def ensurenl(i, e):
''' Ensure element ends with a newline, if not already present.
Passed index in parent and element. '''
if isinstance(e, sfm.Element):
if len(e):
ensurenl(len(e) - 1, e[-1])
else:
e.append(sfm.Text("\n", e.pos, e.parent))
return True
elif not e.endswith("\n"):
e.parent.insert(i + 1, sfm.Text("\n", e.pos, e.parent))
return True
return False
def _insertVerse(self, job, parent, offset, e):
job.parent = parent
if offset != 0:
last = parent[offset - (1 if offset > 0 else 0)]
if isinstance(last, sfm.Text) and last.data[-1] in " \n":
newstr = last.data.rstrip()
lc = last.data[len(newstr):]
last.data = newstr
else:
lc = None
else:
lc = None
if offset >= 0:
parent.insert(offset, job)
offset += 1
else:
parent.append(job)
if lc is not None:
t = sfm.Text(lc, pos=e.pos, parent=parent)
if offset > 0:
parent.insert(offset, t)
offset += 1
else:
parent.append(t)
return offset
def versesToEnd(self):
it = self.iiterel(0, self.doc[0])
currjob = None
lastpara = None
thispara = None
for i, e in it:
if not isinstance(e, sfm.Element):
continue
etype = e.meta.get('texttype', '').lower()
style = e.meta.get('styletype', '').lower()
if style == 'paragraph' and etype == 'versetext':
# if e.parent is not None and e.parent.name == 'p':
# print(e)
thispara = e
offset = 0
elif style == 'paragraph':
if currjob is not None:
self._insertVerse(currjob, lastpara, -1, e)
currjob = None
thispara = None
offset = 0
if e.name == 'v':
if currjob is not None:
if thispara is not None and i > 1:
offset = self._insertVerse(currjob, thispara,
i + offset, e) - i
else:
self._insertVerse(currjob, lastpara, -1, e)
currjob = sfm.Element('vp',
e.pos,
parent=e.parent,
meta=self.sheets['vp'])
currjob.append(sfm.Text(e.args[0], pos=e.pos, parent=currjob))
if thispara is not None and thispara is not e:
lastpara = thispara
if currjob is not None:
lastpara.append(currjob)
lastpara.append(sfm.Text("\n", parent=lastpara))
def fn(e):
s = str(e)
processed = False
for r in regs:
if r[0] is not None and not r[0](e.parent):
continue
ns = r[1].sub(r[2], s)
if ns != s:
processed = True
s = ns
if processed:
return sfm.Text(s, e.pos, e.parent)
else:
return e
def _g(a, e):
if isinstance(e, sfm.Text):
if pred(e):
a.append(sfm.Text(e, e.pos, a or None))
return a
if e.name in removes:
return a
e_ = sfm.Element(e.name,
e.pos,
e.args,
parent=a or None,
meta=e.meta)
reduce(_g, e, e_)
if pred(e):
a.append(e_)
elif len(e_):
a.extend(e_[:])
return a
def fn(e):
if not e.parent or not isScriptureText(e.parent):
return e
done = False
lastspace = id(e.parent[0]) != id(e)
res = []
for (islet, c) in groupby(str(e),
key=lambda x: get_ucd(ord(x), "gc") in
takslc_cats and x != "|"):
chars = "".join(c)
# print("{} = {}".format(chars, islet))
if not len(chars):
continue
if islet:
res.append(("" if lastspace else inschar) +
inschar.join(chars))
done = True
else:
res.append(chars)
lastspace = get_ucd(ord(chars[-1]), "InSC") in ("Invisible_Stacker", "Virama") \
or get_ucd(ord(chars[-1]), "gc") in ("Cf", "WS") \
or chars[-1] in (r"\|")
return sfm.Text("".join(res), e.pos, e.parent) if done else e
def new_element(self, e, name, content):
return sfm.Element(name, e.pos, [], e.parent, content=[sfm.Text("\n", e.pos)] \
+ content, meta=self.sheets[name])
def parse_element(self, e):
if isinstance(e, sfm.Text):
t = self.localise_re.sub(self.localref, str(e))
if t != e:
return [sfm.Text(t, e.pos, e.parent)]
return [e]
elif e.name == "ref" or e.name == "refnp":
res = []
isidparent = e.parent is None or e.parent.name == "id"
reps = []
curr = e.parent.index(e)
while curr + 1 < len(e.parent):
rep = e.parent[curr + 1]
if not isinstance(rep, sfm.Element) or rep.name != "rep":
break
# parse rep
m = re.match("^\s*(.*?)\s*=>\s*(.*?)\s*$", str(rep[0]), re.M)
if m:
reps.append(
(None,
re.compile(r"\b" +
m.group(1).replace("...", "[^\n\r]+") +
"(\\b|(?=\\s)|$)"), m.group(2)))
del e.parent[curr + 1]
for r in RefList.fromStr(str(e[0]), context=self.usfms.booknames):
p = self.get_passage(r,
removes=self.removes,
strippara=e.name == "refnp")
if e.name == "ref":
for i, t in enumerate(p):
if isinstance(t, sfm.Element) and t.meta.get(
'StyleType', '').lower() == 'paragraph':
if i:
p[0:i] = [
self.new_element(
e, "p1" if isidparent else "p", p[0:i])
]
break
else:
p = [
self.new_element(e, "p1" if isidparent else "p", p)
]
res.extend(p)
if len(reps):
res = self.doc.transform_text(*reps, doc=res)
return res
elif e.name == 'inc':
s = "".join(map(str, e)).strip()
for c in s:
if c == "-":
self.removes = set(sum(exclusionmap.values(), []))
else:
self.removes.difference_update(exclusionmap.get(c, []))
elif e.name == 'mod':
mod = Module(e[0].strip(), self.usfms)
return mod.parse()
else:
cs = sum(map(self.parse_element, e), [])
e[:] = cs
return [e]
def iterfn(el, top=False):
if isinstance(el, sfm.Element):
lastv = None
predels = []
for c in el[:]:
if not isinstance(c, sfm.Element) or c.name != "v":
if iterfn(c): # False if deletable ~> empty
if len(predels):
if isinstance(predels[-1], sfm.Element) \
and predels[-1].name == "p" \
and len(predels[-1]) == 1 \
and str(predels[-1][0]).strip() == "...":
predels.pop(-1)
for p in predels:
if isinstance(p, sfm.Element):
p.parent.remove(p)
lastv = None
predels = []
else:
predels.append(c)
elif isinstance(c, sfm.Element) and c.name == "v":
if lastv is not None:
for p in predels:
p.parent.remove(p)
predels = []
if ellipsis:
i = lastv.parent.index(lastv)
ell = sfm.Text("...", parent=lastv.parent)
lastv.parent.insert(i, ell)
predels.append(ell)
lastv.parent.pop(i + 1)
else:
lastv.parent.remove(lastv)
lastv = c
if lastv is not None:
lastv.parent.remove(lastv)
res = len(el) != 0
nonemptypredels = [
p for p in predels if isinstance(p, sfm.Element)
or not re.match(r"^\s*$", str(p))
]
ell = None
if len(nonemptypredels):
if ellipsis:
p = nonemptypredels[0]
i = p.parent.index(p)
st = p.parent.meta.get("styletype", "")
if st is None or st.lower() == "paragraph":
ell = sfm.Text("...", parent=p.parent)
else:
ell = sfm.Element('p',
parent=p.parent,
meta=self.sheets['p'])
ell.append(sfm.Text("...\n", parent=ell))
p.parent.insert(i, ell)
for p in predels:
p.parent.remove(p)
predels = [ell] if ell is not None else []
st = el.meta.get("styletype", "")
if (st is None or st.lower()
== "paragraph") and len(el) == len(predels):
# el.parent.remove(el)
return True if st is None else False # To handle empty markers like \pagebreak
elif re.match(r"^\s*$", str(el)) or re.match(
r"\.{3}\s*$", str(el)):
return False
return True