Attachments, Tags to and from xml, Bracket-access for tags, Recursive SimpleTags #28
Conversation
All my changes at the same place.
|
It seems that the getsizeof() I was doing in some tests about the attachments was not returning the same value as on my machine. It depends on the filesystem I guess. So, instead, I take the integer division by 1000 and compare this approximate value. |
enzyme/mkv.py
Outdated
| @@ -330,7 +338,7 @@ def fromXML(cls, xmlElement): | |||
| :type xmlElement: :class:`xml.etree.ElementTree.Element` | |||
|
|
|||
| """ | |||
| targets = Targets.fromXML(xmlElement.find('Targets')) if xmlElement.find('Targets') else Targets() | |||
| targets = Targets() if xmlElement.find('Targets') is None else Targets.fromXML(xmlElement.find('Targets')) | |||
There was a problem hiding this comment.
What's the difference with the previous code?
There was a problem hiding this comment.
Because of the order of execution of the statement, having xmlElement.find('Targets') == None would raise an error when Targets.fromXML(...) is evaluated. No what we want. Second version tests before evaluating.
|
|
||
| def to_dict(self): | ||
| info_dict = self.__dict__.copy() | ||
| info_dict['duration'] = self.duration.seconds + self.duration.microseconds/1e6 |
There was a problem hiding this comment.
Is self.duration.microseconds/1e6 pep8-valid? I think it needs extra spaces around the /
There was a problem hiding this comment.
I checked on python.org and it says :
If operators with different priorities are used, consider adding whitespace around the operators with the lowest priority(ies).
I guess that applies here : the / has higher priority than +.
| def to_dict(self): | ||
| info_dict = self.__dict__.copy() | ||
| info_dict['duration'] = self.duration.seconds + self.duration.microseconds/1e6 | ||
| info_dict['date_utc'] = ('{0:%Y}-{0:%m}-{0:%d} {0:%H}:{0:%M}:{0:%S}:{0:%f}'.format(self.date_utc))[:-3], # The official Mastroka date standard |
There was a problem hiding this comment.
Why not store the date as datetime object?
There was a problem hiding this comment.
To me, the idea behind the to_dict function was to have a convenient way to serialize the object, like with JSON. And json.dump() does not accept datetime object. Also, this output is the standard for Matroska.
enzyme/mkv.py
Outdated
| @@ -547,6 +555,11 @@ def fromelement(cls, element): | |||
| return cls(start, hidden, enabled, end, uid, string, language) | |||
| return cls(start, hidden, enabled, end, uid) | |||
|
|
|||
| def to_dict(self): | |||
| chap_dict = self.__dict__.copy() | |||
| chap_dict['start'] = self.start.seconds + self.start.microseconds/1e6 | |||
There was a problem hiding this comment.
Operation priorities again.
|
I'm not sure about the bracket access for tags, I find it confusing. mkv[50] isn't something very explicit. However something like mkv.tags[50] looks better. |
|
For the div functionality, I was thinking of the |
|
There. Removed the div's and simply renamed the |
|
Can you remove the last commit and submit a new PR for that change? Also you need to update the unittests accordingly as travis fails. |
This reverts commit 30b3a60. This will be in another pull request.
|
Woups. Forgot about the unittests this time... Sorry. But another pull request it will be. |
Restarting from my old pull requests, here are all my changes in one request.
MKV has now as "attachments" list of Attachment object with all the Mastroka elements: Name, Description, Mime Type and Data. Data is a BytesIO object with the raw binary data of the attachment.
A keyword argument in MKV.init "load_attachments" can turn off the rawData reading, saving some memory space.
Like in the Mastroka docs, SimpleTags can contain children SimpleTags.
Convenience functions using xml.etree.ElementTree Elements
mkv.tags_from_xml(xmlElement): Load Tag data from a xml Element and append it to the mkv.tags.xmlElement = mkv.tags_to_xml(): Write a xml element from the Tags in a mkv.This is useful for saving changed tag data with mkvpropedit. Also, it could be used (as I do) to create a MKV-like object from another type of file (not mkv).
Convenience getitem. MKV object can be bracket-accessed with a targettype (or targettypevalue) and this returns a list of all tags corresponding to this targettype. The sames apply for tags which return a list of simpletags corresponding to the requested name. They both return lists because you could have multiple SimpleTags with the same Name.
Ex.
Return the first tag with a targettypevalue of 50:
mkv[50][0]Return this tag's first simpletag with name 'TITLE':
mkv[50][0]['TITLE'][0]To easily add SimpleTags to a Tag, you can now use the division operator.
Ex. Adding a SimpleTag ACTOR with string "John Smith" and sub-simpletag CHARACTER with string "Foo Bar" to the first tag with targettypevalue of 50.
mkv[50][0] / SimpleTag('ACTOR', string='John Smith') / SimpleTag('CHARACTER', string='Foo Bar')This formalism is, I think, useful when using enzyme interactively.