def buildLarchJar(outputStream,
additionalFilesAsNameBytesPairs,
filterFn=None,
larchJarURL=None):
"""
Build a JAR from an existing Larch JAR, along with additional files to make a packaged program
:param outputStream: The output stream to which the JAR is to be written
:param additionalFilesAsNameBytesPairs: Additional files in the form of a sequence of tuples consisting of (path, bytes)
:param filterFn: (optional) A filter function that can be used to exclude files from the existing Larch JAR; takes the form of function(name) -> boolean, should return True if file should be included
:param larchJarURL: (optional) A URL at which the existing Larch JAR can be obtained. If None, it will use the JAR from which Larch was started. Raises an error if no URL provided and Larch was not started from a JAR
"""
if larchJarURL is None:
larchJarURL = _larchJarURL
if larchJarURL is None:
raise RuntimeError, 'Larch was not loaded from a JAR file and no Larch JAR file was provided'
jarIn = JarInputStream(larchJarURL.openStream())
manifestIn = jarIn.getManifest()
manifestOut = Manifest(manifestIn)
jarOut = JarOutputStream(outputStream, manifestOut)
bytesBuffer = jarray.zeros(_BYTES_BUFFER_SIZE, 'b')
entryIn = jarIn.getNextJarEntry()
while entryIn is not None:
name = entryIn.getName()
if filterFn is None or filterFn(name):
bufferStream = ByteArrayOutputStream()
while True:
bytesRead = jarIn.read(bytesBuffer, 0, _BYTES_BUFFER_SIZE)
if bytesRead == -1:
break
bufferStream.write(bytesBuffer, 0, bytesRead)
entryOut = ZipEntry(name)
entryOut.setSize(bufferStream.size())
jarOut.putNextEntry(entryOut)
bufferStream.writeTo(jarOut)
jarOut.closeEntry()
entryIn = jarIn.getNextJarEntry()
for name, bytes in additionalFilesAsNameBytesPairs:
size = len(bytes)
entryOut = ZipEntry(name)
entryOut.setSize(size)
jarOut.putNextEntry(entryOut)
jarOut.write(bytes, 0, size)
jarOut.closeEntry()
jarOut.finish()
def fake_plugin_jar(name, plugins_config):
output = FileOutputStream(name)
zip = ZipOutputStream(output)
entry = ZipEntry('plugins.config')
zip.putNextEntry(entry)
zip.write(plugins_config)
zip.closeEntry()
zip.close()
def __createPackage(self, outputFile=None):
title = self.__manifest.getString(None, "title")
manifest = self.__createManifest()
context = JAXBContext.newInstance("com.googlecode.fascinator.ims")
m = context.createMarshaller()
m.setProperty(Marshaller.JAXB_FORMATTED_OUTPUT, True)
writer = StringWriter()
jaxbElem = ObjectFactory.createManifest(ObjectFactory(), manifest)
m.marshal(jaxbElem, writer)
writer.close()
if outputFile is not None:
print "writing to %s..." % outputFile
out = FileOutputStream(outputFile)
else:
print "writing to http output stream..."
filename = urllib.quote(title.replace(" ", "_"))
response.setHeader("Content-Disposition",
"attachment; filename=%s.zip" % filename)
out = response.getOutputStream("application/zip")
zipOut = ZipOutputStream(out)
zipOut.putNextEntry(ZipEntry("imsmanifest.xml"))
IOUtils.write(writer.toString(), zipOut)
zipOut.closeEntry()
oidList = self.__manifest.search("id")
for oid in oidList:
obj = Services.getStorage().getObject(oid)
for pid in obj.getPayloadIdList():
payload = obj.getPayload(pid)
if not PayloadType.Annotation.equals(payload.getType()):
zipOut.putNextEntry(
ZipEntry("resources/%s/%s" % (oid, pid)))
IOUtils.copy(payload.open(), zipOut)
payload.close()
zipOut.closeEntry()
obj.close()
zipOut.close()
out.close()
def add_folder(zip, folder):
for file in os.listdir(folder):
file = folder + '/' + file
if os.path.isdir(file):
add_folder(zip, file)
elif os.path.isfile(file):
if verbose:
print file
entry = ZipEntry(file)
zip.putNextEntry(entry)
f = open(file, "rb")
zip.write(f.read())
f.close()
zip.closeEntry()
def add_folder(zos, folder_name, base_folder_name):
f = File(folder_name)
if not f.exists():
return
if f.isDirectory():
for f2 in f.listFiles():
add_folder(zos, f2.absolutePath, base_folder_name)
return
entry_name = folder_name[len(base_folder_name) + 1:len(folder_name)]
ze = ZipEntry(entry_name)
zos.putNextEntry(ze)
input_stream = FileInputStream(folder_name)
buffer = zeros(1024, 'b')
rlen = input_stream.read(buffer)
while (rlen > 0):
zos.write(buffer, 0, rlen)
rlen = input_stream.read(buffer)
input_stream.close()
zos.closeEntry()
def __createEpub(self):
title = self.__manifest.getString(None, "title")
self.vc("response").setHeader(
"Content-Disposition",
"attachment; filename=%s.epub" % urllib.quote(title))
out = self.vc("response").getOutputStream("application/epub+zip")
zipOutputStream = ZipOutputStream(out)
#save mimetype... and the rest of standard files in epub
zipOutputStream.putNextEntry(ZipEntry("mimetype"))
epubMimetypeStream = self.__getResourceAsStream("/epub/mimetype")
IOUtils.copy(epubMimetypeStream, zipOutputStream)
zipOutputStream.closeEntry()
zipOutputStream.putNextEntry(ZipEntry("META-INF/container.xml"))
epubContainerStream = self.__getResourceAsStream("/epub/container.xml")
IOUtils.copy(epubContainerStream, zipOutputStream)
zipOutputStream.closeEntry()
zipOutputStream.putNextEntry(ZipEntry("OEBPS/epub.css"))
epubcss = self.__getResourceAsStream("/epub/epub.css")
IOUtils.copy(epubcss, zipOutputStream)
zipOutputStream.closeEntry()
#### Creating toc.ncx ####
tocXml = ElementTree.Element(
"ncx", {
"version": "2005-1",
"xml:lang": "en",
"xmlns": "http://www.daisy.org/z3986/2005/ncx/"
})
headNode = ElementTree.Element("head")
tocXml.append(headNode)
headNode.append(
ElementTree.Element("meta", {
"name": "dtb:uid",
"content": "1"
}))
headNode.append(
ElementTree.Element("meta", {
"name": "dtb:depth",
"content": "1"
}))
headNode.append(
ElementTree.Element("meta", {
"name": "dtb:totalPageCount",
"content": "1"
}))
headNode.append(
ElementTree.Element("meta", {
"name": "dtb:maxPageNumber",
"content": "1"
}))
headNode.append(
ElementTree.Element("meta", {
"name": "dtb:generator",
"content": "ICE v2"
}))
#docTitle
docTitle = ElementTree.Element("docTitle")
textNode = ElementTree.Element("text")
textNode.text = title
docTitle.append(textNode)
tocXml.append(docTitle)
#docAuthor
docAuthor = ElementTree.Element("docAuthor")
textNode = ElementTree.Element("text")
textNode.text = "ICE v2"
docAuthor.append(textNode)
tocXml.append(docAuthor)
#navMap
navMap = ElementTree.Element("navMap")
tocXml.append(navMap)
#### Creating content.opf ####
contentXml = ElementTree.Element(
"package", {
"version": "2.0",
"xmlns": "http://www.idpf.org/2007/opf",
"unique-identifier": "BookId"
})
metadataNode = ElementTree.Element(
"metadata", {
"xmlns:dc": "http://purl.org/dc/elements/1.1/",
"xmlns:opf": "http://www.idpf.org/2007/opf"
})
contentXml.append(metadataNode)
#metadata information
metadata = ElementTree.Element("dc:title")
metadata.text = title
metadataNode.append(metadata)
metadata = ElementTree.Element("dc:language")
metadata.text = "en-AU"
metadataNode.append(metadata)
metadata = ElementTree.Element("dc:creator", {"opf:role": "aut"})
metadata.text = "ICE"
metadataNode.append(metadata)
metadata = ElementTree.Element("dc:publisher")
metadata.text = "University of Southern Queensland"
metadataNode.append(metadata)
metadata = ElementTree.Element("dc:identifier", {"id": "BookId"})
metadata.text = title
metadataNode.append(metadata)
#manifest
manifest = ElementTree.Element("manifest")
contentXml.append(manifest)
spine = ElementTree.Element("spine", {"toc": "ncx"})
contentXml.append(spine)
item = ElementTree.Element("item", {
"id": "ncx",
"href": "toc.ncx",
"media-type": "text/xml"
})
manifest.append(item)
css = ElementTree.Element("item", {
"id": "style",
"href": "epub.css",
"media-type": "text/css"
})
manifest.append(css)
count = 1
for itemHash in self.__orderedItem:
id, title, htmlFileName, payloadDict, isImage = self.__itemRefDict[
itemHash]
for payloadId in payloadDict:
payload, payloadType = payloadDict[payloadId]
if isinstance(payload, Payload):
payloadId = payloadId.lower()
zipEntryId = payloadId.replace(" ", "_").replace("\\", "/")
if payloadType == "application/xhtml+xml":
zipOutputStream.putNextEntry(
ZipEntry("OEBPS/%s" % zipEntryId))
##process the html....
saxReader = SAXReader(False)
try:
saxDoc = saxReader.read(payload.open())
payload.close()
# ## remove class or style nodes
# classOrStyleNodes = saxDoc.selectNodes("//@class | //@style ")
# for classOrStyleNode in classOrStyleNodes:
# node = classOrStyleNode
# if classOrStyleNode.getParent():
# node = classOrStyleNode.getParent()
# if node.getQualifiedName() == "img":
# attr = node.attribute(QName("class"))
# attr = node.attribute(QName("class"))
# if attr:
# node.remove(attr)
# attr = node.attribute(QName("style"))
# if attr:
# node.remove(attr)
## remove name in a tags
ahrefs = saxDoc.selectNodes(
"//*[local-name()='a' and @name!='']")
for a in ahrefs:
attr = a.attribute(QName("name"))
if attr:
a.remove(attr)
## fix images src name.... replace space with underscore and all lower case
imgs = saxDoc.selectNodes(
"//*[local-name()='img' and contains(@src, '_files')]"
)
for img in imgs:
srcAttr = img.attribute(QName("src"))
if srcAttr:
src = srcAttr.getValue()
#hash the sourcename
filepath, filename = os.path.split(src)
filename, ext = os.path.splitext(filename)
filename = hashlib.md5(
filename).hexdigest()
src = os.path.join(
filepath.lower().replace(" ", "_"),
"node-%s%s" % (filename, ext))
img.addAttribute(QName("src"),
src.replace(" ", "_"))
bodyNode = saxDoc.selectSingleNode(
"//*[local-name()='div' and @class='body']")
bodyNode.setName("div")
out = ByteArrayOutputStream()
format = OutputFormat.createPrettyPrint()
format.setSuppressDeclaration(True)
writer = XMLWriter(out, format)
writer.write(bodyNode)
writer.flush()
contentStr = out.toString("UTF-8")
htmlString = """<?xml version="1.0" encoding="UTF-8"?>
<html xmlns="http://www.w3.org/1999/xhtml"><head><title>%s</title>
<link rel="stylesheet" href="epub.css"/>
</head><body>%s</body></html>"""
htmlString = htmlString % (title, contentStr)
self.__copyString(htmlString, zipOutputStream)
includeFile = False
except:
traceback.print_exc()
else:
#images....
zipOutputStream.putNextEntry(
ZipEntry("OEBPS/%s" % zipEntryId))
IOUtils.copy(payload.open(), zipOutputStream)
payload.close()
zipOutputStream.closeEntry()
else:
zipOutputStream.putNextEntry(
ZipEntry("OEBPS/%s" % zipEntryId))
IOUtils.copy(payload, zipOutputStream)
zipOutputStream.closeEntry()
itemNode = ElementTree.Element("item", {
"media-type": payloadType,
"href": zipEntryId
})
if payloadId == htmlFileName.lower():
itemNode.set("id", itemHash)
else:
itemNode.set("id", payloadId.replace("/", "_"))
manifest.append(itemNode)
if not isImage:
navPoint = ElementTree.Element(
"navPoint", {
"class": "chapter",
"id": "%s" % itemHash,
"playOrder": "%s" % count
})
else:
navPoint = ElementTree.Element(
"navPoint", {
"class": "chapter",
"id": "%s" % htmlFileName,
"playOrder": "%s" % count
})
navMap.append(navPoint)
navLabel = ElementTree.Element("navLabel")
navPoint.append(navLabel)
textNode = ElementTree.Element("text")
textNode.text = title
navLabel.append(textNode)
content = ElementTree.Element("content")
navPoint.append(content)
content.set("src", htmlFileName)
count += 1
itemRefNode = ElementTree.Element("itemref")
spine.append(itemRefNode)
itemRefNode.set("idref", itemHash)
#saving content.opf...
zipOutputStream.putNextEntry(ZipEntry("OEBPS/content.opf"))
self.__copyString(ElementTree.tostring(contentXml), zipOutputStream)
zipOutputStream.closeEntry()
#saving toc.ncx
zipOutputStream.putNextEntry(ZipEntry("OEBPS/toc.ncx"))
self.__copyString(ElementTree.tostring(tocXml), zipOutputStream)
zipOutputStream.closeEntry()
zipOutputStream.close()
def add(self, name, data):
self.zipOutputStream.putNextEntry(ZipEntry(name))
self.__copyString(data, self.zipOutputStream)
self.zipOutputStream.closeEntry()
def addStream(self, name, s):
self.zipOutputStream.putNextEntry(ZipEntry(name))
IOUtils.copy(s, self.zipOutputStream)
self.zipOutputStream.closeEntry()
def getFile(self, name):
fname = self.getName(name)
self.zipfile.putNextEntry(ZipEntry(fname))
return self.zipfile