def get_button(lang_n="", isExtended=False, full=True):
"""Provides a download form.
ARGUMENTS:
- 'lang_n' : the glottocode
- 'ext' : if for the extended corpus
- 'full' : if all fields (or just the button)"""
xml = "<span class='span2' style='margin-left:0px'>no xml</span>"
wav_size = 0
eaf_size = 0
tg_size = 0
xml_size = 0
tsv_size = 0
txt = "<span class='span2' style='margin-left:0px'>Dataset files :</span>"
wav = "<span class='span2' style='margin-left:0px'>" \
"<button class='btn btn-warning' value='wav' name='format' type='input'>WAV</button> " + str(wav_size) + " MB</span>"
eaf = "<span class='span2' style='margin-left:0px'>" \
"<button class='btn btn-primary' value='eaf' name='format' type='input'>eaf</button> " + str(eaf_size) + " MB</span>"
textgrid = "<span class='span2' style='margin-left:0px'>" \
"<button class='btn btn-danger' value='textgrid' name='format' type='input'>TextGrid</button> " + str(tg_size) + " MB</span>"
if xml_size>0:
xml = "<span class='span2' style='margin-left:0px'>" \
"<button class='btn btn-info' value='xml' name='format' type='input'>eaf</button> " + str(xml_size) + " MB</span>"
tsv = "<span class='span2' style='margin-left:0px'>" \
"<button class='btn btn-info' value='tsv' name='format' type='input'>eaf</button> " + str(tsv_size) + " MB</span>"
form = "<form method=\"post\" action=\"/doreLoad\" style='margin:0px;line-height:3em;'>" + \
"<input type=\"hidden\" id=\"f_id\" " "name=\"id\" value=\"" + lang_n + "\">" + \
"<input type=\"hidden\" id=\"f_ext\" name=\"extended\" value=\"" + str(isExtended) + "\">"
if isExtended:
return literal(form + txt + eaf + "</form>")
else:
return literal(form + txt + wav + eaf + textgrid + xml + tsv + "</form>")
def text2html(text, mode='br', sep='\n\n'):
"""
>>> assert 'div' in unicode(text2html('chunk', mode='p'))
"""
if mode == 'p':
return HTML.div(*[HTML.p(literal(newline2br(line))) for line in text.split(sep)])
return HTML.p(literal(newline2br(text)))
def gloss_with_tooltip(gloss):
person_map = {
'1': 'first person',
'2': 'second person',
'3': 'third person',
}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(literal(gloss[end:match.start()]))
abbr = match.group('abbr')
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group('personprefix'):
explanation = '%s %s' % (
person_map[match.group('personprefix')], explanation)
if match.group('personsuffix'):
explanation = '%s %s' % (
explanation, person_map[match.group('personsuffix')])
res.append(HTML.span(
HTML.span(gloss[match.start():match.end()].lower(), class_='sc'),
**{'data-hint': explanation, 'class': 'hint--bottom'}))
else:
res.append(abbr)
end = match.end()
res.append(literal(gloss[end:]))
return filter(None, res)
def text2html(text, mode="br", sep="\n\n"):
"""Turn plain text into simple HTML.
>>> assert 'div' in text_type(text2html('chunk', mode='p'))
"""
if mode == "p":
return HTML.div(*[HTML.p(literal(newline2br(line))) for line in text.split(sep)])
return HTML.p(literal(newline2br(text)))
def text2html(text, mode='br', sep='\n\n'):
"""
>>> assert 'div' in unicode(text2html('chunk', mode='p'))
"""
if mode == 'p':
return HTML.div(
*[HTML.p(literal(newline2br(line))) for line in text.split(sep)])
return HTML.p(literal(newline2br(text)))
def col_defs(self):
return list(
filter(lambda col: not self.semanticfield or col.name != 'sf', [
LWTCodeCol(self, 'lwt_code'),
LinkCol(
self,
'name',
sTitle='Meaning',
sDescription=
"This column shows the labels of the Loanword Typology "
"meanings. By clicking on a meaning label, you get more information "
"about the meaning, as well as a list of all words that are counterparts "
"of that meaning."),
CoreListCol(self, 'core_list'),
Col(self,
'cat',
sTitle='Semantic category',
sDescription=
"Meanings were assigned to semantic categories with "
"word-class-like labels: nouns, verbs, adjectives, adverbs, function "
"words. No claim is made about the grammatical behavior of words "
"corresponding to these meanings. The categories are intended to be "
"purely semantic.",
model_col=Meaning.semantic_category,
choices=get_distinct_values(Meaning.semantic_category)),
SemanticFieldCol(
self,
'sf',
sTitle='Semantic field',
sDescription=
"The first 22 fields are the fields of the Intercontinental "
"Dictionary Series meaning list, proposed by Mary Ritchie Key, and "
"ultimately based on Carl Darling Buck's (1949) Dictionary of selected "
"synonyms in the principal Indo-European languages. The other two fields "
"were added for the Loanword Typology project."),
MeaningScoreCol(self,
'borrowed_score',
sDescription=literal(
'%s' % hb_borrowed_score())),
MeaningScoreCol(self,
'age_score',
sDescription=literal('%s' % hb_age_score())),
MeaningScoreCol(self,
'simplicity_score',
sDescription=literal(
'%s' % hb_simplicity_score())),
Col(self,
'representation',
model_col=Meaning.representation,
sDescription=
"This column shows how many counterparts for this meaning "
"there are in the 41 languages. The number can be higher than 41 because "
"a language may have several counterparts for one meaning (\"synonyms\"),"
" and it may be lower than 41, because not all languages may have a "
"counterpart for a meaning. "),
]))
def col_defs(self):
return [
# ID, Vocabulary, Authors, Number of words, Percentage of loanwords, cite
IntegerIdCol(
self,
'id',
sTitle="ID",
sDescription=literal(
"The vocabulary ID number corresponds to the ordering to the"
" chapters on the book <em>Loanwords in the World's Languages</em>. "
"Languages are listed in rough geographical order from west to east, "
"from Africa via Europe to Asia and the Americas, so that "
"geographically adjacent languages are next to each other."
)),
VocabularyCol(
self,
'vocabulary',
sDescription=literal(
"<p>Each vocabulary of WOLD is a separate electronic "
"publication with a separate author or team of authors. Each vocabulary "
"has a characteristic colour in WOLD.</p><p>Click on a vocabulary to "
"see the words (loanwords and nonloanwords) and their properties.</p>"
)),
ContributorsCol(
self,
'contributor',
sDescription=
"The authors are experts of the language and its history. "
"They also contributed a prose chapter on the borrowing situation in "
"their language that was published in the book "
"Loanwords in the World's Languages."),
Col(self,
'n',
sTitle='Number of words',
model_col=Vocabulary.count_words,
sDescription="There would be 1814 words in each vocabulary, "
"corresponding to the 1814 Loanword Typology meanings, if each meaning "
"had exactly one counterpart, and if all the counterparts were "
"different words. But many (\"polysomous\") words are counterparts of "
"several meanings, many meanings have several word counterparts "
"(\"synonyms\", or \"subcounterparts\"), and many meanings have no "
"counterparts at all, so the number of words in each database varies "
"considerably."),
PercentCol(
self,
'p',
sTitle='Percentage of loanwords',
model_col=Vocabulary.borrowed_score,
sDescription=
"This gives the percentage of words in each language that "
"are \"clearly borrowed\" or \"probably borrowed\"."),
CitationCol(self, 'cite'),
]
def get_legends(self):
items = []
for biome in DBSession.query(Biome)\
.filter(Biome.description != 'ffffff')\
.order_by(as_int(Biome.id)):
items.append(
HTML.label(
HTML.span(
literal(' '),
style='background-color: #%s;' % biome.description,
class_='biome-color'),
literal(biome.name),
style='margin-left: 1em; margin-right: 1em;'))
yield Legend(self, 'categories', items)
def get_legends(self):
items = []
for biome in DBSession.query(Biome)\
.filter(Biome.description != 'ffffff')\
.order_by(as_int(Biome.id)):
items.append(
HTML.label(
HTML.span(
literal(' '),
style='background-color: #%s;' % biome.description,
class_='biome-color'),
literal(biome.name),
style='margin-left: 1em; margin-right: 1em;'))
yield Legend(self, 'categories', items)
def value_li(l):
return (
HTML.img(height="20", width="20", src=self.icon_map[l.pk]),
literal(' '),
languoid_link(self.req, l),
literal(' '),
HTML.label(HTML.input(
type="checkbox",
onclick="GLOTTOLOG3.filterMarkers(this);",
class_="checkbox inline",
checked="checked",
value=str(l.pk)),
class_="checkbox inline",
title="click to toggle markers"),
)
def value_li(l):
return (
HTML.img(height="20", width="20", src=self.icon_map[l.pk]),
literal(' '),
languoid_link(self.req, l),
literal(' '),
HTML.label(
HTML.input(
type="checkbox",
onclick="GLOTTOLOG3.filterMarkers(this);",
class_="checkbox inline",
checked="checked",
value=str(l.pk)),
class_="checkbox inline",
title="click to toggle markers"),
)
def get_attrs(self, item):
label = item.__unicode__()
title = label
if self.dt.parameter:
label = HTML.span(map_marker_img(self.dt.req, item),
literal(' '), label)
return {'label': label, 'title': title}
def newline2br(text):
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def get_audio(ctx):
html_result = "list of audio in nakala"
languages = DBSession.query(doreLanguage).join(Language, Language.pk == doreLanguage.pk).all()
for lang in languages:
if lang.AUDIO is not None and lang.AUDIO != 'na':
html_result += '<li> ' + lang.name + ' <a href="https://test.nakala.fr/' + lang.AUDIO + '" target=_BLANK>view audio</a><br>' + '<a href="./static/' + lang.id + '.zip">add a direct link to a zip archive of the audio ?</a>' + '</li>'
return literal(html_result)
def get_nak(ctx, sour):
html_result = ""
lang = DBSession.query(doreLanguage). \
filter(doreLanguage.id == ctx.id).first()
if lang.NAK is not None and lang.NAK != 'na':
html_result += '<a href="https://test.nakala.fr/' + lang.NAK + '" target=_BLANK>view language information on nakala</a></li>'
return literal(html_result)
def newline2br(text):
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def get_layers(self):
for canton, dialects in itertools.groupby(
DBSession.query(models.Variety).order_by(
models.Variety.canton, as_int(common.Language.id)),
lambda l: l.canton):
dialects = list(dialects)
json = DialectsGeoJson(None).render(dialects, self.req, dump=False)
yield Layer(slug(canton),
canton,
data=json,
marker=HTML.span(
helpers.map_marker_img(self.req,
dialects[0],
marker=self.map_marker),
literal(' '),
dialects[0].canton_img(self.req),
literal(' ')))
def legend_items(req):
for type, label in [('recipient', 'recipient language'),
('donor', 'donor languoid')]:
yield HTML.label(HTML.img(src=icon_url(req, type),
height='20',
width='20'),
literal(' ' + label),
style='margin-left: 1em; margin-right: 1em;')
def feature_description(req, ctx):
desc = ctx.markup_description or ctx.description
desc = re.sub(
"\*\*(?P<id>[0-9]+\-[0-9]+)\*\*",
lambda m: HTML.a('[%s]' % m.group('id'),
href=req.route_url('sentence', id=m.group('id'))),
desc)
desc = re.sub(
"\*\*\<\/span\>(?P<id>[0-9]+\-[0-9]+)\*\*",
lambda m: literal('</span>') + HTML.a(
'[%s]' % m.group('id'),
href=req.route_url('sentence', id=m.group('id'))), desc)
return re.sub(
'\<span style\=\"font-style\: italic;\"\>WALS\<\/span\>\s+feature\s+[0-9]+',
lambda m: HTML.a(literal(desc[m.start():m.end()]),
href=req.route_url('wals', id=ctx.id)), desc)
def linkify_iso_codes(request, text, class_=None, route_name='glottolog.iso'):
def chunks():
start = 0
for match in ISO_PATTERN.finditer(text):
yield text[start:match.start(0)]
url = request.route_url(route_name, id=match.group('iso'))
yield HTML.a(match.group(0), href=url, class_=class_)
start = match.end(0)
yield text[start:]
return literal('').join(chunks())
def newline2br(text):
"""Replace newlines in text with HTML br tags."""
if not text:
return ''
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def js(self):
return Markup(HTML.script(literal("""\
$(document).ready(function() {
$('.%s').clickover({
html: true,
title: 'Download information',
placement: '%s',
trigger: 'click'
});
});""" % (self._opener_class, self.options['doc_position']))))
def js(self):
return Markup(HTML.script(literal("""\
$(document).ready(function() {
$('.%s').clickover({
html: true,
title: 'Download information',
placement: '%s',
trigger: 'click'
});
});""" % (self._opener_class, self.options['doc_position']))))
def feature_description(req, ctx):
desc = ctx.markup_description or ctx.description
desc = re.sub(
"\*\*(?P<id>[0-9]+\-[0-9]+)\*\*",
lambda m: HTML.a(
'[%s]' % m.group('id'), href=req.route_url('sentence', id=m.group('id'))),
desc)
desc = re.sub(
"\*\*\<\/span\>(?P<id>[0-9]+\-[0-9]+)\*\*",
lambda m: literal('</span>') + HTML.a(
'[%s]' % m.group('id'), href=req.route_url('sentence', id=m.group('id'))),
desc)
return re.sub(
'\<span style\=\"font-style\: italic;\"\>WALS\<\/span\>\s+feature\s+[0-9]+',
lambda m: HTML.a(
literal(desc[m.start():m.end()]), href=req.route_url('wals', id=ctx.id)),
desc)
def newline2br(text):
"""Replace newlines in text with HTML br tags."""
if not text:
return ''
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def alt_representations(req, rsc, doc_position='right', exclude=None):
exclude = exclude or []
exclude.extend(['html', 'snippet.html'])
adapters = [
a
for n, a in req.registry.getAdapters([rsc], interfaces.IRepresentation)
if a.extension not in set(exclude)
]
doc = []
for adapter in adapters:
if adapter.__doc__:
doc.append(HTML.dt(adapter.name or adapter.extension))
doc.append(HTML.dd(adapter.__doc__))
doc = HTML.div(
HTML.p(
"""You may download alternative representations of the data on
"%s" by clicking the button """ % rsc.name, icon('download-alt')),
HTML.dl(*doc))
return HTML.div(
HTML.div(
button(
icon('info-sign', inverted=True), **{
'class': ['btn-info'],
'id': 'rsc-dl',
'data-content': unicode(doc)
}),
HTML.a(
icon('download-alt'), HTML.span(class_="caret"), **{
'class_': "btn dropdown-toggle",
'data-toggle': "dropdown",
'href': "#",
'id': "dt-dl-opener",
}),
HTML.ul(
*[
HTML.li(
HTML.
a(a.name or a.extension,
href="#",
onclick="document.location.href = '%s'; return false;"
% req.resource_url(rsc, ext=a.extension),
id='dt-dl-%s' % a.extension)) for a in adapters
], **dict(class_="dropdown-menu")),
class_='btn-group'),
HTML.script(
literal("""\
$(document).ready(function() {
$('#rsc-dl').clickover({
html: true,
title: 'Alternative representations',
placement: '%s',
trigger: 'click'
});
});""" % doc_position)))
def newline2br(text):
"""
>>> assert newline2br(None) == ''
"""
if not text:
return ''
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def newline2br(text):
"""
>>> assert newline2br(None) == ''
"""
if not text:
return ''
chunks = []
for i, line in enumerate(text.split('\n')):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return '\n'.join(chunks)
def get_desc(ctx, sour, dflt=True, us=False, type='html'):
"""Provides a language descriptions."""
data_year = '2022'
isSource = "[click cite button]"
human_creator = get_author(ctx.creator, 'txt')
print(human_creator)
arc_link = literal(ctx.arc_link())
if hasattr(sour, 'year'):
data_year = sour.year
isSource = get_cite(ctx.id, sour, 'bibtex')
style = ("<style>p.cite {padding-left: 25px; text-indent: -15px; "
"font-size: 90%;}</style></style>")
list_marker = ['<p>', '</p>', '<pre class=\"cite\">', '</pre>']
if type == 'markdown':
style = ''
list_marker = ['', '\n\r', '', '\n\r']
arc_link = '[' + ctx.arclink + '](' + ctx.arclink + ' "click to open")'
text = ((list_marker[0] + "The {0} DoReCo data set was compiled by {1} in {2} "
"and further processed by the DoReCo team in {3}.")
.format(ctx.name, human_creator, data_year, "2020-2022"))
arch = ""
if not ctx.arclink == "na":
arch = (" A larger collection of {}'s {} data is archived "
"at {}.".format(human_creator, ctx.name, arc_link))
return literal(text + arch + ((list_marker[1] + list_marker[0] + "A set of files with further information "
"on the {0} DoReCo data set, including metadata and PIDs is "
"automatically included in each download of {0} DoReCo files." +
list_marker[1] +
list_marker[0] + "The {0} DoReCo data set should be cited as follows "
"(a BibTex version of this citation is provided below):" +
list_marker[1] +
list_marker[2] + "{1}" + list_marker[3] + list_marker[
0] + "Please note that when citing "
"this data set, or any number of DoReCo data sets, it is NOT "
"sufficient to refer to DoReCo (and its editors) as a whole, "
"but the full citation for each individual data set must be "
"provided, including the names of the creators of each data "
"set." + list_marker[1]).format(ctx.name, isSource))) # " )))#
def newline2br(text):
"""Replace newlines in text with HTML br tags.
>>> assert newline2br(None) == ''
"""
if not text:
return ""
chunks = []
for i, line in enumerate(text.split("\n")):
if i > 0:
chunks.append(HTML.br())
chunks.append(literal(line))
return "\n".join(chunks)
def alt_representations(req, rsc, doc_position='right', exclude=None):
exclude = exclude or []
exclude.extend(['html', 'snippet.html'])
adapters = [a for n, a in req.registry.getAdapters([rsc], interfaces.IRepresentation)
if a.extension not in set(exclude)]
doc = []
for adapter in adapters:
if adapter.__doc__:
doc.append(HTML.dt(adapter.name or adapter.extension))
doc.append(HTML.dd(adapter.__doc__))
doc = HTML.div(
HTML.p(
"""You may download alternative representations of the data on
"%s" by clicking the button """ % rsc.name,
icon('download-alt')),
HTML.dl(*doc))
return HTML.div(HTML.div(
button(
icon('info-sign', inverted=True),
**{'class': ['btn-info'],
'id': 'rsc-dl',
'data-content': unicode(doc)}),
HTML.a(
icon('download-alt'),
HTML.span(class_="caret"),
**{
'class_': "btn dropdown-toggle",
'data-toggle': "dropdown",
'href': "#",
'id': "dt-dl-opener",
}
),
HTML.ul(
*[HTML.li(HTML.a(
a.name or a.extension,
href="#",
onclick="document.location.href = '%s'; return false;"
% req.resource_url(rsc, ext=a.extension),
id='dt-dl-%s' % a.extension))
for a in adapters],
**dict(class_="dropdown-menu")),
class_='btn-group'),
HTML.script(literal("""\
$(document).ready(function() {
$('#rsc-dl').clickover({
html: true,
title: 'Alternative representations',
placement: '%s',
trigger: 'click'
});
});""" % doc_position)))
def gloss_with_tooltip(gloss):
person_map = {
'1': 'first person',
'2': 'second person',
'3': 'third person',
}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(literal(gloss[end:match.start()]))
abbr = match.group('abbr')
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group('personprefix'):
explanation = '%s %s' % (
person_map[match.group('personprefix')], explanation)
if match.group('personsuffix'):
explanation = '%s %s' % (
explanation, person_map[match.group('personsuffix')])
res.append(
HTML.span(
HTML.span(gloss[match.start():match.end()].lower(),
class_='sc'), **{
'data-hint': explanation,
'class': 'hint--bottom'
}))
else:
res.append(abbr)
end = match.end()
res.append(literal(gloss[end:]))
return filter(None, res)
def ipa_custom(req, segments):
rows = []
for i, data in segments.items():
title, symbol, class_, param, exists, vs = data
if exists and param and (not param.jsondata['core_list'] or i in [15, 74, 77, 84]):
rows.append(HTML.tr(
HTML.td(
parameter_link(req, literal(symbol), vs or p),
title=title, class_=class_),
HTML.th(
title.split('-')[1].strip(),
style="padding-left: 10px; text-align: left;"),
))
return HTML.table(HTML.tbody(*rows)) if rows else ''
def get_meaning_properties(req, ctx):
for attr, info, converter in [
('description', None, lambda s: s),
('typical_context', None, lambda s: s),
('semantic_field', None, lambda sf: link(req, sf)),
('semantic_category', None, lambda s: s),
('borrowed_score', hb_borrowed_score(), lambda f: '{0:.2f}'.format(f)),
('age_score', hb_age_score(), lambda f: '{0:.2f}'.format(f) if f else ''),
('simplicity_score', hb_simplicity_score(), lambda f: '{0:.2f}'.format(f)),
]:
label = attr.capitalize().replace('_', ' ')
if info:
label = HTML.span(
label, literal(' '), infobutton(info, content_type='html'))
yield (label, converter(getattr(ctx, attr)))
def col_defs(self):
return [
IntegerIdCol(
self,
'id',
sDescription=
"The number in this column is the semantic field number. It "
"is the first part of the Loanword Typology Code of the words in the "
"corresponding field."),
LinkCol(
self,
'name',
sDescription=literal(
"The first 22 fields are the fields of the Intercontinental "
"Dictionary Series meaning list, proposed by Mary Ritchie Key, and "
"ultimately based on Carl Darling Buck's (1949) <i>Dictionary of selected"
" synonyms in the principal Indo-European languages</i>. The other two "
"fields were added for the Loanword Typology project.")),
NumberOfMeanings(
self,
'number_of_meanings',
sDescription=
"This gives the number of different meanings in each "
"semantic field."),
SemanticFieldScoreCol(self,
'borrowed_score',
sDescription=literal('%s' %
hb_borrowed_score())),
SemanticFieldScoreCol(self,
'age_score',
sDescription=literal('%s' % hb_age_score())),
SemanticFieldScoreCol(self,
'simplicity_score',
sDescription=literal('%s' %
hb_simplicity_score())),
]
def get_meaning_properties(req, ctx):
for attr, info, converter in [
('description', None, lambda s: s),
('typical_context', None, lambda s: s),
('semantic_field', None, lambda sf: link(req, sf)),
('semantic_category', None, lambda s: s),
('borrowed_score', hb_borrowed_score(), lambda f: '{0:.2f}'.format(f)),
('age_score', hb_age_score(), lambda f: '{0:.2f}'.format(f) if f else ''),
('simplicity_score', hb_simplicity_score(), lambda f: '{0:.2f}'.format(f)),
]:
label = attr.capitalize().replace('_', ' ')
if info:
label = HTML.span(
label, literal(' '), infobutton(info, content_type='html'))
yield (label, converter(getattr(ctx, attr)))
def ipa_custom(req, segments):
rows = []
for i, data in segments.items():
title, symbol, class_, param, exists, vs = data
if exists and param \
and (not param.jsondata['core_list'] or i in [15, 74, 77, 84]):
rows.append(
HTML.tr(
HTML.td(parameter_link(req, literal(symbol), vs or param),
title=title,
class_=class_),
HTML.th(title.split('-', 1)[1].strip(),
style="padding-left: 10px; text-align: left;"),
))
return HTML.table(HTML.tbody(*rows)) if rows else ''
def _media(maintype_, obj, **kw):
label = kw.pop('label', None)
assert maintype_ in ['audio', 'video']
if maintype(obj) != maintype_:
raise ValueError('type mismatch: {0} and {1}'.format(maintype(obj), maintype_))
kw.setdefault('controls', 'controls')
media_element = getattr(HTML, maintype_)(
literal('Your browser does not support the <code>{0}</code> element.'.format(
maintype_)),
HTML.source(src=bitstream_url(obj, type_='web'), type=mimetype(obj)), **kw)
return HTML.div(
media_element,
HTML.br(),
link(obj, label=label),
class_='cdstar_{0}_link'.format(maintype_),
style='margin-top: 10px')
def _media(maintype_, obj, **kw):
label = kw.pop('label', None)
assert maintype_ in ['audio', 'video']
if maintype(obj) != maintype_:
raise ValueError('type mismatch: {0} and {1}'.format(maintype(obj), maintype_))
kw.setdefault('controls', 'controls')
media_element = getattr(HTML, maintype_)(
literal('Your browser does not support the <code>{0}</code> element.'.format(
maintype_)),
HTML.source(src=bitstream_url(obj, type_='web'), type=mimetype(obj)), **kw)
return HTML.div(
media_element,
HTML.br(),
link(obj, label=label),
class_='cdstar_{0}_link'.format(maintype_),
style='margin-top: 10px')
def value_table(ctx, req):
rows = []
langs = {}
for i, de in enumerate(ctx.domain):
exclusive = 0
shared = 0
for v in [
_v for _v in de.values if not _v.valueset.language.language_pk
]:
if len(v.valueset.values) > 1:
shared += 1
else:
exclusive += 1
langs[v.valueset.language_pk] = 1
cells = [
HTML.td(map_marker_img(req, de)),
HTML.td(literal(de.name)),
HTML.td(str(exclusive), class_='right'),
]
if ctx.multivalued:
cells.append(HTML.td(str(shared), class_='right'))
cells.append(HTML.td(str(exclusive + shared), class_='right'))
if exclusive or shared:
rows.append(HTML.tr(*cells))
rows.append(
HTML.tr(
HTML.td('Representation:',
colspan=str(len(cells) - 1),
class_='right'), HTML.td('%s' % len(langs),
class_='right')))
parts = []
if ctx.multivalued:
parts.append(
HTML.thead(
HTML.tr(*[
HTML.th(s, class_='right')
for s in [' ', ' ', 'excl', 'shrd', 'all']
])))
parts.append(HTML.tbody(*rows))
return HTML.table(*parts, class_='table table-condensed')
def value_table(ctx, req):
rows = []
langs = {}
for i, de in enumerate(ctx.domain):
exclusive = 0
shared = 0
for v in [_v for _v in de.values if not _v.valueset.language.language_pk]:
if len(v.valueset.values) > 1:
shared += 1
else:
exclusive += 1
langs[v.valueset.language_pk] = 1
cells = [
HTML.td(map_marker_img(req, de)),
HTML.td(literal(de.name)),
HTML.td(str(exclusive), class_='right'),
]
if ctx.multivalued:
cells.append(HTML.td(str(shared), class_='right'))
cells.append(HTML.td(str(exclusive + shared), class_='right'))
if exclusive or shared:
rows.append(HTML.tr(*cells))
rows.append(HTML.tr(
HTML.td('Representation:', colspan=str(len(cells) - 1), class_='right'),
HTML.td('%s' % len(langs), class_='right')))
parts = []
if ctx.multivalued:
parts.append(HTML.thead(
HTML.tr(*[HTML.th(s, class_='right')
for s in [' ', ' ', 'excl', 'shrd', 'all']])))
parts.append(HTML.tbody(*rows))
return HTML.table(*parts, class_='table table-condensed')
def rendered_sentence(sentence, abbrs=None, fmt='long'):
if sentence.xhtml:
return HTML.div(
HTML.div(Markup(sentence.xhtml), class_='body'), class_="sentence")
if abbrs is None:
q = DBSession.query(models.GlossAbbreviation).filter(
or_(models.GlossAbbreviation.language_pk == sentence.language_pk,
models.GlossAbbreviation.language_pk == None)
)
abbrs = dict((g.id, g.name) for g in q)
def gloss_with_tooltip(gloss):
person_map = {
'1': 'first person',
'2': 'second person',
'3': 'third person',
}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(literal(gloss[end:match.start()]))
abbr = match.group('abbr')
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group('personprefix'):
explanation = '%s %s' % (
person_map[match.group('personprefix')], explanation)
if match.group('personsuffix'):
explanation = '%s %s' % (
explanation, person_map[match.group('personsuffix')])
res.append(HTML.span(
HTML.span(gloss[match.start():match.end()].lower(), class_='sc'),
**{'data-hint': explanation, 'class': 'hint--bottom'}))
else:
res.append(abbr)
end = match.end()
res.append(literal(gloss[end:]))
return filter(None, res)
units = []
if sentence.analyzed and sentence.gloss:
analyzed = sentence.analyzed
glossed = sentence.gloss
for morpheme, gloss in zip(analyzed.split('\t'), glossed.split('\t')):
units.append(HTML.div(
HTML.div(literal(morpheme), class_='morpheme'),
HTML.div(*gloss_with_tooltip(gloss), **{'class': 'gloss'}),
class_='gloss-unit'))
return HTML.p(
HTML.div(
HTML.div(
HTML.div(literal(sentence.markup_text or sentence.name), class_='object-language'),
HTML.div(*units, **{'class': 'gloss-box'}) if units else '',
HTML.div(sentence.description, class_='translation')
if sentence.description else '',
HTML.div(sentence.original_script, class_='original_script')
if sentence.original_script else '',
#HTML.small(literal(sentence.comment)) if sentence.comment and fmt == 'long' else '',
class_='body',
),
class_="sentence",
),
class_="sentence",
)
def get_attrs(self, item):
label = item.__unicode__()
title = label
if self.dt.parameter:
label = HTML.span(map_marker_img(self.dt.req, item), literal(' '), label)
return {'label': label, 'title': title}
def text2html(text, mode='br', sep='\n\n'):
"""Turn plain text into simple HTML."""
if mode == 'p':
return HTML.div(*[HTML.p(literal(newline2br(line))) for line in text.split(sep)])
return HTML.p(literal(newline2br(text)))
def legend_items(req):
for type, label in [('recipient', 'recipient language'), ('donor', 'donor languoid')]:
yield HTML.label(
HTML.img(src=icon_url(req, type), height='20', width='20'),
literal(' ' + label),
style='margin-left: 1em; margin-right: 1em;')
def rendered_sentence(sentence, abbrs=None, fmt='long'):
"""Format a sentence as HTML."""
if sentence.xhtml:
return HTML.div(
HTML.div(Markup(sentence.xhtml), class_='body'), class_="sentence")
if abbrs is None:
q = DBSession.query(models.GlossAbbreviation).filter(
or_(models.GlossAbbreviation.language_pk == sentence.language_pk,
models.GlossAbbreviation.language_pk == None)
)
abbrs = dict((g.id, g.name) for g in q)
def gloss_with_tooltip(gloss):
person_map = {
'1': 'first person',
'2': 'second person',
'3': 'third person',
}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(gloss[end:match.start()])
abbr = match.group('abbr')
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group('personprefix'):
explanation = '%s %s' % (
person_map[match.group('personprefix')], explanation)
if match.group('personsuffix'):
explanation = '%s %s' % (
explanation, person_map[match.group('personsuffix')])
res.append(HTML.span(
HTML.span(gloss[match.start():match.end()].lower(), class_='sc'),
**{'data-hint': explanation, 'class': 'hint--bottom'}))
else:
res.append(abbr)
end = match.end()
res.append(gloss[end:])
return filter(None, res)
def alt_translation(sentence):
res = ''
if sentence.jsondata.get('alt_translation'):
text = sentence.jsondata['alt_translation']
name = ''
if ALT_TRANSLATION_LANGUAGE_PATTERN.match(text):
name, text = [t.strip() for t in text.split(':', 1)]
name = HTML.span(name + ': ')
res = HTML.div(name, HTML.span(text, class_='translation'))
return res
units = []
if sentence.analyzed and sentence.gloss:
analyzed = sentence.analyzed
glossed = sentence.gloss
for morpheme, gloss in zip(analyzed.split('\t'), glossed.split('\t')):
units.append(HTML.div(
HTML.div(morpheme, class_='morpheme'),
HTML.div(*gloss_with_tooltip(gloss), **{'class': 'gloss'}),
class_='gloss-unit'))
return HTML.div(
HTML.div(
HTML.div(
HTML.div(sentence.original_script, class_='original-script')
if sentence.original_script else '',
HTML.div(literal(sentence.markup_text or sentence.name),
class_='object-language'),
HTML.div(*units, **{'class': 'gloss-box'}) if units else '',
HTML.div(sentence.description, class_='translation')
if sentence.description else '',
alt_translation(sentence),
class_='body',
),
class_="sentence",
),
class_="sentence-wrapper",
)
def rendered_sentence(sentence, abbrs=None, fmt="long"):
"""Format a sentence as HTML."""
if sentence.xhtml:
return HTML.div(HTML.div(Markup(sentence.xhtml), class_="body"), class_="sentence")
if abbrs is None:
q = DBSession.query(models.GlossAbbreviation).filter(
or_(
models.GlossAbbreviation.language_pk == sentence.language_pk,
models.GlossAbbreviation.language_pk == None,
)
)
abbrs = dict((g.id, g.name) for g in q)
def gloss_with_tooltip(gloss):
person_map = {"1": "first person", "2": "second person", "3": "third person"}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(gloss[end : match.start()])
abbr = match.group("abbr")
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group("personprefix"):
explanation = "%s %s" % (person_map[match.group("personprefix")], explanation)
if match.group("personsuffix"):
explanation = "%s %s" % (explanation, person_map[match.group("personsuffix")])
res.append(
HTML.span(
HTML.span(gloss[match.start() : match.end()].lower(), class_="sc"),
**{"data-hint": explanation, "class": "hint--bottom"}
)
)
else:
res.append(abbr)
end = match.end()
res.append(gloss[end:])
return filter(None, res)
def alt_translation(sentence):
res = ""
if sentence.jsondata.get("alt_translation"):
text = sentence.jsondata["alt_translation"]
name = ""
if ALT_TRANSLATION_LANGUAGE_PATTERN.match(text):
name, text = [t.strip() for t in text.split(":", 1)]
name = HTML.span(name + ": ")
res = HTML.div(name, HTML.span(text, class_="translation"))
return res
units = []
if sentence.analyzed and sentence.gloss:
analyzed = sentence.analyzed
glossed = sentence.gloss
for morpheme, gloss in zip(analyzed.split("\t"), glossed.split("\t")):
units.append(
HTML.div(
HTML.div(morpheme, class_="morpheme"),
HTML.div(*gloss_with_tooltip(gloss), **{"class": "gloss"}),
class_="gloss-unit",
)
)
return HTML.p(
HTML.div(
HTML.div(
HTML.div(sentence.original_script, class_="original-script") if sentence.original_script else "",
HTML.div(literal(sentence.markup_text or sentence.name), class_="object-language"),
HTML.div(*units, **{"class": "gloss-box"}) if units else "",
HTML.div(sentence.description, class_="translation") if sentence.description else "",
alt_translation(sentence),
class_="body",
),
class_="sentence",
),
class_="sentence",
)
def rendered_sentence(sentence, abbrs=None, fmt='long'):
if sentence.xhtml:
return HTML.div(HTML.div(Markup(sentence.xhtml), class_='body'),
class_="sentence")
if abbrs is None:
q = DBSession.query(models.GlossAbbreviation).filter(
or_(models.GlossAbbreviation.language_pk == sentence.language_pk,
models.GlossAbbreviation.language_pk == None))
abbrs = dict((g.id, g.name) for g in q)
def gloss_with_tooltip(gloss):
person_map = {
'1': 'first person',
'2': 'second person',
'3': 'third person',
}
res = []
end = 0
for match in GLOSS_ABBR_PATTERN.finditer(gloss):
if match.start() > end:
res.append(literal(gloss[end:match.start()]))
abbr = match.group('abbr')
if abbr in abbrs:
explanation = abbrs[abbr]
if match.group('personprefix'):
explanation = '%s %s' % (
person_map[match.group('personprefix')], explanation)
if match.group('personsuffix'):
explanation = '%s %s' % (
explanation, person_map[match.group('personsuffix')])
res.append(
HTML.span(
HTML.span(gloss[match.start():match.end()].lower(),
class_='sc'), **{
'data-hint': explanation,
'class': 'hint--bottom'
}))
else:
res.append(abbr)
end = match.end()
res.append(literal(gloss[end:]))
return filter(None, res)
units = []
if sentence.analyzed and sentence.gloss:
analyzed = sentence.analyzed
glossed = sentence.gloss
for morpheme, gloss in zip(analyzed.split('\t'), glossed.split('\t')):
units.append(
HTML.div(HTML.div(literal(morpheme), class_='morpheme'),
HTML.div(*gloss_with_tooltip(gloss),
**{'class': 'gloss'}),
class_='gloss-unit'))
return HTML.p(
HTML.div(
HTML.div(
HTML.div(literal(sentence.markup_text or sentence.name),
class_='object-language'),
HTML.div(*units, **{'class': 'gloss-box'}) if units else '',
HTML.div(sentence.description, class_='translation')
if sentence.description else '',
HTML.div(sentence.original_script, class_='original_script')
if sentence.original_script else '',
#HTML.small(literal(sentence.comment)) if sentence.comment and fmt == 'long' else '',
class_='body',
),
class_="sentence",
),
class_="sentence",
)
def value_table(ctx, req):
rows = [HTML.tr(
HTML.td(map_marker_img(req, de)),
HTML.td(literal(de.name + ' - ' + de.description)),
HTML.td(str(len(de.values)), class_='right')) for de in ctx.domain]
return HTML.table(HTML.tbody(*rows), class_='table table-condensed')
def value_table(ctx, req):
rows = [HTML.tr(
HTML.td(map_marker_img(req, de)),
HTML.td(literal(de.name + ' - ' + de.description)),
HTML.td(str(len(de.values)), class_='right')) for de in ctx.domain]
return HTML.table(HTML.tbody(*rows), class_='table table-condensed')
def html_label(marker_spec, text):
return HTML.label(
marker_img(self.req.static_url(marker_spec)),
literal(text),
style='margin-left: 1em; margin-right: 1em;')
def value_li(de):
return HTML.label(
map_marker_img(self.req, de),
literal(de.abbr),
style='margin-left: 1em; margin-right: 1em;')
def get_form(lang_n="", isExtended=False, full=True):
"""Provides a download form.
ARGUMENTS:
- 'lang_n' : the glottocode
- 'ext' : if for the extended corpus
- 'full' : if all fields (or just the button)"""
style = ("<style>"
"form {float: left; margin: 0px; padding: 0px;} "
"input {float: left; margin-right: 10px;} "
"select {float: left; margin-right: 10px;} "
"label {float: left; margin-right: 10px;} "
"p.form {float: left; margin-right: 10px; padding: 3px;} "
"p.citeb {float: right;}"
"</style>")
lang = ("<input type=\"hidden\" id=\"f_id\" "
"name=\"id\" value=\"" + lang_n + "\">")
wav = ""
form = ""
text = ""
ext = ""
if full:
if lang_n.lower() == 'languages':
wav = ("<input type=\"hidden\" id=\"f_wav\" "
"name=\"wav\" value=\"False\">")
else:
wav = ("<input type=\"checkbox\" id=\"f_wav\" name=\"wav\" "
"value=\"True\"><label for=\"f_wav\">WAV</label>")
else:
wav = ("<input type=\"hidden\" id=\"f_wav\" "
"name=\"wav\" value=\"True\">")
form = ("<input type=\"hidden\" id=\"f_format\" "
"name=\"format\" value=\"all\">")
if isExtended and not isExtended == "False":
ext = ("<input type=\"hidden\" id=\"f_ext\" "
"name=\"extended\" value=\"True\">")
text = ("<p class=\"form\"><b>Download all files:</b></p>")
form = ("<select id=\f_format\" name=\"format\">"
"<option value=\"elan\">Elan (.eaf)</option></select>")
else:
ext = ("<input type=\"hidden\" id=\"f_ext\" "
"name=\"extended\" value=\"False\">")
text = ("<p class=\"form\"><b>Download all core files:</b></p>")
form = ("<select id=\f_format\" name=\"format\">"
"<option value=\"all\">All types</option>"
"<option value=\"praat\">Praat (.TextGrid)</option>"
"<option value=\"elan\">Elan (.eaf)</option></select>")
# "
# "<option value=\"tabular\">Tabular (.csv)</option>"
# "<option value=\"tei\">TEI (.xml)</option>"
# "<option value=\"MAUS\">MAUS parameters</option>
submit = "<input type=\"submit\" value=\"Download\">"
wav = ""
audio = (
"<p><a href='mailto:[email protected]?subject=[doreco]Audio files for'>contact us</a> if you are interested in downloading all the audio files as a zip archive.</p>")
# TODO audio license, do we exclude the audio for TLA archive too ?
if lang_n == 'languages':
audio = ('<p>Not all audio are available. Please check languages individually</p>')
else:
statement = DBSession.query(doreLanguage) \
.filter(Language.id == lang_n).first()
# print('toto', lang_n, statement)
if statement:
audioLicense = statement.audio_lic
if audioLicense == 'audio linked':
audio = ("<p>See archive for the audio files.</p>")
if statement.AUDIO is not None and statement.AUDIO != 'na':
audio = '<a style="display:block" href="https://test.nakala.fr/' + \
statement.AUDIO + '" target="_BLANK">direct link to audios</a>'
if isExtended==True:
audio = ''
if full:
return literal(style + audio + text + "<form method=\"post\" action=\"/doreLoad\">"
+ ext + lang + wav + form + submit + "</form>")
else:
return literal(style + "<form method=\"post\" action=\"/doreLoad\">"
+ ext + lang + wav + form + submit + "</form>")
def __unicode__(self):
return literal(super(Feature, self).__unicode__())
def legend(req):
return HTML.table(*[
HTML.tr(HTML.td(literal(' '), class_=SEGMENT_VALUES[n][2]),
HTML.td(SEGMENT_VALUES[n][0], style="padding-left: 10px;"))
for n in sorted(SEGMENT_VALUES.keys())
])
def legend(req):
return HTML.table(*[
HTML.tr(
HTML.td(literal(' '), class_=SEGMENT_VALUES[n][2]),
HTML.td(SEGMENT_VALUES[n][0], style="padding-left: 10px;"))
for n in sorted(SEGMENT_VALUES.keys())])
