def process_page(index, page, save, verbose):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
pagemsg("Processing")
parsed = blib.parse(page)
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-conj", "ru-conj-old"] and getparam(t, "1").startswith("pf"):
if tname == "ru-conj":
tempcall = re.sub(r"\{\{ru-conj", "{{ru-generate-verb-forms", unicode(t))
else:
tempcall = re.sub(r"\{\{ru-conj-old", "{{ru-generate-verb-forms|old=y", unicode(t))
result = expand_text(tempcall)
if not result:
pagemsg("WARNING: Error generating forms, skipping")
continue
args = rulib.split_generate_args(result)
for base in ["past_pasv_part", "ppp"]:
for i in ["", "2", "3", "4", "5", "6", "7", "8", "9"]:
val = getparam(t, base + i)
if val and val != "-":
val = re.sub("//.*", "", val)
pagemsg("Found perfective past passive participle: %s" % val)
def process_page(index, page, save, verbose):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
pagemsg("Processing")
manual_ppp_forms = ["past_pasv_part", "past_pasv_part2", "past_pasv_part3",
"past_pasv_part4", "ppp", "ppp2", "ppp3", "ppp4"]
text = unicode(page.text)
parsed = blib.parse(page)
notes = []
for t in parsed.filter_templates():
origt = unicode(t)
tname = unicode(t.name)
if tname == "ru-conj":
manual_ppps = []
for form in manual_ppp_forms:
ppp = getparam(t, form)
if ppp and ppp != "-":
manual_ppps.append(ppp)
if not manual_ppps:
continue
if [x for x in t.params if unicode(x.value) == "or"]:
pagemsg("WARNING: Skipping multi-arg conjugation: %s" % unicode(t))
continue
curvariant = getparam(t, "2")
if "+p" in curvariant or "(7)" in curvariant or "(8)" in curvariant:
pagemsg("WARNING: Found both manual PPP and PPP variant, something wrong: %s" %
unicode(t))
continue
t2 = blib.parse_text(unicode(t)).filter_templates()[0]
for form in manual_ppp_forms:
rmparam(t2, form)
variants_to_try = ["+p"]
if u"ё" in re.sub(u"ённый$", "", manual_ppps[0]):
variants_to_try.append(u"+pё")
if u"жденный" in manual_ppps[0] or u"ждённый" in manual_ppps[0]:
variants_to_try.append(u"+pжд")
notsamemsgs = []
for variant in variants_to_try:
t2.add("2", curvariant + variant)
tempcall = re.sub(r"\{\{ru-conj", "{{ru-generate-verb-forms", unicode(t2))
result = expand_text(tempcall)
if not result:
pagemsg("WARNING: Error generating forms, skipping")
continue
args = rulib.split_generate_args(result)
if "past_pasv_part" not in args:
pagemsg("WARNING: Something wrong, no past passive participle generated: %s" % unicode(t))
continue
auto_ppps = []
for form in manual_ppp_forms:
if form in args:
for ppp in re.split(",", args[form]):
if ppp and ppp != "-":
auto_ppps.append(ppp)
if manual_ppps == auto_ppps:
pagemsg("Manual PPP's %s same as auto-generated PPP's, switching to auto"
% ",".join(manual_ppps))
for form in manual_ppp_forms:
rmparam(t, form)
t.add("2", curvariant + variant)
notes.append("replaced manual PPP's with variant %s" % variant)
break
else:
notsamemsgs.append("WARNING: Manual PPP's %s not same as auto-generated PPP's %s: %s" %
(",".join(manual_ppps), ",".join(auto_ppps), unicode(t)))
else: # no break in for loop
for m in notsamemsgs:
pagemsg(m)
newt = unicode(t)
if origt != newt:
pagemsg("Replaced %s with %s" % (origt, newt))
new_text = unicode(parsed)
if new_text != text:
if verbose:
pagemsg("Replacing <%s> with <%s>" % (text, new_text))
assert notes
comment = "; ".join(notes)
if save:
pagemsg("Saving with comment = %s" % comment)
page.text = new_text
page.save(comment=comment)
else:
pagemsg("Would save with comment = %s" % comment)
def process_page(index, page, save, verbose):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("Processing")
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
parsed = blib.parse(page)
headword_template = None
see_template = None
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-noun+", "ru-proper noun+"]:
if headword_template:
pagemsg("WARNING: Multiple headword templates, skipping")
return
headword_template = t
if unicode(t.name) in ["ru-decl-noun-see"]:
if see_template:
pagemsg("WARNING: Multiple ru-decl-noun-see templates, skipping")
return
see_template = t
if not headword_template:
pagemsg("WARNING: No ru-noun+ or ru-proper noun+ templates, skipping")
return
if not see_template:
pagemsg("WARNING: No ru-decl-noun-see templates, skipping")
return
del see_template.params[:]
for param in headword_template.params:
see_template.add(param.name, param.value)
see_template.name = "ru-noun-table"
if unicode(headword_template.name) == "ru-proper noun+":
# Things are trickier for proper nouns because they default to n=sg, whereas
# ru-noun-table defaults to n=both. We have to expand both templates and
# fetch the value of n, and set it in ru-noun-table if not the same.
# 1. Generate args for headword proper-noun template, using |ndef=sg
# because ru-proper noun+ defaults to sg and ru-generate-noun-args
# would otherwise default to both.
headword_generate_template = re.sub(r"^\{\{ru-proper noun\+", "{{ru-generate-noun-args",
unicode(headword_template))
headword_generate_template = re.sub(r"\}\}$", "|ndef=sg}}", headword_generate_template)
headword_generate_result = expand_text(headword_generate_template)
if not headword_generate_result:
pagemsg("WARNING: Error generating ru-proper noun+ args")
return None
# 2. Fetch actual value of n.
headword_args = ru.split_generate_args(headword_generate_result)
headword_n = headword_args["n"]
# 3. If sg, we always need to set n=sg explicitly in ru-noun-table.
if headword_n == "s":
see_template.add("n", "sg")
# 4. If pl, leave alone, since both will default to plural only if the
# lemma is pl, else n=pl needs to be set for both.
elif headword_n == "p":
pass
# 5. If both, n=both had to have been set explicitly in the headword,
# but it's the default in ru-noun-table unless the lemma is plural.
# So remove n=both, generate the arguments, and see if the actual
# value of args.n is b (for "both"); if not, set n=both.
else:
assert headword_n == "b"
rmparam(see_template, "n")
see_generate_template = re.sub(r"^\{\{ru-noun-table", "{{ru-generate-noun-args",
unicode(see_template))
see_generate_result = expand_text(see_generate_template)
if not see_generate_result:
pagemsg("WARNING: Error generating ru-noun-table args")
return None
see_args = ru.split_generate_args(see_generate_result)
if see_args["n"] != "b":
see_template.add("n", "both")
comment = "Replace ru-decl-noun-see with ru-noun-table, taken from headword template (%s)" % unicode(headword_template.name)
if save:
pagemsg("Saving with comment = %s" % comment)
page.text = unicode(parsed)
page.save(comment=comment)
else:
pagemsg("Would save with comment = %s" % comment)
def process_page_section(index, page, section, verbose):
pagetitle = unicode(page.title())
subpagetitle = re.sub("^.*:", "", pagetitle)
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
if not page.exists():
pagemsg("WARNING: Page doesn't exist, skipping")
return None
parsed = blib.parse_text(section)
noun_table_templates = []
noun_old_templates = []
for t in parsed.filter_templates():
if unicode(t.name) == "ru-decl-noun-see":
pagemsg("Found ru-decl-noun-see, skipping")
return None
for t in parsed.filter_templates():
if unicode(t.name) == "ru-noun-table":
noun_table_templates.append(t)
if unicode(t.name) == "ru-noun-old":
noun_old_templates.append(t)
if len(noun_table_templates) > 1:
pagemsg("WARNING: Found multiple ru-noun-table templates, skipping")
return None
if len(noun_old_templates) > 1:
pagemsg("WARNING: Found multiple ru-noun-old templates, skipping")
return None
if len(noun_table_templates) < 1:
if noun_old_templates:
pagemsg("WARNING: No ru-noun-table templates but found ru-noun-old template(s): %s" %
", ".join(unicode(x) for x in noun_old_templates))
return unicode(parsed), 0, 0, 0, 0
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-noun", "ru-proper noun"]:
pagemsg("Found ru-noun or ru-proper noun, skipping")
return None
headword_templates = []
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-noun+", "ru-proper noun+"]:
headword_templates.append(t)
if len(headword_templates) > 1:
pagemsg("WARNING: Found multiple headword templates, skipping")
return None
if len(headword_templates) < 1:
return unicode(parsed), 0, 0, 0, 0
noun_table_template = noun_table_templates[0]
noun_old_template = noun_old_templates[0] if len(noun_old_templates) == 1 else None
headword_template = headword_templates[0]
decl_templates = [x for x in [noun_table_template, noun_old_template] if x]
if verbose:
pagemsg("Found headword template: %s" % unicode(headword_template))
pagemsg("Found decl template: %s" % unicode(noun_table_template))
if noun_old_template:
pagemsg("Found old decl template: %s" % unicode(noun_old_template))
orig_headword_template = unicode(headword_template)
orig_noun_table_template = unicode(noun_table_template)
genders = blib.fetch_param_chain(headword_template, "g", "g")
masculines = blib.fetch_param_chain(headword_template, "m", "m")
feminines = blib.fetch_param_chain(headword_template, "f", "f")
notrcat = getparam(headword_template, "notrcat")
filtered_headword_params = []
for param in headword_template.params:
name = unicode(param.name)
if re.search("^[gmf][0-9]*$", name) or name == "notrcat":
pass
else:
filtered_headword_params.append((param.name, param.value))
filtered_headword_template = blib.parse_text("{{ru-noun+}}").filter_templates()[0]
for name, value in filtered_headword_params:
filtered_headword_template.add(name, value)
ru_noun_table_cleaned = 0
ru_noun_table_link_copied = 0
ru_noun_changed = 0
ru_proper_noun_changed = 0
new_decl_params = []
for param in noun_table_template.params:
name = unicode(param.name)
if re.search("^[gmf][0-9]*$", name):
pagemsg("WARNING: Found g=, m= or f= in noun-table, removing: %s" %
unicode(noun_table_template))
else:
new_decl_params.append((param.name, param.value))
del noun_table_template.params[:]
for name, value in new_decl_params:
noun_table_template.add(name, value)
if orig_noun_table_template != unicode(noun_table_template):
ru_noun_table_cleaned = 1
modified_noun_table_template = blib.parse_text("{{ru-noun-table}}").filter_templates()[0]
for param in noun_table_template.params:
modified_noun_table_template.add(param.name, param.value)
# If proper noun and n is both then we need to add n=both because
# proper noun+ defaults to n=sg
if unicode(headword_template.name) == "ru-proper noun+":
generate_template = re.sub(r"^\{\{ru-noun-table", "{{ru-generate-noun-args",
unicode(noun_table_template))
generate_result = expand_text(generate_template)
if not generate_result:
pagemsg("WARNING: Error generating noun args, skipping")
return None
args = ru.split_generate_args(generate_result)
# If proper noun and n is both then we need to add n=both because
# proper noun+ defaults to n=sg
if args["n"] == "b" and not getparam(modified_noun_table_template, "n"):
pagemsg("Adding n=both to headword template")
modified_noun_table_template.add("n", "both")
# Correspondingly, if n is sg then we can usually remove n=sg;
# but we need to check that the number is actually sg with n=sg
# removed because of the possibility of plurale tantum lemmas
if args["n"] == "s":
generate_template_with_ndef = generate_template.replace("}}", "|ndef=sg}}")
generate_template_with_ndef = re.sub(r"\|n=s[^=|{}]*", "",
generate_template_with_ndef)
generate_result = expand_text(generate_template_with_ndef)
if not generate_result:
pagemsg("WARNING: Error generating noun args, skipping")
return None
ndef_args = ru.split_generate_args(generate_result)
if ndef_args["n"] == "s":
existing_n = getparam(headword_template, "n")
if existing_n and not re.search(r"^s", existing_n):
pagemsg("WARNING: Something wrong: Found n=%s, not singular" %
existing_n)
pagemsg("Removing n=sg from headword template")
rmparam(modified_noun_table_template, "n")
else:
pagemsg("WARNING: Unable to remove n= from headword template because n=%s" %
ndef_args["n"])
new_headword_template = re.sub(r"^\{\{ru-noun-table", "{{ru-noun+",
unicode(modified_noun_table_template))
existing_filtered_headword_template = unicode(filtered_headword_template)
change_existing_headword = False
if existing_filtered_headword_template != new_headword_template:
if "[" in existing_filtered_headword_template and "[" not in new_headword_template:
if blib.remove_links(existing_filtered_headword_template) == new_headword_template:
pagemsg("Headword has links but decl doesn't and they're otherwise the same, copying headword to decl")
del noun_table_template.params[:]
for param in filtered_headword_template.params:
noun_table_template.add(param.name, param.value)
ru_noun_table_link_copied = 1
ru_noun_table_cleaned = 0
else:
pagemsg("WARNING: Existing headword template %s would be overwritten with %s but links would be erased, not doing it, check manually"
% (existing_filtered_headword_template, new_headword_template))
return None
else:
pagemsg("WARNING: Existing headword template %s will be overwritten with %s"
% (existing_filtered_headword_template, new_headword_template))
change_existing_headword = True
if change_existing_headword and (not lemmas or pagetitle in lemmas):
del headword_template.params[:]
for param in modified_noun_table_template.params:
headword_template.add(param.name, param.value)
blib.set_param_chain(headword_template, genders, "g", "g")
blib.set_param_chain(headword_template, masculines, "m", "m")
blib.set_param_chain(headword_template, feminines, "f", "f")
if notrcat:
headword_template.add("notrcat", notrcat)
#genders = runoun.check_old_noun_headword_forms(headword_template, args,
# subpagetitle, pagemsg)
#if genders == None:
# return None
#new_params = []
#for param in noun_table_template.params:
# new_params.append((param.name, param.value))
#params_to_preserve = runoun.fix_old_headword_params(headword_template,
# new_params, genders, pagemsg)
#if params_to_preserve == None:
# return None
new_noun_table_template = unicode(noun_table_template)
if new_noun_table_template != orig_noun_table_template:
pagemsg("Replacing noun table %s with %s" % (orig_noun_table_template,
new_noun_table_template))
new_headword_template = unicode(headword_template)
if new_headword_template != orig_headword_template:
pagemsg("Replacing headword %s with %s" % (orig_headword_template,
new_headword_template))
if unicode(headword_template.name) == "ru-noun+":
ru_noun_changed = 1
else:
ru_proper_noun_changed = 1
return unicode(parsed), ru_noun_table_cleaned, ru_noun_table_link_copied, ru_noun_changed, ru_proper_noun_changed
def process_page_section(index, page, section, verbose):
pagetitle = unicode(page.title())
subpagetitle = re.sub("^.*:", "", pagetitle)
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
if not page.exists():
pagemsg("WARNING: Page doesn't exist, skipping")
return None
parsed = blib.parse_text(section)
noun_table_templates = []
noun_old_templates = []
for t in parsed.filter_templates():
if unicode(t.name) == "ru-decl-noun-see":
pagemsg("Found ru-decl-noun-see, skipping")
return None
for t in parsed.filter_templates():
if unicode(t.name) == "ru-noun-table":
noun_table_templates.append(t)
if unicode(t.name) == "ru-noun-old":
noun_old_templates.append(t)
if len(noun_table_templates) > 1:
pagemsg("WARNING: Found multiple ru-noun-table templates, skipping")
return None
if len(noun_old_templates) > 1:
pagemsg("WARNING: Found multiple ru-noun-old templates, skipping")
return None
if len(noun_table_templates) < 1:
if noun_old_templates:
pagemsg("WARNING: No ru-noun-table templates but found ru-noun-old template(s): %s" %
", ".join(unicode(x) for x in noun_old_templates))
return unicode(parsed), 0, 0, 0, []
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-noun+", "ru-proper noun+"]:
pagemsg("Found ru-noun+ or ru-proper noun+, skipping")
return None
headword_templates = []
for t in parsed.filter_templates():
if unicode(t.name) in ["ru-noun", "ru-proper noun"]:
headword_templates.append(t)
if len(headword_templates) > 1:
pagemsg("WARNING: Found multiple headword templates, skipping")
return None
if len(headword_templates) < 1:
return unicode(parsed), 0, 0, 0, []
noun_table_template = noun_table_templates[0]
noun_old_template = noun_old_templates[0] if len(noun_old_templates) == 1 else None
headword_template = headword_templates[0]
frobbed_manual_translit = []
decl_templates = [x for x in [noun_table_template, noun_old_template] if x]
if verbose:
pagemsg("Found headword template: %s" % unicode(headword_template))
pagemsg("Found decl template: %s" % unicode(noun_table_template))
if noun_old_template:
pagemsg("Found old decl template: %s" % unicode(noun_old_template))
# Retrieve headword translit and maybe transfer to decl
headword_tr = getparam(headword_template, "tr")
if headword_tr:
if verbose:
pagemsg("Found headword manual translit tr=%s" % headword_tr)
if "," in headword_tr:
pagemsg("WARNING: Comma in headword manual translit, skipping: %s" %
headword_tr)
return None
# Punt if multi-arg-set, can't handle yet
for decl_template in decl_templates:
for param in decl_template.params:
if not param.showkey:
val = unicode(param.value)
if val == "or":
pagemsg("WARNING: Manual translit and multi-decl templates, can't handle, skipping: %s" % unicode(decl_template))
return None
if val == "-" or val == "_" or val.startswith("join:"):
pagemsg("WARNING: Manual translit and multi-word templates, can't handle, skipping: %s" % unicode(decl_template))
return None
for i in xrange(2, 10):
if getparam(headword_template, "tr%s" % i):
pagemsg("WARNING: Headword template has translit param tr%s, can't handle, skipping: %s" % (
i, unicode(headword_template)))
return None
if runoun.arg1_is_stress(getparam(decl_template, "1")):
lemma_arg = "2"
else:
lemma_arg = "1"
lemmaval = getparam(decl_template, lemma_arg)
if not lemmaval:
lemmaval = subpagetitle
if "//" in lemmaval:
m = re.search("^(.*?)//(.*)$", lemmaval)
if m.group(2) != headword_tr:
pagemsg("WARNING: Found existing manual translit in decl template %s, but doesn't match headword translit %s; skipping" % (
lemmaval, headword_tr))
return None
else:
pagemsg("Already found manual translit in decl template %s" %
lemmaval)
else:
lemmaval += "//" + headword_tr
orig_decl_template = unicode(decl_template)
decl_template.add(lemma_arg, lemmaval)
pagemsg("Replacing decl %s with %s" % (orig_decl_template,
unicode(decl_template)))
frobbed_manual_translit = [headword_tr]
genders = blib.fetch_param_chain(headword_template, "2", "g")
bian_replaced = 0
# Change a=bi in decl to a=ia or a=ai, depending on order of anim/inan in
# headword template
for decl_template in decl_templates:
if getparam(decl_template, "a") in ["b", "bi", "bian", "both"]:
saw_in = -1
saw_an = -1
for i,g in enumerate(genders):
if re.search(r"\bin\b", g) and saw_in < 0:
saw_in = i
if re.search(r"\ban\b", g) and saw_an < 0:
saw_an = i
if saw_in >= 0 and saw_an >= 0:
orig_decl_template = unicode(decl_template)
if saw_in < saw_an:
pagemsg("Replacing a=bi with a=ia in decl template")
decl_template.add("a", "ia")
bian_replaced = 1
else:
pagemsg("Replacing a=bi with a=ai in decl template")
decl_template.add("a", "ai")
bian_replaced = 1
pagemsg("Replacing decl %s with %s" % (orig_decl_template,
unicode(decl_template)))
generate_template = re.sub(r"^\{\{ru-noun-table", "{{ru-generate-noun-args",
unicode(noun_table_template))
generate_result = expand_text(generate_template)
if not generate_result:
pagemsg("WARNING: Error generating noun args, skipping")
return None
args = ru.split_generate_args(generate_result)
genders = runoun.check_old_noun_headword_forms(headword_template, args,
subpagetitle, pagemsg)
if genders == None:
return None
new_params = []
for param in noun_table_template.params:
new_params.append((param.name, param.value))
orig_headword_template = unicode(headword_template)
params_to_preserve = runoun.fix_old_headword_params(headword_template,
new_params, genders, pagemsg)
if params_to_preserve == None:
return None
if unicode(headword_template.name) == "ru-proper noun":
# If proper noun and n is both then we need to add n=both because
# proper noun+ defaults to n=sg
if args["n"] == "b" and not getparam(headword_template, "n"):
pagemsg("Adding n=both to headword tempate")
headword_template.add("n", "both")
# Correspondingly, if n is sg then we can usually remove n=sg;
# but we need to check that the number is actually sg with n=sg
# removed because of the possibility of plurale tantum lemmas
if args["n"] == "s":
generate_template_with_ndef = generate_template.replace("}}", "|ndef=sg}}")
generate_template_with_ndef = re.sub(r"\|n=s[^=|{}]*", "",
generate_template_with_ndef)
generate_result = expand_text(generate_template_with_ndef)
if not generate_result:
pagemsg("WARNING: Error generating noun args, skipping")
return None
ndef_args = ru.split_generate_args(generate_result)
if ndef_args["n"] == "s":
existing_n = getparam(headword_template, "n")
if existing_n and not re.search(r"^s", existing_n):
pagemsg("WARNING: Something wrong: Found n=%s, not singular" %
existing_n)
else:
pagemsg("Removing n=sg from headword tempate")
rmparam(headword_template, "n")
else:
pagemsg("WARNING: Unable to remove n= from headword template because n=%s" %
ndef_args["n"])
headword_template.params.extend(params_to_preserve)
ru_noun_changed = 0
ru_proper_noun_changed = 0
if unicode(headword_template.name) == "ru-noun":
headword_template.name = "ru-noun+"
ru_noun_changed = 1
else:
headword_template.name = "ru-proper noun+"
ru_proper_noun_changed = 1
pagemsg("Replacing headword %s with %s" % (orig_headword_template, unicode(headword_template)))
return unicode(parsed), ru_noun_changed, ru_proper_noun_changed, bian_replaced, frobbed_manual_translit
"{{ru-noun+", proposed_template_text)
proposed_decl = blib.parse_text("{{ru-noun-table}}").filter_templates()[0]
for param in generate_template.params:
proposed_decl.add(param.name, param.value)
def pagemsg_with_proposed(text):
pagemsg("Proposed new template (WARNING, omits explicit gender and params to preserve from old template): %s" % proposed_template_text)
pagemsg(text)
if headword_is_proper:
generate_template.add("ndef", "sg")
generate_result = expand_text(unicode(generate_template))
if not generate_result:
pagemsg_with_proposed("WARNING: Error generating noun args, skipping")
return
genargs = ru.split_generate_args(generate_result)
if headword_is_proper and genargs["n"] == "s" and not getparam(proposed_decl, "n"):
proposed_decl.add("n", "sg")
# This will check number mismatch (and animacy mismatch, but that shouldn't
# occur as we've taken the animacy directly from the headword)
new_genders = runoun.check_old_noun_headword_forms(headword_template, genargs,
subpagetitle, pagemsg_with_proposed, laxer_comparison=True)
if new_genders == None:
return None
orig_headword_template = unicode(headword_template)
params_to_preserve = runoun.fix_old_headword_params(headword_template,
params, new_genders, pagemsg_with_proposed)
if params_to_preserve == None:
return None
def process_page(index, page, save, verbose, fix_pages):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
pagemsg("Processing")
text = unicode(page.text)
parsed = blib.parse(page)
notes = []
for t in parsed.filter_templates():
tname = unicode(t.name)
if tname in ["ru-conj", "ru-conj-old"]:
if [x for x in t.params if unicode(x.value) == "or"]:
pagemsg("WARNING: Skipping multi-arg conjugation: %s" % unicode(t))
continue
conjtype = getparam(t, "2")
if tname == "ru-conj":
tempcall = re.sub(r"\{\{ru-conj", "{{ru-generate-verb-forms", unicode(t))
else:
tempcall = re.sub(r"\{\{ru-conj-old", "{{ru-generate-verb-forms|old=y", unicode(t))
result = expand_text(tempcall)
if not result:
pagemsg("WARNING: Error generating forms, skipping")
continue
args = rulib.split_generate_args(result)
for base in ["past_pasv_part", "ppp"]:
forms_to_remove = []
if args[base] == "-":
continue
for form in re.split(",", args[base]):
origform = form
form = re.sub("//.*", "", form)
fix_form = False
if not re.search(ur"([аяеё]́?нный|тый)$", form):
pagemsg("WARNING: Past passive participle doesn't end correctly: %s" % form)
fix_form = True
unstressed_page = rulib.make_unstressed(pagetitle)
unstressed_form = rulib.make_unstressed(form)
warned = False
if unstressed_form[0] != unstressed_page[0]:
pagemsg("WARNING: Past passive participle doesn't begin with same letter, probably for wrong aspect: %s"
% form)
warned = True
fix_form = True
if form.endswith(u"нный"):
if pagetitle.endswith(u"ать"):
good_ending = u"анный"
elif pagetitle.endswith(u"ять"):
good_ending = u"янный"
else:
good_ending = u"енный"
if not unstressed_form.endswith(good_ending):
pagemsg("WARNING: Past passive participle doesn't end right, probably for wrong aspect: %s"
% form)
warned = True
fix_form = True
if not warned:
correct_form = form_ppp(conjtype, pagetitle, args)
if correct_form and unstressed_form != correct_form:
pagemsg("WARNING: Past passive participle not formed according to rule, probably wrong: found %s, expected %s"
% (unstressed_form, correct_form))
fix_form = True
if fix_form:
forms_to_remove.append(origform)
if forms_to_remove and pagetitle in fix_pages:
curvals = []
for i in ["", "2", "3", "4", "5", "6", "7", "8", "9"]:
val = getparam(t, base + i)
if val:
curvals.append(val)
newvals = [x for x in curvals if x not in forms_to_remove]
if len(curvals) - len(newvals) != len(forms_to_remove):
pagemsg("WARNING: Something wrong, couldn't remove all PPP forms %s"
% ",".join(forms_to_remove))
curindex = 1
origt = unicode(t)
for newval in newvals:
t.add(base + ("" if curindex == 1 else str(curindex)), newval)
curindex += 1
for i in xrange(curindex, 10):
rmparam(t, base + ("" if i == 1 else str(i)))
pagemsg("Replacing %s with %s" % (origt, unicode(t)))
notes.append("removed bad past pasv part(s) %s"
% ",".join(forms_to_remove))
def process_page(index, page, direc, delete_bad, fix_verbs, save, verbose):
pagetitle = unicode(page.title())
def pagemsg(txt):
msg("Page %s %s: %s" % (index, pagetitle, txt))
pagemsg("WARNING: Script no longer applies and would need fixing up")
return
def expand_text(tempcall):
return blib.expand_text(tempcall, pagetitle, pagemsg, verbose)
pagemsg("Processing")
text = unicode(page.text)
parsed = blib.parse(page)
notes = []
direc = direc.replace("3oa", u"3°a")
for t in parsed.filter_templates():
origt = unicode(t)
if unicode(t.name) in ["ru-conj"]:
conjtype = getparam(t, "1")
if not conjtype.startswith("3olda"):
continue
if conjtype.startswith("3olda") and conjtype != "3olda":
pagemsg("WARNING: Found 3a-old with variant, can't process: %s" % unicode(t))
continue
tempcall = re.sub(r"\{\{ru-conj", "{{ru-generate-verb-forms", unicode(t))
result = expand_text(tempcall)
if not result:
pagemsg("WARNING: Error generating forms, skipping")
continue
oldargs = rulib.split_generate_args(result)
rmparam(t, "6")
rmparam(t, "5")
rmparam(t, "4")
t.add("1", direc)
tempcall = re.sub(r"\{\{ru-conj", "{{ru-generate-verb-forms", unicode(t))
result = expand_text(tempcall)
if not result:
pagemsg("WARNING: Error generating forms, skipping")
continue
if delete_bad:
newargs = rulib.split_generate_args(result)
for form in ["past_m", "past_f", "past_n", "past_pl", "past_m_short",
"past_f_short", "past_n_short", "past_pl_short"]:
oldforms = re.split(",", oldargs[form]) if form in oldargs else []
newforms = re.split(",", newargs[form]) if form in newargs else []
for oldform in oldforms:
if oldform not in newforms:
formpagename = rulib.remove_accents(oldform)
formpage = pywikibot.Page(site, formpagename)
if not formpage.exists():
pagemsg("WARNING: Form page %s doesn't exist, skipping" % formpagename)
elif formpagename == pagetitle:
pagemsg("WARNING: Attempt to delete dictionary form, skipping")
else:
text = unicode(formpage.text)
if "Etymology 1" in text:
pagemsg("WARNING: Found 'Etymology 1', skipping form %s" % formpagename)
elif "----" in text:
pagemsg("WARNING: Multiple languages apparently in form, skippin form %s" % formpagename)
else:
numinfls = len(re.findall(r"\{\{inflection of\|", text))
if numinfls < 1:
pagemsg("WARNING: Something wrong, no 'inflection of' templates on page for form %s" % formpagename)
elif numinfls > 1:
pagemsg("WARNING: Multiple 'inflection of' templates on page for form %s, skipping" % formpagename)
else:
comment = "Delete erroneously created long form of %s" % pagetitle
pagemsg("Existing text for form %s: [[%s]]" % (
formpagename, text))
if save:
formpage.delete(comment)
else:
pagemsg("Would delete page %s with comment=%s" %
(formpagename, comment))
notes.append("fix 3olda -> %s" % direc)
newt = unicode(t)
if origt != newt:
pagemsg("Replaced %s with %s" % (origt, newt))
new_text = unicode(parsed)
if new_text != text and fix_verbs:
if verbose:
pagemsg("Replacing <%s> with <%s>" % (text, new_text))
assert notes
comment = "; ".join(notes)
if save:
pagemsg("Saving with comment = %s" % comment)
page.text = new_text
page.save(comment=comment)
else:
pagemsg("Would save with comment = %s" % comment)
