def process_page(page, index, parsed):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
if blib.page_should_be_ignored(pagetitle):
pagemsg("Skipping ignored page")
return None, None
def hack_templates(parsed, langname):
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in request_templates:
if getparam(t, "lang"):
continue
if langname and langname != "English":
pagemsg(
"WARNING: Would default to English but in %s section, skipping: %s"
% (langname, origt))
continue
notes.append("add lang=en for {{%s}} with missing lang code" %
tn)
rmparam(t, "lang") # in case it's blank
# Fetch all params.
params = []
for param in t.params:
pname = unicode(param.name)
params.append((pname, param.value, param.showkey))
# Erase all params.
del t.params[:]
newline = "\n" if "\n" in unicode(t.name) else ""
t.add("lang", "en" + newline, preserve_spacing=False)
# Put remaining parameters in order.
for name, value, showkey in params:
t.add(name, value, showkey=showkey, preserve_spacing=False)
pagemsg("Replaced <%s> with <%s>" % (origt, unicode(t)))
pagemsg("Processing")
text = unicode(page.text)
notes = []
sections = re.split("(^==[^=]*==\n)", text, 0, re.M)
for j in xrange(0, len(sections), 2):
if j == 0:
langname = None
else:
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
parsed = blib.parse_text(sections[j])
hack_templates(parsed, langname)
sections[j] = unicode(parsed)
newtext = "".join(sections)
return newtext, notes
def process_page(page, index, parsed):
pagetitle = unicode(page.title())
text = unicode(page.text)
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
notes = []
if blib.page_should_be_ignored(pagetitle):
pagemsg("WARNING: Page should be ignored")
return None, None
parsed = blib.parse_text(text)
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in templates:
infl_params = templates[tn]
lang = getparam(t, "lang")
if lang:
has_lang = True
term = getparam(t, "1")
alt = getparam(t, "2")
gloss = getparam(t, "3")
else:
has_lang = False
lang = getparam(t, "1")
term = getparam(t, "2")
alt = getparam(t, "3")
gloss = getparam(t, "4")
params = []
for param in t.params:
pname = unicode(param.name).strip()
pval = unicode(param.value).strip()
if pname in ["lang", "1", "2", "3"] or (pname == "4"
and not has_lang):
continue
pagemsg("WARNING: Unrecognized param %s, skipping" % pname)
return None, None
# Erase all params.
del t.params[:]
# Put back new params.
blib.set_template_name(t, "inflection of")
t.add("1", lang)
t.add("2", term)
t.add("3", alt)
for index, tag in enumerate(infl_params):
t.add(str(index + 4), tag)
if gloss:
t.add("t", gloss)
if origt != unicode(t):
pagemsg("Replaced %s with %s" % (origt, unicode(t)))
notes.append("replace {{%s}} with {{inflection of}}" % tn)
return unicode(parsed), notes
def process_page(page, index, parsed):
pagetitle = unicode(page.title())
text = unicode(page.text)
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
notes = []
if blib.page_should_be_ignored(pagetitle):
pagemsg("WARNING: Page should be ignored")
return None, None
parsed = blib.parse_text(text)
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn == "doublet":
params = []
for param in t.params:
pname = unicode(param.name).strip()
pval = unicode(param.value).strip()
showkey = param.showkey
if not pval:
continue
if pname == "3":
pname = "alt1"
showkey = True
elif pname == "4":
pname = "t1"
showkey = True
elif pname in [
"t", "gloss", "tr", "ts", "pos", "lit", "alt", "sc",
"id", "g"
]:
pname = pname + "1"
elif pname in ["1", "2", "notext", "nocap", "nocat"]:
pass
else:
pagemsg(
"WARNING: Unrecognized param %s=%s in %s, skipping" %
(pname, pval, origt))
break
params.append((pname, pval, showkey))
else: # No break
# Erase all params.
del t.params[:]
# Put back new params.
for pname, pval, showkey in params:
t.add(pname, pval, showkey=showkey, preserve_spacing=False)
if origt != unicode(t):
pagemsg("Replaced %s with %s" % (origt, unicode(t)))
notes.append("restructure {{doublet}} for new syntax")
return unicode(parsed), notes
def process_text_on_page(index, pagetitle, text):
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
#pagemsg("Processing")
if blib.page_should_be_ignored(pagetitle):
#pagemsg("WARNING: Page should be ignored")
return
sections = re.split("(^==[^=\n]+==\n)", text, 0, re.M)
langs = []
for j in xrange(1, len(sections), 2):
m = re.search("^==(.*)==$", sections[j])
langs.append(m.group(1))
pagemsg("Languages = %s" % ",".join(langs))
def process_text_on_page(pagetitle, index, text):
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
if blib.page_should_be_ignored(pagetitle):
pagemsg("WARNING: Page should be ignored")
return None, None
parsed = blib.parse_text(text)
for t in parsed.filter_templates():
tn = tname(t)
if tn == "form of":
lang = getparam(t, "lang")
if lang:
form = getparam(t, "1")
else:
form = getparam(t, "2")
form_of_forms[form] += 1
def process_text_on_page(index, pagetitle, text):
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
#pagemsg("Processing")
if blib.page_should_be_ignored(pagetitle):
#pagemsg("WARNING: Page should be ignored")
return
if all(x not in text for x in inflection_of_templates):
return
subsections = re.split("(^==+[^=\n]+==+\n)", text, 0, re.M)
for j in xrange(2, len(subsections), 2):
for template in inflection_of_templates:
if re.search(
r"^[#*]+ \{\{%s.*\n[#*]+ \{\{%s.*" % (template, template),
subsections[j], re.M):
pagemsg("Found subsection with combinable %s:\n%s" %
(template, subsections[j].strip()))
def process_text_on_page(index, pagetitle, text):
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
#pagemsg("Processing")
if blib.page_should_be_ignored(pagetitle):
#pagemsg("WARNING: Page should be ignored")
return
if "inflection of" not in text:
return
parsed = blib.parse_text(text)
templates_to_replace = []
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in ["inflection of"]:
if getparam(t, "lang"):
term_param = 1
else:
term_param = 2
for param in t.params:
pname = unicode(param.name).strip()
pval = unicode(param.value).strip()
if re.search("^[0-9]+$", pname):
if int(pname) >= term_param + 2:
if pval in ["and", "or", ";", ";<!--\n-->"
] or "/" in pval or "," in pval:
pagemsg("Found template: %s" % origt)
break
return
def process_page(page, index, parsed):
pagetitle = unicode(page.title())
text = unicode(page.text)
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
notes = []
if blib.page_should_be_ignored(pagetitle):
pagemsg("WARNING: Page should be ignored")
return None, None
def combine_doublets(m):
first = blib.parse_text(m.group(1))
rest = blib.parse_text(m.group(2))
t1 = list(first.filter_templates())[0]
if getparam(t1, "3") or getparam(t1, "4") or getparam(t1, "alt2") or getparam(t1, "alt3"):
pagemsg("WARNING: Can't combine %s, first template already has multiple terms" %
m.group(0))
return m.group(0)
next_index = 2
lang = getparam(t1, "1")
for t in rest.filter_templates(recursive=False):
tlang = getparam(t, "1")
if lang != tlang:
pagemsg("WARNING: Lang %s in continuation template %s not same as lang %s in first template %s" % (
tlang, unicode(t), lang, unicode(t1)))
return m.group(0)
for param in t.params:
pname = unicode(param.name).strip()
pval = unicode(param.value).strip()
if not pval:
continue
if pname == "2":
t1.add(str(next_index + 1), pval)
elif pname == "3":
t1.add("alt%s" % next_index, pval)
elif pname == "4":
t1.add("t%s" % next_index, pval)
elif pname in ["t", "gloss", "tr", "ts", "pos", "lit", "alt", "sc",
"id", "g"]:
t1.add("%s%s" % (pname, next_index), pval)
elif pname in ["t1", "gloss1", "tr1", "ts1", "pos1", "lit1", "alt1", "sc1",
"id1", "g1"]:
t1.add("%s%s" % (pname[:-1], next_index), pval)
elif pname in ["1", "notext", "nocap", "nocat"]:
pass
else:
pagemsg("WARNING: Unrecognized param %s=%s in %s, skipping" %
(pname, pval, unicode(t)))
return m.group(0)
next_index += 1
for param in ["notext", "nocap", "nocat"]:
val = getparam(t1, param)
rmparam(t1, param)
if val:
t1.add(param, val)
newtext = unicode(t1)
pagemsg("Replaced %s with %s" % (m.group(0), newtext))
return newtext
newtext = re.sub(r"(\{\{doublet\|(?:[^{}\n]|\{\{[^{}\n]*\}\})*\}\})((?:(?:, *|,? *and *)\{\{(?:m|l|doublet)\|(?:[^{}\n]|\{\{[^{}\n]*\}\})*\}\})+)", combine_doublets, text)
if newtext != text:
notes.append("combine adjacent doublets")
text = newtext
return text, notes
def process_page(page, index, parsed):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
if blib.page_should_be_ignored(pagetitle):
pagemsg("Skipping ignored page")
return None, None
def hack_templates(parsed, langname, subsectitle, langnamecode=None,
is_citation=False):
if langname not in blib.languages_byCanonicalName:
if not is_citation:
langnamecode = None
else:
langnamecode = blib.languages_byCanonicalName[langname]["code"]
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in ["citation", "citations"] and is_citation:
langnamecode = getparam(t, "lang")
elif tn in quote_templates:
if getparam(t, "lang"):
continue
lang = getparam(t, "language")
if lang:
notes.append("Convert language=%s to lang=%s in %s" % (lang, lang, tn))
else:
if subsectitle.startswith("Etymology") or subsectitle.startswith("Pronunciation"):
pagemsg("WARNING: Found template in %s section for language %s, might be different language, skipping: %s" % (
subsectitle, langname, origt))
continue
if not langnamecode:
pagemsg("WARNING: Unrecognized language %s, unable to add language to %s" % (langname, tn))
continue
if langnamecode == "en" and (getparam(t, "translation") or getparam(t, "t")):
pagemsg("WARNING: Translation section in putative English quote, skipping: %s" % origt)
continue
if langnamecode == "mul":
notes.append("infer lang=en for %s in Translingual section and add termlang=mul" % tn)
else:
notes.append("infer lang=%s for %s based on section it's in" % (langnamecode, tn))
rmparam(t, "language")
# Fetch all params.
params = []
for param in t.params:
pname = unicode(param.name)
params.append((pname, param.value, param.showkey))
# Erase all params.
del t.params[:]
if langnamecode == "mul":
termlang = langnamecode
langnamecode = "en"
else:
termlang = None
# Put lang parameter.
newline = "\n" if "\n" in unicode(t.name) else ""
t.add("lang", langnamecode + newline, preserve_spacing=False)
if termlang:
t.add("termlang", termlang + newline, preserve_spacing=False)
# Put remaining parameters in order.
for name, value, showkey in params:
t.add(name, value, showkey=showkey, preserve_spacing=False)
pagemsg("Replaced <%s> with <%s>" % (origt, unicode(t)))
return langnamecode
pagemsg("Processing")
text = unicode(page.text)
notes = []
sections = re.split("(^==[^=]*==\n)", text, 0, re.M)
if not pagetitle.startswith("Citations"):
for j in xrange(2, len(sections), 2):
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
subsections = re.split("(^==.*==\n)", sections[j], 0, re.M)
for k in xrange(2, len(subsections), 2):
m = re.search("^===*(.*?)=*==\n$", subsections[k - 1])
assert m
subsectitle = m.group(1)
parsed = blib.parse_text(subsections[k])
hack_templates(parsed, langname, subsectitle)
subsections[k] = unicode(parsed)
sections[j] = "".join(subsections)
else:
# Citation section?
langnamecode = None
for j in xrange(0, len(sections), 2):
if j == 0:
langname = "Unknown"
else:
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
parsed = blib.parse_text(sections[j])
langnamecode = hack_templates(parsed, langname, "Unknown",
langnamecode=langnamecode, is_citation=True)
sections[j] = unicode(parsed)
newtext = "".join(sections)
return newtext, notes
def process_page(page, index, parsed):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
if blib.page_should_be_ignored(pagetitle):
pagemsg("Skipping ignored page")
return None, ""
def hack_templates(parsed, subsectitle):
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in quote_templates:
if not getparam(t, "nocat"):
continue
if getparam(t, "lang").strip() != "en":
continue
notes.append(
"convert nocat=1 in lang=en Translingual section to termlang=mul"
)
# Fetch all params.
params = []
for param in t.params:
pname = unicode(param.name)
if pname.strip() != "nocat":
params.append((pname, param.value, param.showkey))
# Erase all params.
del t.params[:]
# Put lang and termlang parameters.
newline = "\n" if "\n" in unicode(t.name) else ""
t.add("lang", "en" + newline, preserve_spacing=False)
t.add("termlang", "mul" + newline, preserve_spacing=False)
# Put remaining parameters in order.
for name, value, showkey in params:
t.add(name, value, showkey=showkey, preserve_spacing=False)
pagemsg("Replaced <%s> with <%s>" % (origt, unicode(t)))
pagemsg("Processing")
text = unicode(page.text)
notes = []
sections = re.split("(^==[^=]*==\n)", text, 0, re.M)
for j in xrange(2, len(sections), 2):
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
if langname != "Translingual":
continue
subsections = re.split("(^==.*==\n)", sections[j], 0, re.M)
for k in xrange(2, len(subsections), 2):
m = re.search("^===*(.*?)=*==\n$", subsections[k - 1])
assert m
subsectitle = m.group(1)
parsed = blib.parse_text(subsections[k])
hack_templates(parsed, subsectitle)
subsections[k] = unicode(parsed)
sections[j] = "".join(subsections)
newtext = "".join(sections)
return newtext, notes
def process_page(page, index, parsed, lang_in_1):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
if blib.page_should_be_ignored(pagetitle):
pagemsg("Skipping ignored page")
return None, None
langparam = "1" if lang_in_1 else "lang"
def hack_templates(parsed, langname, langnamecode=None, is_citation=False):
if langname not in blib.languages_byCanonicalName:
if not is_citation:
langnamecode = None
else:
langnamecode = blib.languages_byCanonicalName[langname]["code"]
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn in ["citation", "citations"] and is_citation:
langnamecode = getparam(t, "lang") or getparam(t, "1")
if tn in templates_to_process:
if getparam(t, langparam):
pass
elif not langnamecode:
pagemsg(
"WARNING: Unrecognized language %s, unable to add language to %s"
% (langname, origt))
else:
notes.append(
"infer %s=%s for {{%s}} based on section it's in" %
(langparam, langnamecode, tn))
newline = "\n" if "\n" in unicode(t.name) else ""
if langparam == "1":
if t.has("lang"):
pagemsg(
"WARNING: Template has lang=, removing: %s" %
origt)
notes.append("remove lang= from {{%s}}" % tn)
rmparam(t, "lang")
t.add(langparam,
langnamecode + newline,
preserve_spacing=False)
else:
# Fetch all params.
params = []
for param in t.params:
pname = unicode(param.name)
params.append((pname, param.value, param.showkey))
# Erase all params.
del t.params[:]
t.add(langparam,
langnamecode + newline,
preserve_spacing=False)
# Put remaining parameters in order.
for name, value, showkey in params:
t.add(name,
value,
showkey=showkey,
preserve_spacing=False)
if tn in templates_to_rename:
blib.set_template_name(t, templates_to_rename[tn])
notes.append("rename {{%s}} to {{%s}}" %
(tn, templates_to_rename[tn]))
newt = unicode(t)
if newt != origt:
pagemsg("Replaced <%s> with <%s>" % (origt, newt))
return langnamecode
pagemsg("Processing")
text = unicode(page.text)
notes = []
sections = re.split("(^==[^=]*==\n)", text, 0, re.M)
if not pagetitle.startswith("Citations"):
for j in xrange(2, len(sections), 2):
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
parsed = blib.parse_text(sections[j])
hack_templates(parsed, langname)
sections[j] = unicode(parsed)
else:
# Citation section?
langnamecode = None
for j in xrange(0, len(sections), 2):
if j == 0:
langname = "Unknown"
else:
m = re.search("^==(.*)==\n$", sections[j - 1])
assert m
langname = m.group(1)
parsed = blib.parse_text(sections[j])
langnamecode = hack_templates(parsed,
langname,
langnamecode=langnamecode,
is_citation=True)
sections[j] = unicode(parsed)
newtext = "".join(sections)
return newtext, notes
def process_text_on_page(pagetitle, index, text):
global args
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
notes = []
if blib.page_should_be_ignored(pagetitle):
pagemsg("WARNING: Page should be ignored")
return None, None
parsed = blib.parse_text(text)
templates_to_replace = []
for t in parsed.filter_templates():
origt = unicode(t)
tn = tname(t)
if tn == "inflection of":
params = []
if getparam(t, "lang"):
lang = getparam(t, "lang")
term_param = 1
notes.append("moved lang= in {{%s}} to 1=" % tn)
else:
lang = getparam(t, "1")
term_param = 2
tr = getparam(t, "tr")
term = getparam(t, str(term_param))
alt = getparam(t, "alt") or getparam(t, str(term_param + 1))
tags = []
for param in t.params:
pname = unicode(param.name).strip()
pval = unicode(param.value).strip()
if re.search("^[0-9]+$", pname):
if int(pname) >= term_param + 2:
if pval:
tags.append(pval)
else:
notes.append("removed empty tags from {{%s}}" % tn)
elif pname not in ["lang", "tr", "alt"]:
params.append((pname, pval, param.showkey))
if lang == "pl":
newtags = ["nv" if tag == "other" else tag for tag in tags]
if tags != newtags:
notes.append(
"replaced 'other' with 'nv' in Polish {{%s}}" % tn)
tags = newtags
# Erase all params.
del t.params[:]
# Put back new params.
# Strip comment continuations and line breaks. Such cases generally have linebreaks after semicolons
# as well, but we remove those. (FIXME, consider preserving them.)
t.add("1", remove_comment_continuations(lang))
t.add("2", remove_comment_continuations(term))
tr = remove_comment_continuations(tr)
if tr:
t.add("tr", tr)
t.add("3", remove_comment_continuations(alt))
next_tag_param = 4
# Put back the tags into the template and note stats on bad tags
for tag in tags:
t.add(str(next_tag_param), tag)
next_tag_param += 1
for pname, pval, showkey in params:
t.add(pname, pval, showkey=showkey, preserve_spacing=False)
if origt != unicode(t):
pagemsg("Replaced %s with %s" % (origt, unicode(t)))
return unicode(parsed), notes
