v1.45.0

Fixed

• Fixes a bug where special characters from certain word lists weren't correctly escaped when matched with a regex. This resulted in eggs being incorrectly matched as the transition word e.g., for example.
• Fixes a crash when loading the createWorker module because the window was accessed in the global scope immediately.

Added

• When the recalibration feature flag is switched on:
  • The single title assessment is added. This assessment makes sure that you don't use superfluous H1s in your text.
  • The following assessments are not used anymore:
    • The assessment checking the length or your URL.
    • The assessment checking whether your URL contains stopwords.
  • Assessments changes:
    • Keyphrase density: changes scoring schema to account for the length of the keyphrase and changes feedback strings so that we give feedback about the number of occurrences rather than a percentage.
    • Outbound links assessment: changes the scoring schema so that red bullet instead of an orange bullet is shown when you have no outbound links.
    • Image alt attributes: if there are at least 5 images, checks whether the alt tags contain the keyphrase or synoynyms in 30-70% of all images. If there are less than 5 images, 1 image with the keyphrase or synonym in the alt tag is still scored as good.
    • Keyphrase in title: function words preceding the exact match keyphrase are ignored when determining the position of the keyphrase in the title.
    • Keyphrase length: makes the scoring scheme less strict for languages that don't have function word support, so that for these languages keyphrases with 1-6 words are scored as green, 7-9 as orange, and more than 9 as red.
    • Keyphrase in subheading: only takes H2 and H3 level subheadings into account and changes the scoring schema so that 30%-75% of these subheadings need to include the keyphrase or its synonyms. In languages without function word support, a match is only counted if all the words from the keyphrase/synonym appear in the subheading.
    • Text length: on taxonomy pages, the recommended minimum text length is increased from 150 to 250 words.
• The browser console now shows more descriptive error messages when something went wrong during analyses in the web worker.

Changed

• Improved README Usage to detail the Web Worker API.