Excel Integration - Internal Documentation and Implementation Notes
Contents
- Excel Integration - Internal Documentation and Implementation Notes
- Definitions
- Client-Server Interface
- Server-Side Implementation
- Client-Side Implementation
Definitions
header structure
key set
layout sheet
record header
record update count
RevA
RevB
RevC
revision sequence
Revision Sequences
Legacy browser data is problematic for the new Excel tool. The Excel tool expects each document that can be managed to be one of the following revision sequences:
RevA
RevA Display WorkflowTypeID
RevB
RevB Display WorkflowTypeID
RevC
RevC Display WorkflowType ID
Type
Notes
VA
105
ZF
106
FR
107
Standard
Could be invalid if code of FR revision code is >= 'A'
VA
105
ZF
106
GA
107
Standard
Could be invalid if code of GA revision code is < 'A'
VA
105
Partially Standard
VA
105
ZF
106
Partially Standard
ZF
106
Partially Standard
ZF
106
FR
107
Partially Standard
ZF
106
GA
107
Partially Standard
Non Standard
Create new revision sequence based on last revision. Depending on the last revision code, this may force the RevC to be a GA
XX
Non Standard
Document linked to unknown revisionTypeId
Notes:
- See Specification, page ?, for a description of the revision coding rules
The vorabzugFlag must be set/cleared depending on whether a VA revision is present in the sequence
The revCType must be set depending on whether the RevC is a FR or GA
Client-Server Interface
The following client-server transactions exist:
- Login
- Logout
- Refresh Metadata
- Refresh User Role Metadata
- Document Search
- Get Document Detail
- Get Revision Detail
- Get List of Document Links
- Add Document Link
- Delete Document Link
- Upsert Document and Revisions
- Log
Notes:
- Transactions are always initiated by the Excel client
- Requests are HTTP requests, consisting of a URL and CGI encoded variables
- Responses are delivered as HTTP responses and contain data encoded in XML
- Character set:
- Requests: UTF-8
- Responses: UTF-8
The semantics, requests and responses associated with these transactions are described in the following sections.
Error Messages
to be completed
Response
to be completed
Login
to be completed
Logout
to be completed
Refresh Metadata
to be completed
Refresh User Role Metadata
to be completed
Document Search
Request
Using sql, we have produced a prototype query that could be used as a design spec for a new rest transaction. The tool currently uses the following rest api to return a list of drawings
The following is an example query:
/dev/linthal115dev/revisionseq/list? &maxRecords=50 &contractId=310 &siteClassification2Id=2 &allRevisions=1 &revisionTypeIds=8,14,13,15 &maxRecords=50
new rest api /doucuments/revisionseq (replacement for /documents/list?)
Example:
The following URL: http://.../documents/revisionseq?contractId=120&maxRecords=200&siteClassification2Id=2 gives the response:
<?xml version='1.0' encoding='UTF-8'?> <recordSet name="revisionSequences" idFieldName="drawingId"> <record type="" key="drawingId"> <drawingCode>LA2-E-6-00-200-211-10_0001</drawingCode> <drawingId>1261</drawingId> <updateCountDoc>0</updateCountDoc> <revisionIdA>740</revisionIdA> <revisionIdB/> <revisionIdC/> </record> <record type="" key="drawingId"> <drawingCode>LA2-E-6-00-200-211-10_0002</drawingCode> <drawingId>1264</drawingId> <updateCountDoc>0</updateCountDoc> <revisionIdA>741</revisionIdA> <revisionIdB/> <revisionIdC/> </record> ... <record type="" key="drawingId"> <drawingCode>LA2-E-6-03-300-211-10_2056</drawingCode> <drawingId>39143</drawingId> <updateCountDoc>0</updateCountDoc> <revisionIdA>33657</revisionIdA> <revisionIdB/> <revisionIdC/> </record> </recordSet>
Change request (2012-01-25) to rest transaction /documents/revisionseq? sql
Currently the server returns data with matching revision types as per our standard data and partially standard data. Non standard data can be considered as data that does not match the pattern described in the revision section at the beginning of this page.
- The reqular expression constraint should be additionally constrained by the workflowTypes specified in the revision section.
- To make the search consistent with the browser search, we should modify the sql to return the non standard data as well.
Change request (2012-01-25) to rest transaction /documents/revisionseq? xml response
The rest transaction should be extended to include a new element called revisionLast.
- The existing reva,revb and revc elements should be retained. However, they need some additional child elements specified below.
The parameter revisionTypeIds should be removed.
Result set contents: (basically the fields in the key set)
- Document part
- documentId
- document record update count
- classification fields -- are they necessary?
- title -- how to fill in title? From last revision?
- Last revision part: (only needed when a blank-blank-blank sequence is returned)
- revisionCode
- revisionNo
- title0
- title1
- title2
- title3
- scale
- size
- RevA part: (could be empty)
- revisionId
- revisionTypeId
- revisionCode -- not currently in the key set
- revisionNo -- not current set by the VBA code
- revision record update count
- displayWorkflowTypeId
- RevB part: (could be empty)
- revisionId
- revisionTypeId
- revisionCode -- not currently in the key set
- revisionNo -- not current set by the VBA code
- revision record update count
- displayWorkflowTypeId
- RevC part: (could be empty)
- revisionId
- revisionTypeId
- revisionCode -- not currently in the key set
- revisionNo -- not current set by the VBA code
- revision record update count
- displayWorkflowTypeId
The proposed XML would look like this
<?xml version='1.0' encoding='UTF-8'?> <recordSet name="revisionSequences" idFieldName="drawingId"> <record type="" key="drawingId"> <drawingCode>LA2-E-6-00-200-211-10_0001</drawingCode> <drawingId>1261</drawingId> <updateCountDoc>0</updateCountDoc> <revisionLast key="revisionId"> <revisionId>234</revisionId> <revisionCode>v00</revisionCode> <revisionNo>100</revisionNo> <title0>blah</title0> <title1>more blah</title1> <title2>more blah</title2> <title3>more blah</title3> <scale>100:1</scale> <size>blah</size> </revisionLast> <revisionIdA key="revisionId"> <revisionId>740</revisionId> <revisionTypeId></revisionTypeId> <revisionCode></revisionCode> <revisionNo></revisionNo> <updateCountRev></updateCountRev> <displayWorkflowTypeId></displayWorkflowTypeId> </revisionIdA> <revisionIdB key="revisionId"> <revisionId>740</revisionId> <revisionTypeId></revisionTypeId> <revisionCode></revisionCode> <revisionNo></revisionNo> <updateCountRev></updateCountRev> <displayWorkflowTypeId></displayWorkflowTypeId> </revisionIdB> <revisionIdC key="revisionId"> <revisionId>740</revisionId> <revisionTypeId></revisionTypeId> <revisionCode></revisionCode> <revisionNo></revisionNo> <updateCountRev></updateCountRev> <displayWorkflowTypeId></displayWorkflowTypeId> </revisionIdC> </record> </recordSet>
Get Document Detail
to be completed
Get Revision Detail
to be completed
Get List of Document Links
to be completed
Add Document Link
to be completed
Delete Document Link
to be completed
Upsert Document and Revisions
to be completed
Log
to be completed
Server-Side Implementation
- The server-side implementation consists of a series of REST transactions
You can find the actual code that implements each transaction by looking up the request URLs in the RestMap.
Document Search
Proposal for processing valid revisions server side
to be completed
The prototype query consists of an outer query with 4 sub queries.
An improved server side solution should help performance and simplify the Excel VBA code.
Example SQL
-- Create temporary table to contain the results to be returned to Excel drop table if exists t1; create temporary table t1 ( drawingId int, allRevisionIds varchar(255), allRevTypeCodes varchar(255), revisionIds varchar(255), revTypeCodes varchar(255), -- 'VA', 'VA,ZG', 'VA,ZF,GA', etc. seqType varchar(3) -- 'A', 'BC', etc. ); -- Select all documents that have revision sequences that conform to the Excel requirements -- and set the revisionIds and revisiontype codes insert into t1 ( drawingId, allRevisionIds, allRevTypeCodes ) select drw.drawingId, group_concat(rev.revisionId order by rev.revisionNo separator ','), -- allRevisionIds group_concat(revType.code order by rev.revisionNo separator ',') -- allRevTypeCodes from Drawings drw, Revisions rev, RevisionTypeRef revType where rev.drawingId = drw.drawingId and revType.revisionTypeId = rev.revisionTypeId and drw.contractId = 120 and drw.siteClassification2Id = 2 group by drw.drawingId having group_concat(revType.code order by rev.revisionNo separator ',') regexp 'VA$|VA,ZF$|VA,ZF,FR$|VA,ZF,GA$|ZF$|ZF,FR$|ZF,GA$'; -- Update the sequence types update t1 set seqType = case when allRevTypeCodes regexp 'VA$' then 'A' when allRevTypeCodes regexp 'VA,ZF$' then 'AB' when allRevTypeCodes regexp 'VA,ZF,FR$|VA,ZF,GA$' then 'ABC' when allRevTypeCodes regexp 'ZF$' then 'B' when allRevTypeCodes regexp 'ZF,FR$|ZF,GA$' then 'BC' end; -- Not sure what this does update t1 set revTypeCodes = substring_index( allRevTypeCodes, ',', -length(seqType) ), revisionIds = substring_index( allRevisionIds, ',', -length(seqType) ); -- Deliver result set to Excel select * from t1;
Client-Side Implementation
Requirements Notes
Requirement Notes for VBA functions and routines:
- Probably need to store (in the key set) the last revision code and number (at least for blank-blank-blank revision sequences), depending on the implement details. The client does noeed to know the last revision number, but may not need to store it
- Recognise the delivery of a document with a blank-blank-blank revision sequence
Recognise delivery of documents with open (e.g. undefined) tail-end revisions. E.g. where the RevC or RevB and RevB are missing
- Be able to initialise RevB user defined fields/flags, when not already present
- Be able to initialise RevC user defined fields/flags, when not already present
- Be able to initialise user defined fields (mainly the vorabzugFlag) when the RevA is not present in an otherwise valid revision sequence. But also be able to flag an error if the revision code/number does not conform to what is describe in the specification
- Routine to create a new revision sequence based on the latest revision's revision number and code
- Search form: Checkbox to only return valid revisions? By default returns all documents, even those that do not conform to accepted revision sequences
Handling of non and partially standard data in the Excel Client
- Non standard data can be considered as data that does not match the pattern described in the revision section at the beginning of this page.
- Partially standard data can be considered as a non complete revision sequence, with one or more of the expected revisionTypes.
- Standard data contains all three expected revisionTypes.
Requirement
Parameters
Solution
To save data revtypes must be a sequence
revCode, revNo
To save data worktypes have to match expected values
displayworkflowTypeId
Vorabzug flag and revCType handling could require post processing depending on revision sequence
One or two revs missing
Excel Worksheets
to be completed
The Excel client is mainly data-driven and relies on a number of worksheets for defining the processing of incoming and outgoing data. The primary controlling worksheets are:
Header Structure
Layout Sheet
Record Headers
Header Structure
Formula definitions
Example formula:
_^^revCT0_^^row_-_^^revCVersandDelta1_^^row
The formula is split on the '_' character, which results in the following array:
Formula Component
Meaning
Example Value
^^revCT0
The column for the field revCT0
AB
^^row
The current row number
27
-
Minus sign
-
^^revCVersandDelta1
The column for the field revCVersandDelta1
BD
^^row
The current row number
27
For the example values given above, the result is the following formula:
AB27-BD27
Layout Sheet
to be completed
Record Headers
to be completed
Forms
to be completed
VBA Modules
to be completed