Interventional'
if terms[count] == 'Longitudinal Cohort Study':
terms[count] = 'Longitudinal'
if terms[count] == 'Mendelian Randomized':
terms[count] = 'Mendelian'
if terms[count] == 'Multicenter Study':
terms[count] = 'Multicenter'
if terms[count] == 'Nested Case Control Study':
terms[count] = 'Nested Case Control'
if terms[count] == 'Observational Studies':
terms[count] = 'Observational'
if terms[count] == 'Partial Factorial Randomized Trial':
terms[count] = 'Partial Factorial Randomized'
if terms[count] == 'Placebo Controlled Study':
terms[count] = 'Placebo Controlled'
if terms[count] == 'Population Based Study':
terms[count] = 'Population Based Control'
if terms[count] == 'Prospective Cohort Study':
terms[count] = 'Prospective'
if terms[count] == 'Quantitative Cross Sectional Study':
terms[count] = 'Quantitative Cross Sectional'
if terms[count] == 'Randomized Trial':
terms[count] = 'Randomized'
if terms[count] == 'Phase 1':
terms[count] = 'Phase1'
if terms[count] == 'Phase 2':
terms[count] = 'Phase2'
if terms[count] == 'Phase 3':
terms[count] = 'Phase3'
if count <= len(terms) - 2:
#MainQuery+= ' Type:' + terms[count].strip() + ' ' +conj[count]
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (Type:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Type:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Type:' + terms[count].strip(
) + ') ' + conj[count]
else:
#MainQuery+= ' Type:' + terms[count].strip() + ' '
if fields[count] == fields[count - 1]:
MainQuery += ' Type:' + terms[count].strip() + ') '
else:
MainQuery += ' Type:' + terms[count].strip() + ' '
#print MainQuery
# Work with Race
if fields[count].find('Race') >= 0:
#if terms[count]=='White':
# terms[count]='white OR Demographics:cacausian '
if terms[count] == 'Mixed Race':
terms[count] = 'Multiple'
# Combine all queries together
if count <= len(terms) - 2:
#MainQuery+= ' Demographics:' + terms[count].strip() + ' ' +conj[count]
if fields[count] == fields[count + 1]:
if fields[count - 1] != fields[count]:
MainQuery += ' (Demographics:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += ' Demographics:' + terms[count].strip(
) + ' ' + conj[count]
else:
MainQuery += 'Demographics:' + terms[count].strip(
) + ') ' + conj[count]
else:
if fields[count] == fields[count - 1]:
MainQuery += ' Demographics:' + terms[count].strip() + ') '
else:
MainQuery += ' Demographics:' + terms[count].strip() + ' '
#else:
# MainQuery+= ' Demographics:' + terms[count].strip()
#print MainQuery
# ======================================
# Work with Study
if fields[count].strip() == 'Study':
mparser = MultifieldParser(["IDName", "path", "title", "desc"],
schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# Work with Study Name
if fields[count].find('Study Name') >= 0:
mparser = MultifieldParser(["title"], schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# Work with Study ID
if fields[count].find('Study ID') >= 0:
mparser = MultifieldParser(["IDName"], schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# ======================================
# Work with Variable
if fields[count] == 'Variable':
mparser = MultifieldParser(
["phenID", "phenName", "phenType", "phenDesc"], schema)
query = mparser.parse(unicode(iQuery))
if count < len(terms) - 1:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# Work with Variable Description
if fields[count].find('Variable Desc') >= 0:
mparser = MultifieldParser(["phenDesc"], schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# Work with Variable Name
if fields[count].find('Variable Name') >= 0:
mparser = MultifieldParser(["phenName"], schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# Work with Variable ID
if fields[count].find('Variable ID') >= 0:
mparser = MultifieldParser(["phenID"], schema)
query = mparser.parse(unicode(iQuery))
# Combine all queries together
if count <= len(terms) - 2:
MainQuery += str(query) + ' ' + conj[count] + ' '
else:
MainQuery += str(query)
# ======================================
# Work with All Fields
if fields[count].find('All Fields') >= 0:
# Added on May 21, 2013
query = iQuery
if concept == True:
query_temp = QueryAnalys(query, concept)
else:
query_temp = query
# Combine all queries together
#if count<=len(terms)-2:
# MainQuery+=str(query) + ' ' +conj[count] + ' '
#else:
# MainQuery+=str(query)
if count <= len(terms) - 2:
MainQuery += query_temp + ' ' + conj[count] + ' '
else:
MainQuery += query_temp + ' '
count += 1
return (MainQuery, terms)
def Map(Qinput,concept):
# Check if not correct syntax
# IF THERE ARE NOT FIELDS
if Qinput.find('[')==-1 or Qinput.find(']')==-1:
return ('top',['top'])
# ==========
# Replace 'and','or'
Qinput=Qinput.replace(' and',' ')
Qinput=Qinput.replace(' or',' ')
# Replace 'AND', 'OR'
if Qinput.strip().split()[-1]=='AND' or Qinput.strip().split()[-1]=='OR':
Qinput = ' '.join(Qinput.strip().split()[0:-1])
#print Qinput
map1 = Qinput
map1 = map1.replace('[','|')
map1 = map1.replace(']','|')
map2 = map1.split('|')
#print map2
#print len(map2)
terms=[]
fields=[]
conj = []
for index1 in range(0,len(map2),2):
if len(map2[index1].strip())>0:
text = map2[index1]
if text.find('AND')>=0:
conj.append('AND')
text_str = text.replace('AND','')
elif text.find('OR')>=0:
conj.append('OR')
text_str = text.replace('OR','')
else:
text_str = text
terms.append(text_str.strip())
if index1+1<len(map2):
fields.append(map2[index1+1])
#print terms
#print fields
#print conj
# Fix parenthesis, first remove all parenthesis
for i in range(0,len(terms)):
terms[i]=terms[i].replace("(",'').strip()
terms[i]=terms[i].replace(")",'').strip()
if terms[-1]=='':
terms=terms[0:-1]
#print "=========="
#print terms
#print fields
#print conj
# ==============================================
# Convert to Whoosh query
# ==============================================
schema = Schema(IDName=TEXT(stored=True),path=ID(stored=True), title=TEXT(stored=True), desc=TEXT(stored=True), Type=TEXT(stored=True), cohort=NUMERIC(stored=True), inexclude=TEXT(stored=True), platform=TEXT(stored=True), MESHterm=TEXT(stored=True), history=TEXT(stored=True), attributes=TEXT(stored=True), topic=TEXT(stored=True),disease=TEXT(stored=True),measurement=TEXT(stored=True),demographics=TEXT(stored=True),geography=TEXT(stored=True),age=TEXT(stored=True),gender=TEXT(stored=True),category=TEXT(stored=True),IRB=TEXT(stored=True),ConsentType=TEXT(stored=True),phen=TEXT(stored=True),phenID=TEXT(stored=True),phenDesc=TEXT(stored=True),phenName=TEXT(stored=True),phenCUI=TEXT(stored=True),phenMap=TEXT(stored=True), AgeMin=NUMERIC(stored=True), AgeMax=NUMERIC(stored=True), MaleNum=NUMERIC(stored=True), FemaleNum=NUMERIC(stored=True), OtherGenderNum=NUMERIC(stored=True), UnknownGenderNum=NUMERIC(stored=True), Demographics=TEXT(stored=True), phenType=TEXT(stored=True))
## Convert into Advanced Search
MainQuery=''
count = 0
for iQuery in terms:
# Default MParser
mparser = MultifieldParser(["IDName","path","title","desc","Type","cohort","platform","topic","disease","measurement","demographics","geography","age","gender","category","phenID","phenName","phenDesc","phenCUI","phenMap","Demographics","phenType"], schema)
# Work with SampleSize
if fields[count].find('SampleSize')>=0:
MinSize = terms[count].split(',')[0].strip('(')
MaxSize = terms[count].split(',')[1].strip(')')
if MinSize.find('*')>=0:
MinSizeN=0
else:
MinSizeN=int(MinSize)
if MaxSize.find('*')>=0:
MaxSizeN=100000
else:
MaxSizeN=int(MaxSize)
if count<len(terms)-1:
MainQuery+= 'cohort:' + '[' + str(MinSizeN) + ' to ' + str(MaxSizeN) + '] ' + conj[count] + ' '
else:
MainQuery+= 'cohort:' + '[' + str(MinSizeN) + ' to ' + str(MaxSizeN) + ']'
# Work with Age
if fields[count]=='Age':
#print terms
MinAge = terms[count].strip().split(',')[0].strip('(')
MaxAge = terms[count].strip().split(',')[1].strip(')')
if MinAge.find('*')>=0:
MinAgeN=0
else:
MinAgeN=int(MinAge)
if MaxAge.find('*')>=0:
MaxAgeN=150
else:
MaxAgeN=int(MaxAge)
if count<len(terms)-1:
MainQuery+= 'AgeMin:' + '[' + str(MinAgeN) + ' to 150] ' + ' AND AgeMax:' + '[0 to ' + str(MaxAgeN) + '] ' + conj[count] + ' '
else:
MainQuery+= 'AgeMin:' + '[' + str(MinAgeN) + ' to 150] ' + ' AND AgeMax:' + '[0 to ' + str(MaxAgeN) + ']'
# Work with StudySubject
if fields[count].find('StudySubject')>=0:
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (Type:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Type:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Type:' + terms[count].strip() + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' Type:' + terms[count].strip() + ') '
else:
MainQuery+=' Type:' + terms[count].strip() + ' '
# Work with Ethnicity
if fields[count].find('Ethnicity')>=0:
#print terms[count]
if terms[count].find('Hispanic')==0:
terms[count]='Hispanic'
elif terms[count].find('Not Hispanic')==0:
terms[count]='NonLatino'
else:
terms[count]='Hispanic OR Demographics:NonLatino'
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (Demographics:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Demographics:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Demographics:' + terms[count].strip() + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' Demographics:' + terms[count].strip() + ') '
else:
MainQuery+=' Demographics:' + terms[count].strip() + ' '
# Work with Platform
if fields[count].find('Platform')>=0:
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '") ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' platform:"' + terms[count].strip() + '") '
else:
MainQuery+=' platform:"' + terms[count].strip() + '" '
# Work with DataAnalysisMethod
if fields[count].find('DataAnalysisMethod')>=0:
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '") ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' platform:"' + terms[count].strip() + '") '
else:
MainQuery+=' platform:"' + terms[count].strip() + '" '
# Work with Machine
if fields[count].find('Machine')>=0:
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '") ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' platform:"' + terms[count].strip() + '") '
else:
MainQuery+=' platform:"' + terms[count].strip() + '" '
# Work with SenquencingTechnique
if fields[count].find('SequencingTechnique')>=0:
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '" ' +conj[count]
else:
MainQuery+=' platform:"' + terms[count].strip() + '") ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' platform:"' + terms[count].strip() + '") '
else:
MainQuery+=' platform:"' + terms[count].strip() + '" '
# Work with IRB
if fields[count].find('IRB')>=0:
if terms[count]=='Not Required':
terms[count]='No'
if count<len(terms)-1:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= '(IRB:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+= ' IRB:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+= ' IRB:' + terms[count].strip() + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' IRB:' + terms[count].strip() + ') '
else:
MainQuery+= ' IRB:' + terms[count].strip() + ' '
# Work with Consent Type
if fields[count].find('Consent')>=0:
#print terms
if terms[count]=='Unrestricted':
terms[count]='No'
if terms[count]=='Restricted':
terms[count]='Restricted'
if terms[count]=='Unspecified':
terms[count]='Unspecified'
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (ConsentType:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' ConsentType:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' ConsentType:' + terms[count].strip() + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' ConsentType:' + terms[count].strip() + ') '
else:
MainQuery+=' ConsentType:' + terms[count].strip() + ' '
#if count<len(terms)-1:
# MainQuery+= ' ConsentType:' + terms[count].strip() + ' ' +conj[count]
#else:
# MainQuery+= ' ConsentType:' + terms[count].strip()
# Work with Nationality
if fields[count].find('Nationality')>=0:
mparser = MultifieldParser(["Demographics","demographics"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
#if count<=len(terms)-2:
# MainQuery+=str(query) + ' ' +conj[count] + ' '
#else:
# MainQuery+=str(query)
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' ( ' + str(query) + ' ' +conj[count]
else:
MainQuery+= ' ' + str(query) + ' ' +conj[count]
else:
MainQuery+=' ' + str(query) + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' ' + str(query) + ') '
else:
MainQuery+=' ' + str(query) + ' '
# Work with Gender
if fields[count].find('Sex')>=0:
if terms[count]=='Male':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] '
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+=' (MaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to]) ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+=' MaleNum:[1 to]) '
else:
MainQuery+=' MaleNum:[1 to] '
if terms[count]=='Female':
#if count<len(terms)-1:
# MainQuery+= ' FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' FemaleNum:[1 to] '
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+=' (FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' FemaleNum:[1 to]) ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+=' FemaleNum:[1 to]) '
else:
MainQuery+=' FemaleNum:[1 to] '
if terms[count]=='Both':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] AND FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] AND FemaleNum:[1 to] '
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+=' (MaleNum:[1 to] AND FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to] AND FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to] AND FemaleNum:[1 to]) ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+=' MaleNum:[1 to] AND FemaleNum:[1 to]) '
else:
MainQuery+=' MaleNum:[1 to] AND FemaleNum:[1 to] '
if terms[count]=='Either':
#if count<len(terms)-1:
# MainQuery+= ' MaleNum:[1 to] OR FemaleNum:[1 to] ' +conj[count]
#else:
# MainQuery+= ' MaleNum:[1 to] OR FemaleNum:[1 to] '
if count<=len(terms)-2:
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+=' (MaleNum:[1 to] OR FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to] OR FemaleNum:[1 to] ' +conj[count]
else:
MainQuery+=' MaleNum:[1 to] OR FemaleNum:[1 to]) ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+=' MaleNum:[1 to] OR FemaleNum:[1 to]) '
else:
MainQuery+=' MaleNum:[1 to] OR FemaleNum:[1 to] '
# Work with Study Design
if fields[count].find('StudyDesign')>=0:
if terms[count].find('Genome Wide Association Study')>=0:
terms[count]= 'gwas'
if terms[count]=='Case-Control Study':
terms[count]= 'Case-Control'
if terms[count]=='Cross Sectional Study':
terms[count]= 'Cross Sectional'
if terms[count]=='Double Blind Study':
terms[count]= 'Double Blind'
if terms[count]=='Interventional Studies':
terms[count]='Interventional'
if terms[count]=='Longitudinal Cohort Study':
terms[count]='Longitudinal'
if terms[count]=='Mendelian Randomized':
terms[count]='Mendelian'
if terms[count]=='Multicenter Study':
terms[count]='Multicenter'
if terms[count]=='Nested Case Control Study':
terms[count]='Nested Case Control'
if terms[count]=='Observational Studies':
terms[count]='Observational'
if terms[count]=='Partial Factorial Randomized Trial':
terms[count]='Partial Factorial Randomized'
if terms[count]=='Placebo Controlled Study':
terms[count]='Placebo Controlled'
if terms[count]=='Population Based Study':
terms[count]='Population Based Control'
if terms[count]=='Prospective Cohort Study':
terms[count]='Prospective'
if terms[count]=='Quantitative Cross Sectional Study':
terms[count]='Quantitative Cross Sectional'
if terms[count]=='Randomized Trial':
terms[count]='Randomized'
if terms[count]=='Phase 1':
terms[count]='Phase1'
if terms[count]=='Phase 2':
terms[count]='Phase2'
if terms[count]=='Phase 3':
terms[count]='Phase3'
if count<=len(terms)-2:
#MainQuery+= ' Type:' + terms[count].strip() + ' ' +conj[count]
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (Type:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Type:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Type:' + terms[count].strip() + ') ' +conj[count]
else:
#MainQuery+= ' Type:' + terms[count].strip() + ' '
if fields[count]==fields[count-1]:
MainQuery+= ' Type:' + terms[count].strip() + ') '
else:
MainQuery+=' Type:' + terms[count].strip() + ' '
#print MainQuery
# Work with Race
if fields[count].find('Race')>=0:
#if terms[count]=='White':
# terms[count]='white OR Demographics:cacausian '
if terms[count]=='Mixed Race':
terms[count]='Multiple'
# Combine all queries together
if count<=len(terms)-2:
#MainQuery+= ' Demographics:' + terms[count].strip() + ' ' +conj[count]
if fields[count]==fields[count+1]:
if fields[count-1]!=fields[count]:
MainQuery+= ' (Demographics:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+=' Demographics:' + terms[count].strip() + ' ' +conj[count]
else:
MainQuery+='Demographics:' + terms[count].strip() + ') ' +conj[count]
else:
if fields[count]==fields[count-1]:
MainQuery+= ' Demographics:' + terms[count].strip() + ') '
else:
MainQuery+=' Demographics:' + terms[count].strip() + ' '
#else:
# MainQuery+= ' Demographics:' + terms[count].strip()
#print MainQuery
# ======================================
# Work with Study
if fields[count].strip()=='Study':
mparser = MultifieldParser(["IDName","path","title","desc"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# Work with Study Name
if fields[count].find('Study Name')>=0:
mparser = MultifieldParser(["title"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# Work with Study ID
if fields[count].find('Study ID')>=0:
mparser = MultifieldParser(["IDName"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# ======================================
# Work with Variable
if fields[count]=='Variable':
mparser = MultifieldParser(["phenID","phenName","phenType","phenDesc"], schema)
query= mparser.parse(unicode(iQuery))
if count<len(terms)-1:
MainQuery+= str(query) + ' ' +conj[count] + ' '
else:
MainQuery+= str(query)
# Work with Variable Description
if fields[count].find('Variable Desc')>=0:
mparser = MultifieldParser(["phenDesc"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# Work with Variable Name
if fields[count].find('Variable Name')>=0:
mparser = MultifieldParser(["phenName"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# Work with Variable ID
if fields[count].find('Variable ID')>=0:
mparser = MultifieldParser(["phenID"], schema)
query= mparser.parse(unicode(iQuery))
# Combine all queries together
if count<=len(terms)-2:
MainQuery+=str(query) + ' ' +conj[count] + ' '
else:
MainQuery+=str(query)
# ======================================
# Work with All Fields
if fields[count].find('All Fields')>=0:
# Added on May 21, 2013
query = iQuery
if concept==True:
query_temp = QueryAnalys(query, concept)
else:
query_temp = query
# Combine all queries together
#if count<=len(terms)-2:
# MainQuery+=str(query) + ' ' +conj[count] + ' '
#else:
# MainQuery+=str(query)
if count<=len(terms)-2:
MainQuery+=query_temp + ' ' + conj[count] + ' '
else:
MainQuery+=query_temp + ' '
count+=1
return (MainQuery,terms)
