def createTargetInstance(modelXbrl, targetUrl, targetDocumentSchemaRefs, filingFiles, baseXmlLang=None, defaultXmlLang=None): def addLocallyReferencedFile(elt, filingFiles): if elt.tag in ("a", "img"): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src") and not isHttpUrl( attrValue) and not os.path.isabs(attrValue): attrValue = attrValue.partition('#')[0] # remove anchor if attrValue: # ignore anchor references to base document attrValue = os.path.normpath( attrValue ) # change url path separators to host separators file = os.path.join(sourceDir, attrValue) if modelXbrl.fileSource.isInArchive( file, checkExistence=True) or os.path.exists(file): filingFiles.add(file) targetInstance = ModelXbrl.create( modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True, discover=False) # don't attempt to load DTS ixTargetRootElt = modelXbrl.ixTargetRootElements[getattr( modelXbrl, "ixdsTarget", None)] langIsSet = False # copy ix resources target root attributes for attrName, attrValue in ixTargetRootElt.items(): if attrName != "target": # ix:references target is not mapped to xbrli:xbrl targetInstance.modelDocument.xmlRootElement.set( attrName, attrValue) if attrName == "{http://www.w3.org/XML/1998/namespace}lang": langIsSet = True defaultXmlLang = attrValue if attrName.startswith("{"): ns, _sep, ln = attrName[1:].rpartition("}") if ns: prefix = xmlnsprefix(ixTargetRootElt, ns) if prefix not in (None, "xml"): setXmlns(targetInstance.modelDocument, prefix, ns) if not langIsSet and baseXmlLang: targetInstance.modelDocument.xmlRootElement.set( "{http://www.w3.org/XML/1998/namespace}lang", baseXmlLang) if defaultXmlLang is None: defaultXmlLang = baseXmlLang # allows facts/footnotes to override baseXmlLang ValidateXbrlDimensions.loadDimensionDefaults( targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in sorted(modelXbrl.contexts.values(), key=lambda c: c.objectIndex ): # contexts may come from multiple IXDS files ignore = targetInstance.createContext( context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in sorted(modelXbrl.units.values(), key=lambda u: u.objectIndex ): # units may come from multiple IXDS files measures = unit.measures ignore = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: # HF does not de-duplicate, which is currently-desired behavior attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue for attrName, attrValue in fact.items(): if attrName.startswith("{"): attrs[qname( attrName, fact.nsmap )] = attrValue # using qname allows setting prefix in extracted instance newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) # if fact.isFraction, create numerator and denominator newFactForOldObjId[fact.objectIndex] = newFact if filingFiles is not None and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references so that referenced files are included in the zip. for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML("<body>\n{0}\n</body>\n".format( xmltext)).iter(): addLocallyReferencedFile(elt, filingFiles) except (XMLSyntaxError, UnicodeDecodeError): pass # TODO: Why ignore UnicodeDecodeError? elif fact.isTuple: attrs = {} if fact.id: attrs["id"] = fact.id if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" for attrName, attrValue in fact.items(): if attrName.startswith("{"): attrs[qname(attrName, fact.nsmap)] = attrValue newTuple = targetInstance.createFact(fact.qname, attributes=attrs, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) modelXbrl.modelManager.showStatus( _("Creating and validating footnotes and relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) footnoteIdCount = {} for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any( lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format( attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) xmlLang = linkChild.xmlLang if xmlLang is not None and xmlLang != defaultXmlLang: # default newChild.set( "{http://www.w3.org/XML/1998/namespace}lang", xmlLang) copyIxFootnoteHtml( linkChild, newChild, targetModelDocument=targetInstance.modelDocument, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): addLocallyReferencedFile(elt, filingFiles) return targetInstance
def saveTargetDocument(modelXbrl, targetDocumentFilename, targetDocumentSchemaRefs, outputZip=None, filingFiles=None, *args, **kwargs): targetUrl = modelXbrl.modelManager.cntlr.webCache.normalizeUrl( targetDocumentFilename, modelXbrl.modelDocument.filepath) targetUrlParts = targetUrl.rpartition(".") targetUrl = targetUrlParts[0] + "_extracted." + targetUrlParts[2] modelXbrl.modelManager.showStatus( _("Extracting instance ") + os.path.basename(targetUrl)) targetInstance = ModelXbrl.create(modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True) ValidateXbrlDimensions.loadDimensionDefaults( targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in modelXbrl.contexts.values(): newCntx = targetInstance.createContext( context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in modelXbrl.units.values(): measures = unit.measures newUnit = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) newFactForOldObjId[fact.objectIndex] = newFact if filingFiles and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML( "<body>\n{0}\n</body>\n".format(xmltext)): if elt.tag in ("a", "img") and not isHttpUrl( attrValue) and not os.path.isabs( attrvalue): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src"): filingFiles.add(attrValue) except (XMLSyntaxError, UnicodeDecodeError): pass elif fact.isTuple: newTuple = targetInstance.createFact(fact.qname, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) # footnote links footnoteIdCount = {} modelXbrl.modelManager.showStatus( _("Creating and validating footnotes & relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any( lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format( attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) copyIxFootnoteHtml( linkChild, newChild, targetModelDocument=targetInstance.modelDocument, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): if elt.tag in ("a", "img"): for attrTag, attrValue in elt.items(): if attrTag in ( "href", "src") and not isHttpUrl( attrValue ) and not os.path.isabs(attrvalue): filingFiles.add(attrValue) targetInstance.saveInstance(overrideFilepath=targetUrl, outputZip=outputZip) if getattr(modelXbrl, "isTestcaseVariation", False): modelXbrl.extractedInlineInstance = True # for validation comparison modelXbrl.modelManager.showStatus(_("Saved extracted instance"), 5000)
def saveTargetDocument(modelXbrl, targetDocumentFilename, targetDocumentSchemaRefs, outputZip=None, filingFiles=None, *args, **kwargs): targetUrl = modelXbrl.modelManager.cntlr.webCache.normalizeUrl( targetDocumentFilename, modelXbrl.modelDocument.filepath) def addLocallyReferencedFile(elt, filingFiles): if elt.tag in ("a", "img"): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src") and not isHttpUrl( attrValue) and not os.path.isabs(attrValue): attrValue = attrValue.partition('#')[0] # remove anchor if attrValue: # ignore anchor references to base document attrValue = os.path.normpath( attrValue ) # change url path separators to host separators file = os.path.join(sourceDir, attrValue) if modelXbrl.fileSource.isInArchive( file, checkExistence=True) or os.path.exists(file): filingFiles.add(file) targetUrlParts = targetUrl.rpartition(".") targetUrl = targetUrlParts[0] + "_extracted." + targetUrlParts[2] modelXbrl.modelManager.showStatus( _("Extracting instance ") + os.path.basename(targetUrl)) rootElt = modelXbrl.modelDocument.xmlRootElement # take baseXmlLang from <html> or <base> baseXmlLang = rootElt.get( "{http://www.w3.org/XML/1998/namespace}lang") or rootElt.get("lang") for ixElt in modelXbrl.modelDocument.xmlRootElement.iterdescendants( tag="{http://www.w3.org/1999/xhtml}body"): baseXmlLang = ixElt.get("{http://www.w3.org/XML/1998/namespace}lang" ) or rootElt.get("lang") or baseXmlLang targetInstance = ModelXbrl.create( modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True, discover=False) # don't attempt to load DTS if baseXmlLang: targetInstance.modelDocument.xmlRootElement.set( "{http://www.w3.org/XML/1998/namespace}lang", baseXmlLang) ValidateXbrlDimensions.loadDimensionDefaults( targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in sorted(modelXbrl.contexts.values(), key=lambda c: elementChildSequence(c)): ignore = targetInstance.createContext( context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in modelXbrl.units.values(): measures = unit.measures ignore = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: # HF does not de-duplicate, which is currently-desired behavior attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue if fact.concept is not None and fact.concept.baseXsdType in ( "string", "normalizedString"): # default xmlLang = fact.xmlLang if xmlLang is not None and xmlLang != baseXmlLang: attrs[ "{http://www.w3.org/XML/1998/namespace}lang"] = xmlLang newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) # if fact.isFraction, create numerator and denominator newFactForOldObjId[fact.objectIndex] = newFact if filingFiles is not None and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references so that referenced files are included in the zip. for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML("<body>\n{0}\n</body>\n".format( xmltext)).iter(): addLocallyReferencedFile(elt, filingFiles) except (XMLSyntaxError, UnicodeDecodeError): pass # TODO: Why ignore UnicodeDecodeError? elif fact.isTuple: newTuple = targetInstance.createFact(fact.qname, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) modelXbrl.modelManager.showStatus( _("Creating and validating footnotes and relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) footnoteIdCount = {} for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any( lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format( attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) xmlLang = linkChild.xmlLang if xmlLang is not None and xmlLang != baseXmlLang: # default newChild.set( "{http://www.w3.org/XML/1998/namespace}lang", xmlLang) copyIxFootnoteHtml( linkChild, newChild, targetModelDocument=targetInstance.modelDocument, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): addLocallyReferencedFile(elt, filingFiles) targetInstance.saveInstance(overrideFilepath=targetUrl, outputZip=outputZip) if getattr(modelXbrl, "isTestcaseVariation", False): modelXbrl.extractedInlineInstance = True # for validation comparison modelXbrl.modelManager.showStatus(_("Saved extracted instance"), 5000)
def createTargetInstance(modelXbrl, targetUrl, targetDocumentSchemaRefs, filingFiles, baseXmlLang=None, defaultXmlLang=None): targetInstance = ModelXbrl.create(modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True, discover=False) # don't attempt to load DTS if baseXmlLang: targetInstance.modelDocument.xmlRootElement.set("{http://www.w3.org/XML/1998/namespace}lang", baseXmlLang) if defaultXmlLang is None: defaultXmlLang = baseXmlLang # allows facts/footnotes to override baseXmlLang ValidateXbrlDimensions.loadDimensionDefaults(targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in sorted(modelXbrl.contexts.values(), key=lambda c: c.objectIndex): # contexts may come from multiple IXDS files ignore = targetInstance.createContext(context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in sorted(modelXbrl.units.values(), key=lambda u: u.objectIndex): # units may come from multiple IXDS files measures = unit.measures ignore = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: # HF does not de-duplicate, which is currently-desired behavior attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue if fact.concept is not None and fact.concept.baseXsdType in ("string", "normalizedString"): # default xmlLang = fact.xmlLang if xmlLang is not None and xmlLang != defaultXmlLang: attrs["{http://www.w3.org/XML/1998/namespace}lang"] = xmlLang newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) # if fact.isFraction, create numerator and denominator newFactForOldObjId[fact.objectIndex] = newFact if filingFiles is not None and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references so that referenced files are included in the zip. for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML("<body>\n{0}\n</body>\n".format(xmltext)).iter(): addLocallyReferencedFile(elt, filingFiles) except (XMLSyntaxError, UnicodeDecodeError): pass # TODO: Why ignore UnicodeDecodeError? elif fact.isTuple: newTuple = targetInstance.createFact(fact.qname, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) modelXbrl.modelManager.showStatus(_("Creating and validating footnotes and relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) footnoteIdCount = {} for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any(lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format(attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) xmlLang = linkChild.xmlLang if xmlLang is not None and xmlLang != defaultXmlLang: # default newChild.set("{http://www.w3.org/XML/1998/namespace}lang", xmlLang) copyIxFootnoteHtml(linkChild, newChild, targetModelDocument=targetInstance.modelDocument, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): addLocallyReferencedFile(elt,filingFiles) return targetInstance
def saveTargetDocument(modelXbrl, targetDocumentFilename, targetDocumentSchemaRefs, outputZip=None, filingFiles=None): targetUrl = modelXbrl.modelManager.cntlr.webCache.normalizeUrl(targetDocumentFilename, modelXbrl.modelDocument.filepath) targetUrlParts = targetUrl.rpartition(".") targetUrl = targetUrlParts[0] + "_extracted." + targetUrlParts[2] modelXbrl.modelManager.showStatus(_("Extracting instance ") + os.path.basename(targetUrl)) targetInstance = ModelXbrl.create(modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True) ValidateXbrlDimensions.loadDimensionDefaults(targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in modelXbrl.contexts.values(): newCntx = targetInstance.createContext(context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in modelXbrl.units.values(): measures = unit.measures newUnit = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) newFactForOldObjId[fact.objectIndex] = newFact if filingFiles and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML("<body>\n{0}\n</body>\n".format(xmltext)): if elt.tag in ("a", "img") and not isHttpUrl(attrValue) and not os.path.isabs(attrvalue): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src"): filingFiles.add(attrValue) except (XMLSyntaxError, UnicodeDecodeError): pass elif fact.isTuple: newTuple = targetInstance.createFact(fact.qname, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) # footnote links footnoteIdCount = {} modelXbrl.modelManager.showStatus(_("Creating and validating footnotes & relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any(lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format(attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) copyIxFootnoteHtml(linkChild, newChild, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): if elt.tag in ("a", "img"): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src") and not isHttpUrl(attrValue) and not os.path.isabs(attrvalue): filingFiles.add(attrValue) targetInstance.saveInstance(overrideFilepath=targetUrl, outputZip=outputZip) modelXbrl.modelManager.showStatus(_("Saved extracted instance"), 5000)
def saveTargetDocument(modelXbrl, targetDocumentFilename, targetDocumentSchemaRefs, outputZip=None, filingFiles=None, *args, **kwargs): targetUrl = modelXbrl.modelManager.cntlr.webCache.normalizeUrl(targetDocumentFilename, modelXbrl.modelDocument.filepath) def addLocallyReferencedFile(elt,filingFiles): if elt.tag in ("a", "img"): for attrTag, attrValue in elt.items(): if attrTag in ("href", "src") and not isHttpUrl(attrValue) and not os.path.isabs(attrValue): attrValue = attrValue.partition('#')[0] # remove anchor if attrValue: # ignore anchor references to base document attrValue = os.path.normpath(attrValue) # change url path separators to host separators file = os.path.join(sourceDir,attrValue) if modelXbrl.fileSource.isInArchive(file, checkExistence=True) or os.path.exists(file): filingFiles.add(file) targetUrlParts = targetUrl.rpartition(".") targetUrl = targetUrlParts[0] + "_extracted." + targetUrlParts[2] modelXbrl.modelManager.showStatus(_("Extracting instance ") + os.path.basename(targetUrl)) rootElt = modelXbrl.modelDocument.xmlRootElement # take baseXmlLang from <html> or <base> baseXmlLang = rootElt.get("{http://www.w3.org/XML/1998/namespace}lang") or rootElt.get("lang") for ixElt in modelXbrl.modelDocument.xmlRootElement.iterdescendants(tag="{http://www.w3.org/1999/xhtml}body"): baseXmlLang = ixElt.get("{http://www.w3.org/XML/1998/namespace}lang") or rootElt.get("lang") or baseXmlLang targetInstance = ModelXbrl.create(modelXbrl.modelManager, newDocumentType=Type.INSTANCE, url=targetUrl, schemaRefs=targetDocumentSchemaRefs, isEntry=True, discover=False) # don't attempt to load DTS if baseXmlLang: targetInstance.modelDocument.xmlRootElement.set("{http://www.w3.org/XML/1998/namespace}lang", baseXmlLang) ValidateXbrlDimensions.loadDimensionDefaults(targetInstance) # need dimension defaults # roleRef and arcroleRef (of each inline document) for sourceRefs in (modelXbrl.targetRoleRefs, modelXbrl.targetArcroleRefs): for roleRefElt in sourceRefs.values(): addChild(targetInstance.modelDocument.xmlRootElement, roleRefElt.qname, attributes=roleRefElt.items()) # contexts for context in sorted(modelXbrl.contexts.values(), key=lambda c: elementChildSequence(c)): ignore = targetInstance.createContext(context.entityIdentifier[0], context.entityIdentifier[1], 'instant' if context.isInstantPeriod else 'duration' if context.isStartEndPeriod else 'forever', context.startDatetime, context.endDatetime, None, context.qnameDims, [], [], id=context.id) for unit in modelXbrl.units.values(): measures = unit.measures ignore = targetInstance.createUnit(measures[0], measures[1], id=unit.id) modelXbrl.modelManager.showStatus(_("Creating and validating facts")) newFactForOldObjId = {} def createFacts(facts, parent): for fact in facts: if fact.isItem: # HF does not de-duplicate, which is currently-desired behavior attrs = {"contextRef": fact.contextID} if fact.id: attrs["id"] = fact.id if fact.isNumeric: attrs["unitRef"] = fact.unitID if fact.get("decimals"): attrs["decimals"] = fact.get("decimals") if fact.get("precision"): attrs["precision"] = fact.get("precision") if fact.isNil: attrs[XbrlConst.qnXsiNil] = "true" text = None else: text = fact.xValue if fact.xValid else fact.textValue if fact.concept is not None and fact.concept.baseXsdType in ("string", "normalizedString"): # default xmlLang = fact.xmlLang if xmlLang is not None and xmlLang != baseXmlLang: attrs["{http://www.w3.org/XML/1998/namespace}lang"] = xmlLang newFact = targetInstance.createFact(fact.qname, attributes=attrs, text=text, parent=parent) # if fact.isFraction, create numerator and denominator newFactForOldObjId[fact.objectIndex] = newFact if filingFiles is not None and fact.concept is not None and fact.concept.isTextBlock: # check for img and other filing references so that referenced files are included in the zip. for xmltext in [text] + CDATApattern.findall(text): try: for elt in XML("<body>\n{0}\n</body>\n".format(xmltext)).iter(): addLocallyReferencedFile(elt, filingFiles) except (XMLSyntaxError, UnicodeDecodeError): pass # TODO: Why ignore UnicodeDecodeError? elif fact.isTuple: newTuple = targetInstance.createFact(fact.qname, parent=parent) newFactForOldObjId[fact.objectIndex] = newTuple createFacts(fact.modelTupleFacts, newTuple) createFacts(modelXbrl.facts, None) modelXbrl.modelManager.showStatus(_("Creating and validating footnotes and relationships")) HREF = "{http://www.w3.org/1999/xlink}href" footnoteLinks = defaultdict(list) footnoteIdCount = {} for linkKey, linkPrototypes in modelXbrl.baseSets.items(): arcrole, linkrole, linkqname, arcqname = linkKey if (linkrole and linkqname and arcqname and # fully specified roles arcrole != "XBRL-footnotes" and any(lP.modelDocument.type == Type.INLINEXBRL for lP in linkPrototypes)): for linkPrototype in linkPrototypes: if linkPrototype not in footnoteLinks[linkrole]: footnoteLinks[linkrole].append(linkPrototype) for linkrole in sorted(footnoteLinks.keys()): for linkPrototype in footnoteLinks[linkrole]: newLink = addChild(targetInstance.modelDocument.xmlRootElement, linkPrototype.qname, attributes=linkPrototype.attributes) for linkChild in linkPrototype: attributes = linkChild.attributes if isinstance(linkChild, LocPrototype): if HREF not in linkChild.attributes: linkChild.attributes[HREF] = \ "#" + elementFragmentIdentifier(newFactForOldObjId[linkChild.dereference().objectIndex]) addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ArcPrototype): addChild(newLink, linkChild.qname, attributes=attributes) elif isinstance(linkChild, ModelInlineFootnote): idUseCount = footnoteIdCount.get(linkChild.footnoteID, 0) + 1 if idUseCount > 1: # if footnote with id in other links bump the id number attributes = linkChild.attributes.copy() attributes["id"] = "{}_{}".format(attributes["id"], idUseCount) footnoteIdCount[linkChild.footnoteID] = idUseCount newChild = addChild(newLink, linkChild.qname, attributes=attributes) xmlLang = linkChild.xmlLang if xmlLang is not None and xmlLang != baseXmlLang: # default newChild.set("{http://www.w3.org/XML/1998/namespace}lang", xmlLang) copyIxFootnoteHtml(linkChild, newChild, targetModelDocument=targetInstance.modelDocument, withText=True) if filingFiles and linkChild.textValue: footnoteHtml = XML("<body/>") copyIxFootnoteHtml(linkChild, footnoteHtml) for elt in footnoteHtml.iter(): addLocallyReferencedFile(elt,filingFiles) targetInstance.saveInstance(overrideFilepath=targetUrl, outputZip=outputZip) if getattr(modelXbrl, "isTestcaseVariation", False): modelXbrl.extractedInlineInstance = True # for validation comparison modelXbrl.modelManager.showStatus(_("Saved extracted instance"), 5000)