X-Git-Url: https://git.mdrn.pl/redakcja.git/blobdiff_plain/6e88635c94b36eaff84bd4274983af977986b699..3acdb5c1daebfd692344bb93cd8d0594607caa9b:/apps/catalogue/xml_tools.py?ds=inline diff --git a/apps/catalogue/xml_tools.py b/apps/catalogue/xml_tools.py old mode 100755 new mode 100644 index 522806b6..242714b6 --- a/apps/catalogue/xml_tools.py +++ b/apps/catalogue/xml_tools.py @@ -1,6 +1,5 @@ # -*- coding: utf-8 -*- from copy import deepcopy -from functools import wraps import re from lxml import etree @@ -14,110 +13,6 @@ class ParseError(BaseException): pass -def obj_memoized(f): - """ - A decorator that caches return value of object methods. - The cache is kept with the object, in a _obj_memoized property. - """ - @wraps(f) - def wrapper(self, *args, **kwargs): - if not hasattr(self, '_obj_memoized'): - self._obj_memoized = {} - key = (f.__name__,) + args + tuple(sorted(kwargs.iteritems())) - try: - return self._obj_memoized[key] - except TypeError: - return f(self, *args, **kwargs) - except KeyError: - self._obj_memoized[key] = f(self, *args, **kwargs) - return self._obj_memoized[key] - return wrapper - - -class GradedText(object): - _edoc = None - - ROOT = 'utwor' - RDF = '{http://www.w3.org/1999/02/22-rdf-syntax-ns#}RDF' - - def __init__(self, text): - self._text = text - - @obj_memoized - def is_xml(self): - """ - Determines if it's a well-formed XML. - - >>> GradedText("").is_xml() - True - >>> GradedText("").is_xml() - False - """ - try: - self._edoc = etree.fromstring(self._text) - except etree.XMLSyntaxError: - return False - return True - - @obj_memoized - def is_wl(self): - """ - Determines if it's an XML with a and a master tag. - - >>> GradedText("").is_wl() - True - >>> GradedText("").is_wl() - False - """ - if self.is_xml(): - e = self._edoc - # FIXME: there could be comments - ret = e.tag == self.ROOT and ( - len(e) == 1 and e[0].tag in MASTERS or - len(e) == 2 and e[0].tag == self.RDF - and e[1].tag in MASTERS) - if ret: - self._master = e[-1].tag - del self._edoc - return ret - else: - return False - - @obj_memoized - def is_broken_wl(self): - """ - Determines if it at least looks like broken WL file - and not just some untagged text. - - >>> GradedText("<").is_broken_wl() - True - >>> GradedText("some text").is_broken_wl() - False - """ - if self.is_wl(): - return True - text = self._text.strip() - return text.startswith('') and text.endswith('') - - def master(self): - """ - Gets the master tag. - - >>> GradedText("").master() - 'powiesc' - """ - assert self.is_wl() - return self._master - - @obj_memoized - def has_trim_begin(self): - return RE_TRIM_BEGIN.search(self._text) - - @obj_memoized - def has_trim_end(self): - return RE_TRIM_END.search(self._text) - - def _trim(text, trim_begin=True, trim_end=True): """ Cut off everything before RE_TRIM_BEGIN and after RE_TRIM_END, so @@ -153,26 +48,6 @@ def compile_text(parts): return "".join(texts) -def change_master(text, master): - """ - Changes the master tag in a WL document. - """ - e = etree.fromstring(text) - e[-1].tag = master - return unicode(etree.tostring(e, encoding="utf-8"), 'utf-8') - - -def basic_structure(text, master): - e = etree.fromstring(''' - - - -''' % (TRIM_BEGIN, TRIM_END)) - e[0].tag = master - e[0][0].tail = "\n"*3 + text + "\n"*3 - return unicode(etree.tostring(e, encoding="utf-8"), 'utf-8') - - def add_trim_begin(text): trim_tag = etree.Comment(TRIM_BEGIN) e = etree.fromstring(text) @@ -230,7 +105,7 @@ def split_xml(text): for a in name_elem.findall('.//' + tag): a.text='' del a[:] - name = etree.tostring(name_elem, method='text', encoding='utf-8') + name = etree.tostring(name_elem, method='text', encoding='utf-8').strip() # in the original, remove everything from the start of the last chapter parent = element.getparent()