Code review comment for lp:~dosage-dev/dosage/bunch-of-comics

  1. There are a number of coding style infractions:
    * Only 2 lines between top-level suites.
    * Lines (not regular expressions) longer than 80 columns.
  2. The regular expression for HateSong.prevSearch can probably be simplified to use exact lengths.
  3. Bellen.imageSearch has an odd regular expression, spaces seem like the kind of thing that would appear in a "src" attribute.
  4. Effbot has a reasonable HTML entity decoder implementation[1] that could be the start of a better normalizeUrl implementation.

[1] http://effbot.org/zone/re-sub.htm#unescape-html