AQL Parser
Hello, What is the best (comfortable, easiest to use, most often used, most interoperable, etc.) implementation of an AQL parser currently available ? After some googleling I found a Git repository of JacSoyYo. Alternatively I could cut the parser from the EtherCIS-Server. Currently I am only interested in a runtime object model of AQL queries so I do not need an interpreter engine. Greetings Georg -- - Dipl.-Inf. Georg Fette Raum: B001 Universität WürzburgTel.: +49-(0)931-31-85516 Am Hubland Fax.: +49-(0)931-31-86732 97074 Würzburg mail: georg.fe...@uni-wuerzburg.de - ___ openEHR-technical mailing list openEHR-technical@lists.openehr.org http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org
AQL-parser in Java? + AQL response format (using the AS operator)
Hi! On Wed, Mar 4, 2009 at 02:54, Heath Frankel heath.frankel at oceaninformatics.com wrote: Response to 2, the Ocean EhrGate Web Service provides a ResultSet as defined in the following WSDAL fragment: ? ?xs:complexType name=ResultSet ... Thanks, I know the service interface and WSDL for Ocean EhrGate [1a,b] since we got Ocean's permission to use it in an experimental research application [2] I created. (For those not familiar with EhrGate an example AQL query and response is provided further down after my signature in this mail.) On Wed, Mar 4, 2009 at 03:04, Heath Frankel heath.frankel at oceaninformatics.com wrote: Sorry, I didn't realise when I responded to this that it was such an old post No problem, I'm glad the topic comes to life again since we still should resolve it. I think it would be wonderful to have standardised AQL response formats for e.g. XML and JSON. The usecase I am thinking of is not the EHR extract exchange but rather the ability for e.g a GUI implementation to be used with any future AQL-capable persistence implementation (PatientOS, OSHIP, Zilics, Ocean, LiU and others maybe?). By the way, does AQL cater for querying CONTRIBUTIONS and FOLDERS too? We (LiU) are working on a REST-based [3] service implementation that we'll contribute as an inspiration for an official specification when we have tested it. I guess the Ocean EhrGate implementation could be an inspiration for an official SOAP based service specification. No matter if an implementer wants to go for SOAP or REST the XML response format could be the same (except that it in the SOAP response will be wrapped in some extra SOAPy-envelope XML that a REST based response won't need). It is the response format (without SOAP envelope) that I'd be interested in further discussions about. On Wed, Mar 4, 2009 at 02:54, Heath Frankel heath.frankel at oceaninformatics.com wrote: I am very interested in the idea of basing a query result class on the ITEM_TABLE class ... In addition, it may also be necessary to have a query result class that is based on the ITEM_TREE class to support more complex hierarchical query results ... W3C has specified a response format [4] for SPARQL queries that I think could be used as an inspiration for an AQL-response specification. Interested readers, please have a look at it and give some feedback. I have not managed to figure out how to get the result of using the AS operator in AQL-queries visible in the result from EhrGate, the variable binding approach in [4] might be useful in AQL responses. In the case where AS has not been used then the path expression used (e.g. c/context/start_time) could be returned as variable name. So let's discuss a bit an then make a draft specification for XML based AQL-responses :-) (JSON- and YAML-based response specifications can follow later...) References: [1a] http://demo.oceanehr.com/EhrBank14/EhrService.asmx [1b] http://demo.oceanehr.com/EhrBank14/EhrService.asmx?WSDL [2] http://www.imt.liu.se/~erisu/2008/ehrbank/EHRBankTool.html [3] http://en.wikipedia.org/wiki/REST [4] http://www.w3.org/TR/rdf-sparql-XMLres/ Best regards, Erik Sundvall erisu at imt.liu.sehttp://www.imt.liu.se/~erisu/Tel: +46-13-227579 This query used in [2] connected to [1]... SELECT c/context/start_time , c/name AS my_variable, c/composer/name, v/uid/value FROM EHR[uid='fe619372-25ed-4549-a108-6ddbd92a96a7'] CONTAINS VERSION v CONTAINS COMPOSITION c WHERE c/category/defining_code/code_string='433' ..will get a response like this (don't worry, this is a fake patient)... RetreiveResultsResponse xmlns:xsd=http://www.w3.org/2001/XMLSchema; xmlns=http://www.oceaninformatics.biz/OceanEhr/EhrBank/EhrService; xmlns:soap=http://schemas.xmlsoap.org/soap/envelope/; xmlns:xsi=http://www.w3.org/2001/XMLSchema-instance; RetreiveResultsResult TotalResults35/TotalResults Rows ResultRow Items anyType xsi:type=q1:DV_DATE_TIME xmlns:q1=http://schemas.openehr.org/v1; q1:value2006-12-05T13:51:55/q1:value /anyType anyType xsi:type=q2:DV_TEXT xmlns:q2=http://schemas.openehr.org/v1; q2:valueBlood Glucose/q2:value /anyType anyType xsi:type=xsd:stringEhrGateUnit/anyType anyType xsi:type=xsd:string05c0762d-2044-4ce6-a6f3-f7bb959cda63::10aec661-5458-4ff6-8e63-c2265537196d::1/anyType /Items /ResultRow [...other results removed from example...] /Rows /RetreiveResultsResult /RetreiveResultsResponse -Original Message- From: openehr-technical-bounces at openehr.org [mailto:openehr-technical-bounces at openehr.org] On Behalf Of Erik Sundvall Sent: Tuesday, September 30, 2008 6:35 PM To: For openEHR technical discussions Subject: AQL-parser in Java? + AQL response format (using the AS operator) ... 2. Are there any suggestions for standardised response formats? It would be interesting if we could query each others' implementations (Python
AQL-parser in Java? + AQL response format (using the AS operator)
Hi Erik, Response to 2, the Ocean EhrGate Web Service provides a ResultSet as defined in the following WSDAL fragment: xs:complexType name=ResultSet xs:sequence xs:element minOccurs=0 name=name type=xs:string / xs:element name=totalResults type=xs:int / xs:element minOccurs=0 maxOccurs=unbounded name=columns nillable=true xs:complexType xs:sequence xs:element minOccurs=0 name=path type=xs:string / xs:element minOccurs=0 name=name type=xs:string / /xs:sequence /xs:complexType /xs:element xs:element minOccurs=0 name=rows xs:complexType xs:sequence xs:element minOccurs=0 maxOccurs=unbounded name=row xs:complexType xs:sequence xs:element minOccurs=0 maxOccurs=unbounded name=items nillable=true / /xs:sequence /xs:complexType /xs:element /xs:sequence /xs:complexType /xs:element /xs:sequence /xs:complexType I am very interested in the idea of basing a query result class on the ITEM_TABLE class for consistency reasons, however ITEM_TABLE cannot be used directly because it is limited to rows of ELEMENTs. A query result class needs to support rows of ANY items as is the case above. In addition, it may also be necessary to have a query result class that is based on the ITEM_TREE class to support more complex hierarchical query results that are difficult to unambiguously represent using a traditional table structure, which requires cross-join logic to be applied as mentioned recently (sorry we have not had time to respond to that post, day-job priorities). Heath -Original Message- From: openehr-technical-boun...@openehr.org [mailto:openehr-technical-bounces at openehr.org] On Behalf Of Erik Sundvall Sent: Tuesday, September 30, 2008 6:35 PM To: For openEHR technical discussions Subject: AQL-parser in Java? + AQL response format (using the AS operator) Hi! 1. Has anybody planned, or started, to create an AQL-parser in Java? Maybe the basics from AQL to parse tree could be common for several different persistence implementations and then the rest of the implementation will be different for different persistence mechanisms. Our first interest will be in transforming AQL queries to queries against an XML-database containing versioned Compositions. (It's for an educational system where performance isn't crucial.) 2. Are there any suggestions for standardised response formats? It would be interesting if we could query each others' implementations (Python and various Java based ones) and interpret the response :-) When returning entire openEHR subtrees (an entire Composition or Entry for example) I guess the current XML-format could be used to start with. The response is likely to often be a list of such subtrees and the format of that list Would need to be standardised. Another case is when using the AS operator, how do we in a standardised way identify the named parts. Best regards, Erik Sundvall erisu at imt.liu.sehttp://www.imt.liu.se/~erisu/Tel: +46-13-227579 ___ openEHR-technical mailing list openEHR-technical at openehr.org http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-technical
AQL-parser in Java? + AQL response format (using the AS operator)
Sorry, I didn't realise when I responded to this that it was such an old post (having email management problems). Anyway, it seems to be current hot topic so it probably won't go to waste. Heath -Original Message- From: Heath Frankel [mailto:heath.fran...@oceaninformatics.com] Sent: Wednesday, March 04, 2009 12:25 PM To: 'erisu at imt.liu.se'; 'For openEHR technical discussions' Subject: RE: AQL-parser in Java? + AQL response format (using the AS operator) Hi Erik, Response to 2, the Ocean EhrGate Web Service provides a ResultSet as defined in the following WSDAL fragment: xs:complexType name=ResultSet xs:sequence xs:element minOccurs=0 name=name type=xs:string / xs:element name=totalResults type=xs:int / xs:element minOccurs=0 maxOccurs=unbounded name=columns nillable=true xs:complexType xs:sequence xs:element minOccurs=0 name=path type=xs:string / xs:element minOccurs=0 name=name type=xs:string / /xs:sequence /xs:complexType /xs:element xs:element minOccurs=0 name=rows xs:complexType xs:sequence xs:element minOccurs=0 maxOccurs=unbounded name=row xs:complexType xs:sequence xs:element minOccurs=0 maxOccurs=unbounded name=items nillable=true / /xs:sequence /xs:complexType /xs:element /xs:sequence /xs:complexType /xs:element /xs:sequence /xs:complexType I am very interested in the idea of basing a query result class on the ITEM_TABLE class for consistency reasons, however ITEM_TABLE cannot be used directly because it is limited to rows of ELEMENTs. A query result class needs to support rows of ANY items as is the case above. In addition, it may also be necessary to have a query result class that is based on the ITEM_TREE class to support more complex hierarchical query results that are difficult to unambiguously represent using a traditional table structure, which requires cross-join logic to be applied as mentioned recently (sorry we have not had time to respond to that post, day-job priorities). Heath -Original Message- From: openehr-technical-boun...@openehr.org [mailto:openehr-technical-bounces at openehr.org] On Behalf Of Erik Sundvall Sent: Tuesday, September 30, 2008 6:35 PM To: For openEHR technical discussions Subject: AQL-parser in Java? + AQL response format (using the AS operator) Hi! 1. Has anybody planned, or started, to create an AQL-parser in Java? Maybe the basics from AQL to parse tree could be common for several different persistence implementations and then the rest of the implementation will be different for different persistence mechanisms. Our first interest will be in transforming AQL queries to queries against an XML-database containing versioned Compositions. (It's for an educational system where performance isn't crucial.) 2. Are there any suggestions for standardised response formats? It would be interesting if we could query each others' implementations (Python and various Java based ones) and interpret the response :-) When returning entire openEHR subtrees (an entire Composition or Entry for example) I guess the current XML-format could be used to start with. The response is likely to often be a list of such subtrees and the format of that list Would need to be standardised. Another case is when using the AS operator, how do we in a standardised way identify the named parts. Best regards, Erik Sundvall erisu at imt.liu.sehttp://www.imt.liu.se/~erisu/Tel: +46-13-227579 ___ openEHR-technical mailing list openEHR-technical at openehr.org http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-technical
AQL-parser in Java? + AQL response format (using the AS operator)
Hi! 1. Has anybody planned, or started, to create an AQL-parser in Java? Maybe the basics from AQL to parse tree could be common for several different persistence implementations and then the rest of the implementation will be different for different persistence mechanisms. Our first interest will be in transforming AQL queries to queries against an XML-database containing versioned Compositions. (It's for an educational system where performance isn't crucial.) 2. Are there any suggestions for standardised response formats? It would be interesting if we could query each others' implementations (Python and various Java based ones) and interpret the response :-) When returning entire openEHR subtrees (an entire Composition or Entry for example) I guess the current XML-format could be used to start with. The response is likely to often be a list of such subtrees and the format of that list Would need to be standardised. Another case is when using the AS operator, how do we in a standardised way identify the named parts. Best regards, Erik Sundvall erisu at imt.liu.sehttp://www.imt.liu.se/~erisu/Tel: +46-13-227579
AQL-parser in Java? + AQL response format (using the AS operator)
Hi Eric, I suggest that you read the Extract Information Model document. Though it still needs some vetting, this should answer many of your questions. Regards, Tim On Tue, 2008-09-30 at 11:04 +0200, Erik Sundvall wrote: Hi! 1. Has anybody planned, or started, to create an AQL-parser in Java? Maybe the basics from AQL to parse tree could be common for several different persistence implementations and then the rest of the implementation will be different for different persistence mechanisms. Our first interest will be in transforming AQL queries to queries against an XML-database containing versioned Compositions. (It's for an educational system where performance isn't crucial.) 2. Are there any suggestions for standardised response formats? It would be interesting if we could query each others' implementations (Python and various Java based ones) and interpret the response :-) When returning entire openEHR subtrees (an entire Composition or Entry for example) I guess the current XML-format could be used to start with. The response is likely to often be a list of such subtrees and the format of that list Would need to be standardised. Another case is when using the AS operator, how do we in a standardised way identify the named parts. Best regards, Erik Sundvall erisu at imt.liu.sehttp://www.imt.liu.se/~erisu/Tel: +46-13-227579 ___ openEHR-technical mailing list openEHR-technical at openehr.org http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-technical -- Timothy Cook, MSc Health Informatics Research Development Services LinkedIn Profile:http://www.linkedin.com/in/timothywaynecook Skype ID == timothy.cook ** *You may get my Public GPG key from popular keyservers or * *from this link http://timothywayne.cook.googlepages.com/home* **
