This is a parser/searcher library for OpenEdge xref files.
Searching the sources which contain creates and deletes of in 'Klantbes' table is as easy as:
import { Parser, Searcher } from 'xrefparser';
const parser = new Parser();
const xrefdata = parser.parseDir('c:/usr/xref', 'D:\\wintmp\\build492943069\\devmain\\');
const searcher = new Searcher(xrefdata);
const sources = searcher
.getTableReferences('Klantbes', true, undefined, true)
.map(table => table.sourcefile);
console.log(sources);
When instantiated there's basically one relevant method; parseDir
parseDir(dirname: string, sourcebasedir?: string): XrefFile[]
dirname
is the root directory where all the .xref files are located.
sourcebasedir
is the optional parameter stating what the root directory of the sources when they were compiled. This is substracted from the sourcefile
properties.
The only constructor takes XrefFile[]
as input.
add(xreffiles: Xreffile[]);
This method add/merges the xreffiles
into the existing Xreffile[]'s.
getDatabaseNames(sources?: string[]): string[]
Returns an array of database names found in sources
. If sources
is undefined, all the database names of all sources are returned.
getDatabaseReferences(databaseName: string): XrefFile[]
Returns an array with all the XrefFile
contains references to the databaseName
getFieldReferences(fieldname: string, tablename?: string, hasUpdates?: boolean): XrefFile[]
Looks for occurences of fieldname
, optionally in table tablename
. The optional hasUpdates
parameter gives the possibility the search for sources which update a certain field (or not). If tablename
or hasUpdates
are not to be taken into consideration, omit them or pass undefined
.
getTableReferences(tablename: string, hasCreates?: boolean, hasUpdates?: boolean, hasDeletes?: boolean): Xreffile[]
Looks for uses of tablename
in the all the Xreffile
objects and returns XrefFile[]
. Use map
as in the example to turn it into an array of source names (or whatever you want). If for the has*
parameters undefined
is used, the particular action is not searched for. This contrary to hasCreated = false
where sources are returned that do not create the particular table.
Above case takes on a Dell XPS15 9560 around 8s to parse all the xref files and 3ms to search for create/delete references in the 'Klantbes' table. There are around 3850 xref files which are combined 350MB on disk.
The release notes can be found at github/releasenotes.md
The Parser
turns in the .xref
files into an array of Xreffile
objects. Beneath an example which is pretty self explanatory:
{
"xreffile": "C:/devoe/sandbox/src/xref_out/xref/CustomerBE.cls.xref",
"sourcefile": "xref/CustomerBE.cls",
"cpInternal": "UTF-8",
"cpStream": "UTF-8",
"includes": [
"xref/findfirstcustomer.i"
],
"tablenames": [
"Customer"
],
"classes": [],
"invokes": [
"xref.BaseClass:Cleanup",
"xref.BusinessEntity:BusinessEntity",
"xref.CustomerBE:Cleanup"
],
"annotations": [
"todo(version=12.x, message=refactor for efficiency)",
"linecheck(lineno=44)",
"linecheck(lineno=46)"
],
"tables": [
{
"name": "Customer",
"database": "sports2000",
"isCreated": false,
"isDeleted": false,
"isUpdated": true,
"fields": [
{
"name": "CustNum",
"isUpdated": false
},
{
"name": "CreditLimit",
"isUpdated": true
},
{
"name": "Name",
"isUpdated": false
}
]
}
],
"class": {
"name": "xref.CustomerBE",
"inherits": "xref.BusinessEntity",
"implements": [
"xref.IDisposable"
]
}
}