def setup_context(self, context):
if not sc.config.discourse["forum_url"]:
return
query = "%22{}%22|{}".format(uid_to_acro(self.uid).replace(" ", " "), self.uid)
context.discourse_link = "{}?search={}".format(sc.config.discourse["forum_url"], query)
context.discourse_url = sc.config.discourse["forum_url"]
context.embed = self.embed
try:
context.discourse_results = sc.search.discourse.search(self.uid)
except:
logger.exception("Failed to retrieve discourse_results")
def setup_context(self, context):
if not sc.config.discourse['forum_url']:
return
query = '%22{}%20%22'.format(
uid_to_acro(self.uid).replace(' ', ' '), self.uid)
context.discourse_link = '{}search?q={}'.format(
sc.config.discourse['forum_url'], query)
context.discourse_url = sc.config.discourse['forum_url']
context.embed = self.embed
try:
context.discourse_results = sc.search.discourse.search(self.uid)
except:
logger.exception('Failed to retrieve discourse_results')
def build_jinja2_environment():
env = jinja2.Environment(loader=jinja2.FileSystemLoader(
str(sc.templates_dir)),
extensions=[AssetsExtension],
trim_blocks=True,
lstrip_blocks=True,
auto_reload=not config.compile_assets)
env.assets_environment = assets.get_env()
env.filters['date'] = util.format_date
env.filters['time'] = util.format_time
env.filters['max'] = max
env.filters['min'] = min
env.filters['datetime'] = util.format_datetime
env.filters['timedelta'] = util.format_timedelta
env.filters['uid_to_name'] = lambda uid: uid_to_name(uid)
env.filters['uid_to_acro'] = lambda uid: uid_to_acro(uid)
env.filters['json'] = json.dumps
def sub_filter(string, pattern, repl):
return regex.sub(pattern, repl, string)
env.filters['sub'] = sub_filter
def sht_expansion(string):
"""Add links from SHT vol/page references to the sht-lookup page."""
if 'SHT' in string:
# Replace with spaces
string = string.replace(' ', ' ')
baseurl = '/sht-lookup/'
def replacement(m):
# Ignore 'also cf. ix p. 393ff' and ' A'
first, second = m[1], m[2]
if regex.match(r'.+p\.\s*', first) or \
regex.match(r'.+A', first):
return '{}{}'.format(first, second)
else:
# replace n-dash with dash
path = second.replace('−', '-')
return '{}<a href="{}{}" target="_blank">{}</a>'.format(
first, baseurl, path, second)
string = regex.sub(
r'([^0-9]+)([0-9]{1,4}(?:\.?[\−\+0-9a-zA-Z]+)?)', replacement,
string)
return string
env.filters['sht_expansion'] = sht_expansion
return env
def build_jinja2_environment():
env = jinja2.Environment(
loader=jinja2.FileSystemLoader(str(sc.templates_dir)),
extensions=[AssetsExtension],
trim_blocks=True,
lstrip_blocks=True,
auto_reload=not config.compile_assets,
)
env.assets_environment = assets.get_env()
env.filters["date"] = util.format_date
env.filters["time"] = util.format_time
env.filters["max"] = max
env.filters["min"] = min
env.filters["datetime"] = util.format_datetime
env.filters["timedelta"] = util.format_timedelta
env.filters["uid_to_name"] = lambda uid: uid_to_name(uid)
env.filters["uid_to_acro"] = lambda uid: uid_to_acro(uid)
env.filters["json"] = json.dumps
def sub_filter(string, pattern, repl):
return regex.sub(pattern, repl, string)
env.filters["sub"] = sub_filter
def sht_expansion(string):
"""Add links from SHT vol/page references to the sht-lookup page."""
if "SHT" in string:
# Replace with spaces
string = string.replace(" ", " ")
baseurl = "/sht-lookup/"
def replacement(m):
# Ignore 'also cf. ix p. 393ff' and ' A'
first, second = m[1], m[2]
if regex.match(r".+p\.\s*", first) or regex.match(r".+A", first):
return "{}{}".format(first, second)
else:
# replace n-dash with dash
path = second.replace("−", "-")
return '{}<a href="{}{}" target="_blank">{}</a>'.format(first, baseurl, path, second)
string = regex.sub(r"([^0-9]+)([0-9]{1,4}(?:\.?[\−\+0-9a-zA-Z]+)?)", replacement, string)
return string
env.filters["sht_expansion"] = sht_expansion
return env
def setup_context(self, context):
from sc.tools import html
m = self.content_regex.search(self.get_html())
m.detach_string() # Free up memory now.
imm = scimm.imm()
context.uid = self.uid
context.sutta = imm.suttas.get(self.uid)
context.subdivision = imm.subdivisions.get(self.uid)
context.division = imm.divisions.get(self.uid)
context.canonical = self.canonical
context.textdata = textdata = imm.get_text_data(self.uid, self.lang_code)
context.title = textdata.name if textdata else "?"
context.text = m["content"]
if self.lang_code in imm.font_data["css_font_class"]:
context.font_class = imm.font_data["css_font_class"][self.lang_code]
context.discourse_link = None
context.discourse_results = None
if (
self.uid in imm.suttas
and sc.config.discourse["forum_url"]
and self.lang_code == "en"
or imm.languages[self.lang_code].isroot
):
sutta = imm.suttas
query = "%22{}%22|{}".format(uid_to_acro(self.uid).replace(" ", " "), self.uid)
context.discourse_link = "{}?search={}".format(sc.config.discourse["forum_url"], query)
context.discourse_url = sc.config.discourse["forum_url"]
try:
context.discourse_results = sc.search.discourse.search(self.uid)
except:
logger.exception("Failed to retrieve discourse_results")
if context.embed:
context.text = self.shorter_text(context.text)
context.has_quotes = "‘" in context.text or "“" in context.text
try:
context.snippet = self.get_snippet(context.text)
except Exception as e:
logger.error("Failed to generated snippet for {} ({})".format(self.uid, str(e)))
context.snippet = ""
# Eliminate newlines from Full-width-glyph languages like Chinese
# because they convert into spaces when rendered.
# TODO: This check should use 'language' table
if self.lang_code in {"zh", "lzh", "ko", "jp"}:
context.text = self.massage_cjk(context.text)
context.lang_code = self.lang_code
context.root_lang = None
if context.sutta:
context.root_lang = context.sutta.lang
elif context.division:
context.root_lang = context.division.collection.lang
elif context.subdivision:
context.root_lang = context.subdivision.division.collection.lang
context.text_refs = imm.get_text_refs(context.uid)
nextprev = imm.get_next_prev(uid=self.uid, lang_uid=self.lang_code)
context.next_data = nextprev["next"]
context.prev_data = nextprev["prev"]
context.exports["uid"] = self.uid
context.exports["lang"] = self.lang_code
def setup_context(self, context):
from sc.tools import html
m = self.content_regex.search(self.get_html())
m.detach_string() # Free up memory now.
imm = scimm.imm()
context.uid = self.uid
context.sutta = imm.suttas.get(self.uid)
context.subdivision = imm.subdivisions.get(self.uid)
context.division = imm.divisions.get(self.uid)
context.canonical = self.canonical
context.textdata = textdata = imm.get_text_data(
self.uid, self.lang_code)
context.title = textdata.name if textdata else '?'
context.text = m['content']
if self.lang_code in imm.font_data['css_font_class']:
context.font_class = imm.font_data['css_font_class'][
self.lang_code]
context.discourse_link = None
context.discourse_results = None
if (self.uid in imm.suttas and sc.config.discourse['forum_url']
and self.lang_code == 'en'
or imm.languages[self.lang_code].isroot):
sutta = imm.suttas
query = '%22{}%20%22'.format(
uid_to_acro(self.uid).replace(' ', ' '), self.uid)
context.discourse_link = '{}search?q={}'.format(
sc.config.discourse['forum_url'], query)
context.discourse_url = sc.config.discourse['forum_url']
try:
context.discourse_results = sc.search.discourse.search(
self.uid)
except:
logger.exception('Failed to retrieve discourse_results')
if context.embed:
context.text = self.shorter_text(context.text)
context.has_quotes = '‘' in context.text or '“' in context.text
try:
context.snippet = self.get_snippet(context.text)
except Exception as e:
logger.error('Failed to generated snippet for {} ({})'.format(
self.uid, str(e)))
context.snippet = ''
# Eliminate newlines from Full-width-glyph languages like Chinese
# because they convert into spaces when rendered.
# TODO: This check should use 'language' table
if self.lang_code in {'zh', 'lzh', 'ko', 'jp'}:
context.text = self.massage_cjk(context.text)
context.lang_code = self.lang_code
context.root_lang = None
if context.sutta:
context.root_lang = context.sutta.lang
elif context.division:
context.root_lang = context.division.collection.lang
elif context.subdivision:
context.root_lang = context.subdivision.division.collection.lang
context.text_refs = imm.get_text_refs(context.uid)
nextprev = imm.get_next_prev(uid=self.uid, lang_uid=self.lang_code)
context.next_data = nextprev['next']
context.prev_data = nextprev['prev']
context.exports['uid'] = self.uid
context.exports['lang'] = self.lang_code
def build(self):
""" Build the sutta central In Memory Model
This starts from the highest level (i.e. collection) and works to
the lowest level (i.e. parallels - the relationship between suttas)
Since it is fully navigitable both up and down this means some
elements can't be populated initially. This means that suttas
insert themselves into the subdivision where they belong.
Some tables are indexed as dicts, with the key being the uid.
These include:
collection, division, subdivision, sutta, language
When classes are contained within a class, for example, suttas in
a subdivision, this is always represented by a list. That list will
be sorted appropriately and can be directly outputted, generally
without any need for filtering or sorting.
When an attribute is a list or dict, the name always end in an 's'
for example:
imm.suttas['sn1.1'].subdivision.division.subdivisions[0].suttas[0]
Some things, such as parallels, are not indexed at all, and are
only accessable as attributes of the relevant suttas.
The imm also examines the file system. The fully qualified path to a
text can be acquired using:
imm.text_paths[lang][uid]
"""
# Build Pitakas
self.pitakas = OrderedDict()
for row in table_reader('pitaka'):
self.pitakas[row.uid] = Pitaka(uid=row.uid, name=row.name, always_full=row.always_full)
# Build Sects
self.sects = OrderedDict()
for row in table_reader('sect'):
self.sects[row.uid] = Sect(uid=row.uid, name=row.name)
# Build Languages (indexed by id)
self.languages = OrderedDict()
for row in table_reader('language'):
self.languages[row.uid] = Language(
uid=row.uid,
name=row.name,
iso_code=row.iso_code,
isroot=row.isroot,
priority=int(row.priority),
search_priority=float(row.search_priority),
collections=[],
)
# Note that one isocode can map to multiple languages
# for example zh modern/ancient
self.isocode_to_language = {}
for language in self.languages.values():
if language.iso_code not in self.isocode_to_language:
self.isocode_to_language[language.iso_code] = []
self.isocode_to_language[language.iso_code].append(language)
# From external_text table
text_refs = defaultdict(list)
for row in table_reader('external_text'):
text_refs[row.sutta_uid].append( TextRef(lang=self.languages[row.language], name=None, abstract=row.abstract, url=row.url, priority=row.priority) )
self._external_text_refs = text_refs.copy()
collections = []
for i, row in enumerate(table_reader('collection')):
if row.sect_uid:
sect = self.sects[row.sect_uid]
else:
sect = None
collection = Collection(
uid=row.uid,
name=row.name,
abbrev_name=row.abbrev_name,
lang=self.languages[row.language],
sect=sect,
pitaka=self.pitakas[row.pitaka_uid],
menu_seq=i,
divisions=[] # Populate later
)
collections.append(collection)
# Sort collections by menu_seq
collections.sort(key=Collection.sort_key)
self.collections = OrderedDict()
for collection in collections:
self.collections[collection.uid] = collection
self.languages[collection.lang.uid].collections.append(collection)
# Build divisions (indexed by uid)
self.divisions = OrderedDict()
for i, row in enumerate(table_reader('division')):
collection = self.collections[row.collection_uid]
text_ref = self.get_text_ref(uid=row.uid, lang_uid=collection.lang.uid);
division = Division(
uid=row.uid,
name=row.name,
alt_name=row.alt_name,
text_ref=text_ref,
acronym=row.acronym or uid_to_acro(row.uid),
subdiv_ind=row.subdiv_ind,
menu_seq=i,
menu_gwn_ind=bool(row.menu_gwn_ind),
collection=collection,
subdivisions=[], # Populate later
)
self.divisions[row.uid] = division
# Populate collections
collection.divisions.append(division)
# Sort divisions within collections by menu_seq
for collection in self.collections.values():
collection.divisions.sort(key=Division.sort_key)
# Build subdivisions (indexed by uid)
self.subdivisions = OrderedDict()
self.nosubs = set()
for i, row in enumerate(table_reader('subdivision')):
subdivision = Subdivision(
uid=row.uid,
acronym=row.acronym,
division=self.divisions[row.division_uid],
name=row.name,
vagga_numbering_ind=row.vagga_numbering_ind,
order=i,
vaggas=[], # Populate later
suttas=[] # Populate later
)
self.subdivisions[row.uid] = subdivision
if row.uid.endswith('-nosub'):
self.nosubs.add(row.uid[:-6])
# populate divisions.subdivisions
self.divisions[row.division_uid].subdivisions.append(subdivision)
for division in self.divisions.values():
if not division.subdivisions:
subdivision = Subdivision(
uid=division.uid,
acronym=None,
division=division,
name=None,
vagga_numbering_ind=False,
order=9000,
vaggas=[],
suttas=[])
division.subdivisions.append(subdivision)
self.subdivisions[division.uid] = subdivision
# Build vaggas
self.vaggas = OrderedDict()
for row in table_reader('vagga'):
vagga = Vagga(
subdivision=self.subdivisions[row.subdivision_uid],
number=row.number,
name=row.name,
suttas=[], # Populate later
)
self.vaggas[(row.subdivision_uid, row.number)] = vagga
# Populate subdivision.vaggas
vagga.subdivision.vaggas.append(vagga)
for subdivision in self.subdivisions.values():
if not subdivision.vaggas:
subdivision.vaggas.append(Vagga(
subdivision=subdivision,
number=0,
name=None,
suttas=[]))
# Load biblio entries (Not into an instance variable)
biblios = {}
for row in table_reader('biblio'):
biblios[row.uid] = BiblioEntry(
uid=row.uid,
name=row.name,
text=row.text)
# Build suttas (indexed by uid)
suttas = []
for row in table_reader('sutta'):
uid = row.uid
volpage = row.volpage.split('//')
acro = row.acronym.split('//')
if not acro[0]:
acro[0] = uid_to_acro(uid)
lang = self.languages[row.language]
subdivision = self.subdivisions[row.subdivision_uid]
if row.vagga_number:
vagga_number = int(row.vagga_number)
vagga = subdivision.vaggas[vagga_number - 1]
else:
vagga_number = 0
vagga = subdivision.vaggas[0]
m = regex.search(r'(?r)\d+', row.uid)
if m:
number = int(m[0])
else:
number = 9999
biblio_entry = None
if row.biblio_uid:
biblio_entry = biblios.get(row.biblio_uid)
sutta = Sutta(
uid=row.uid,
acronym=acro[0],
alt_acronym=acro[1] if len(acro) > 1 else None,
name=row.name,
vagga_number=vagga_number,
lang=lang,
subdivision=subdivision,
vagga=vagga,
number=number,
number_in_vagga=row.number_in_vagga,
volpage=volpage[0],
alt_volpage_info=volpage[1] if len(volpage) > 1 else None,
biblio_entry=biblio_entry,
parallels=[],
imm=self,
)
suttas.append( (uid, sutta) )
suttas = sorted(suttas, key=numsortkey)
self.suttas = OrderedDict(suttas)
# Populate subdivisions.suttas
for sutta in self.suttas.values():
sutta.subdivision.suttas.append(sutta)
sutta.vagga.suttas.append(sutta)
def acronym(self):
return uid_to_acro(self.uid)
def build(self):
""" Build the sutta central In Memory Model
This starts from the highest level (i.e. collection) and works to
the lowest level (i.e. parallels - the relationship between suttas)
Since it is fully navigitable both up and down this means some
elements can't be populated initially. This means that suttas
insert themselves into the subdivision where they belong.
Some tables are indexed as dicts, with the key being the uid.
These include:
collection, division, subdivision, sutta, language
When classes are contained within a class, for example, suttas in
a subdivision, this is always represented by a list. That list will
be sorted appropriately and can be directly outputted, generally
without any need for filtering or sorting.
When an attribute is a list or dict, the name always end in an 's'
for example:
imm.suttas['sn1.1'].subdivision.division.subdivisions[0].suttas[0]
Some things, such as parallels, are not indexed at all, and are
only accessable as attributes of the relevant suttas.
The imm also examines the file system. The fully qualified path to a
text can be acquired using:
imm.text_paths[lang][uid]
"""
# Build Pitakas
self.pitakas = OrderedDict()
for row in table_reader('pitaka'):
self.pitakas[row.uid] = Pitaka(uid=row.uid, name=row.name, always_full=row.always_full)
# Build Sects
self.sects = OrderedDict()
for row in table_reader('sect'):
self.sects[row.uid] = Sect(uid=row.uid, name=row.name)
# Build Languages (indexed by id)
self.languages = OrderedDict()
for row in table_reader('language'):
self.languages[row.uid] = Language(
uid=row.uid,
name=row.name,
iso_code=row.iso_code,
isroot=row.isroot,
priority=int(row.priority),
search_priority=float(row.search_priority),
collections=[],
)
# Note that one isocode can map to multiple languages
# for example zh modern/ancient
self.isocode_to_language = {}
for language in self.languages.values():
if language.iso_code not in self.isocode_to_language:
self.isocode_to_language[language.iso_code] = []
self.isocode_to_language[language.iso_code].append(language)
# From external_text table
text_refs = defaultdict(list)
for row in table_reader('external_text'):
text_refs[row.sutta_uid].append( TextRef(lang=self.languages[row.language], name=None, abstract=row.abstract, url=row.url, priority=row.priority) )
self._external_text_refs = text_refs.copy()
collections = []
for i, row in enumerate(table_reader('collection')):
if row.sect_uid:
sect = self.sects[row.sect_uid]
else:
sect = None
collection = Collection(
uid=row.uid,
name=row.name,
abbrev_name=row.abbrev_name,
lang=self.languages[row.language],
sect=sect,
pitaka=self.pitakas[row.pitaka_uid],
menu_seq=i,
divisions=[] # Populate later
)
collections.append(collection)
# Sort collections by menu_seq
collections.sort(key=Collection.sort_key)
self.collections = OrderedDict()
for collection in collections:
self.collections[collection.uid] = collection
self.languages[collection.lang.uid].collections.append(collection)
# Build divisions (indexed by uid)
self.divisions = OrderedDict()
for i, row in enumerate(table_reader('division')):
collection = self.collections[row.collection_uid]
text_ref = self.get_text_ref(uid=row.uid, lang_uid=collection.lang.uid);
division = Division(
uid=row.uid,
name=row.name,
alt_name=row.alt_name,
text_ref=text_ref,
acronym=row.acronym or uid_to_acro(row.uid),
subdiv_ind=row.subdiv_ind,
menu_seq=i,
menu_gwn_ind=bool(row.menu_gwn_ind),
collection=collection,
subdivisions=[], # Populate later
)
self.divisions[row.uid] = division
# Populate collections
collection.divisions.append(division)
# Sort divisions within collections by menu_seq
for collection in self.collections.values():
collection.divisions.sort(key=Division.sort_key)
# Build subdivisions (indexed by uid)
self.subdivisions = OrderedDict()
self.nosubs = set()
for i, row in enumerate(table_reader('subdivision')):
subdivision = Subdivision(
uid=row.uid,
acronym=row.acronym,
division=self.divisions[row.division_uid],
name=row.name,
vagga_numbering_ind=row.vagga_numbering_ind,
order=i,
vaggas=[], # Populate later
suttas=[] # Populate later
)
self.subdivisions[row.uid] = subdivision
if row.uid.endswith('-nosub'):
self.nosubs.add(row.uid[:-6])
# populate divisions.subdivisions
self.divisions[row.division_uid].subdivisions.append(subdivision)
for division in self.divisions.values():
if not division.subdivisions:
subdivision = Subdivision(
uid=division.uid,
acronym=None,
division=division,
name=None,
vagga_numbering_ind=False,
order=9000,
vaggas=[],
suttas=[])
division.subdivisions.append(subdivision)
self.subdivisions[division.uid] = subdivision
# Build vaggas
self.vaggas = OrderedDict()
for row in table_reader('vagga'):
vagga = Vagga(
subdivision=self.subdivisions[row.subdivision_uid],
number=row.number,
name=row.name,
suttas=[], # Populate later
)
self.vaggas[(row.subdivision_uid, row.number)] = vagga
# Populate subdivision.vaggas
vagga.subdivision.vaggas.append(vagga)
for subdivision in self.subdivisions.values():
if not subdivision.vaggas:
subdivision.vaggas.append(Vagga(
subdivision=subdivision,
number=0,
name=None,
suttas=[]))
# Load biblio entries (Not into an instance variable)
biblios = {}
for row in table_reader('biblio'):
biblios[row.uid] = BiblioEntry(
uid=row.uid,
name=row.name,
text=row.text)
# Build suttas (indexed by uid)
suttas = []
for row in table_reader('sutta'):
uid = row.uid
volpage = row.volpage.split('//')
acro = row.acronym.split('//')
if not acro[0]:
acro[0] = uid_to_acro(uid)
lang = self.languages[row.language]
subdivision = self.subdivisions[row.subdivision_uid]
if row.vagga_number:
vagga_number = int(row.vagga_number)
vagga = subdivision.vaggas[vagga_number - 1]
else:
vagga_number = 0
vagga = subdivision.vaggas[0]
m = regex.search(r'(?r)\d+', row.uid)
if m:
number = int(m[0])
else:
number = 9999
biblio_entry = None
if row.biblio_uid:
biblio_entry = biblios.get(row.biblio_uid)
sutta = Sutta(
uid=row.uid,
acronym=acro[0],
alt_acronym=acro[1] if len(acro) > 1 else None,
name=row.name,
vagga_number=vagga_number,
lang=lang,
subdivision=subdivision,
vagga=vagga,
number=number,
number_in_vagga=row.number_in_vagga,
volpage=volpage[0],
alt_volpage_info=volpage[1] if len(volpage) > 1 else None,
biblio_entry=biblio_entry,
parallels=[],
imm=self,
)
suttas.append( (uid, sutta) )
suttas = sorted(suttas, key=numsortkey)
self.suttas = OrderedDict(suttas)
# Populate subdivisions.suttas
for sutta in self.suttas.values():
sutta.subdivision.suttas.append(sutta)
sutta.vagga.suttas.append(sutta)
def acronym(self):
return uid_to_acro(self.uid)
