dirty epub fixes

["Reviczky, Adam" <adam.reviczky@kcl.ac.uk>]

[-- Attachment #1: Type: text/plain, Size: 1464 bytes --]

Hans,

Since I didn't got a working epub for the first few tries, I did some further inspections with epubcheck[1].

Using this minimal example:
\setupbackend[export=test.xml,xhtml=test.xhtml,css={test.css,mathml.css}]
\starttext
Test
\stoptext

The main errors are:
* opf: id has to be either "stylesheet" for css or the xhtml filename without the suffix
* opf: itemref needs to be the name of the xhtml without the suffix

Here's a very dirty quick fix I did to get a working epub:
add after line 112 or so:
+             local id = file.removesuffix(filename)

line 116:
-                 used[#used+1] = format("<item id='%s' href='%s' media-type='%s'/>",i,filename,mime)
+                 if suffix == "css" then
+                     id = "stylesheet"
+                 elseif suffix == "ncx" then
+                     id = file.suffix(filename)
+                 end
+                 used[#used+1] = format("<item id='%s' href='%s' media-type='%s'/>",id,filename,mime)

line 121:
-         package   = format(package,identifier,identifier,concat(used,"\n"),root)
+         package   = format(package,identifier,identifier,concat(used,"\n"),file.removesuffix(root))

I'm sure you make a better fix (I've attached the patched one in case it's easier to diff).
Not sure if you want to do something about the other errors, but it'd be nice to have a valid epub in the end.

Adam


[1] http://code.google.com/p/epubcheck/

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: epub.log --]
[-- Type: text/x-log; name="epub.log", Size: 990 bytes --]

Epubcheck Version 1.2

ERROR: test.tree/test.epub: extra field length for first filename must be 0, but was 28
ERROR: test.tree/test.epub/OPS/test.opf(3): value of attribute "unique-identifier" is invalid; must be an XML name without colons
ERROR: test.tree/test.epub/OPS/test.opf(8): value of attribute "id" is invalid; must be an XML name without colons
ERROR: test.tree/test.epub/OPS/test.opf(13): value of attribute "id" is invalid; must be an XML name without colons
ERROR: test.tree/test.epub/OPS/test.opf(14): value of attribute "id" is invalid; must be an XML name without colons
ERROR: test.tree/test.epub/OPS/test.opf(15): value of attribute "id" is invalid; must be an XML name without colons
ERROR: test.tree/test.epub/OPS/test.opf(18): item with id 'ncx' not found
ERROR: test.tree/test.epub/OPS/test.opf(19): item with id 'test.xhtml' not found
ERROR: test.tree/test.epub/OPS/test.xhtml(12): elements from namespace "" are not allowed

Check finished with warnings or errors!

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #3: my-mtx-epub.lua --]
[-- Type: text/x-lua; name="my-mtx-epub.lua", Size: 4279 bytes --]

if not modules then modules = { } end modules ['mtx-epub'] = {
    version   = 1.001,
    comment   = "companion to mtxrun.lua",
    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
    copyright = "PRAGMA ADE / ConTeXt Development Team",
    license   = "see context related readme files"
}

local format = string.format
local concat = table.concat

local helpinfo = [[
--make                create epub zip file

example:

mtxrun --script epub --make mydocument
]]

local application = logs.application {
    name     = "mtx-epub",
    banner   = "ConTeXt EPUB Helpers 0.10",
    helpinfo = helpinfo,
}

-- script code

scripts      = scripts      or { }
scripts.epub = scripts.epub or { }

local mimetype = "application/epub+zip"

local container = [[
<?xml version="1.0" encoding="UTF-8" ?>

<container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
    <rootfiles>
        <rootfile full-path="OPS/%s" media-type="application/oebps-package+xml"/>
    </rootfiles>
</container>
]]

local package = [[
<?xml version="1.0"?>

<package version="2.0" xmlns="http://www.idpf.org/2007/opf" unique-identifier="%s">

    <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf">
        <dc:title>My Title</dc:title>
        <dc:language>en</dc:language>
        <dc:identifier id="%s" />
        <dc:creator opf:file-as="Self, My" opf:role="aut">MySelf</dc:creator>
    </metadata>

    <manifest>
        %s
    </manifest>

    <spine toc="ncx">
        <itemref idref="%s" />
    </spine>

</package>
]]

local mimetypes = {
    xhtml = "application/xhtml+xml",
    css   = "text/css",
}

-- specification = {
--     name = "document",
--     identifier = "123",
--     root = "a.xhtml",
--     files = {
--         "a.xhtml",
--         "b.css",
--         "c.png",
--     }
-- }

function scripts.epub.make()

    local filename = environment.files[1]

    if filename and filename ~= "" then

        filename = file.basename(filename)
        local specfile = file.replacesuffix(filename,"specification")
        local specification = lfs.isfile(specfile) and dofile(specfile) or { }

     -- inspect(specification)

        local name       = specification.name       or file.removesuffix(filename)
        local identifier = specification.identifier or os.uuid()
        local files      = specification.files      or { file.addsuffix(filename,"xhtml") }
        local root       = specification.root       or files[1]

        local epubname   = name
        local epubpath   = file.replacesuffix(name,"tree")
        local epubfile   = file.replacesuffix(name,"epub")
        local epubroot   = file.replacesuffix(name,"opf")

        lfs.mkdir(epubpath)
        lfs.mkdir(file.join(epubpath,"META-INF"))
        lfs.mkdir(file.join(epubpath,"OPS"))

        local used  = { }

        for i=1,#files do
            local filename = files[i]
            local suffix = file.suffix(filename)
            local id = file.removesuffix(filename)
            local mime = mimetypes[suffix]
            if mime then
                file.copy(filename,file.join(epubpath,"OPS",filename))
                if suffix == "css" then
                    id = "stylesheet"
                elseif suffix == "ncx" then
                    id = file.suffix(filename)
                end
                used[#used+1] = format("<item id='%s' href='%s' media-type='%s'/>",id,filename,mime)
            end
        end

        container = format(container,epubroot)
        package   = format(package,identifier,identifier,concat(used,"\n"),file.removesuffix(root))

        io.savedata(file.join(epubpath,"mimetype"),mimetype)
        io.savedata(file.join(epubpath,"META-INF","container.xml"),container)
        io.savedata(file.join(epubpath,"OPS",epubroot),package)

        lfs.chdir(epubpath)

        os.remove(epubfile)

        os.execute(format("zip %s -0 %s",epubfile,"mimetype"))
        os.execute(format("zip %s -r %s",epubfile,"META-INF"))
        os.execute(format("zip %s -r %s",epubfile,"OPS"))

        lfs.chdir("..")

        application.report("epub archive: %s",file.join(epubpath,epubfile))

    end

end

--

if environment.argument("make") then
    scripts.epub.make()
else
    application.help()
end

[-- Attachment #4: Type: text/plain, Size: 485 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________