Skip to content
This repository has been archived by the owner on Oct 17, 2019. It is now read-only.

Commit

Permalink
Must Support (#86)
Browse files Browse the repository at this point in the history 
* WIP add content profile grammar and begin listener

* Add data to content profile specifications (except path)

* Add content profile listener and add config file field for content profile

* Remove extraneous comment

* Add handling for invalid content profile file in config

* Add support for Value in chain, fix styling, add extra error handling

* Add null checks for configuration

* Clear currentDef and currentRule when appropriate

When the parser exits the CP definition or a CP rule definition, clear the corresponding member property.  Otherwise old rules can get applied to new CPs!

* Add support for inherited attributes in content profiles

* Address comments based on review feedback

* Content Profile paths must use the effectiveIdentifier

After discussion w/ Mark, we determined that CPs should always use the "effective identifiers" in the paths.  For example, if field Foo is constrained to SonOfFoo, then the CP path should use SonOfFoo.

In addition fixed an issue where the ChoiceValue options were being iterated as identifiers when they are really Values (that have identifiers).

* Minor tweak for clarity (based on code review)

* Update error numbers

* Upgrade shr-models to 5.8.0

Take in the new models for ContentProfiles.

* v5.7.0
  • Loading branch information
@cmoesel
Dylan Mahalingam authored and cmoesel committed Mar 27, 2019
1 parent fced66d commit a371487
Show file tree
Hide file tree
Showing 10 changed files with 2,065 additions and 7 deletions.
216 changes: 216 additions & 0 deletions lib/contentProfileListener.js
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,216 @@
const bunyan = require('bunyan');
const {FileStream, CommonTokenStream} = require('antlr4/index');
const {ParseTreeWalker} = require('antlr4/tree');
const {SHRContentProfileLexer} = require('./parsers/SHRContentProfileLexer');
const {SHRContentProfileParser} = require('./parsers/SHRContentProfileParser');
const {SHRContentProfileParserListener} = require('./parsers/SHRContentProfileParserListener');
const {SHRErrorListener} = require('./errorListener.js');
const {Specifications, Version, ContentProfile, ContentProfileRule, Identifier, IdentifiableValue, ChoiceValue, PrimitiveIdentifier, PRIMITIVES} = require('shr-models');

var rootLogger = bunyan.createLogger({name: 'shr-text-import'});
var logger = rootLogger;
function setLogger(bunyanLogger) {
rootLogger = logger = bunyanLogger;
}

class ContentProfileImporter extends SHRContentProfileParserListener {
constructor(specifications = new Specifications()) {
super();
// The specifications it collects
this._specs = specifications;
// The currently active namespace
this._currentNs = '';
// The currently active grammar version
this._currentGrammarVersion = '';
// The currently active content profile definition
this._currentDef = null;
// The currently active content profile rule
this._currentRule = null;
}

get specifications() { return this._specs; }

importFile(file) {
// Setup a child logger to associate logs with the current file
const lastLogger = logger;
logger = rootLogger.child({ file: file });
logger.debug('Start importing content profile file');
try {
const errListener = new SHRErrorListener(logger);
const chars = new FileStream(file);
const lexer = new SHRContentProfileLexer(chars);
lexer.removeErrorListeners();
lexer.addErrorListener(errListener);
const tokens = new CommonTokenStream(lexer);
const parser = new SHRContentProfileParser(tokens);
parser.removeErrorListeners();
parser.addErrorListener(errListener);
parser.buildParseTrees = true;
const tree = parser.doc();
const walker = new ParseTreeWalker();
walker.walk(this, tree);
} finally {
logger.debug('Done importing content profile file');
this.logger = lastLogger;
}
}

enterDoc(ctx) {
// set grammar version
const version = ctx.docHeader().version();
const major = parseInt(version.WHOLE_NUMBER()[0], 10);
const minor = parseInt(version.WHOLE_NUMBER()[1], 10);
this._currentGrammarVersion = new Version(major, minor);

logger.debug({version: this._currentGrammarVersion.toString()}, 'Entered content profile file');
}

exitDoc(ctx) {
// clear current namespace, current content, current rule, and grammar version
logger.debug('Exiting content profile file');
this._currentNs = null;
this._currentDef = null;
this._currentRule = null;
this._currentGrammarVersion = null;
}

enterNamespaceHeader(ctx) {
// set current namespace
this._currentNs = ctx.namespace().getText();
}

enterContentHeader(ctx) {
// set current content
const name = ctx.simpleName().getText();
const identifier = new Identifier(this._currentNs, name);
this._currentDef = new ContentProfile(identifier);
this._currentDef.grammarVersion = this._currentGrammarVersion;
}

enterCpRule(ctx) {
// find identifier for each data element in path,
// then create and set current rule with that path

const pathStr = ctx.simpleOrPathName().getText();
const names = pathStr.split('.');
let path = [];
let currentElement = this._specs.dataElements.findByIdentifier(this._currentDef.identifier);

if (currentElement) {
for (const name of names) {
let parentElements = this.getRecursiveBasedOns(currentElement.identifier);
if (PRIMITIVES.includes(name)) {
path.push(new PrimitiveIdentifier(name));
break;
} else if (name === 'Value' || name === '_Value') {
path.push(new Identifier('', '_Value'));

// find value from self or most recent ancestor
let value = null;
for (const id of parentElements) {
const el = this._specs.dataElements.findByIdentifier(id);
value = el ? el.value : null;

if (value) {
break;
}
}

if (value && !(value instanceof ChoiceValue)) {
currentElement = this._specs.dataElements.findByIdentifier(value.effectiveIdentifier);
} else {
break; // Exit loop to reach error condition below
}
} else {
// Collect fields from current elements and all parents
let fields = new Map();
for (const id of parentElements) {
const el = this._specs.dataElements.findByIdentifier(id);
[el.value, ...el.fields].forEach(field => {
const key = (field && field.identifier) ? field.identifier : 'value';
if (field && !fields.has(key)) fields.set(key, field);
});
}

let value = Array.from(fields.values()).find(field => {
if (field instanceof IdentifiableValue) {
// match name on effectiveIdentifier since CP requires author to use constrained type name
return field.effectiveIdentifier.name === name;
} else if (field instanceof ChoiceValue) {
// match name on one of the choice option's effectiveIdentifier
return field.aggregateOptions.some(o => o.effectiveIdentifier && o.effectiveIdentifier.name === name);
}

return false;
});

let element;
if (value && value.effectiveIdentifier) {
element = this._specs.dataElements.findByIdentifier(value.effectiveIdentifier);
}

if (element) {
path.push(element.identifier);
currentElement = element;
} else {
break; // Exit loop to reach error condition below
}
}
}
} else {
logger.error(
'Definition not found for data element in content profile path: %s. ERROR_CODE:11035',
this._currentDef.identifier.fqn
);
}

if (path.length === names.length) {
this._currentRule = new ContentProfileRule(path);
} else {
// TODO: We may be able to help the author by suggesting fixes when the problem is that they referred to the old identifier instead of the effectiveIdentifier.
// This will require some rework of the above code to detect and remember this situation.
logger.error('Path not found for %s: %s. ERROR_CODE:11036', this._currentDef.identifier.fqn, pathStr);
}
}

enterFlag(ctx) {
if (this._currentRule) {
this._currentRule.mustSupport = (ctx.KW_MUST_SUPPORT() != null);
this._currentDef.addRule(this._currentRule);
}
}

exitCpRule(ctx) {
this._currentRule = null;
}

exitContentDef(ctx) {
this._specs.contentProfiles.add(this._currentDef);
this._currentDef = null;
}

// NOTE: This function "borrowed" from shr-expand
getRecursiveBasedOns(identifier, alreadyProcessed = []) {
// If it's primitive or we've already processed this one, don't go further (avoid circular dependencies)
if (identifier.isPrimitive || alreadyProcessed.some(id => id.equals(identifier))) {
return alreadyProcessed;
}

// We haven't processed it, so look it up
const element = this._specs.dataElements.findByIdentifier(identifier);
if (typeof element === 'undefined') {
logger.error('Cannot resolve element definition for %s. ERROR_CODE:13023', identifier.fqn);
return alreadyProcessed;
}
// Add it to the already processed list (again, to avoid circular dependencies)
alreadyProcessed.push(identifier);
// Now recursively get the BasedOns for each of the BasedOns
for (const basedOn of element.basedOn) {
alreadyProcessed = this.getRecursiveBasedOns(basedOn, alreadyProcessed);
}

return alreadyProcessed;
}
}

module.exports = {ContentProfileImporter, setLogger};
13 changes: 13 additions & 0 deletions lib/import.js
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@ const {Preprocessor, VERSION, GRAMMAR_VERSION} = require('./preprocessor');
const {DataElementImporter} = require('./dataElementListener');
const {ValueSetImporter} = require('./valueSetListener');
const {MappingImporter} = require('./mappingListener');
const {ContentProfileImporter} = require('./contentProfileListener');
const {CimcoreImporter} = require('./cimcore/cimcoreImport');

var logger = bunyan.createLogger({name: 'shr-text-import'});
Expand All @@ -15,6 +16,7 @@ function setLogger(bunyanLogger) {
require('./dataElementListener').setLogger(logger);
require('./valueSetListener').setLogger(logger);
require('./mappingListener').setLogger(logger);
require('./contentProfileListener').setLogger(logger);
require('./cimcore/cimcoreImport').setLogger(logger);
}

Expand All @@ -37,6 +39,17 @@ function importFromFilePath(filePath, configuration=[], specifications = new Spe
for (const file of filesByType.map) {
mappingImporter.importFile(file);
}
const contentProfileImporter = new ContentProfileImporter(specifications);
let contentProfileFound = false;
for (const file of filesByType.contentProfile) {
if (configuration && (path.basename(file) === configuration.contentProfile)) {
contentProfileFound = true;
contentProfileImporter.importFile(file);
}
}
if (configuration && configuration.contentProfile && !contentProfileFound) {
logger.error('Could not find content profile file: %s. ERROR_CODE:11037', configuration.contentProfile);
}
return specifications;
}

Expand Down
131 changes: 131 additions & 0 deletions lib/parsers/SHRContentProfileLexer.js
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,131 @@
// Generated from SHRContentProfileLexer.g4 by ANTLR 4.5
// jshint ignore: start
var antlr4 = require('antlr4/index');


var serializedATN = ["\3\u0430\ud6d1\u8206\uad2d\u4417\uaef1\u8d80\uaadd",
"\2\23\u00c2\b\1\4\2\t\2\4\3\t\3\4\4\t\4\4\5\t\5\4\6\t\6\4\7\t\7\4\b",
"\t\b\4\t\t\t\4\n\t\n\4\13\t\13\4\f\t\f\4\r\t\r\4\16\t\16\4\17\t\17\4",
"\20\t\20\4\21\t\21\4\22\t\22\3\2\3\2\3\2\3\2\3\2\3\2\3\2\3\2\3\2\3\3",
"\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\3\4\3\4\3\4",
"\3\4\3\4\3\4\3\4\3\4\3\4\3\4\3\4\3\5\3\5\3\5\3\6\3\6\3\7\3\7\3\b\6\b",
"Q\n\b\r\b\16\bR\3\t\3\t\7\tW\n\t\f\t\16\tZ\13\t\3\n\3\n\7\n^\n\n\f\n",
"\16\na\13\n\3\13\3\13\7\13e\n\13\f\13\16\13h\13\13\3\f\3\f\7\fl\n\f",
"\f\f\16\fo\13\f\3\f\3\f\3\f\7\ft\n\f\f\f\16\fw\13\f\6\fy\n\f\r\f\16",
"\fz\3\r\3\r\7\r\177\n\r\f\r\16\r\u0082\13\r\3\r\3\r\3\r\7\r\u0087\n",
"\r\f\r\16\r\u008a\13\r\7\r\u008c\n\r\f\r\16\r\u008f\13\r\3\r\3\r\3\r",
"\7\r\u0094\n\r\f\r\16\r\u0097\13\r\3\16\3\16\7\16\u009b\n\16\f\16\16",
"\16\u009e\13\16\3\16\3\16\3\17\3\17\3\17\3\17\3\20\3\20\3\20\3\20\3",
"\21\3\21\3\21\3\21\7\21\u00ae\n\21\f\21\16\21\u00b1\13\21\3\21\3\21",
"\3\21\3\21\3\21\3\22\3\22\3\22\3\22\7\22\u00bc\n\22\f\22\16\22\u00bf",
"\13\22\3\22\3\22\3\u00af\2\23\3\3\5\4\7\5\t\6\13\7\r\b\17\t\21\n\23",
"\13\25\f\27\r\31\16\33\17\35\20\37\21!\22#\23\3\2\f\3\2\62;\3\2C\\\5",
"\2\62;C\\aa\7\2//\62;C\\aac|\3\2c|\6\2//\62;C\\c|\5\2//\62;C|\4\2$$",
"^^\5\2\13\13\17\17\"\"\4\2\f\f\17\17\u00cf\2\3\3\2\2\2\2\5\3\2\2\2\2",
"\7\3\2\2\2\2\t\3\2\2\2\2\13\3\2\2\2\2\r\3\2\2\2\2\17\3\2\2\2\2\21\3",
"\2\2\2\2\23\3\2\2\2\2\25\3\2\2\2\2\27\3\2\2\2\2\31\3\2\2\2\2\33\3\2",
"\2\2\2\35\3\2\2\2\2\37\3\2\2\2\2!\3\2\2\2\2#\3\2\2\2\3%\3\2\2\2\5.\3",
"\2\2\2\7=\3\2\2\2\tH\3\2\2\2\13K\3\2\2\2\rM\3\2\2\2\17P\3\2\2\2\21T",
"\3\2\2\2\23[\3\2\2\2\25b\3\2\2\2\27i\3\2\2\2\31|\3\2\2\2\33\u0098\3",
"\2\2\2\35\u00a1\3\2\2\2\37\u00a5\3\2\2\2!\u00a9\3\2\2\2#\u00b7\3\2\2",
"\2%&\7I\2\2&\'\7t\2\2\'(\7c\2\2()\7o\2\2)*\7o\2\2*+\7c\2\2+,\7t\2\2",
",-\7<\2\2-\4\3\2\2\2./\7E\2\2/\60\7q\2\2\60\61\7p\2\2\61\62\7v\2\2\62",
"\63\7g\2\2\63\64\7p\2\2\64\65\7v\2\2\65\66\7R\2\2\66\67\7t\2\2\678\7",
"q\2\289\7h\2\29:\7k\2\2:;\7n\2\2;<\7g\2\2<\6\3\2\2\2=>\7P\2\2>?\7c\2",
"\2?@\7o\2\2@A\7g\2\2AB\7u\2\2BC\7r\2\2CD\7c\2\2DE\7e\2\2EF\7g\2\2FG",
"\7<\2\2G\b\3\2\2\2HI\7O\2\2IJ\7U\2\2J\n\3\2\2\2KL\7\60\2\2L\f\3\2\2",
"\2MN\7<\2\2N\16\3\2\2\2OQ\t\2\2\2PO\3\2\2\2QR\3\2\2\2RP\3\2\2\2RS\3",
"\2\2\2S\20\3\2\2\2TX\t\3\2\2UW\t\4\2\2VU\3\2\2\2WZ\3\2\2\2XV\3\2\2\2",
"XY\3\2\2\2Y\22\3\2\2\2ZX\3\2\2\2[_\t\3\2\2\\^\t\5\2\2]\\\3\2\2\2^a\3",
"\2\2\2_]\3\2\2\2_`\3\2\2\2`\24\3\2\2\2a_\3\2\2\2bf\t\6\2\2ce\t\7\2\2",
"dc\3\2\2\2eh\3\2\2\2fd\3\2\2\2fg\3\2\2\2g\26\3\2\2\2hf\3\2\2\2im\t\6",
"\2\2jl\t\7\2\2kj\3\2\2\2lo\3\2\2\2mk\3\2\2\2mn\3\2\2\2nx\3\2\2\2om\3",
"\2\2\2pq\7\60\2\2qu\t\6\2\2rt\t\b\2\2sr\3\2\2\2tw\3\2\2\2us\3\2\2\2",
"uv\3\2\2\2vy\3\2\2\2wu\3\2\2\2xp\3\2\2\2yz\3\2\2\2zx\3\2\2\2z{\3\2\2",
"\2{\30\3\2\2\2|\u0080\t\6\2\2}\177\t\7\2\2~}\3\2\2\2\177\u0082\3\2\2",
"\2\u0080~\3\2\2\2\u0080\u0081\3\2\2\2\u0081\u008d\3\2\2\2\u0082\u0080",
"\3\2\2\2\u0083\u0084\7\60\2\2\u0084\u0088\t\6\2\2\u0085\u0087\t\b\2",
"\2\u0086\u0085\3\2\2\2\u0087\u008a\3\2\2\2\u0088\u0086\3\2\2\2\u0088",
"\u0089\3\2\2\2\u0089\u008c\3\2\2\2\u008a\u0088\3\2\2\2\u008b\u0083\3",
"\2\2\2\u008c\u008f\3\2\2\2\u008d\u008b\3\2\2\2\u008d\u008e\3\2\2\2\u008e",
"\u0090\3\2\2\2\u008f\u008d\3\2\2\2\u0090\u0091\7\60\2\2\u0091\u0095",
"\t\3\2\2\u0092\u0094\t\b\2\2\u0093\u0092\3\2\2\2\u0094\u0097\3\2\2\2",
"\u0095\u0093\3\2\2\2\u0095\u0096\3\2\2\2\u0096\32\3\2\2\2\u0097\u0095",
"\3\2\2\2\u0098\u009c\7$\2\2\u0099\u009b\n\t\2\2\u009a\u0099\3\2\2\2",
"\u009b\u009e\3\2\2\2\u009c\u009a\3\2\2\2\u009c\u009d\3\2\2\2\u009d\u009f",
"\3\2\2\2\u009e\u009c\3\2\2\2\u009f\u00a0\7$\2\2\u00a0\34\3\2\2\2\u00a1",
"\u00a2\t\n\2\2\u00a2\u00a3\3\2\2\2\u00a3\u00a4\b\17\2\2\u00a4\36\3\2",
"\2\2\u00a5\u00a6\7\f\2\2\u00a6\u00a7\3\2\2\2\u00a7\u00a8\b\20\2\2\u00a8",
" \3\2\2\2\u00a9\u00aa\7\61\2\2\u00aa\u00ab\7,\2\2\u00ab\u00af\3\2\2",
"\2\u00ac\u00ae\13\2\2\2\u00ad\u00ac\3\2\2\2\u00ae\u00b1\3\2\2\2\u00af",
"\u00b0\3\2\2\2\u00af\u00ad\3\2\2\2\u00b0\u00b2\3\2\2\2\u00b1\u00af\3",
"\2\2\2\u00b2\u00b3\7,\2\2\u00b3\u00b4\7\61\2\2\u00b4\u00b5\3\2\2\2\u00b5",
"\u00b6\b\21\3\2\u00b6\"\3\2\2\2\u00b7\u00b8\7\61\2\2\u00b8\u00b9\7\61",
"\2\2\u00b9\u00bd\3\2\2\2\u00ba\u00bc\n\13\2\2\u00bb\u00ba\3\2\2\2\u00bc",
"\u00bf\3\2\2\2\u00bd\u00bb\3\2\2\2\u00bd\u00be\3\2\2\2\u00be\u00c0\3",
"\2\2\2\u00bf\u00bd\3\2\2\2\u00c0\u00c1\b\22\3\2\u00c1$\3\2\2\2\21\2",
"RX_fmuz\u0080\u0088\u008d\u0095\u009c\u00af\u00bd\4\2\3\2\b\2\2"].join("");


var atn = new antlr4.atn.ATNDeserializer().deserialize(serializedATN);

var decisionsToDFA = atn.decisionToState.map( function(ds, index) { return new antlr4.dfa.DFA(ds, index); });

function SHRContentProfileLexer(input) {
antlr4.Lexer.call(this, input);
this._interp = new antlr4.atn.LexerATNSimulator(this, atn, decisionsToDFA, new antlr4.PredictionContextCache());
return this;
}

SHRContentProfileLexer.prototype = Object.create(antlr4.Lexer.prototype);
SHRContentProfileLexer.prototype.constructor = SHRContentProfileLexer;

SHRContentProfileLexer.EOF = antlr4.Token.EOF;
SHRContentProfileLexer.KW_GRAMMAR = 1;
SHRContentProfileLexer.KW_G_CONTENT_PROFILE = 2;
SHRContentProfileLexer.KW_NAMESPACE = 3;
SHRContentProfileLexer.KW_MUST_SUPPORT = 4;
SHRContentProfileLexer.DOT = 5;
SHRContentProfileLexer.COLON = 6;
SHRContentProfileLexer.WHOLE_NUMBER = 7;
SHRContentProfileLexer.ALL_CAPS = 8;
SHRContentProfileLexer.UPPER_WORD = 9;
SHRContentProfileLexer.LOWER_WORD = 10;
SHRContentProfileLexer.DOT_SEPARATED_LW = 11;
SHRContentProfileLexer.DOT_SEPARATED_UW = 12;
SHRContentProfileLexer.STRING = 13;
SHRContentProfileLexer.WS = 14;
SHRContentProfileLexer.NEWLINE = 15;
SHRContentProfileLexer.COMMENT = 16;
SHRContentProfileLexer.LINE_COMMENT = 17;


SHRContentProfileLexer.modeNames = [ "DEFAULT_MODE" ];

SHRContentProfileLexer.literalNames = [ 'null', "'Grammar:'", "'ContentProfile'",
"'Namespace:'", "'MS'", "'.'", "':'",
'null', 'null', 'null', 'null',
'null', 'null', 'null', 'null',
"'\n'" ];

SHRContentProfileLexer.symbolicNames = [ 'null', "KW_GRAMMAR", "KW_G_CONTENT_PROFILE",
"KW_NAMESPACE", "KW_MUST_SUPPORT",
"DOT", "COLON", "WHOLE_NUMBER",
"ALL_CAPS", "UPPER_WORD", "LOWER_WORD",
"DOT_SEPARATED_LW", "DOT_SEPARATED_UW",
"STRING", "WS", "NEWLINE", "COMMENT",
"LINE_COMMENT" ];

SHRContentProfileLexer.ruleNames = [ "KW_GRAMMAR", "KW_G_CONTENT_PROFILE",
"KW_NAMESPACE", "KW_MUST_SUPPORT",
"DOT", "COLON", "WHOLE_NUMBER", "ALL_CAPS",
"UPPER_WORD", "LOWER_WORD", "DOT_SEPARATED_LW",
"DOT_SEPARATED_UW", "STRING", "WS",
"NEWLINE", "COMMENT", "LINE_COMMENT" ];

SHRContentProfileLexer.grammarFileName = "SHRContentProfileLexer.g4";



exports.SHRContentProfileLexer = SHRContentProfileLexer;

Loading

0 comments on commit a371487

Please sign in to comment.