+ def __iter__(self):
+ return iter(self.chunk_set.all())
+
+ def __getitem__(self, chunk):
+ return self.chunk_set.all()[chunk]
+
+ def __len__(self):
+ return self.chunk_set.count()
+
+ @staticmethod
+ def trim(text, trim_begin=True, trim_end=True):
+ """
+ Cut off everything before RE_TRIM_BEGIN and after RE_TRIM_END, so
+ that eg. one big XML file can be compiled from many small XML files.
+ """
+ if trim_begin:
+ text = RE_TRIM_BEGIN.split(text, maxsplit=1)[-1]
+ if trim_end:
+ text = RE_TRIM_END.split(text, maxsplit=1)[0]
+ return text
+
+ def materialize(self):
+ """
+ Get full text of the document compiled from chunks.
+ Takes the current versions of all texts for now, but it should
+ be possible to specify a tag or a point in time for compiling.
+
+ First non-empty text's beginning isn't trimmed,
+ and last non-empty text's end isn't trimmed.
+ """
+ texts = []
+ trim_begin = False
+ text = ''
+ for chunk in self:
+ next_text = chunk.materialize()
+ if not next_text:
+ continue
+ if text:
+ # trim the end, because there's more non-empty text
+ # don't trim beginning, if `text' is the first non-empty part
+ texts.append(self.trim(text, trim_begin=trim_begin))
+ trim_begin = True
+ text = next_text
+ # don't trim the end, because there's no more text coming after `text'
+ # only trim beginning if it's not still the first non-empty
+ texts.append(self.trim(text, trim_begin=trim_begin, trim_end=False))
+ return "".join(texts)
+