[SinbadEV's Crazy Ideas] Comic Book Reader + XML
Posted: 09 Nov 2010, 18:54
I came up with an idea that is loosely based on digital comics I've seen online and would be an extension of the existing "CBR" file specification... a "cbx" file or something... anyways...
***
Comic Book Reader + XML
Archive will contain ÔÇ£Sequential Image FilesÔÇØ along with 1 or more XML file providing the information typically contained within the .txt and .sfv files along with additional frame/text data, for backwards compatibility the presence of old style txt and sfv will be tolerated.
Frame/Text Data:
Each ÔÇ£pageÔÇØ will be defined in the xml file with things like itÔÇÖs facing page match, if it is a multi-page spanning page, if it is a cover or ad of ÔÇ£errataÔÇØ or whatever else I can come up with, for comics containing more than one story it may define which story a page is part of and the title of this story etc.
There will also be a list of ÔÇ£framesÔÇØ which are co-ordinates defining a rectangle on a specific page, these will be the ÔÇ£viewsÔÇØ that are presented to users if they are viewing the comic in ÔÇ£zoomÔÇØ mode.
There will be a list of ÔÇ£bubblesÔÇØ which are the co-ordinates of a square (or perhaps other shape) containing a text box or speech bubble or sound effect splash or other textual data. In addition the content text and an ÔÇ£read-orderÔÇØ can be specified which could be useful for making the text of the page easier to read at lower zoom/resolution but could also be used for translations (aka an original Japanese comic could have the read-order of pages, frames and bubbles reversed and the bubbles and sound effects could be clicked to reveal the English equivalent)
In addition to the new specification and apps to read comics using it a suite of tools for processing images or existing comic archives and producing these XML files and archives would be useful. As a long term goal coming up with an intelligent algorithm for creating the XML files automatically and OCRing the text would be great, though I imagine someone would need to confirm and manually correct processed comics.
***
Comic Book Reader + XML
Archive will contain ÔÇ£Sequential Image FilesÔÇØ along with 1 or more XML file providing the information typically contained within the .txt and .sfv files along with additional frame/text data, for backwards compatibility the presence of old style txt and sfv will be tolerated.
Frame/Text Data:
Each ÔÇ£pageÔÇØ will be defined in the xml file with things like itÔÇÖs facing page match, if it is a multi-page spanning page, if it is a cover or ad of ÔÇ£errataÔÇØ or whatever else I can come up with, for comics containing more than one story it may define which story a page is part of and the title of this story etc.
There will also be a list of ÔÇ£framesÔÇØ which are co-ordinates defining a rectangle on a specific page, these will be the ÔÇ£viewsÔÇØ that are presented to users if they are viewing the comic in ÔÇ£zoomÔÇØ mode.
There will be a list of ÔÇ£bubblesÔÇØ which are the co-ordinates of a square (or perhaps other shape) containing a text box or speech bubble or sound effect splash or other textual data. In addition the content text and an ÔÇ£read-orderÔÇØ can be specified which could be useful for making the text of the page easier to read at lower zoom/resolution but could also be used for translations (aka an original Japanese comic could have the read-order of pages, frames and bubbles reversed and the bubbles and sound effects could be clicked to reveal the English equivalent)
In addition to the new specification and apps to read comics using it a suite of tools for processing images or existing comic archives and producing these XML files and archives would be useful. As a long term goal coming up with an intelligent algorithm for creating the XML files automatically and OCRing the text would be great, though I imagine someone would need to confirm and manually correct processed comics.