Martin Paul Eve bio photo

Martin Paul Eve

Professor of Literature, Technology and Publishing at Birkbeck, University of London

Email Books Twitter Github Stackoverflow MLA CORE Institutional Repo Hypothes.is ORCID ID  ORCID iD Wikipedia Pictures for Re-Use

Rockaby refactoring and abstraction

Rockaby Project

I've used the snow so far this morning to start some pythonic refactoring of Rockaby.

As I mentioned in my project announcement, Rockaby started life several years ago and it was a quick morning's worth of hacking about in .NET to build something that I thought nobody would ever see. As such, the code was hideous and messy. For the first code-drop, I simply transliterated the .NET code into Python/Django. This made for some grim view manipulation:

    # build an index of where this character occurs
    CharacterPages = []
    Parts = []

    # create a list of lists (to store page numbers)
    for counter in range(0,len(pynchon.parts.Parts.FormattedPartNames)):
        Parts.append([])

    for episode in episodes:
        if character_name in episode.characters:
            Parts[episode.part - 1].append(episode.episode)
            
            epChars = episode.characters.replace('\r\n', '\n').split('\n')
            for s in epChars:
                epCharsSplit = s.split('->')
                
                if epCharsSplit[0].lower() == character_name.lower():
                    splitPages = epCharsSplit[1].split(',')
                    
                    for pageNum in splitPages:
                        CharacterPages.append(pageNum)
                        
    ret = ''

    for counter in range(0,len(pynchon.parts.Parts.FormattedPartNames)):
        part = Parts[counter]
        
        part.sort()
        
        if (len(part) > 0):
            ret = ret + '<p class="p1"><span><a href="/parts/' + pynchon.parts.Parts.SanitizedPartNames[counter + 1] + '/" title="' + pynchon.parts.Parts.FormattedPartNames[counter + 1] + '">' + pynchon.parts.Parts.FormattedPartNames[counter + 1] + '</a>:'
            
            for i in part:
                ret = ret + ' [<a href="/episodes/' + pynchon.parts.Parts.SanitizedPartNames[counter + 1] + '/synopsis/episode_' + str(i) + '/" title="Synopsis of ' + pynchon.parts.Parts.FormattedPartNames[counter + 1] + ' Episode ' + str(i) + '">' + str(i) + '</a>],'


            ret = ret[0:len(ret) - 1]
            
            ret = ret + '<br/></span></p>'

Anyway, the latest commit replaces that mess with the following view:

episodes_used = {}
presentation = []

for charsepinfo in charsinepisodes:
	if not charsepinfo.episode.part in episodes_used:
		episodes_used[charsepinfo.episode.part] = []
	episodes_used[charsepinfo.episode.part].append(charsepinfo.episode)

The models have now also been updated in the move towards abstraction from its original Pynchon-specific context. See the latest commit for details.